Proceedings of CSIR & IEEE Sponsored National Conference on Power Electronics and Controllers for Renewable Energy Systems,

18& 19 March 2011

Leveling PV Power Output Fluctuations of a PV Diesel Hybrid System Connected to an Isolated Power Utility
Gobind G. (Student) Power Systems Division, DEEE, College of Engineering, Guindy, Anna University Chennai gobind.g@gmail.com A. Karthikeyan (Research Scholar) Power Systems Division, DEEE, College of Engineering, Guindy, Anna University Chennai contactbpci@gmail.com P. Somasundaram (Assistant Professor) Power Systems Division, DEEE, College of Engineering, Guindy, Anna University Chennai mpsomasundaram@yahoo.com

AbstractThe use of Photovoltaic (PV) generators in isolated sites, where conventional generation is not practical, can pose a problem because PV output power depends on insolation and ambient temperature, making it an uncontrollable source. Furthermore, it is not available during the night. The resolution of this problem is to install a hybrid system using alternative sources such as diesel generator. The extracted maximum PV power is given to the utility without leveling and the response of the hybrid system following common disturbances such as solar insolation variations and load changes is investigated. Fluctuating PV power causes frequency deviations and reduction in reliability of the isolated power utility or microgrid when large output power from several PV systems is penetrated in the utility. In this paper, to overcome these problems, an Energy Storage System (ESS) based control method is proposed. The ESS control model introduced here will maintain the energy storage ratio of the ESS near 50%. This will increase the life time of the ESS as well as will decrease the maintenance cost of the ESS. Simulation results show that the proposed method is effective for leveling output power fluctuations and that it is feasible to reduce the frequency deviations of the isolated power utility to maintain reliability. Keywords- Energy Storage System, Frequency deviation, insolation, photovoltaic (PV) output power fluctuations.

and large frequency deviation in electric power system operation [1][3]. Therefore, for the penetration of large PV systems output power in the utility without reduction of the reliability of utility power systems, suitable measures must be applied to the PV systems side. Power characteristics of PV ensembles are presented in [4] where monitored data from 100 PV systems were used to study effects of combined power generation of these systems, compared to the characteristics of an individual system. It was claimed that a significant amount of power fluctuations disappeared; however, large amount of short-term power fluctuations remained. In addition, when the numbers of PV power generation systems were decreased, the power fluctuations increased. On the PV system side, energy storage devices like batteries can be used as smoothing devices for a PV systems output. There have been investigations aimed at improving the performance of PV systems equipped with batteries [7]-[12]. However, energy storage device increases capital cost, as it needs maintenance. The maintenance cost mainly depends on charge/discharge action of battery. Therefore, controlling the charge/discharge action along with maintaining the energy storage ratio near 50% is a issue need to be addressed. Besides, optimal capacity of the ESS needed to reduce the PV output power fluctuations should also be investigated. However, in previous investigations, the optimal size of battery and battery parameters for charge/discharge action are not considered. In this paper, to address the above given issue, a control methodology for ESS along with PV generator is presented, which will maintain storage energy ratio of the battery up to 50% and will select optimal size of the battery needed for smoothing PV output power fluctuations. From the simulation results, it has been found that the proposed method works well to reduce the PV output power fluctuations and also selects the optimal ESS capacity needed for it maintaining the energy storage ratio near 50%. The paper is organized as follows: Section II provides the concept of an isolated power system, description of PV power generation system, control of power converter, and a brief

INTRODUCTION Diesel generators are the main source of electricity in most of the isolated islands and the electricity demand is constantly increasing in these isolated islands. Heavy oil for diesel generators involves buying, transport, and storage costs. To avoid global warming, the consumption of fossil fuels must be reduced. Hence, the installation of renewable energy power production plants has become a burning issue not only for isolated islands but also all over the world. Among various renewable energy systems, photovoltaic power generation systems (PV systems) are expected to play an important role as a clean electricity power source in meeting future electricity demands. PV systems output power depends on weather conditions. In the future, when a significant number of PV systems will be connected to the grids of power utilities, power output fluctuations may cause problems like voltage fluctuation

37

Proceedings of CSIR & IEEE Sponsored National Conference on Power Electronics and Controllers for Renewable Energy Systems, 18& 19 March 2011

review of solar module characteristics. Section III describes the ESS control method. In Section IV, effectiveness and feasibility of the proposed method is demonstrated by simulation results. Conclusions are drawn in Section V. ISOLATED POWER SYSTEM The concept of the power utility used in this paper is shown in Fig. 1. This is actually a PVdiesel hybrid power system consisting of diesel generator and PV systems that generate power to supply the load demand. In addition, it is assumed that the isolated power utility is not connected to large power utility and is always operated independently as a standalone system. The isolated power system model consisting of a diesel generator, PV power generation systems, Energy storage system and a load is shown in detail in Fig. 2,where Si is the insolation, P*max is the maximum power point tracking (MPPT) command power, Pbat is the battery power, Ppvsys is the PV power by PV supplied to the power system, P d is generated power by diesel generators, R is the speed regulation, Tg is the governor time constant, Td is the diesel generator time constant, PL is the load, M is the inertia constant, D is the damping constant, u is the input to the governor, f is the frequency deviation of the isolated power utility. The control of diesel generator described in [5] is also used in this paper to fully exploit the renewable energy.

Fig.3. Single PV power generation system

To reduce the steady-state losses caused by the oscillation of the array operating point around the MPP, the amplitude of the duty cycle perturbation d has to be lowered. A low value of makes the algorithm less efficient in case of rapidly changing atmospheric conditions. The optimal choice of d in these situations given by [18] is used so that there will not be much wider oscillation once MPP is reached. The MPPT algorithm is shown in Fig.4 and the duty cycle perturbation in Fig.5.

Fig.1. Concept of isolated power utility.

Fig.4. Three pointed weighted P&O algorithm. Fig.2. Isolated power system model.

As shown in Fig.3 a single PV power generation system including solar array, DC-DC converter and inverter. The DCDC boost converter acts as an interface between the load and the PV Array. Maximum Power Point tracking is achieved by adjusting the Duty cycle of the DC-DC converter. For simple structure and less costly implementation, a 3 point weighted P&O algorithm was chosen in the present structure to generate the MPPT power command.

38

Proceedings of CSIR & IEEE Sponsored National Conference on Power Electronics and Controllers for Renewable Energy Systems, 18& 19 March 2011

Fig.5. Duty cycle perturbation. Fig. 7. Solar module characteristics curves. (a) and (c) Current-Voltage curves. (b) and (d) Power-Voltage curves.

As the design of power converter and the control system is significantly influenced by the solar module characteristics, these will be reviewed here briefly. The solar module is a nonlinear device and can be represented as a current source model, as shown in Fig.6. The traditional IV characteristic of a solar module, neglecting the internal series resistance, is given by the following equation [13]: where Io and Vo are the output current and output voltage of the solar module, respectively, Ig is the generated current under a given insolation, Isat is the reverse saturation current, q is the charge of an electron, K is the Boltzmanns constant, A is the ideality factor, T is the temperature (in Kelvin), Np is the number of cells in parallel, and Irsh is the current due to intrinsic shunt resistance of the solar module. The saturation current (Isat ) of the solar module varies with temperature according to the following equation [13]:

For the solar module, (1)(4) are used in the development of MATLAB/SIMULINK based computer simulations. Fig. 7(a) and (b) shows the simulated amperevolt and powervolt curves for the solar module for different insolation at constant temperature. Fig. 7(c) and (d) shows the simulated amperevolt and powervolt curves for the solar module for different temperature at constant insolation. Here, the discrete data points shown are taken from the manufacturers data sheet [6] for validating the model. From these curves, it is observed that the output characteristics of the solar array is nonlinear and is vitally affected by the variation of insolation. ENERGY STORAGE SYSTEM CONTROL METHOD In order to smooth PV output power fluctuations, average PV power is generated from MPPT generated PV power through a low pass filter. The average PV power is given by the following equation Pavg = Pmaxf(s) Where f(s )= This average power is used as command power for the battery control system. Therefore, this command power will be achieved from the PV system by charging/discharging action of the Battery. The proposed battery control model is shown in Fig. 8. At first the proposed model calculates PV output power fluctuations by the given below equation. Ppvsys = Pavg Pmax (6) (5)

Fig.6. Equivalent circuit of a solar module.

where Ior is the saturation current at Tr , T is the temperature of the solar array (in Kelvin), Tr is the reference temperature, Eg is the band gap energy, It is the short-circuit current temperature coefficient, Isc is the short-circuit current of solar module, and Si is the insolation in watts per square meter. The current due to the shunt resistance is given by the following equation [13]:

where Ns is the number of cells in series and R sh is the internal shunt resistance of the solar module.

where Ppvsys is PV output power fluctuations. To reduce the charge/discharge action of the battery, a definite value for the dead zone Dzone is set. Dead zone will activate when Ppvsys is greater than the dead zone value. Therefore, if the dead value is set big, small PV power fluctuations will remain in the system. On the other hand, if the dead zone value is set small, PV output power fluctuations will be smoothed well, however, battery will experience rapid charge/discharge action. So, choosing the optimal parameter for battery control system is a vital issue. From Fig. 8, it can be seen that correction power Pc is added with dead zone. Then, battery output Pbat is computed through the limiter whose maximum range is the converter capacity CI .Battery experiences charge/discharge action when Pbat becomes positive/negative. The batterys remaining capacity Wbat is calculated through integration. This integrator is also considered as a limiter whose maximum range is the battery capacity Cbat . The remaining battery capacity Wbat is passed through a 2-D lookup table to maintain the storage ratio ( 100%) bat bat C W = to near 50%.

39

Proceedings of CSIR & IEEE Sponsored National Conference on Power Electronics and Controllers for Renewable Energy Systems, 18& 19 March 2011

Fig. 8. Proposed battery control model

The lookup table is shown in Fig. 9. The lookup table produces a correction value M. The product of correct value M and converter capacity CI becomes the correct power Pc.

Fig. 9. Lookup table for correction value of M.

SIMULATION RESULTS In this paper, the effectiveness of output power leveling of PV array and frequency deviation reduction of power system using the proposed method is examined by simulation with system model and parameters as mentioned in [13]-[17]. In order to use parameters of real PV system given in [16], [17], the rated output power of the PV array is 241kW. The total number of PV power generation systems used in this paper is two and the combined rated output power of two PV power generation systems is 481kW. Simulation parameters of power system, PV array and power converter are shown in TABLE I. TABLE I. Simulation Parameters

Different isolations are used for PV power generation systems. Insolations for PV plants 1,and 2 are shown in Fig. 10(a),(b). Load is shown in Fig. 10(e). The simulation results without using control i.e. output power of PV plant 1 and 2 are shown in Fig. 10(c),(d). However, from Fig. 10(f), it can be said that when these powers are added together, the combined output power of two PV plants is not significantly smoothed. Fig. 10(h) shows frequency deviation f, which deviates by more than 2 Hz frequently. This is a severe problem for maintaining power system reliability. Therefore, the combined output power of two PV plants has harmful effects on the power system. Output power of diesel generator Pd is shown in Fig. 10(g). From Fig. 10(g), it is observed that the diesel generator output power Pd fluctuates in order to cancel out fluctuations of output power PA and load PL. Fig. 10(i) shows the PV power produced by the proposed control. From, Fig. 10(i), it can be said that PV power produced by proposed method is leveled by battery charging/discharging action. Fig. 10(l) shows diesel power where diesel power produced by proposed method fluctuates less than the diesel power produced without control. Fig. 10 (j) shows the frequency deviations where frequency deviations produced by the proposed method is almost near to zero. Therefore, it can be said that the proposed method is effective to reduce the frequency deviation of the utility. Fig. 10(k) shows battery charging/discharging action. Fig. 10(m) shows the storage energy ratio which is maintained below 50%, thus, it will reduce maintenance cost for the battery.

(a) Insolation of PV generator 1

40

Proceedings of CSIR & IEEE Sponsored National Conference on Power Electronics and Controllers for Renewable Energy Systems, 18& 19 March 2011

(e) Load

(b) Insolation of PV generator 2

(f) Combined PV Output power

(c) Output of PV generator 1

(g) Diesel power output without PV leveling

(d) Output of PV generator 2

(h) Frequency deviation without PV leveling

41

Proceedings of CSIR & IEEE Sponsored National Conference on Power Electronics and Controllers for Renewable Energy Systems, 18& 19 March 2011

(m) Energy storage ratio

(i) Leveled combined PV power

After the text edit has been completed, the paper is ready for the template. Duplicate the template file by using the Save As command, and use the naming convention prescribed by your conference for the name of your paper. In this newly created file, highlight all of the contents and import your prepared text file. V. CONCLUSION In this paper, PV output power fluctuations are leveled using the proposed methodology through charge/discharge action of Energy Storage System (ESS). This addition of battery (ESS) will increase the system cost, more specifically, it will increase installation cost of the system. On the other hand, this addition will offer some technical, economical, and environmental benefits in the long run. The cost of battery system is small compared with the cost of PV power generation system. So, in the long run, the proposed method with battery will be cost-effective considering the fact that it is supplying all the available maximum PV power. Moreover, the proposed method with battery will reduce great amount of carbon dioxide emission as well as fuel cost of diesel generator compared to modified MPPT considering the long time span. Considering the situation when insolation will decrease suddenly or no insolation will be present (some hours in a day and in all night), battery power can be used for load leveling, frequency regulation, etc. Therefore, the proposed method can be a good tradeoff between supplying available maximum power and maintaining power quality. From the simulation results, it has been found that the proposed method is able to achieve the required control parameters for ESS and the optimal battery and converter capacities to minimize the frequency deviations.

(j) Frequency deviation after PV leveling

(k) Battery charging/discharging [1]

REFERENCES
S. Yanagawa, T. Kato, K. Wu, A. Tabata, and Y. Suzuoki, Evaluation of LFC capacity for output fluctuation of photovoltaic generation systems based on multi-point observation of insolation, in Proc. IEEE Power Eng.Soc. Summer Meeting, 2001, pp. 16521657. A. Woyte, V. V. Thong, R. Belmans, and J. Nijs, Voltage fluctuations on distribution level introduced by photovoltaic systems, IEEE Trans.Energy Convers., vol. 21, no. 1, pp. 202209, Mar. 2006. H. Asano, K. Yajima, and Y. Kaya, Influence of photovoltaic power generation of required capacity for load frequency control, IEEE Trans.Energy Convers., vol. 11, no. 1, pp. 188193, Mar. 1996. E. Wiemken, H. G. Beyer, W. Heydenreich, and K. Kiefer, Power characteristics of PV ensembles: Experiences from the combined power production of 100 grid connected PV systems distributed over the area of Germany, Solar Energy, vol. 70, no. 6, pp. 513518, 2001. D. Canever, G. J.W. Dudgeon, S. Massucco, J. R.McDonald, and F. Silvestro, Model validation and coordinated operation of a photovoltaic array and a diesel power plant for distributed generation, in Proc. IEEE Power Eng. Soc. Summer Meeting, 2001, vol. 1, pp. 626631. Sharp Electronics Corp. (2008). [Online]. Available: http://solar.sharpusa.com/files/SESG-216U1F.pdf T. Kinjo, T. Senjyu, N. Urasaki, and H. Fujita, Output levelling of renewable energy by electric double layer capacitor applied for energy

[2]

[3]

[4]

[5] (l) Diesel power output after PV leveling [6] [7]

42

Proceedings of CSIR & IEEE Sponsored National Conference on Power Electronics and Controllers for Renewable Energy Systems, 18& 19 March 2011

storage system, IEEE Trans.Energy Convers., vol. 21, no.1, pp. 221227, March 2006. [8] Md. H. Rahman and S. Yamashiro, Novel distributed power generating system of PV-ECaSS using solar energy estimation, IEEE Trans. Energy Convers., vol. 22, no.2, pp. 358-367, June 2007. [9] J. P. Barton and D. G. Infield, A probabilistic method for calculating the usefulness of a store with finite energy capacity for smoothing electricity generation from wind and solar power, Journal of Power Sources., vol. 162, pp. 943-948, 2006. [10] R. Wanger, Large lead/acid batteries for frequency regulation, load levelling and solar power applications, Journal of Power Sources, vo. 67, pp. 163-172, 1997. [11] H. Sugihara, S. Nishikawa, and Y. Kimura, Observation of the hybrid system using photovoltaic and sodium-sulphur battery,in Proc. JSES/JWEA Joint Conference, 2001, pp.13-16. [12] H. Miyauchi, K. Eguchi, and H. Hayashi, SEMS to power quality improvement, in Proc. IEEJ Conference of Powerand Energy Society, 2002, pp. 110-115

[13] C. Hua, J. Lin, and C. Shen, Implementation of a DSP controlled photovoltaic system with peak power tracking, IEEE Trans. Ind. Electron., vol. 45, no. 1, pp. 99-107, Feb.1998. [14] [Online]. Available: http://www.electricitystoarge.com [15] T. Senjyu, T. Kaneko, A. Yona, N. Urasaki, T. Funabashi,F. Yamada, and S. Sugimoto, Output power control for large wind power penetration in small power system, in Proc. IEEE PES General Meeting, session windpower,CD-ROM, 2007. [16] [Online]. Available: http://www.powerlight.com/ success/pdf/ LVVWDRonzone Reservoir.pdf. [17] [Online]. Available: http://www.xantrex.com/web/ id/ 214/p/1840/pt/23/product.asp. [18] N. Femia, G. Petrone, and G. Spagnuolo, and M. Vitelli, Optimization of perturb and observe maximum power point tracking method, IEEE Trans. Power Electron., vol. 20, no. 4, pp. 963973, Jul. 2005.

43