Complete Report

Modeling and Simulation of a Novel Photovoltaic Energy Conversion System Using Fuzzy Controller
Chapter 1
INTRODUCTION
Photovoltaic System is getting popular by day as the crude oil price increases and is unstable in the global market. Furthermore with green peace movement, and the consciousness of mankind has heightened up regarding green energy, photovoltaic maybe one of the solution for better as well cleaner energy as it is naturally harnessed from the Suns energy. Although the technology is mainly well known in the space mission, yet its still an alien for domestic usages. This is due to the high initial cost, generation efficiency and reliability [1]. On the other hand, to answer the cry for alternative energy has made the PV system again popular among the researchers. Having said so, the rural areas where the grid connection is extremely expensive, PV Systems have been implied to give hope to these areas, while for the urban life, the PV Water Heater is common and can be found on the roof of the houses. Currently, more than 3500MW of photovoltaic system have been installed all over the world [2]. Referring to the results from Earth Policy Institute (EPI), the world production of solar PV cells increased 32% in 2003, compared to the most recent 5-year average of 27% a year. Production increased to 742 MW, with cumulative global production at 3145 MW at the end of year 2003, enough to meet the electricity niche of one million homes. Referring to the EPI, this extraordinary growth is driven to some degree by improvements in materials and technology, but primarily by market introduction programs and government incentives [2].This fact can clearly conclude that this solar energy (photovoltaic) is a very promising as next generation energy source.
1.1 Problem Formulation

PV generation systems has a major problem: the conversion efficiency of electric power generation is very low, especially under low irradiation conditions and the amount of electric power generated by solar arrays changes continuously with weather conditions. The efficiency of a PV plant is affected mainly by three factors: the efficiency of the PV panel (in commercial PV panels it is between 8-15% [3]), the efficiency of the inverter (95-98 % [4]) and the efficiency of the maximum power point tracking (MPPT) algorithm
Dept. of Electrical and Electronics, NMAMIT, Nitte 1
(which is over 98% [5]). Improving the efficiency of the PV panel and the inverter is not easy as it depends on the technology available, it may require better components, which can increase drastically the cost of the installation. Instead, improving the tracking of the maximum power point (MPP) with new control algorithms is easier, not expensive and can be done even in plants which are already in use by updating their control algorithms, which would lead to an immediate increase in PV power generation and consequently a reduction in its price. MPPT algorithms are necessary because PV arrays have a non linear voltagecurrent characteristic with a unique point where the power produced is maximum [6]. This point depends on the temperature of the panels and on the irradiance conditions. Both conditions change during the day and are also different depending on the season of the year. Furthermore, irradiation can change rapidly due to changing atmospheric conditions such as clouds. It is very important to track the MPP accurately under all possible conditions so that the maximum available power is always obtained. In the past years numerous MPPT algorithms have been published [7]. They differ in many aspects such as complexity, sensors required, cost or efficiency. However, it is pointless to use a more expensive or more complicated method if with a simpler and less expensive one similar results can be obtained. This is the reason why some of the proposed techniques are not used. Measuring the efficiency of MPPT algorithms has not been standardized until the European Standard EN 50530 was published at the end of May, 2010 [8]. It specifies how to test the efficiency of MPPT methods both statically and dynamically. In any case, there are no publications comparing the results of the different MPPT algorithms under the conditions proposed in the standard. The objective of this thesis is firstly to review different MPPT algorithms. Then the most popular, perturb and observe (P&O) and fuzzy logic control (FLC) are analyzed and the results are compared.
1.2 Objective of Work

In the literature, many maximum power point tracking (MPPT) techniques are proposed and implemented. These techniques include look-up table methods, perturbation
and observation (P&O) methods and computational methods. One of the computational methods which have demonstrated fine performances under different environmental operating conditions is the fuzzy based MPPT technique. The fuzzy theory based on fuzzy sets and fuzzy algorithms provides a general method of expressing linguistic rules so that they may be processed quickly by a computer. In this project circuit based simulation model for a PV panel in order to estimate the electrical behavior of the panel with respect to changes in environmental parameters of temperature and irradiation is presented, along with a method to track maximum power of PV system using Fuzzy Logic Controller (FLC) and P&O. Photovoltaic (PV) energy utilization has increased interest in electrical power applications. Hence its very essential to develop a simulation model of the PV panel for simulation studies. Also, it is crucial to operate the PV energy conversion systems near the Maximum Power Point to increase the efficiency of the PV system. However, the nonlinear nature of PV system is apparent, i.e. the current and power of the PV array depends on the array terminal operating voltage. In addition, the maximum power operating point varies with insolation level and temperature. Therefore, the tracking control of the maximum power point is a complicated problem. A more efficient method to solve this problem becomes crucially important. This project proposes the use of Fuzzy Logic Control for maximum power generation. The FLC is appropriate for non-linear control and it does not use complex mathematics. An accurate PV module electrical model is presented based on Shockley diode equation for simulation studies.
1.3 Methodology Adapted

The approach that has been applied in this project can be divided into two major segments. The first segment is the literature review which involves the study of different MPPT techniques and work carried out in this regard and the second part is on the circuitry modeling and simulation. In the beginning, the literature review will help to understand the photovoltaic cell modeling and different maximum power point tracking algorithms. In the second part, the photovoltaic cell will be simulated using MATLAB/Simulink. Boost converter designing is done required for the simulation of the whole photovoltaic system. Perturb and observe algorithm will be implemented using Simulink. Fuzzy logic
Dept. of Electrical and Electronics, NMAMIT, Nitte
controller is designed using the simulink. Finally the output power of the PV array obtained using the two techniques are compared,
1.4 Organization of the Report

Chapter 1 of the report will tell about the project collectively. This chapter explains the crucial aspect of the project work such as problem formulation, objectives and methodology as well the report outline will also be discussed finally. Chapter2 is completely dedicated to literature review about the PV system. This chapter will be solely theoretical, in detail discussing about the solar energy, PV characteristics, standard test conditions, different materials used in the manufacturing of PV cell and details about research carried out by others in maximum power point tracking. Chapter 3 will explain about different maximum power point tracking algorithms. Chapter 4 will explain the proposed system and each component will be explained in brief. And discussion in depth about the mathematical modeling of PV array and boost converter. Will also give details regarding the designing of the fuzzy logic control and perturb and observe algorithm for maximum power point tracking. Chapter 5 will give details of modeling of the PV array and the simulation of the whole system using MATLAB/Simulink. In Chapter 6 the obtained results will be analyzed in terms of power output of the maximum power point tracking methods i.e. of perturb and observe algorithm and fuzzy logic control will be done. Conclusion and suggestion for improving this work is detailed out in Chapter 7.
CHAPTER 2
PRELIMINARIES
Photovoltaics generate electric power when illuminated by sunlight or artificial light. They directly convert the sun's energy into electricity which can be easily transported and converted to other forms for the benefit of society. In terms of power they come in various sizes from mW to MW ranges. This modular structure allows one to scale them depending on application. Practically, they may consist of one or more solar modules in combination with other balance of system (BOS) components covered by the National Electric Code (NEC). The balance of system components include mounting materials for the modules, wire and all wiring components, lightning protectors, grounding connections, power converters and battery storage.
2.1 Solar resource

2.1.1 Introduction
Knowledge of the sun is very important in the optimization of photovoltaic systems [3]. Solar energy is the most abundant renewable resource. The electromagnetic waves emitted by the sun are referred to as solar radiation. The amount of sunlight received by any surface on earth will depend on several factors including; geographical location, time of the day, season, local landscape and local weather. The light's angle of incidence on a given surface will depend on the orientation since the Earth's surface is round and the intensity will depend on the distance that the light has to travel to reach the respective surface. The radiation received by a surface will have two components one which is direct and will depend on the distance the rays travel (air mass). The other component is called diffuse radiation and is illustrated in Figure 2.1. The range of wavelengths of light that reach the earth varies for 300nm to 400nm approximately [4]. This is significantly different from the spectrum outside the atmosphere, which closely resembles 'black body' radiation, since the atmosphere selectively absorbs certain wavelengths.
Figure 2.1: Types of radiation from the sun
2.1.2 Spectrum of the Sun

Two different spectral distributions have been defined for the sun. The AM0 spectrum relates to radiation in outer space and the AM 1.5 G spectrum is at sea level at certain standard conditions. The photovoltaic (PV) industry and the American Society for Testing and Materials (ASTM), American government research and development laboratories have developed and defined two standard terrestrial solar spectral irradiance distributions: a standard direct normal and a standard total spectral irradiance. An instrument called the pyranometer is used to measure global radiation. This instrument is designed to respond to all wavelengths and therefore gives an accurate value of the total power in any incident spectrum. The solar spectrum is shown in Figure 2.2. Important terms are defined as follows.
i. Spectral irradiance, I - has units of surface area in a wavelength differential d. ii. Irradiance - has units of all the respective wavelengths.
and refers to the power received by a unit
and refers to the integral of the spectral irradiance over
iii. Radiation - refers to the integral of Irradiance over a specified time period.
2.1.3 Standard Test Conditions (STC)

Uniform conditions are usually specified so that a performance comparison can be made between different PV units (cell, modules). The parameters obtained from the testing are usually provided on the manufacturer's datasheet. Measurements are performed under these standard test conditions and the electrical characteristics obtained characterize the module accurately under these conditions. The conditions are specified as follows:
Figure 2.2: American Society for Testing and Materials (ASTM) Terrestrial Reference Spectra for Photovoltaic Performance Evaluation i. The reference vertical irradiance with a typical value of 1000 with a typical value of and ii. Reference cell temperature for performance rating, a tolerance of ;
iii. A specified light spectral distribution with an air mass, AM =1.5. Air mass figures provide a relative measure of the path the sun must travel through the atmosphere.
2.2 Solar Cell

2.2.1 Operating principle
Solar cells are the basic components of photovoltaic panels. Most are made from silicon even though other materials are also used. Solar cells take advantage of the photoelectric effect: the ability of some semiconductors to convert electromagnetic radiation directly into electrical current. The charged particles generated by the incident radiation are separated conveniently to create an electrical current by an appropriate design of the structure of the solar cell, as will be explained in brief below. A solar cell is basically a p-n junction which is made from two different layers of silicon doped with a small quantity of impurity atoms: in the case of the n-layer, atoms with one more valence electron, called donors, and in the case of the p-layer, with one less valence electron, known as acceptors. When the two layers are joined together, near the interface the free electrons of the n-layer are diffused in the p-side, leaving behind an area positively charged by the donors. Similarly, the free holes in the p-layer are diffused in the n-side, leaving behind a region negatively charged by the acceptors. This creates an electrical field between the two sides that is a potential barrier to further flow. The equilibrium is reached in the junction when the electrons and holes cannot surpass that potential barrier and consequently they cannot move. This electric field pulls the electrons and holes in opposite directions so the current can flow in one way only: electrons can move from the p-side to the n-side and the holes in the opposite direction. A diagram of the p-n junction showing the effect of the mentioned electric field is illustrated in Figure 2.3. Metallic contacts are added at both sides to collect the electrons and holes so the current can flow. In the case of the n-layer, which is facing the solar irradiance, the contacts are several metallic strips, as they must allow the light to pass to the solar cell, called finger.
Figure 2.3: Solar Cell The structure of the solar cell has been described so far and the operating principle is next. The photons of the solar radiation shine on the cell. Three different cases can happen: some of the photons are reflected from the top surface of the cell and metal fingers. Those that are not reflected penetrate in the substrate. Some of them, usually the ones with less energy, pass through the cell without causing any effect. Only those with energy level above the band gap of the silicon can create an electron-hole pair. These pairs are generated at both sides of the p-n junction. The minority charges (electrons in the p-side, holes in the n-side) are diffused to the junction and swept away in opposite directions (electrons towards the n-side, holes towards the p-side) by the electric field, generating a current in the cell, which is collected by the metal contacts at both sides. This can be seen in the figure above, Figure 2.3. This is the light-generated current which depends directly on the irradiation: if it is higher, then it contains more photons with enough energy to create more electron-hole pairs and consequently more current is generated by the solar cell.
2.2.2 Equivalent Circuit of a Solar Cell

The solar cell can be represented by the electrical model shown in Figure 2.4. Its current voltage characteristic is expressed by the following equation (1): (2.1) where I and V are the solar cell output current and voltage respectively, I0 is the dark saturation current, vt= kT/q, q is the charge of an electron, a is the diode quality (ideality) factor, k is the Boltzmann constant, T is the absolute temperature and Rse and Rpa are the series and shunt resistances of the solar cell. Rse is the resistance offered by the contacts and the bulk semiconductor material of the solar cell.
Figure 2.4: Equivalent circuit of a solar cell. The origin of the shunt resistance Rp is more difficult to explain. It is related to the non ideal nature of the pn junction and the presence of impurities near the edges of the cell that provide a short-circuit path around the junction [3]. In an ideal case Rs would be zero and Rp infinite. However, this ideal scenario is not possible and manufacturers try to minimize the effect of both resistances to improve their products.
2.2.3 Open Circuit voltage, short circuit current and maximum power point
There are three classic parameters that are very important on the PV characteristics namely short-circuit current maximum power point value at the points , open-circuit voltage and the . The power delivered by a PV cell attains a maximum . The classical points are shown in Figure 2.5 and are
usually given as part of a manufacturer's data sheet for a PV module as shown in Table 2.1. This information is enough to build a simple model of the module to test power converters, but for a more accurate model more information is required.
Figure 2.5: Important points in the characteristic curves of a solar panel.

Table 2.1: Electrical characteristics provided on a datasheet General Specs Open Circuit Voltage ( Short Circuit Current ( Maximum Power, Watts ( Maximum Power Voltage ( Maximum Power Current ( Maximum System Voltage ) ) ) ) ) Thermal Characteristics Temp. coefficient of Temp. coefficient of (V/0C) (A/0C)
2.2.4 Temperature and Irradiance Effects

Two important factors that have to be taken into account are the irradiation and the temperature. They strongly affect the characteristics of solar modules. As a result, the MPP varies during the day and that is the main reason why the MPP must constantly be tracked and ensure that the maximum available power is obtained from the panel. The effect of the irradiance on the voltage-current (V-I) and voltage-power (V-P) characteristics is depicted in Figure 2.5. As was previously mentioned, the photogenerated current is directly proportional to the irradiance level, so an increment in the irradiation leads to a higher photo-generated current. Moreover, the short circuit current is directly proportional to the photo generated current, therefore it is directly proportional to the irradiance. When the operating point is not the short circuit, in which no power is generated, the photo generated current is also the main factor in the PV current, as is expressed by equation (2.1). For this reason the voltage-current characteristic varies with
the irradiation. In contrast, the effect in the open circuit voltage is relatively small, as the dependence of the light generated current is logarithmic Figure 2.6 shows that the change in the current is greater than in the voltage. In practice, the voltage dependency on the irradiation is often neglected [9]. As the effect on both the current and voltage is positive, i.e. both increase when the irradiation rises, the effect on the power is also positive: the more irradiation, the more power is generated.
Figure 2.6: V-I and V-P curves at constant temperature (25C) and four different insolation values. The temperature, on the other hand, affects mostly the voltage. The open circuit voltage is linearly dependent on the temperature, as shown in the following equation: (2.2) According to (2.2), the effect of the temperature on VOC is negative, because Kv is negative, i.e. when the temperature rises, the voltage decreases. The current increases with the temperature but very little and it does not compensate the decrease in the voltage caused by a given temperature rise. That is why the power also decreases. PV panel manufacturers provide in their data sheets the temperature coefficients, which are the parameters that specify how the open circuit voltage, the short circuit current and the
maximum power vary when the temperature changes. Figure 2.7 shows how the voltagecurrent and the voltage-power characteristics change with temperature. As was mentioned before, the temperature and the irradiation depend on the atmospheric conditions, which are not constant during the year and not even during a single day; they can vary rapidly due to fast changing conditions such as clouds. This causes the MPP to move constantly, depending on the irradiation and temperature conditions. If the operating point is not close to the MPP, great power losses occur. Hence it is essential to track the MPP in any conditions to assure that the maximum available power is obtained from the PV panel. In a modern solar power converter, this task is entrusted to the MPPT algorithms.
Figure 2.7: V-I and V-P curves at constant irradiation (1 kW/m2) and three different temperatures.
2.2.5 Types of Solar Cells

Over the past decades, silicon has been almost the only material used for manufacturing solar cells. Although other materials and techniques have been developed, silicon is used in more than the 80% of the production [3]. Silicon is so popular because it is one of the most abundant materials in the Earths crust, in the form of silicon dioxide, and it is not toxic. Monocrystalline and polycrystalline silicon solar cells are the two
major types of silicon solar cells. There is a third type, amorphous silicon, but the efficiency is worse than with the previous types so it is less used. Other new solar cells are made of copper indium gallium (di)selenide (CIGS) or cadmium telluride (CdTe). Much research and development (R&D) effort is being made to develop new materials, but nowadays there are no commercial substitutes to the above types of solar cells. In this section these different solar cells are reviewed. One of the most important characteristics of solar cells is the efficiency, which is the percentage of solar radiation that is transformed into electricity. It is measured under Standard Test Conditions (STC), irradiance of 1000 W/m, air mass coefficient (it characterizes the solar spectrum after the solar radiation has travelled through the atmosphere) A.M 1.5, and a cell junction temperature of 25C. The higher efficiency, the smaller surface is needed for a given power. This is important because in some applications the space is limited and other costs and parameters of the installation depend on the installed PV surface.
Monocrystalline silicon
Monocrystalline silicon solar cells are the most efficient ones. They are made from wafers (very thin slices) of single crystals obtained from pure molten silicon. These single crystal wafers have uniform and predictable properties as the structure of the crystal is highly ordered. However the manufacturing process must be really careful and occurs at high temperatures, which is expensive. The efficiency of these cells is around 15-18% [10] and the surface needed to get 1 kW in STC is about 7 m2.
Polycrystalline silicon
These cells are also made from wafers of pure molten silicon. However, the crystal structure is random as the silicon cools; it crystallizes simultaneously in many different points producing an irregular crystal structure of random sizes, shapes and orientation. These structures are not as ideal as in the monocrystalline cells so the efficiency is lower, around 11-15% [10]. However the manufacturing process is less expensive, so the lower efficiency is compensated in some way. The surface needed to obtain 1 kW in STC is about 8m2.
Amorphous and thin-film silicon

Amorphous silicon is the non-crystalline form of the silicon and it can be deposited as thin-films onto different substrates. The deposition can be made at low temperatures. The manufacturing process is simpler, easier and cheaper than in the crystalline cells. The weak point of these cells is their lower efficiency, around 6-8% [10]. This efficiency is measured under STC. However, the performance under weaker or diffuse irradiation, such as that in cloudy days, can be higher than in crystalline cells and their temperature coefficient is smaller [3]. Amorphous silicon is also a better light absorber than crystalline, so despite having low efficiency, the thin film is a competitive and promising technology. The first solar cells were of thin-film technology. They have been used since the 1980s in consumer electronics applications, such as calculators. In recent years it has also begun to be used in high power applications due to the characteristics mentioned above. One common use nowadays is as building cladding, for example in facades, as its price is competitive compared with other high quality cladding materials and it offer the advantage of electricity generation. The main advantages of thin film technologies are the ease of manufacturing at low temperatures using inexpensive substrates and continuous production methods, avoiding the need for mounting individual wafers and the potential for lightweight and flexible solar cells. These advantages are common to most of the thin-film solar cells, not only the ones made from amorphous silicon. Over recent years, one more type of silicon has been developed, microcrystalline silicon [3]. It can also be deposited as thin-films onto different substrates, minimizing the quantities of crystalline silicon needed and improving the efficiency of amorphous silicon. However, the light absorption of microcrystalline silicon compared to amorphous silicon is poor. The solution can be an effective light trapping to keep the incident light within the film. This type of silicon is not a commercial technology yet and more R&D is needed.
Other cells and materials

As was mention in the introduction of this chapter, there are other materials apart from silicon that can be used for manufacturing solar cells. These compounds are also
thin film deposited, so they have the same advantages as the silicon thin film solar cells but with a better efficiency. Among these compounds, two are already used in commercial solar cells. They are CIGS and CdTe. The efficiency is around 10-13% [10] and it will rise in the following years as the technologies are improved. It is commonly said that thin film technology is the way to achieve the grid parity, i.e. the point at which the cost of generating electricity is equal, or cheaper than grid power [3]. The main disadvantages of these technologies are the toxicity of some of the compounds and the shortage of some of the elements used. In the case of the CIGS, indium is used. This element is not as abundant as silicon in the Earths crust and it is in high demand for other electronics products such as liquid-crystal display (LCD) monitors, which has generated a shortage and consequently a high price rise in the recent years. Moreover, to create the p-n junction, CIGS is interfaced with cadmium sulphide (CdS), another semiconductor. The problem is that cadmium is a heavy metal which is cumulatively poisonous. In the case of CdTe, the other compound used in commercial thin film solar cells, it is not as toxic as its individual components, but some precautions must be taken during the manufacturing process. Gallium Arsenide (GaAs) has been used for space applications mainly for two reasons: firstly, it is less susceptible to suffer damage from the space radiation than silicon, and secondly, due to its direct bandgap of 1.42 eV, it can take advantage of a greater part of the solar spectrum. Despite being a more expensive material, space projects can afford it as cost is not the most important factor to decide the components. Nowadays it is being investigated to be used in terrestrial PV applications using light concentrators (mirror or lenses) to focus the light onto small cells, reducing the price as less material is required. Triple junction GaAs cells have already passed 40% efficiency in the laboratory using light concentrators [3]. The main handicap at present for this technology is that concentration systems are expensive as they have to track the Sun along the day. One other technology that is being actively researched is dye-sensitized cells [3]. These cells are made from artificial organic materials and are seen as part of the third generation of solar cells. The efficiency of these cells is above that of amorphous silicon
and within the thin-film ones. The main advantage is that they work well under low and diffuse light and their temperature coefficients are lower. The materials used are nontoxic and abundant and their manufacturing processes are relatively simple. Flexible modules can easily be made using flexible substrates and they can be used for building integrated PV: roofs, windows, as they can be manufactured in many shapes, sizes and design criteria. These last two paragraphs illustrate technologies that are being currently investigated. They are non commercial technologies yet, but it is expected that in the following years they will become competitive and will be also used, increasing the possibilities of PV power generation. The silicon and thin film solar cells described before are currently the technologies used in commercial PV applications. Nevertheless, what is important for this work is that all the different solar cells presented above have similar non-linear voltagecurrent characteristics and are affected by irradiation and temperature in a similar way. The only difference is that different type of cells have different levels of sensitivity, nevertheless the same algorithms can be used to track the MPP.
2.3 Literature Review

Chihchiang Hua and Chihming Shen, have done analysis and design of different types of DC/DC converters in a solar energy system to investigate the performance of the converters [35]. A simple method which combines a discrete time control and a PI compensator is used to track the Maximum power points (MPP's) of the solar array. The system is kept to operate close to the MPPT's, thus the maximum possible power transfer from the solar array is achieved. The implementation of the proposed converter system was based on a digital signal processor (DSP). Experimental tests were carried out for buck, boost and buck-boost converters using a simple maximum power point tracking (MPPT) algorithm. The efficiencies for the system with different converters are compared. The paper is use full in evaluating the response of step up, step down converter for the MPPT system. Paper proposes that the Step down converter is the best option for the use in the MPPT system as it give higher efficiency. K.H. Hussein et al, have developed a new MPPT algorithm based on the fact that the MPOP (maximum peak operating point) of a PV generator can be tracked accurately
by comparing the incremental and instantaneous conductances of the PV array [36]. The work was carried out by both simulation and experiment, with results showing that the developed incremental conductance (InCond) algorithm has successfully tracked the MPOP, even in cases of rapidly changing atmospheric conditions, and has higher efficiency than ordinary algorithms in terms of total PV energy transferred to the load. Charles.R.Sullivan and M.J Powers have developed a robust oscillation method for implementing the maximum power point tracking for the solar arrays [37]. The method uses only one variable that is load current for detecting the maximum power. This method is suitable for the battery charging application where MPPT is to be implemented. The algorithm is implemented through a simple circuit. The paper gives detailed discussion about design of a step up converter used for the MPPT. D. P. Hohm, M. E. Ropp, The authors have compared all the different kinds of algorithm that are used for the maximum power point tracking [38]. This helps in proper selection of the algorithm. Preliminary results indicate that perturb and observe compares favorably with incremental conductance and constant voltage. Although incremental conductance is able to provide marginally better performance in case of rapidly varying atmospheric conditions, the increased complexity of the algorithm will require more expensive hardware, and therefore may have an advantage over perturb and observe only in large PV arrays. Mohammad A. S. Masoum et al, give a detailed theoretical and experimental analyses of two simple, fast and reliable maximum power-point tracking (MPPT) techniques for photovoltaic (PV) systems are presented. Voltage-based (VMPPT) and the Current-based (CMPPT) approaches [39]. A microprocessor-controlled tracker capable of online voltage and current measurements and programmed with VMPPT and CMPPT algorithms is constructed. The load of the solar system is either a water pump or resistance. The paper has given a simulink model of the Dc/Dc converter and a solar PV module. J.H.R Enslin employed an integrated maximum power point tracker with a soft switching technique to obtain a low cost high efficiency system [40], [41]. A feed forward maximum power point tracking scheme is developed for the interleaved dual boost converter fed PV system with fuzzy controller. The tracking algorithm changes the duty ratio of the converter such that solar cell array voltage equals the voltage corresponding to the MPP at the solar radiation. This is achieved by the feed forward loop, which generate
an error signal by comparing the voltage of PV array with a reference voltage [42], [43]. A modular converter using Insulated Gate Bipolar Transistor (IGBT) for PV applications was developed. Depending upon the configuration the same may be used as either an inverter or DC-DC converter for tracking. IGBTs operated at switching frequency of 18 kHz claims the advantage of minimum emission, noise and reduced harmonics. High efficiency and the intrinsic safety in case of over load were the considerations in this work [44]. To operate the PV cell close to the maximum power operating point, a simple panel load matching technique can be used[45],[46]. This technique is simple and inexpensive because no additional circuitry is used. The risk of component failure is kept low. However, this system does not take into consideration any changes in irradiance and temperature. In addition, the aging of the PV cell material and the accumulation of dust and dirt on the panel surface may also cause a variation in the maximum power point and efficiency. The perturbation and observation method (P&O), which moves the operating point toward the maximum power point by periodically increasing or decreasing the array voltage is often used in many photovoltaic systems. This approach does not track the maximum power points rapidly and accurately. The incremental conductance method is used in photovoltaic systems. The incremental conductance method tracks the maximum power point by comparing the incremental and instantaneous conductance of the solar array. The incremental conductance is estimated by measuring small changes in array voltage and current. These small changes may be induced by deliberate control action. A method which improves the incremental conductance method and can identify the incremental conductance of the array more rapidly has been proposed. However, the harmonic components measurement of the array voltage and current is necessary to adjust the array reference voltage. This chapter explains about the basics of solar cell, types of materials used in the manufacturing of the solar cells, temperature and irradiance effects and the listed few of the research work done in this field. Next chapter will explain the algorithms used to track maximum power point.
19
CHAPTER 3
ALGORITHMS TO TRACK THE MAXIMUM POWER POINT

As was previously explained, MPPT algorithms are necessary in PV applications because the MPP of a solar panel varies with the irradiation and temperature, so the use of MPPT algorithms is required in order to obtain the maximum power from a solar array. Over the past decades many methods to find the MPP have been developed and published. These techniques differ in many aspects such as required sensors, complexity, cost, range of effectiveness, convergence speed, correct tracking when irradiation and/or temperature change, hardware needed for the implementation or popularity, among others. A complete review of 19 different MPPT algorithms can be found in [7]. Among these techniques, the P&O and the InCond algorithms are the most common. These techniques have the advantage of an easy implementation but they also have drawbacks, as will be shown later. Other techniques based on different principles are fuzzy logic control, neural network, fractional open circuit voltage or short circuit current, current sweep, etc. Most of these methods yield a local maximum and some, like the fractional open circuit voltage or short circuit current, give an approximated MPP, not the exact one. In normal conditions the V-P curve has only one maximum, so it is not a problem. However, if the PV array is partially shaded, there are multiple maxima in these curves. In order to relieve this problem, some algorithms have been implemented as in [11]. In the next section the most popular MPPT techniques are discussed.
3.1 Hill-Climbing Techniques

Both P&O and InCond algorithms are based on the hill-climbing principle, which consists of moving the operation point of the PV array in the direction in which power increases [11] and [12]. Hill-climbing techniques are the most popular MPPT methods due to their ease of implementation and good performance when the irradiation is constant [12]. The advantages of both methods are the simplicity and low computational power they need. The shortcomings are also well-known: oscillations around the MPP and they can get lost and track the MPP in the wrong direction during
rapidly changing atmospheric conditions [13]-[16]. These drawbacks will be explained later.
3.1.1 Perturb and Observe

The P&O algorithm is also called hill-climbing, but both names refer to the same algorithm depending on how it is implemented. Hill-climbing involves a perturbation on the duty cycle of the power converter and P&O a perturbation in the operating voltage of the DC link between the PV array and the power converter [7]. In the case of the Hill-climbing, perturbing the duty cycle of the power converter implies modifying the voltage of the DC link between the PV array and the power converter, so both names refer to the same technique. In this method, the sign of the last perturbation and the sign of the last increment in the power are used to decide what the next perturbation should be. As can be seen in Figure 3.1, on the left of the MPP incrementing the voltage increases the power whereas on the right decrementing the voltage increases the power.
Figure 3.1: PV panel characteristic curves. If there is an increment in the power, the perturbation should be kept in the same direction and if the power decreases, then the next perturbation should be in the opposite direction. Based on these facts, the algorithm is implemented [7]. The process is repeated until the MPP is reached. Then the operating point oscillates around the MPP. This
problem is common also to the InCond method, as was mention earlier. A scheme of the algorithm is shown in Figure 3.2.
Figure 3.2: The flowchart of the P&O Algorithm
3.1.2 Incremental conductance

The incremental conductance algorithm is based on the fact that the slope of the curve power vs. voltage (current) of the PV module is zero at the MPP, positive (negative) on the left of it and negative (positive) on the right, as can be seen in Figure 3.3: at the MPP on the left on the right By comparing the increment of the power vs. the increment of the voltage (current) between two consecutives samples, the change in the MPP voltage can be determined. In
both P&O and InCond schemes, how fast the MPP is reached depends on the size of the increment of the reference voltage.
Figure 3.3: Incremental Conductance algorithm. The drawbacks of these techniques are mainly two. The first and main one is that they can easily lose track of the MPP if the irradiation changes rapidly [6], [12]-[14]. In case of step changes they track the MPP very well, because the change is instantaneous and the curve does not keep on changing. However, when the irradiation changes following a slope, the curve in which the algorithms are based changes continuously with the irradiation, as can be seen in Figure 14, so the changes in the voltage and current are not only due to the perturbation of the voltage. As a consequence it is not possible for the algorithms to determine whether the change in the power is due to its own voltage increment or due to the change in the irradiation.
23
Figure 3.4: P-V curve depending on the irradiation. The other handicap of both methods is the oscillations of the voltage and current around the MPP in the steady state [6], [13], [15] and [16]. This is due to the fact that the control is discrete and the voltage and current are not constantly at the MPP but oscillating around it. The size of the oscillations depends on the size of the rate of change of the reference voltage. The greater it is, the higher is the amplitude of the oscillations. However, how fast the MPP is reached also depends on this rate of change and this dependence is inversely proportional to the size of the voltage increments. The traditional solution is a trade off: if the increment is small so that the oscillations decrease, then the MPP is reached slowly and vice versa, so a compromise solution has to be found. To overcome these drawbacks some solutions have been published in recent years. Regarding the rapid change of the irradiation conditions, Sera et al. published in [12] an improved P&O method, called dP-P&O, in which an additional measurement is performed without perturbation in the voltage and current. In this way, every three consecutive samples the effect of the perturbation in the voltage (current) and the effect of the change in the atmospheric conditions can be evaluated so that the increment in the power used in the algorithm only contains the effect caused purely by the MPPT algorithm. Then the correct decision about the direction of the next perturbation can be
taken. The efficiency of the tracking is improved. Although the method was tested using irradiation slopes, they were not the ones proposed in the new European Standard EN 50530 [8]. A different solution is suggested in [16], which considers the traditional P&O algorithm, in which the perturbation amplitude is tuned constantly taking into account the previous changes in the power. It also includes a stage in which the latest increment in the power is compared with the latest perturbation amplitude to determine if the power increment was due to a change in the irradiation. If this is the case, then the voltage perturbation is set to the same direction as the change in the power condition. The steady state error and the tracking speed are improved, but the algorithm has only been tested with irradiation step changes and not with the irradiation slopes proposed in [8]. In relationship with the oscillations around the MPP in steady state, Zhang et al. proposed in [15] a variable perturbation step for the P&O algorithm to reduce the oscillation around it. This modified P&O method determines also if the operating point is near to or far from the MPP and adjusts the size of the perturbation according to that: if the operating point is near to the MPP, the perturbation size is reduced and if the point is far, then it is increased. This technique improves the convergence speed and reduces the oscillation around the MPP. A similar technique is found in [17]: a variation of the traditional P&O algorithm in which the amplitude of the voltage perturbation is adapted to the actual operating conditions: large perturbation amplitudes are chosen far from the maximum whereas small ones are used near the MPP. The proposed algorithm requires initial panel identification and has to be tuned for each plant. With this technique the dynamic response and the steady state stability are improved. Unfortunately, the last two algorithms do not improve the tracking under changing irradiance conditions. Although the authors claim the performance is better, the algorithms have only been tested with irradiation step changes but not with irradiation ramps as proposed in the European Standard mentioned above [8]. Many papers have been published about optimizing the parameters of these algorithms for different hardware configurations. In [6] the sample frequency for P&O is optimized and in [13] it is shown how the P&O MPPT parameters must be customized to
the dynamic behaviour of the specific converter adopted. It has been traditionally said that the performance of InCond algorithm is better than the P&O. However, according to [6] and [13] the performance is similar if the parameters of the P&O method are optimized. In any case, both algorithms are based on the same principle and have the same problem so they have been analyzed together.
3.1.3 Other Hill Climbing Maximum Power Point Tracking Methods

There are other three techniques revised in [7] that can be grouped with the hill climbing algorithms: ripple correlation control (RCC), dP/dV or dP/dI Feedback control and slide control. RCC uses the ripple imposed by the power converter on the PV array to track the MPP. It correlates dp/dt with di/dt or dv/dt, to drive the power gradient to zero, which happens when the MPP is reached. According to [18] or are positive to the
left of the MPP, negative to the right and zero at the MPP. Actually the same criteria is used by the InCond algorithm but expressed in a different form, thus it will suffer the same problems. In fact, it has been only tested with irradiation steps, which are not appropriate to test the dynamic performance. Besides, it needs low switching frequencies to have enough ripple so the correct decisions can be made and it is an analog technique. On the contrary, inverters are nowadays controlled digitally with DSPs, so this method does not show any advantage to the P&O or InCond. dP/dV or dP/dI Feedback control is a technique which computes the slope of the P-V or P-I characteristic curve and feeds it back to the controller in order to drive it to zero, as they are zero at the MPP. Again this is another implementation of the InCond algorithm, so it has the same advantages and disadvantages. Finally, in the slide control, the switching function used is again dP/dV, thus the same problems as with the InCond algorithm can be expected under changing irradiation. To summarise, the last three MPPT methods are based on the same principles as the P&O and the InCond algorithms, so they have the same advantages and disadvantages. All hillDept. of Electrical and Electronics, NMAMIT, Nitte 26
climbing MPPT methods depend on the PV arrays V-P or I-P characteristics, which vary with temperature and irradiation, therefore these MPPT methods can be confused when the irradiation or temperature are changing, as it is explained in [12]. Finally, the other hill-climbing MPPT methods do not offer any improvement to the original P&O and InCond algorithms.
3.2 Fuzzy Logic Control

The use of fuzzy logic control has become popular over the last decade because it can deal with imprecise inputs, does not need an accurate mathematical model and can handle nonlinearity. Microcontrollers have also helped in the popularization of fuzzy logic control [7]. The fuzzy logic consists of three stages: fuzzification, inference system and defuzzification. Fuzzification comprises the process of transforming numerical crisp inputs into linguistic variables based on the degree of membership to certain sets. Membership functions, like the ones in Figure 3.5, are used to associate a grade to each linguistic term. The number of membership functions used depends on the accuracy of the controller, but it usually varies between 5 and 7 [7], [19]-[21]. In Figure 15 seven fuzzy levels are used: NB (Negative Big), NM (Negative Medium), NS (Negative Small), ZE (Zero), PS (Positive Small), PM (Positive Medium) and PB (Positive Big). The values a, b and c are based on the range values of the numerical variable. In some cases the membership functions are chosen less symmetric or even optimized for the application for better accuracy [8], [25]. The inputs of the fuzzy controller are usually an error, E, and the change in the error, E. The error can be chosen by the designer, but usually it is chosen as because it is zero at the MPP. Then E and E are defined as follows: (3.1) (3.2) In other cases U and P are used. The output of the fuzzy logic controller is usually a change in the duty ratio of the power converter, D, or a change in the reference voltage of the DC-link, V. The rule base, also known as rule base lookup table or fuzzy rule algorithm, associates the fuzzy
is used as error [23] or other inputs are considered, as in [25], where
output to the fuzzy inputs based on the power converter used and on the knowledge of the user.
Figure 3.5: Membership functions. The last stage of the fuzzy logic control is the defuzzification. In this stage the output is converted from a linguistic variable to a numerical crisp one again using membership functions as those in Figure 3.5. There are different methods to transform the linguistic variables into crisp values. It can be said that the most popular is the center of gravity method. The advantages of these controllers, besides dealing with imprecise inputs, not needing an accurate mathematical model and handling nonlinearity, are fast convergence and minimal oscillations around the MPP. Furthermore, they have been shown to perform well under step changes in the irradiation. Another disadvantage is that their effectiveness depends a lot on the skills of the designer; not only on choosing the right error computation, but also in coming up with an appropriate rule base [7].
3.3 Neural networks

Another MPPT method well adapted to microcontrollers is Neural Networks [7]. They came along with Fuzzy Logic and both are part of the so called Soft Computing. The simplest example of a Neural Network (NN) has three layers called the input layer, hidden layer and output layer, as shown in Figure 3.6. More complicated NNs are built adding more hidden layers. The number of layers and the number of nodes in each layer as well as the function used in each layer vary and depend on the user knowledge. The input variables can be parameters of the PV array such as VOC and ISC, atmospheric data as irradiation and temperature or a combination of these. The output is usually one or more reference signals like the duty cycle or the DC-link reference voltage.
Figure 3.6: Neural network. The performance of the NN depends on the functions used by the hidden layer and how well the neural network has been trained. The links between the nodes are all weighted. In Figure 3.6 the weight between the nodes i and j is labelled as wij. The weights are adjusted in the training process. To execute this training process, data of the patterns between inputs and outputs of the neural network are recorded over a lengthy period of time, so that the MPP can be tracked accurately. The main disadvantage of this MPPT technique is the fact that the data needed for the training process has to be specifically acquired for every PV array and location, as the characteristics of the PV array vary depending on the model and the atmospheric conditions depend on the location. These characteristics also change with time, so the neural network has to be periodically trained.
3.4 Fractional open circuit voltage

This method uses the approximately linear relationship between the MPP voltage (VMPP) and the open circuit voltage (VOC), which varies with the irradiance and temperature [7]: where k1 is a constant depending on the characteristics of the PV array and it has to be determined beforehand by determining the VMPP and VOC for different levels of irradiation and different temperatures. According to [7] the constant k1 has been reported to be between 0.71 and 0.78.
Once the constant of proportionality, k1, is known, the MPP voltage VMPP can be determined periodically by measuring VOC. To measure VOC the power converter has to be shut down momentarily so in each measurement a loss of power occurs. Another problem of this method is that it is incapable of tracking the MPP under irradiation slopes, because the determination of VMPP is not continuous. One more disadvantage is that the MPP reached is not the real one because the relationship is only an approximation. To overcome these drawbacks, some solutions have been proposed, as is reported in [7]. For example, pilot cells can be used to obtain VOC. They are solar cells that represent the PV arrays cells and which are not used to produce electricity but to obtain characteristics parameters such as VOC without interfering with the power converters. These pilot cells have to be carefully chosen and placed to represent the PV array characteristics and the irradiation conditions. One drawback of using these pilot cells is that the cost of the system is increased. Depending on the application, this technique can be used because it is very easy to implement and it is cheap - it does not require DSP or microcontroller control and just one voltage sensor is used [7]. However, according to [7] this method is not valid under partial shading of the PV array because then the constant k1 changes. To update then k1 a voltage sweep is proposed though this increases the complexity of the system, the cost increases and there are more power losses during the sweep.
3.5 Fractional short circuit current

Just like in the fractional open circuit voltage method, there is a relationship, under varying atmospheric conditions, between the short circuit current ISC and the MPP current, IMPP, as is shown by: The coefficient of proportionality k2 has to be determined according to each PV array, as in the previous method happened with k1. According to [7] the constant k2 has been reported to be between 0.78 and 0.92. Measuring the short circuit current while the system is operating is a problem. It usually requires adding an additional switch to the power converter to periodically short the PV array and measure ISC. In [22] ISC is measured by shorting the PV array with an additional field-effect transistor added between the PV array and the DC link capacitor.
One other option is shown in [23]: a boost converter is used and the switch of the converter is used to short the PV array. Short circuiting the PV array also leads to a loss of power. One last handicap is that the real MPP is not reached because the proportional relationship is an approximation. Furthermore, k2 changes if the PV array is partially shaded, which happens due to shades or surface contamination. To overcome this problem, [22] proposes an online tuning of k2 and [24] a periodical sweep of the PV voltage from open circuit to short circuit to update k2 and guarantee that the real MPP is reached in the presence of multiple maxima which obviously increases the complexity of the system. Most of the literature using this MPPT technique uses a DSP as controller [7].
3.6 Current sweep

In this method the I-V characteristic curve is obtained using a sweep waveform for the PV array current. The sweep is repeated at fixed time intervals so the I-V curve is updated periodically and the MPP voltage (VMPP) can be determined from it at these same intervals. How the I-V curve is determined and the function chosen for the sweep waveform can be found in [25]. With this method the real MPP is obtained. On the other hand, the sweep takes certain time during which the operating point is not the MPP, which implies some loss of available power. Strictly speaking, it is not possible to track the MPP under irradiation slopes, because the MPP varies continuously. Only if the sweep is instantaneous the global MPP could be found, but that is impossible. Furthermore, the implementation complexity is high, the convergence speed is slow and both voltage and current measurements are required. As pointed out in [25] a MPPT method is worth using only if its power consumption is lower than the increase in power it brings to the entire PV system. Due to the drawbacks and complexity exposed above, this MPPT method is not the best option to track the MPP continuously. However, it can be used as a complement to other methods, for example when initializing the PV system in the morning, to begin the tracking in the real MPP and then change to another algorithm, or to check sometimes during the day if the system is operating at the real MPP. One more application can be checking if there are multiple maxima due to shading conditions.
31
3.7 Maximum Power Point Current and Voltage Computation

IMPP & VMPP computation is a technique in which the MPP is calculated based on the measurements of the irradiance and the temperature using a model of the PV module [7]. The drawbacks are the extra measurements needed, which are sometimes difficult to obtain, and the necessity of an accurate model of the PV array. On the other hand, the MPP is correctly tracked even under changing atmospheric conditions. It can be used in large plants, where the economic investment is huge and a perfect tracking is needed to obtain the maximum available power from the solar arrays.
3.8 State Based Maximum Power Point Tracking Technique

The state based MPP technique is based on a state-space representation of the plant and a nonlinear time-varying dynamic feedback controller. This technique is argued to be robust and tracks the MPP even under changing irradiation and in the presence of multiple maxima. However, no experimental results are given in [26], the implementation complexity is high, as the state-space representation has to be built for each PV plant, and the performance under changing irradiance has not been tested according to the standard.
3.9 Multiple Maxima Search

It has not been considered in this thesis, but when the PV array is shaded the P-V curve presents multiple maxima and most MPPT algorithms including P&O, InCond and fuzzy logic control, cannot determine the global maximum. Usually a local MPP is found, depending on the starting point of the algorithm [27]. In recent years, some algorithms have been proposed to overcome this limitation. The most relevant are reviewed in the introduction of [27], which claims that the most effective is the direct search technique that is based on the dividing rectangles algorithm. If this method is continuously used to track the MPP, the maximum reported efficiency is 97% [27]. However, it can be used periodically to determine where the global maximum is and then change to a traditional algorithm whose efficiency can be over 99% [17].This could be effective as the shades move slowly during the day. In this way, the losses that occur due to convergence to a local instead of the global MPP, which is a handicap of most traditional algorithms, could be avoided.
32
Most of the MPPT algorithms developed over the past years have been reviewed in the previous sections. Some of them are very similar and use the same principle but expressed in different ways, like the last three algorithms listed in the hill-climbing techniques. The most popular MPPT algorithms according to the number of publications are P&O and Fuzzy Logic. It makes sense because they are the simplest algorithms capable of finding the real MPP. However, they have some disadvantages, as discussed earlier. They were selected because of their simplicity and popularity. The FLC is designed according to the references and its efficiency is tested and compared to P&O method.
33
CHAPTER 4
SYSTEM SPECIFICATION AND MATHEMATICAL MODELING

This chapter explains the proposed system for maximum power point tracker in detail. This system comprises of various blocks such as Photovoltaic array, Pulse width modulator, Fuzzy controller and the Boost converter, which are combined together. The overview of all the above mentioned blocks is given in this particular session.
4.1 Block Diagram

The Figure 4.1 shows the overall block diagram of proposed MPPT system using Fuzzy Controller.
PV ARRAY
DC/DC CONVERTER
LOAD
FUZZY CONTROLLER
PWM
Figure 4.1: Block diagram of proposed MPPT system using fuzzy controller. The block diagram consists of the following blocks: PV array Fuzzy Controller Pulse Width Modulator Boost Converter
4.2 Photovoltaic System

The history of PV dates back to 1839 when a French physicist, Edmund Becquerel, discovered the first photovoltaic effect when he illuminated a metal electrode in an electrolytic solution [16]. Thirty-seven years later British physicist, William Adams,
with his student, Richard Day, discovered a photovoltaic material, selenium, and made solid cells with 1~2% efficiency which were soon widely adopted in the exposure meters of camera [16]. In 1954 the first generation of semiconductor silicon-based PV cells was born, with efficiency of 6% [3], and adopted in space applications. Today, the production of PV cells is following an exponential growth curve since technological advancement of late 80s that has started to rapidly improve efficiency and reduce cost. As conventional sources of energy are rapidly depleting and cost of energy is rising, photovoltaic energy becomes a promising alternative source, since it has some of the advantages i.e. available in bulk, free of cost, pollution free and distributed throughout the earth. Photovoltaic or solar cells as they are often referred to are semiconductor devices that convert sunlight into direct current (DC) electricity. The main drawbacks of these systems are that the initial installation cost is considerably high and the energy conversion efficiency is relatively low. To overcome these problems, the following two essential ways can be used. 1) Increase the efficiency of conversion for the solar array and 2) Maximize the output power from the solar array. Photovoltaic systems are the nonlinear variation of output voltage and current with solar radiation levels, operating temperature, ageing and load current. Due to variation of voltage and current levels the constant power cannot be delivered to the load. To overcome these problems, the maximum power point of the PV system is tracked using online and offline algorithms and the system operating point is forced toward this optimal condition.
4.3 Modeling of PV Array

Modeling of PV modules or array is required in order to design and monitor these systems. Usually, a Grey Model process is used to model PV arrays. In such models, physical parameters are determined using the measured data given in the datasheets (by manufacturers). This chapter, explains the basic models of PV array and their advantages and disadvantages over one another.
4.3.1 Representation of PV Devices

The PV devices are basically represented in two different models viz.
Single diode model Double diode model Single Diode Model: In a single diode model, there is a current source parallel to a diode. The current source represents light-generated current, that varies linearly with solar irradiation. This is the simplest and most widely used model as it offers a good compromise between simplicity and accuracy [28], [29]. Figure below shows the single diode model circuit. The characteristic equation for single diode model is given by [30]: (4.1)
Figure 4.2: Single diode model
Double Diode Model: In this model an extra diode attached in parallel to the circuit of single diode model. This diode is included to provide an even more accurate I-V characteristic curve that considers for the difference in flow of current at low current values due to charge recombination in the semiconductor's depletion region [31].
36
Figure 4.3: Double Diode Model

The accuracy of this model more than the single diode model but because of the difficulty to solve the equation, single diode model is preferred.
The characteristic equation for double diode model is given by [30]: (4.2)
Where, Iph - the photo-generated current Io - the dark saturation current Is1 - saturation current due to diffusion Is2 - is the saturation current due to recombination in the space charge layer Rs - cell series resistance Rp - the cell (shunt) resistance a - the diode quality factor q - the electronic charge, k - the Boltzmann's constant, T - the ambient temperature, in Kelvin For simplicity, single diode model is studied in this project. The Diode saturation current, can be expressed by [28], [32]:
37
(4.3) Where, is band gap energy of semiconductor ( = 1.12 eV for the polycrystalline Si at
25 C [28], [32], [33]) And, is diode saturation current at nominal test conditions given by [28]: (4.4) Where, and are short-circuit current, open circuit voltage and junction
thermal voltage at nominal conditions respectively.

Now, the single diode model of PV device can be improved by modifying the above equation by [28]:
(4.5) Since, the saturation current has strong temperature dependence; equation (8) results in a linear variation of with respect to temperature, T. The models validity with the new
equation has been tested experimentally [28].
4.3.2 Method to determine the Unknown Parameters

There are five unknown parameters in the characteristic equation (7) of a single diode model which are, ( , , a, Rs, Rp)
In this method, we assume diode ideality factor Now, we remain with four unknown parameters can be calculated using equation (10) It is trivial that there can be only one pair of (
.
.
) that warranties
at maximum power-point condition. From [28], the relationship can be found between as shown by making
38
(4.6) (4.7) In order to find out , we need to do certain number of iterations so that equals experimental MPP
the peak value of power at maximum power-point, power,
. It can be easily done using a numerical method. It is required to choose can be initialized by 0 and to be the
proper initial conditions for these unknown parameters. the inverse of slope at
then slowly incremented. As mentioned earlier, parallel resistance can be approximated to . So, we can easily assume the initial condition for
negated inverse of slope at . Thus, we get (4.8) Previously, for simplicity, authors used to assume but since due to iterative
updates and, the model developed can be further improved by [28] (4.9) Since, there are five unknown parameters, if we are able to get five equations in terms of these unknowns, they can be solved easily to get the solution [30]. There are three remarkable points on the I,V curve. These are at short-circuit condition, open-circuit condition and maximum power-point condition. At short circuit condition, we obtain (4.10) Similarly, for Open-circuit condition ( and ), we obtain (4.11) and V=0. After substituting these values in equation (7),
39
And for Maximum power-point condition (
and
), we obtain (4.12)
These three equations are obtained using these key points. In order to get another two equations, we can differentiate equation (7) with respect to V . Thus, we get (4.13) Again, by substituting short-circuit condition in equation (19), we obtain, (4.14) Again, by substituting open-circuit condition in equation (19), we obtain, (4.15) The power transferred from the PV device is given by (4.16) Differentiating power equation (19) with respect to V, we get (4.17) To find the value at value of power at maximum power point we need to equate (4.18) Substituting , we get following equation, (4.19)
Thus, the equations obtained in the preceding analysis are independent and can be easily solved in MATLAB using iterative method to solve system of non-linear equations .
40
4.4 Dc-Dc Converters

DC-DC converters can be used as switching mode regulators to convert an unregulated dc voltage to a regulated dc output voltage. The regulation is normally achieved by PWM at a fixed frequency and the switching device is generally BJT, MOSFET or IGBT. The minimum oscillator frequency should be about 100 times longer than the transistor switching time to maximize efficiency. This limitation is due to the switching loss in the transistor. The transistor switching loss increases with the switching frequency and thereby, the efficiency decreases. The core loss of the inductors limits the high frequency operation. Control voltage Vc is obtained by comparing the output voltage with its desired value. Then the output voltage can be compared with its desired value to obtain the contol voltage Vcr. The PWM control signal for the dc converter is generated by comparing Vcr with a sawtooth voltage Vr.[8]. There are four topologies for the switching regulators: buck converter, boost converter, buck-boost converter, ck converter. However my project work deals with the boost regulator and further discussions will be concentrated towards this one.
4.4.1 Boost Converter and its Operation

The figure (4.4) below shows a step up or PWM boost converter. It consists of a dc input voltage source Vg, boost inductor L, controlled switch S, diode D, filter capacitor C, and the load resistance R. When the switch S is in the on state, the current in the boost inductor increases linearly and the diode D is off at that time. When the switch S is turned off, the energy stored in the inductor is released through the diode to the output RC circuit [34].
Figure 4.4: Circuit diagram of boost converter

4.4.2 Steady State Analysis of the Boost Converter

Off State: In the OFF state, the circuit becomes as shown in the Figure 4.5 below
Figure 4.5: The OFF state diagram of the boost converter When the switch is off, the sum total of inductor voltage and input voltage appear as the load voltage. On State: In the ON state, the circuit diagram is as shown below in Figure (4.6)
Figure 4.6: The ON state diagram of the boost converter When the switch is ON, the inductor is charged from the input voltage source Vg and the capacitor discharges across the load. The duty cycle, where
Figure 4.7: Inductor current and Inductor voltage waveform

From the inductor voltage balance equation, we have:-
(4.20) Conversion ratio, From inductor current ripple analysis, change in inductor current,
The boost converter operates in CCM (continuous conducting mode) for
where
The current supplied to the output RC circuit is discontinuous. Thus a large filter capacitor is used to limit the output voltage ripple. The filter capacitor must provide the output dc current to the load when the diode D is off. The minimum value of the filter capacitance that results in the voltage ripple Vr[8] is given by:
4.4.3 Design of the Boost Converter

(1) CURRENT RIPPLE FACTOR (CRF): According to IEC harmonics standard, CRP should be bounded within 30%. i.e (2) VOLTAGE RIPPLE FACTOR (VRF): i.e
(3) SWITCHING FREQUENCY (fs): Fs= 20 KHz

GIVEN DATA: Input voltage, Vg 30V
Output voltage, Vo 120V Output load current,
Step 1 : Calculation of Duty cycle (D):
Io 1A
Step 2: Calculation of Ripple Current (
Step 3: Calculation of Inductor value (L):
Step 4: Calculation of capacitor value(C) :

We have,
= 225
4.5 Fuzzy Logic Controller

Fuzzy sets originated in the year 1965 and this concept was proposed by Lofti A. Zadeh. Since then it has grown and is found in several application areas. According to Zadeh, The notion of a fuzzy set provides a convenient point of departure for the construction of a conceptual framework which parallels in many respects of the framework used in the case of ordinary sets, but is more general than the latter and potentially, may prove to have a much wider scope of applicability, specifically in the fields of pattern classification and information processing. Fuzzy logics are multivalued logics that form a suitable basis for logical systems reasoning under uncertainty or vagueness that allows intermediate values to be defined between conventional evaluations like true/false, yes/no, high/low, etc. These evaluations can be formulated mathematically and processed by computers, in order to apply a more human-like way of thinking in the programming of computers. Fuzzy logic provides an inference morphology that enables approximate human reasoning capabilities to be applied to knowledge-based systems. The theory of fuzzy logic provides a mathematical strength to capture the uncertainties associated with human cognitive processes, such as thinking and reasoning. Fuzzy systems are suitable for uncertain or approximate reasoning, especially for the system with a mathematical model that is difficult to derive. Fuzzy logic allows decision making with estimated values under incomplete or uncertain information.
4.5.1 Fuzzy Sets and Membership Functions

The concept of a fuzzy set is an extension of the concept of a crisp set. Similar to a crisp set a universe set U is defined by its membership function from U to [0, 1]. Consider U to be a non-empty set also known as the universal set or universe of discourse or domain. A fuzzy set on U is defined as . Here is known as the membership function, and is known as the membership grade of x. Membership function is the degree of truth or degree of compatibility. The membership function is the crucial component of a fuzzy set. Therefore all the operations on fuzzy sets are defined based on their membership functions. The membership function is a graphical representation of the magnitude of participation of each input. It associates a weighting with each of the inputs that are
processed, define functional overlap between inputs, and ultimately determines an output response. The rules use the input membership values as weighting factors to determine their influence on the fuzzy output sets of the final output conclusion. Once the functions are inferred, scaled, and combined, they are defuzzified into a crisp output which drives the system. There are different memberships functions associated with each input and output response. Reasonable functions are often piecewise linear function, such as triangular or trapezoidal functions. The value for the membership function can be taken in the interval [0, 1]. When the functions are nonlinear the Gaussian membership function will be taken for the smooth operation.
4.5.2. Fuzzy Inference System

A fuzzy inference system is the process of formulating the mapping from a given input to an output using fuzzy logic. To compute the output of this FIS for the given the inputs, we must go through six steps: Determining a set of fuzzy rules. Fuzzifying the inputs using the input membership functions. Combining the fuzzified inputs according to the fuzzy rules to establish rule strength. Finding the consequence of the rule by combining the rule strength and the output membership function. Combining the consequences to get an output distribution, and Defuzzifying the output distribution. Fuzzy inference systems have been successfully applied in fields such as automatic control, fault diagnosis, data classification, decision analysis, expert systems, and computer vision.
4.5.3. Fuzzy Linguistic Variables

Just like an algebraic variable takes numbers as values, a linguistic variable takes words or sentences as values. The set of values that it can take is called its term set. Each
value in the term set is a fuzzy variable defined over a base variable. The base variable defines the universe of discourse for all the fuzzy variables in short. In short the hierarchy is as follows: Linguistic variable Fuzzy variable Base variable.
Figure: 4.8: Schematic Diagram of Operation of a Fuzzy Interface System In 1973, Professor Lotfi Zadeh [51] proposed the concept of linguistic or "fuzzy" variables. Think of them as linguistic objects or words, rather than numbers. Suppose that X = "age." Then we can define fuzzy sets "young," "middle aged," and "old" that are characterized by MFs ,young,middleaged and old, respectively. Just as a linguistic variable "Age can assume different linguistic values, such as "young," "middle aged and old" in this case. If "age" assumes the value of "young," then we have the expression "age is young," and so forth for the other values.
4.5.4. Fuzzy Controller/ Fuzzy if-then Rule

Fuzzy logic controllers are based on the combination of Fuzzy set theory and fuzzy logic. Systems are controlled by fuzzy logic controllers based on rules instead of equations. This collection of rules is known as the rule base usually in the form of IF-THEN-ELSE statements. Here the IF part is known as the antecedent and the THEN part is the consequent. The antecedents are connected with simple Boolean functions like AND, OR, NOT etc., Figure 4.2 outlines a simple architecture for a fuzzy logic controller [52]. The outputs from a
system are converted into a suitable form by the fuzzification block. Once all the rules have been defined based on the application, the control process starts with the computation of the rule consequences. The computation of the rule consequences takes place within the computational unit. Finally, the fuzzy set is defuzzified into one crisp control action using the defuzzification module.
A fuzzy if-then rule (also known as fuzzy rule, fuzzy implication or fuzzy conditional statement) assumes the form if x is A then y is B. Where A and B are linguistic values defined by fuzzy sets on universes of discourse x and y respectively. Often x is A is called the antecedent or premise, while y is B is called the consequence or conclusion. (Some of the linguistic terms used are shown in Table: 4.1).
4.5.5 Creating Fuzzy Rules

Fuzzy rules are a collection of linguistic statements that describe how the FIS should make a decision regarding classifying an input or controlling an output. Fuzzy rules are always written in the following form: if (input1 is membership function1) and/or (input2 is membership function2) and/or . then (outputn is output membership functionn). For example: one could make up a rule that says: if temperature is high and humidity is high then room is hot. There would have to be membership functions that define what we mean by high temperature (input1), high humidity (input2) and a hot room (output1). This process of taking an input such as temperature and processing it through a membership function to
determine what we mean by "high" temperature is called fuzzification. Also, we must define what we mean by "and" / "or" in the fuzzy rule. This is called fuzzy combination.
Figure: 4.9: Fuzzy Controller Architecture

4.5.6. Fuzzification
The purpose of fuzzification is to map the inputs from a set of devices (for example sensors or features of those sensors such as amplitude or spectrum) to values from 0 to 1 using a set of input membership functions. In the schematic diagram shown in Figure 4.1, there are two inputs, x0 and y0 shown at the lower left corner. These inputs are mapped into fuzzy numbers by drawing a line up from the inputs to the input membership functions above and marking the intersection point. These input membership functions, can represent fuzzy concepts such as "large" or "small", "old" or "young", "hot" or "cold", etc. The membership functions could then represent "large" amounts of tension coming from a muscle or "small" amounts of tension. When choosing the input membership functions, the definition of what we mean by "large" and "small" may be different for each input.
4.5.7 Defuzzification of Output Distribution

In many situations, for a system whose output is fuzzy, it is easier to take a crisp decision if the output is represented as a single scalar quantity. This conversion of a fuzzy set to single crisp value is called defuzzification and is the reverse process of fuzzification. There are two common methods for defuzzification generally followed: Centroid Method: Also known as the centre of gravity or the centre of area method, it obtains the centre of area (x*) occupied by the fuzzy set. It is given by the expression; (For a continuous membership function) (For a discrete membership function) Here, n represents the numbers of element in the sample, xis are the elements, and (xi) is its membership function. Mamdani method was used for fuzzification and for defuzzification centroid method was used.Designing of Fuzzy controller is based on the flow chart shown in Figure 4.10 Five linguistic variables are considered NB, NS, Z, PS and PB. Inputs to the fuzzifier is E and E as specified by equation (3) and (4). The rule base [47] is given in
Table 4.1. Output of the defuzzification block is the duty cycle. The fuzzy subsets used in both fuzzification and defuzzification stages are represented by triangular membership functions as given by Figure 4.11, Figure 4.12 and Figure 4.13.
4.10: A control flowchart for MPPT with fuzzy control
50
Table 4.1: Rule base for the Fuzzy Controller E/dE NB NS Z PS PB NB NB NS NS Z Z NS NB NS Z Z Z Z NB NS Z PS PB PS Z Z Z PS PB PB Z Z PS PS PB
Figure 4.11: Membership function plots of
51
4.6 Pulse Width Modulation

Pulse-width modulation (PWM), or pulse-duration modulation (PDM), is a commonly used technique for controlling power to inertial electrical devices made practical by modern electronic power switches.The average value of voltage (and current) fed to the load is controlled by turning the switch between supply and load on and off at a fast pace. The longer the switch is on compared to the off periods, the higher the power supplied to the load is. The PWM switching frequency has to be much faster than what would affect the load, which is to say the device that uses the power. Typically switchings have to be done several times a minute in an electric stove, 120 Hz in a lamp dimmer, from few kilohertz (kHz) to tens of kHz for a motor drive and well into the tens or hundreds of kHz in audio amplifiers and computer power supplies.The term duty cycle describes the proportion of 'on' time to the regular interval or 'period' of time; a low duty cycle corresponds to low power, because the power is off for most of the time. Duty cycle is expressed in percent, 100% being fully on. The main advantage of PWM is that power loss in the switching devices is very low. When a switch is off there is practically no current, and when it is on, there is almost no voltage drop across the switch. Power loss, being the product of voltage and current, is thus in both cases close to zero. PWM also works well with digital controls, which, because of their on/off nature, can easily set the needed duty cycle.
52
Figure. 4.14: PWM Signal A simple method to generate the PWM pulse train corresponding to a given signal is the intersective PWM: the signal (sinusoidal) is compared with a sawtooth waveform. When the latter is less than the former, the PWM signal is in high state (1). Otherwise it is in the low state (0).
PWM has also been used in certain communication systems where its duty cycle has been used to convey information over a communications channel.The simplest way to generate a PWM signal is the intersective method, which requires only a sawtooth or a triangle waveform (easily generated using a simple oscillator) and a comparator. When the value of the reference signal (the green sine wave in Fig 3.12) is more than the modulation waveform (blue), the PWM signal (magenta) is in the high state, otherwise it is in the low state.
53
CHAPTER 5
MODELING USING MATLAB/Simulink

The MATLAB/Simulink software will be used for the modeling and simulation purposes. This software prepares all the electrical and mathematical blocks that needed in the project under Power System Blockset, Signal Routing and Math Operations (Simulink).This software is easy to use as it is more on graphical user interface pertaining to building or modeling any circuits or mathematical equations.
In this chapter, the method and steps in modeling the solar cell up to PV array are shown clearly. The modeling was done by stages. The first stage was modeling the mathematical equation for the Shockley diode current and the light generated photovoltaic In the second stage modeling is done for the fuzzy controller, third stage PWM and the boost converter is done
. 5.1 Building the Mathematical Model and Circuit

The components being used in developing and modeling the mathematical equation and also the circuitry are taken from the MATLAB/SIMULINK library. In order to model Shockley diode current, the light generated photovoltaic current, the components that being use are the voltage measurement block, current measurement block, go to block, from block and control current source block.
Figure 5.1 shows the mathematical modeling for the reverse current saturation (Io) at the reference temperature which given by the equation 31 as below.
(5.1)
Figure 5.2 shows the mathematical modeling for the light generated current of the photovoltaic cell which depends linearly on the influence of temperature and solar radiation as given by the equation 3.2 below. (5.2)
Figure 5.1: Mathematical Modeling Implementation for Io
Figure 5.2: Mathematical Modeling Implementation for Ipv
Figure 5.3 shows the mathematical modeling for the model current Im referring to the appropriate model circuit as in Figure 4.2 for which given by the equation 33 below. (5.3)
55
Figure 5.3: Mathematical Modeling Implementation for model current Im Figure 5.4 shows the circuitry modeling part. Figure 5.5 depicts PV array after the mask process.
Figure 5.4: Circuitry Design for PV
Figure 5.5: PV Model subsystem
56
5.2 Modeling of P&O Based MPPT

The P&O based MPPT system is designed as per the flowchart shown in Figure 5.6. Converter used for Fuzzy control and P&O MPPT techniques are same. Simulink model for the P&O based MPPT is shown in Figure 5.7 and the whole system is shown in Figure 5.8.
START
Read V(k) and I(k) from panel and calculate P(k)=I(k)*V(k)
Delay P(k) and V(k) by k-1 Instant, P(k-1),V(k-1)
P=P(k-1)-P(k) V=V(k-1)-V(k)
Yes
P>0 No
No
Yes
Yes
V<0
V<0
No
D=D+D
D=D-D
D=D+D
D=D-D
To Switch
Figure 5.6: The flowchart of P and O MPPT algorithm

Figure 5.7: Model of the P and O based MPPT
Figure 5.8: Model of the P&O based MPPT system in simulink
5.3 Modeling of fuzzy Controller Based MPPT

The FLC requires the expert knowledge of the process operation for the FLC parameter setting, and the controller can be only as good as the expertise involved in the design. The modeling of fuzzy controller using the simulink is as given below. The simulink model of the whole system is as shown below in the Figure5.7
Figure 5.9: Model of the fuzzy controller for MPPT
Figure 5.10: Model of the proposed Fuzzy based MPPT system in simulink. The Figure 5.7 shows the proposed Fuzzy based MPPT system developed in simulink. The inputs to the system are temperature and different illumination levels. Input to the fuzzy controller are Voltage and Current from the PV array. Perturb and observe algorithm is also implemented with the same model replacing the fuzzy controller with the P&O MPPT system. The tracking of the maximum power after simulation of the model is shown in the following chapter.
59
CHAPTER 6
SIMULATION RESULTS AND DISCUSSION

Values of the parameter used in developing the MATLAB code for the Photovoltaic array have been tabled below
Table 6.1: Parameters Used in the MATLAB Code

PARAMETERS Short circuit current, Isc Open circuit voltage ,Voc Array current @ maximum power point ,Impp Array voltage @ maximum power point ,Vmpp Maximum Power, Pmax Voltage/Temperature coeff ,Kv Current/Temperature coeff Ki Number of cell in series, Ns Nominal irradiance Nominal operating Temp 8.21A 32.9V 7.61A 26.3V 200 W -0.123 V/K 3.18 54 1000 W/m2 25oC A/K
VALUES
6.1 Variation of Characteristics with Temperature

The performance of a photovoltaic module at a constant level of irradiance (1000 W/m or 1 Sun in this case) is depicted in the graph shown in Figure 6.1. It provides a clear view on how the curves vary with temperature and constitutes a typical set of curves that are displayed on a manufacturer's datasheet. There is significant reduction in the power output of the photovoltaic system as cell temperature increases. This relationship is clearly depicted in Figure 6.2 where the power is plotted as a function of voltage for 3 different temperatures.
2
60
Figure 6.1: IV curves for a photovoltaic module at different temperatures
6.2 Variation of Characteristics with Irradiance

To show the effect of irradiance on the performance of a module the temperature is kept fixed at 250C and the values of irradiance are changed to different values. The variation of the current-voltage characteristics with irradiance are shown in Figure 6.3. It is quite clear that irradiance has a major effect on the short circuit current and indeed the relationship between irradiance and the short circuit current is a linear one. Power was also calculated at different levels of irradiation as shown in Figure 6.4. To obtain the maximum possible output from the photovoltaic module, it has to operate at the voltage corresponding to maximum power.
61
Figure 6.2: Variation of power output with temperature for a photovoltaic module
Figure 6.3: Variation of power and voltage with irradiance. One Sun is defined as incident irradiance of 1000W/m2. Therefore 0.8 Sun = 800W/m2, 0.6 Sun = 600W/m2 and 0.4 Sun = 400W/m2
62
Figure 6.4: Variation of current and voltage with irradiance. One Sun is defined as incident irradiance of 1000W/m2. Therefore 0.8 Sun = 800W/m2, 0.6 Sun = 600W/m2 and 0.4 Sun = 400W/m2.
6.3 Simulation Results of the System Simulation model for the whole system is developed in MATLAB/Simulink. Simulation is done in three different modes in first mode the PV system without the maximum power point tracker is simulated and the results are shown below. In the second mode a maximum power point tracker is implemented with perturb and observe algorithm and finally in the third mode maximum power point tracker implemented with fuzzy logic is done. The obtained simulation results for the Voltage, Current and Power of the PV array are listed below. Simulation is performed when illumination changes from. 400W/m2 to 700W/m2 at 0.5s and from 700W/m2 to 1000W/m2 at 0.15s. Finally the Comparison is done with the tracking time of the maximum power of the perturb and observe algorithm with that of fuzzy logic control and it is summarized in the Table 6.2.
63
6.3.1 Without Maximum Power Point Tracker.

The developed system with only boost converter is simulated using MATLAB and the resulting waveforms for voltage, current and power of the PV array are shown in
the figure 6.5,Figure 6.6 and Figure 6.7 respectively. The maximum power is around 150W. Efficiency is around 75%. To improve the efficiency, MPPT methods based on P&O and Fuzzy control are implemented.
Figure 6.5: The Change of solar illumination in simulation
Figure 6.6: The current of PV array without MPPT
64
Figure 6.7: The voltage of PV array without MPPT
Figure 6.8: The power of PV array without MPPT
65
6.3.2 Simulation Results of PV Array Based on MPPT with Perturb and Observe
The PV array based on MPPT with Perturb and Observe is simulated using matlab and the resulting waveforms for voltage, current and power of the PV array are
shown in the Figure 6.9, Figure 6.10 and Figure 6.11 respectively. The maximum power obtained is 200W but there is too much fluctuation around the maximum power. To reduce this fluctuation and enable fast tracking fuzzy control method is implemented.
Figure 6.9: The Voltage of PV array based on MPPT with perturb and observe
66
Figure 6.10: The Current of PV array based on MPPT with perturb and observe
Figure 6.11: The power of PV array based on MPPT with perturb and observe
67
6.3.3 Simulation Results of PV Array Based on MPPT with Fuzzy Control

The PV array based on MPPT with Fuzzy control is simulated using MATLAB and the resulting waveforms for voltage, current and power of the PV array are shown in
the Figure 6.12, Figure 6.13 and Figure 6.14 respectively. The maximum power obtained is 200W, there is less fluctuation around the maximum power as compared to the P&O method. Tracking is also fast compared to P&O method.
Figure 6.12: The Voltage of PV array based on MPPT with Fuzzy control
68
Figure 6.13: The Current of PV array based on MPPT with Fuzzy control
Figure 6.14: The Power of PV array based on MPPT with Fuzzy control
69
TABLE 6.2 The Tracking Performance Comparison Between Perturb & Observe and Fuzzy Control MPPT Methods MPPT Method P&O Fuzzy Control Tracking time of PV power with circumstance step change 400 700W/m2 700 1000W/m2 at 0.5s at 0.15s 0.016s 0.015s 0.02s 0.015s
From Figure 6.9, Figure 6.10, F Figure 6.11, Figure 6.12, Figure 6.13, Figure 6.14 and Table 6.2, Compared to perturb and observe method, it can be known that fuzzy control MPPT algorithm can fast track MPP in voltage, current and power. So the fuzzy control method is able to improve the dynamic and steady state performance of the PV system simultaneously. MPPT fuzzy logic controllers have been shown to perform well under varying illumination condition.
70
CONCLUSION AND FUTURE SCOPE

The need for maximum power point tracking has been discussed in this thesis. Different methods involved are explained and the reason for selecting fuzzy control was also explained. The main purpose of this project is to develop efficient MPPT techniques which will fast track the maximum power, which is done by using the fuzzy control. Perturb and observe method of MPPT tracking is also implemented. Comparison is done based on these two methods and Fuzzy control methods proves to fast track the maximum power compared to the perturb and observe method. So the fuzzy control method is able to improve the dynamic and steady state performance of the PV system simultaneously. MPPT fuzzy logic controllers have been shown to perform well under varying illumination condition. A complete fuzzy logic solar array maximum power tracking controller has been designed and simulated in the software in this project. Simulation results show fast convergence to the MPP and minimal fluctuation about it. Fuzzy controller tracks the maximum power point of a PV module at given atmospheric conditions very fast and efficiently. The sudden change in atmospheric conditions shifts the maximum power point abruptly which is tracked accurately by this controller. If practically implemented, this method can increase the efficiency of the PV system by quite a large scale. Since the proposed approach requires only the measurement of PV array output current and voltage, not the measurement of solar irradiation level and temperature, it decreases resources as well as the design complexity. So the proposed algorithm is simple and can be easily implemented on any fast controller such as the digital signal processor. The advantages of the fuzzy controller are that the control algorithm gives fast convergence and robust performance against parameter variation, and can accept noisy and inaccurate signals. It is recommended for a design of fuzzy controller, for minimal fluctuation around the MPP. Also, this work can be extended for any DC/DC converters. Hardware implementation of the proposed method can be done using microcontroller or DSP.
71
REFERENCES
[1] A Photovoltaic Array Simulation Model for Matlab-Simulink GUI Environment, by Yun Tiam Tan, Student Member, IEEE, Daniel S. Kirschen, Senior Member, IEEE, and Nicholas Jenkins, Senior Member, IEEE [2] PVSIM : A Simulation Program for Photovoltaic Cells, Modules, and Arrays by David L King, James K. Dudley, and William E. Boyson Sandia National Laboratories [3] [4] [5] P. A. Lynn, Electricity from Sunlight: An Introduction to Photovoltaics, John Wiley & Sons, 2010, p. 238. Sunny Family 2010/2011 - The Future of Solar Technology, SMA product catalogue, 2010. L. Piegari, R. Rizzo, "Adaptive perturb and observe algorithm for photovoltaic maximum power point tracking," Renewable Power Generation, IET, vol. 4, no. 4, pp. 317-328, July 2010. [6] [7] N. Femia, G. Petrone, G. Spagnuolo, M. Vitelli, "Optimizing sampling rate of P&O MPPT technique," in Proc. IEEE PESC, 2004, pp. 1945- 1949. T. Esram, P.L. Chapman, "Comparison of Photovoltaic Array Maximum Power Point Tracking Techniques," IEEE Transactions on Energy Conversion, vol. 22, no. 2, pp. 439- 449, June 2007. [8]
[9]
Overall efficiency of grid connected photovoltaic inverters, European Standard EN50530, 2010.
T. Markvart, Solar electricity, Wiley, 2000, p. 280.
[10]
Trends in photovoltaic applications. Survey report of selected IEA countries between 1992 and 2009, International Energy Agency, Report IEA-PVPS Task 1 T1-19:2010.
[11]
Tat Luat Nguyen, Kay-Soon Low, "A Global Maximum Power Point Tracking Scheme Employing DIRECT Search Algorithm for Photovoltaic Systems," IEEE Transactions on Industrial Electronics, vol. 57, no. 10, pp. 3456-3467, Oct. 2010.
[12]
D. Sera, T. Kerekes, R. Teodorescu, F. Blaabjerg, "Improved MPPT Algorithms for Rapidly Changing Environmental Conditions," in Proc. 12th International Conference on Power Electronics and Motion Control, 2006, pp. 1614-1619.
72
[13]
N. Femia, G. Petrone, G. Spagnuolo, M. Vitelli, "Optimization of perturb and observe maximum power point tracking method," IEEE Transactions on Power Electronics, vol. 20, no. 4, pp. 963-973, July 2005.
[14]
K.H. Hussein, I. Muta, T. Hoshino, M. Osakada, "Maximum photovoltaic power tracking: an algorithm for rapidly changing atmospheric conditions," IEE Proceedings on Generation, Transmission and Distribution, vol. 142, no. 1, pp. 59-64, Jan 1995.
[15]
C. Zhang, D. Zhao, J. Wang, G. Chen, "A modified MPPT method with variable perturbation step for photovoltaic system," in Power Electronics and Motion Control Conference, 2009, pp. 2096-2099.
[16] [17]
W. Xiao, W. G. Dunford, "A modified adaptive hill climbing MPPT method for photovoltaic power systems," in Proc. IEEE PESC, 2004, pp. 1957-1963. L. Piegari, R. Rizzo, "Adaptive perturb and observe algorithm for photovoltaic maximum power point tracking," Renewable Power Generation, IET, vol. 4, no. 4, pp. 317-328, July 2010.
[18]
P. Midya, P. T. Krein, R. J. Turnbull, R. Reppa, and J. Kimball, "Dynamic m a xi m um power point tracker for photovoltaic applications," in Proc. 27th Annual IEEE Power Electronics Specialists Conference, 1996, pp. 1710-1716.
[19]
N. Patcharaprakiti, S. Premrudeepreechacharn, "Maximum power point tracking using adaptive fuzzy logic control for grid-connected photovoltaic system," in Proc. IEEE Power Engineering Society Winter Meeting, 2002, pp. 372-377.
[20]
Chao Zhang, Dean Zhao, "MPPT with asymmetric fuzzy control for photovoltaic system," in Proc. 4th IEEE Conference on Industrial Electronics and Applications, 2009, pp. 2180-2183.
[21]
J. Li, H. Wang, "Maximum power point tracking of photovoltaic generation based on the fuzzy control method," in Proc. International Conference on Sustainable Power Generation and Supply, 2009, pp. 1-6
[22]
T. Noguchi, S. Togashi, R. Nakamoto, "Short-current pulse-based maximumpower-point tracking method for multiple photovoltaic-and-converter module system," IEEE Transactions on Industrial Electronics, vol. 49, no. 1, pp. 217-223, Feb 2002.
73
[23]
S. Yuvarajan, S. Xu, "Photo-voltaic power converter with a simple maximumpowerpoint-tracker," in Proc. International Symposium on Circuits and Systems, 2003, vol. 3, pp. 399-402.
[24]
B. Bekker, H. J. Beukes, "Finding an optimal PV panel maximum power point tracking method," in Proc. 7th AFRICON Conference in Africa, 2004, vol. 2, pp. 1125-1129.
[25]
M. Bodur, M. Ermis, "Maximum power point tracking for low power photovoltaic solar panels," in Proc. 7th Mediterranean Electrotechnical Conference, 1994, vol. 2, pp. 758761.
[26]
E. V. Solodovnik, S. Liu, and R. A. Dougal, "Power Controller Design for Maximum Power Tracking in Solar Installations," IEEE Transactions in Power Electronics, vol. 19, pp. 1295-1304, Sept. 2004.
[27]
Tat Luat Nguyen, Kay-Soon Low, "A Global Maximum Power Point Tracking Scheme Employing DIRECT Search Algorithm for Photovoltaic Systems," IEEE Transactions on Industrial Electronics, vol. 57, no. 10, pp. 3456-3467, Oct. 2010.
[28]
Marcelo Gradella Villalva, Jonas Rafael Gazoli and Ernesto Ruppert Filho Comprehensive Approach to Modeling and Simulation of Photovoltaic Arrays, IEEE Transaction on Power Electronics, Vol. 24, No. 5, Pg.1198-1204, May 2009
[29]
C. Carrero, J.Amador, and S. Arnaltes, A single procedure for helping PV designers to select silicon PV module and evaluate the loss resistances, Renewable Energy, vol. 32, no. 15, pp. 25792589, Dec. 2007.
[30] [31] [32]
Gwinyai Dzimano, B.S. Modeling of Photovoltaic Systems, The Ohio State University, Pg.1-18,48-52, 2008. Ryan C. Campbell, A Circuit-based Photovoltaic Array Model for Power System Studies, Pg.1 Student Member, IEEE. W. De Soto, S.A.Klein, andW. A. Beckman, Improvement and validation of a model for photovoltaic array performance, Solar Energy, vol. 80, no. 1, pp. 78 88, Jan. 2006.
[33]
[34]
G. Walker, Evaluating MPPT converter topologies using a matlab PV model, J. Elect. Electron. Eng., Australia, vol. 21, no. 1, pp. 4555, 2001.
Muhammad H. Rashid, Power Electronics Circuits, Devices and Applications, Third Edition.
74
[35]
Chihchiang Hua, Jongrong Lin, and Chihming Shen, Implementation of a DSPControlled Photovoltaic System with Peak Power Tracking, IEEE Transactions On Industrial Electronics, Vol. 45, No. 1, February 1998 Pp 99-107.
[36]
K.H. Hussein, I. Muta, I. Hoshino &, M. Osakada, Maximum photovoltaic power tracking: an algorithm for rapidly changing atmospheric conditions, Proc. Inst.
Elect. Eng. vol. 142, Jan. 1995.
[37]
Charles.R.Sullivan and M.J Powers, High Efficiency Maximum Power Tracker for Solar Arrays in a Solar Powered Race Vehicle, IEEE PESC93, 1993, pp.574580.
[38]
D. P. Hohm, M. E. Ropp, Comparative Study of Maximum Power Point Tracking Algorithms Using an Experimental, Programmable, Maximum Power Point Tracking Test Bed, IEEE,2000.pp.1699-1702
[39]
Mohammad A. S. Masoum, Hooman Dehbonei, and Ewald F. Fuchs, Theoretical and Experimental Analyses of Photovoltaic Systems With Voltage and CurrentBased Maximum Power-Point Tracking, IEEE Transactions On Energy Conversion, Vol. 17, No. 4, December 2002.
[40]
J.H.R Enslin, M.S.Woif, D.B Synman and W.Seigus Integrated photovoltaic Maximum power point tracking converter IEE transactions on industrial electronics, Vol 44 pp769-773, Dec 1997.
[41] [42]
T.J.Liang, Y.C.Kuo and J.F.Chen, Single stage photovoltaic energy conversion system, IEEE proceedings, vol 148, no.4, pp 339-344, July 2001. M.Veerachamy, T.Senjyu and Uezato K, maximum power point tracking control of IDB inverter supplied PV systemsIEE Proceedings of power applications,vol 148,pp 494-502,Nov 2001.
[43]
Veerachary M,Senjyu,T,Uezato,K., Feed forward maximum power point tracking of PV systems using fussy controller,IEEE transactions on Aerospace and Electronic Systems Vol.38,Issue 3, May 2001, Page{s}:969_981,July 2002.
[44]
A.Lindermann, D.Hackstein, A modular converter with IGBTS for PV applications, 11th E.C.Phtovoltaic Solar energy conference, 12-16 October 1992, Montreux, Switerland, pp 1147-1150.
[45]
S. Rahmam, M. A. Khallat, and B. H. Chowdhury, A discussion on the diversity in the applications of photovoltaic system, IEEE Trans. Energy Conversion, vol. 3, pp. 738746, Dec. 1988.
75
[46]
B. K. Bose, P. M. Szczesny, and R. L. Steigerwald, Microcomputer control of a residential photovoltaic power conditioning system, IEEE Trans. Ind. Application., vol. IA-21, pp. 11821191, Sept. 1985.
[47]
Mohsen Taherbaneh, Hassan Ghafori fard, Amir Hosein Rezaie and Shahab Karbasian, Combination of Fuzzy-Based Maximum Power Point Tracker and
Sun Tracker for Deployable Solar Panels in Photovoltaic Systems, IEEE Transaction, 2007
76

Complete Report

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Complete Report

Uploaded by

Copyright:

Available Formats

Modeling and Simulation of a Novel Photovoltaic Energy Conversion System Using Fuzzy Controller

1.1 Problem Formulation

1.2 Objective of Work

1.3 Methodology Adapted

Dept. of Electrical and Electronics, NMAMIT, Nitte

1.4 Organization of the Report

Dept. of Electrical and Electronics, NMAMIT, Nitte

2.1 Solar resource

Dept. of Electrical and Electronics, NMAMIT, Nitte

Figure 2.1: Types of radiation from the sun

2.1.2 Spectrum of the Sun

Dept. of Electrical and Electronics, NMAMIT, Nitte

and refers to the power received by a unit

and refers to the integral of the spectral irradiance over

2.1.3 Standard Test Conditions (STC)

Dept. of Electrical and Electronics, NMAMIT, Nitte

2.2 Solar Cell

Dept. of Electrical and Electronics, NMAMIT, Nitte

2.2.2 Equivalent Circuit of a Solar Cell

Figure 2.5: Important points in the characteristic curves of a solar panel.

2.2.4 Temperature and Irradiance Effects

2.2.5 Types of Solar Cells

Amorphous and thin-film silicon

Other cells and materials

2.3 Literature Review

Dept. of Electrical and Electronics, NMAMIT, Nitte

ALGORITHMS TO TRACK THE MAXIMUM POWER POINT

3.1 Hill-Climbing Techniques

3.1.1 Perturb and Observe

Figure 3.2: The flowchart of the P&O Algorithm

3.1.2 Incremental conductance

Dept. of Electrical and Electronics, NMAMIT, Nitte

3.1.3 Other Hill Climbing Maximum Power Point Tracking Methods

3.2 Fuzzy Logic Control

is used as error [23] or other inputs are considered, as in [25], where

3.3 Neural networks

3.4 Fractional open circuit voltage

3.5 Fractional short circuit current

3.6 Current sweep

Dept. of Electrical and Electronics, NMAMIT, Nitte

3.7 Maximum Power Point Current and Voltage Computation

3.8 State Based Maximum Power Point Tracking Technique

3.9 Multiple Maxima Search

Dept. of Electrical and Electronics, NMAMIT, Nitte

Dept. of Electrical and Electronics, NMAMIT, Nitte

SYSTEM SPECIFICATION AND MATHEMATICAL MODELING

4.1 Block Diagram

4.2 Photovoltaic System

4.3 Modeling of PV Array

4.3.1 Representation of PV Devices

Figure 4.2: Single diode model

Dept. of Electrical and Electronics, NMAMIT, Nitte

Figure 4.3: Double Diode Model

Dept. of Electrical and Electronics, NMAMIT, Nitte

thermal voltage at nominal conditions respectively.

equation has been tested experimentally [28].

4.3.2 Method to determine the Unknown Parameters

Dept. of Electrical and Electronics, NMAMIT, Nitte

the peak value of power at maximum power-point, power,

Dept. of Electrical and Electronics, NMAMIT, Nitte

And for Maximum power-point condition (

Dept. of Electrical and Electronics, NMAMIT, Nitte

4.4 Dc-Dc Converters