1596

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 60, NO. 4, APRIL 2013

A Maximum Power Point Tracking Technique for Partially Shaded Photovoltaic Systems in Microgrids
Bader N. Alajmi, Khaled H. Ahmed, Member, IEEE, Stephen J. Finney, and Barry W. Williams

AbstractA modied fuzzy-logic controller for maximum power point (MPP) tracking is proposed to increase photovoltaic (PV) system performance during partially shaded conditions. Instead of perturbing and observing the PV system MPP, the controller scans and stores the maximum power during the perturbing and observing procedures. The controller offers accurate convergence to the global maximum operating point under different partial shadowing conditions. A mathematical model of the PV system under partial shadowing conditions is derived. To validate the proposed modied fuzzy-logic-based controller, simulation and experimentation results are provided. Index TermsBoost converter, fuzzy-logic controller (FLC), maximum power point (MPP) tracker (MPPT), partial shadowing, photovoltaic (PV).

I. I NTRODUCTION HE limitation of conventional energy sources and environmental issues is motivating the world toward the development of microgrid systems. Microgrids can be dened as a system with at least one distributed generation source, associated loads, and storage systems and should be able to operate in grid-connected and island modes [1][5]. The operation of microgrids offers different advantages to customers and utilities, such as improved service reliability, better economics, minimization of energy consumption, and environmental impact [6], [7]. Renewable energies like solar and wind are commonly used as distributed generation sources in microgrid systems, because they are environmentally friendly, a sustainable source of energy, and constructed on either the consumer or utility side [8][10]. Photovoltaic (PV) systems are considered to be one of the most efcient and well-accepted renewable energy sources because of their suitability in distributed generation, mobile applications, transportation, and satellite systems [11][14]. However, PV systems suffer from a major drawback which is the nonlinearity between the output voltage and current particularly under partially shaded conditions [15]. During partially shaded

Manuscript received April 28, 2011; accepted August 18, 2011. Date of publication September 19, 2011; date of current version November 22, 2012. B. N. Alajmi, S. J. Finney, and B. W. Williams are with the Department of Electronic and Electrical Engineering, University of Strathclyde, G11XW Glasgow, U.K. (e-mail: Bader.alajmi@eee.strath.ac.uk; s.nney@eee.strath. ac.uk; b.w.williams@eee.strath.ac.uk). K. H. Ahmed is with the Electrical Engineering Department, Faculty of Engineering, Alexandria University, Alexandria 21544, Egypt (e-mail: khaled.ahmed@ieee.org). Color versions of one or more of the gures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identier 10.1109/TIE.2011.2168796

conditions, the system P V characteristic curve has multiple peaks. Therefore, a conventional maximum power point (MPP) tracker (MPPT) such as hill climbing, incremental conductance, and ripple correlation could miss the global maximum point [16][19]. A study of partial shadowing conditions in [20] shows that using a conventional MPPT during partial shadowing could result in signicant losses of PV output power. Therefore, different researchers have investigated the limitation to improve tracking efciency. Noguchi et al. [21] propose a short-circuit pulse-based MPPT with fast scan on the P V curve to identify the proportional parameter which is commonly used in a current-based MPPT [22]. The global maximum point is found; however, an additional switch in parallel with the PV source is required to compute the shortcircuit current every few minutes. Therefore, such a method causes momentary power losses and requires additional cost. To avoid using an extra switch, the authors in [23] propose a controller that swings the converters duty cycle from zero to one to measure the open-circuit voltage and the short-circuit current and then computes the optimum voltage and current. From the computed values, the operating point is moved in one step to the optimal operating point. The conventional hill climbing algorithm is used to maintain operation around the maximum point. Local maxima are avoided; however, signicant loss in power is experienced during the computation of the open-circuit voltage and the short-circuit current. Based on observation and investigation of the P V characteristic, the authors in [24] claim that, on either side of the global maximum point, the local maxima consistently decrease. Therefore, an MPPT scheme for a PV system under partial shadowing conditions is proposed based on this observation. In [25], a simulation of a partially shaded PV system rejects the observation in [24], and an MPPT based on conventional hill climbing and a partial shadowing identier has been proposed. Also, it is observed [25] that the global maximum point of a PV system under nonuniform conditions is always located to the left of maximum power at normal weather conditions. Therefore, a trajectory line of the PV system under different isolation levels is stored in a data-based memory to identify the partially shaded conditions. Such a technique could save the controller from having unnecessary global scan; however, the trajectory line is different between PV systems, and also, the PV parameters vary with time. A line search algorithm with Fibonacci sequence has been employed to track the global power point under partial shadowing conditions. However, this method can miss the global MPP

0278-0046/$26.00 2011 IEEE

ALAJMI et al.: MPP TRACKING TECHNIQUE FOR PARTIALLY SHADED PHOTOVOLTAIC SYSTEMS IN MICROGRIDS

1597

in some partial shadowing conditions [26]. In [27], a direct search algorithm is employed to search the Lipschitz function which describes the PV power and voltage relationship in an interval. In order for this method to ensure nding the global maximum power, the initial point must be carefully selected; otherwise, the controller may be trapped at a local MPP. Practical swarm optimization has been implied to track the maximum global operating point under abnormal weather conditions. A large time delay is required to allow agents to compute the global maximum point, resulting in a long computation time to reach the maximum operating point [28]. A two-stage MPPT using monitoring cells is proposed in [29], where the operating point in the rst stage moves to the MPP by assuming that the P V curve is uniform. In the second stage, the increment resistor method is used to locate the real MPP. It could overlook the global MPP in some partial shadowing conditions, and also, open-circuit voltage and shortcircuit current measurements are required. To reduce the partial shadowing effect, an adaptive solar PV array is proposed [30]. A model-based control algorithm is used to control a switching matrix that connects a solar adaptive bank to a xed part of the PV array. Similarly, dynamic electrical array reconguration is proposed [31] to improve the PV energy production during partial shadowing conditions. A controllable switching matrix is inserted between the PV generator and the central inverter to allow electrical reconnection of the available PV modules. In these approaches, conventional MPPT can be applied to extract the MPP; however, power stage complexity and cost are increased. In this paper, a fuzzy-logic-based hill climbing technique is proposed to track the global maximum point in a nonuniform P I curve characteristic. The proposed technique identies the global maximum among any number of local maxima. Unlike conventional MPPT where the PV system operating power is perturbed and observed to track the MPP, the proposed controller scans and stores the maximum power during the perturbation and observation stages. Moreover, PV modeling during partial shading conditions is proposed. To demonstrate the effectiveness and robustness of the proposed controller method, computer-aided simulation and experimentation are used to validate the results.

Fig. 1. PV array system.

Fig. 2. Equivalent circuit of a PV cell.

the mathematical model of the generated current in a PV system is represented by Ipv = np Iph Io e
q (Vpv +Rs Ipv ) AkT ns

(Vpv + Rs Ipv ) ns Rsh (1)

II. S YSTEM S TRUCTURE A PV system consists of a solar array which is a group of series-/parallel-connected modules, where the basic building block within the module is a solar cell. Commercially, the solar cell rated power varies between 1 and 2 W depending on the solar cell material and the surface area; therefore, to design a solar module, the solar cells power is measured, and then, the modules are connected in series based on the desired output. Fig. 1 shows a PV system. Generally, the electrical characteristics of the PV system are represented by power versus voltage (current/duty cycle) and by current versus voltage. The characteristic curves of the solar cell are nonlinear because of operational physical phenomena. By using the equivalent circuit of the solar cell shown in Fig. 2,

where Vpv and Ipv represent the PV array output voltage and current, respectively. Rs and Rsh are the solar cell series and shunt resistances, respectively. q is the electron charge (1.6 1019 C), Iph is the light generated current, Io is the reverse saturation current, A is a dimensionless junction material factor, k is Boltzmanns constant (1.38 1023 J/K), T is the temperature (in kelvins), and np and ns are the numbers of cells connected in parallel and series, respectively. The characteristic curves of the PV array system depend on the radiation and temperature of the PV system. For a given system, during normal conditions where the radiation is equally distributed among the PV modules, the powerdutycycle (P D) characteristics under varying weather conditions are shown in Fig. 3. However, when the radiation is not equally distributed, local and global maxima are introduced in the characteristic curves. In order to understand such phenomena, a PV array system with nine modules connected in series and parallel is considered, as shown in Fig. 4. There are different possibilities for the radiation distribution among the PV modules; the following ve cases are randomly considered. Case 1) One module in each column is completely shaded (viz., modules 1, 4, and 7). Case 2) One module in each column is partially shaded with equal radiation levels (viz., modules 2, 5, and 8).

1598

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 60, NO. 4, APRIL 2013

Fig. 5.

PV output power characteristics under the ve cases.

shaded modules result in a reduction of the PV output power, creating local maxima, where the number of local maxima increases as the variation of the radiation levels on each module increases. The PV output power characteristics for the ve different cases are shown in Fig. 5. III. PV M ODELING U NDER PARTIAL S HADING During shadowing conditions, the PV system mathematical model in (1) is no longer valid because different radiation levels are distributed around the PV system. Therefore, a new mathematical model is required to represent the PV system under partial shadowing conditions. An extensive study has been undertaken for different PV module connections to derive a general PV mathematical model under shadowing conditions. For simplicity, three series-connected PV modules are considered. Based on the data for the Shell SP150-PC, the shortcircuit current and the open-circuit voltage for each PV module under rated radiation level are 4.4 A and 43.4 V, respectively. One PV module is partially shaded, and it receives a radiation of 500 W/m2 , while the other two modules receive the rated radiation, which is 1000 W/m2 . The I V characteristic and PV system block diagram are shown in Fig. 6. The three points in Fig. 6(b) are considered: point 1 where the current equals Isc and the voltage equals zero, point 2 where the current equals Istep and the voltage equals Vstep , and point 3 where the current equals zero and the voltage equals Voc . By inspection, the previous variables can be dened as follows. 1) Isc is the short-circuit current of the unshaded PV modules. 2) Istep is the short-circuit current of the shaded PV module. 3) Vstep is the summation of the open-circuit voltages of the unshaded modules. 4) Voc is the summation of the open-circuit voltages of the shaded and unshaded modules. From these observations, the mathematical model of the given PV system is (2), shown at the bottom of the next page, s where nus s and ns are the numbers of series unshaded and shaded modules, respectively, and us and s are the radiation levels on the unshaded and shaded modules, respectively. Equation (2) is valid for two radiation levels distributed around the series-connected PV modules. Therefore, (2) is

Fig. 3. Inuences of (a) temperature (T ) and (b) solar radiation (G) on the P D characteristics.

Fig. 4. PV array system with nine modules connected in series and parallel.

Case 3) One module in each column is partially shaded with unequal radiation levels (viz., modules 3, 6, and 9). Case 4) Two modules in the rst column and one module in each other column are partially shaded with equal radiation levels (viz., modules 1, 2, 4, and 7). Case 5) All modules are partially shaded with different radiation levels. Simulation results for the ve cases indicate that a completely shaded module causes a reduction of the PV output power without creating local maxima. However, partially

ALAJMI et al.: MPP TRACKING TECHNIQUE FOR PARTIALLY SHADED PHOTOVOLTAIC SYSTEMS IN MICROGRIDS

1599

Fig. 7. Characteristic curve of the PV output current and voltage.

Fig. 6. (a) Block diagram of the three-series-connected-PV-module system. (b) Characteristic curve of the PV output current and voltage.

extended to handle three radiation levels, as in (3). When the radiation distribution levels increase, the number of voltage/ current steps increases. Fig. 7 shows the I V characteristic of three series-connected PV modules with different shadowing levels. The PV system shown in Fig. 6(a) is tested under different partial shadowing conditions. Two PV modules are partially shaded and receive two different radiation levels, which are 500 and 300 W/m2 , and the third module receives rated radiation, which is 1000 W/m2 . Two more observations are added to the previous observations as follows. 1) Istep2 is the short-circuit current of the shaded PV modules with the highest radiation level.

2) Vstep1 is the open-circuit voltage of the unshaded modules plus the summation of the shaded module opencircuit voltages without the open-circuit voltage of the lowest radiation module. Therefore, the mathematical model for three different radiation levels on three series-connected PV modules is (3), shown 1 at the bottom of the page, where ns s is the number of partially 2 shaded PV modules with the lowest radiation level and ns s is the number of partially shaded PV modules with the highest radiation level. s1 is the lowest radiation level, and s2 is the highest radiation level. From (2) and (3), the general mathematical model of n seriesconnected PV modules in a PV system is (4), shown at the bottom of the next page, where nsN s is the number of partially shaded PV modules with the highest radiation level and sN is the highest radiation level. N is the number of distributed radiation levels. Usually, the PV system consists of parallel-/series-connected PV modules. Therefore, to drive a mathematical model for the overall PV system, the previous three series-connected PV modules are connected in parallel with another three seriesconnected PV modules, as shown in Fig. 8(a). For simplicity, the radiation levels are distributed as follows. In the rst branch, modules 1 and 2 receive a radiation of 1000 W/m2 , and the shaded module receives 500 W/m2 . The

AkT nus s ln q Vpv = s AkT nus ln q

Isc us Ipv Io Isc us Ipv Io

, +
AkT ns s q

Ipv > Istep ln

Isc s Ipv Io

(2) , Ipv < Istep

Vpv =

AkT nus s q AkT nus s q AkT nus s q

ln ln ln

Isc un Ipv , Io 1 Isc un Ipv I s I ns s + AkT ln sc Io pv Io q 1 Isc Ipv I s1 I ns s + AkT ln sc Io pv Io q

Ipv > I2step , +

2 AkT ns s q

I1step < Ipv < I2step ln

Isc s2 Ipv Io

(3)

Ipv < I1step

1600

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 60, NO. 4, APRIL 2013

IV. P ROPOSED M ETHOD Unlike conventional MPPTs where the PV system operating power is perturbed and observed to track the MPP, scanning, storing, perturbing, and observing the operating power of the PV system are used for the proposed MPPT. The proposed method is able to track the MPP under any weather conditions, particularly partial shadowing where local and global maximum points exist. During the initial condition or varying weather conditions, the proposed MPPT makes a wide range search to scan and store the maximum power value on the PV system. A preset value which represents the accepted difference between the identied maximum power and the operated power is stored to decide the controller rules. If the difference between the identied maximum power and the operated power is greater than the preset value, the duty cycle is increased; otherwise, fuzzy-logic-based MPPT is applied. In this case, the algorithm ensures that the MPPT is not trapped by local maxima and quickly recovers the new global maximum point during varying weather conditions. Unlike conventional scanning MPPT, the essentiality of using a long time delay is not required because the controller scans the P V curve while perturbation and observation are carried out. Fig. 9 shows the owchart of the proposed method, where Vpv and Ipv are the PV output voltage and current, respectively, D is the duty cycle, Pm is the global MPP, and Pm is a constant that identies the allowable difference between the global maximum point and the operating power point. Three scanning and storing techniques are proposed to identify the global maximum power during initial conditions or varying weather conditions. The rst technique is to initialize the system with maximum duty cycle since the PV output power usually takes some samples before reaching the operating point at maximum duty cycle. The P V characteristic curve is scanned, and the global MPP is stored. The second technique is to increase the duty cycle from a minimum to a maximum value with a xed step. In this case, the P V curve is scanned, and the global MPP is stored. The last technique is to apply a large initial perturbation step to make a wide search range on the PV power locus. Unlike the previous two proposed techniques, scanning and storing the PV power are accomplished during perturbation and observation. The three techniques guarantee to nd and store the global MPP; however, the global identication time is different from one to another. Moreover, the duty cycle must

Fig. 8. (a) Block diagram of the series-/parallel-connected-PV-module system. (b) Characteristic curve of the PV output current and voltage.

three PV modules in the second branch are partially shaded to receive a radiation of 300 W/m2 . The I V characteristic of the PV system is shown in Fig. 8(b). Curve A represents the I V characteristic of the whole PV system, curve B represents the I V characteristic of branch 1, and curve C represents the I V characteristic of branch 2. Clearly, the PV output current is the summation of the instant currents from each branch IpvTotal = Ibranch1 + Ibranch2 . (5)

From (5), the general mathematical model of N parallelconnected PV modules in the PV system is IpvTotal = Ibranch1 + Ibranch2 + + IbranchN . (6)

Vpv =

AkT nus s q

ln

Isc un Ipv Io

AkT nus s q AkT nus s q

. . .

ln ln

1 Isc un Ipv I s I ns s + AkT ln sc Io pv Io q 1 Isc Ipv I s1 I ns s + AkT ln sc Io pv Io q

, + +
AkT nsN s q

Ipv > IN step . . . I1step < Ipv < I2step ln

Isc sN Ipv Io

Ipv < I1step (4)

ALAJMI et al.: MPP TRACKING TECHNIQUE FOR PARTIALLY SHADED PHOTOVOLTAIC SYSTEMS IN MICROGRIDS

1601

TABLE I F UZZY-L OGIC RULES

is considered in the proposed MPPT design. The inputs to the fuzzy-logic controller (FLC) are P = P (k ) P (k 1) I = I (k ) I (k 1) PM = Pm (k ) P (k ) and the output equation is
Fig. 9. Proposed method owchart.

(7) (8) (9)

D = D(k ) D(k 1)

(10)

Fig. 10. PV array system block diagram and the proposed MPPT controller.

return to a minimum value whenever it exceeds the maximum value to track the global MPP. Any conventional MPPT method can be used with the proposed method; however, fuzzy-logic-based MPPT [32] is preferred specially when using the third technique because the tracking speed is not constant. Therefore, during initial conditions or varying weather conditions, the initial tracking speed should be fast enough to make a wide range power scan and store the maximum available power. On the other side, when the operating point reaches the global maximum, the tracking speed decreases to minimize any oscillation around the global maximum point. The PV system block diagram, along with the proposed controller, is shown in Fig. 10.

where P and I are the PV array output power change and current change, respectively, PM is the difference between the stored global maximum power (PM ) and the current power, and D is the boost converter duty cycle change. To ensure that the PV global maximum power is stored during the scanning procedure, a fast initial tracking speed is used. The variable inputs P and I are divided into four fuzzy subsets: positive big (PB), positive small (PS), negative big (NB), and negative small (NS). The variable input PM is divided into two fuzzy subsets: PB and PS. The output variable D is divided into six fuzzy subsets: PB, positive medium (PM), PS, NB, negative medium (NM), and NS. Therefore, the fuzzy algorithm requires 32 fuzzy control rules; these rules are based on the regulation of a hill climbing algorithm along with the reference power. To operate the fuzzy combination, Mamdanis method with maxmin is used. The fuzzy rules are shown in Table I. After simulating the PV system and studying the behavior of the controller inputs and output, the shapes and fuzzy subset partitions of the membership function in both of the inputs and output are shown in Fig. 11. The last fuzzy controller stage is defuzzication where the center-of-the-area algorithm is used to convert the fuzzy subset duty cycle changes to real numbers
n

(Di )Di D =
i n i

(Di )

(11)

where D is the fuzzy controller output and Di is the center of the maxmin composition at the output membership function. VI. S IMULATION R ESULTS AND D ISCUSSION The tested PV array is composed of ten series modules with a rated power of 850 W, where the design specication and circuit parameters are shown in Table II. The simulation results are

V. FLC D ESIGN Modication to the fuzzy-logic-based MPPT algorithm, using the scanning and storing procedures, is proposed to quickly locate the global MPP. Fuzzication of the owchart in Fig. 9

1602

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 60, NO. 4, APRIL 2013

Fig. 11. Membership functions: (a) Input P , (b) input I , (c) input PM , and (d) output D. TABLE II D ESIGN S PECIFICATION AND C IRCUIT

Fig. 13. PV output power for the proposed MPPT under partially shaded conditions along with the PV power locus.

Fig. 14. Output power for a partially shaded PV system under two radiation levels along with the PV power locus.

Fig. 12. PV output power characteristic for the proposed MPPT under two radiation levels.

carried out using Matlab/Simulink to validate the performance of the proposed MPPT. A large initial perturbation step is used to scan and store the global MPP. A. Normal Weather Conditions The proposed system is tested under two equally distributed radiation levels: 500 and 1000 W/m2 . As shown in Fig. 12, the global maximum scan causes some power loss during initial and varying weather conditions. Nevertheless, the proposed MPPT still attains the MPP in a relatively short time, with small oscillation in steady state. Partial Shadowing Condition: To validate the performance of the proposed MPPT during partial shading, the PV system is tested under different distributed radiation levels; seven unshaded modules receive 1000 W/m2 while the other three modules are partially shaded with radiation levels of 800, 500, and 100 W/m2 . The PV output power, along with the system

power locus, is shown in Fig. 13. Clearly, the local maxima do not prevent the proposed controller from reaching the global MPP. In addition, the tracking of the global maximum is fast with small oscillations at steady state. For further verication, the same partial shadowing conditions are repeated with different radiation levels on the unshaded modules; the radiation level varies from 500 to 1000 W/m2 at 0.5 s and then from 1000 to 500 W/m2 at 1 s. From Fig. 14, the proposed MPPT scans and then tracks the global maximum point in a relatively short time even under varying weather conditions. To clarify and analyze the behavior of the proposed MPPT, Fig. 15 shows the duty cycle of the PV system converter under varying weather conditions of three partially shaded modules. In the initial stage, a wide duty cycle range search is applied to scan for the global maximum point. Once the global maximum power is found, the controller stores the value and compares it with the current operation power. If the difference between the stored global maximum power and the current operating power is greater than a preset value, the duty cycle increases. On the other side, the fuzzy-logic control perturbs and observes the PV system to keep the operation at the optimum duty cycle. During varying weather conditions, the controller resets the stored global maximum value and repeats the same process to nd the new global MPP.

ALAJMI et al.: MPP TRACKING TECHNIQUE FOR PARTIALLY SHADED PHOTOVOLTAIC SYSTEMS IN MICROGRIDS

1603

Fig. 15. Duty cycle behavior of the proposed MPPT.

Fig. 16. PV output power for a partially shaded PV system under the three scanning techniques.

Fig. 17. (a) Hardware diagram arrangement. (b) Test rig photograph.

B. Comparison Between the Three Proposed Scanning Techniques As mentioned, three scanning techniques can be used to identify the global maximum power during the initial condition or varying weather conditions. 1) Initialize the system with a duty cycle close to the maximum allowable duty cycle. 2) Increase the duty cycle from minimum to maximum with a xed step. 3) Apply a large perturbation step during initial or varying weather conditions. A comparison between these scanning techniques is tested for a PV system under different distributed radiation levels; seven unshaded modules receive 1000 W/m2 while the other three modules are partially shaded with radiation levels of 800, 500, and 100 W/m2 . As shown in Fig. 16, the global maximum is attained for the three proposed scanning techniques. However, the techniques offer different scanning ranges and speeds. For the rst two techniques, the PV power locus is fully scanned to identify the global maximum, while the third technique scans only a wide range of the PV power locus. Therefore, the third technique is slightly faster with less power losses during the initial and varying weather conditions. In the three techniques, the duty cycle must return to a minimum value whenever it exceeds the maximum value to track the global MPP.

VII. E XPERIMENTAL R ESULTS Experimental verication of the proposed MPPT is achieved with the conguration shown in Fig. 17. The experimental setup consists of an Agilent modular solar array simulator to generate the PV system I V curves, a boost converter with 4-kHz switching frequency to boost the output voltage and track the MPP, and a battery load to x the boost converter output voltage and store the PV system energy. An Inneon TriCore TC1796 is used to realize the proposed MPPT. Different I V curves are programmed into the PV source simulator to test the experimental system under different weather conditions. The results of the proposed MPPT under two uniform insolation conditions are shown in Fig. 18. The MPP is attained after scanning and tracking the I V curve in a relatively short time, with small oscillation in steady state. When the radiation level varies, the proposed MPPT is reset and repeats the scanning and tracking procedures. To conrm the effectiveness of the proposed MPPT, (4) is programmed into the PV source simulator to emulate the PV system characteristic curve under partial shadowing conditions. As shown in Fig. 19, the proposed fuzzy-logic-based MPPT with a large initial perturbation step scans and stores the power locus during perturbation and observation. The local maxima in the P V characteristic do not prevent the proposed MPPT from successfully capturing the global MPP in a relatively short time, with small oscillation around the MPP.

1604

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 60, NO. 4, APRIL 2013

Fig. 18. levels.

Experimental results of the proposed MPPT under two radiation

Fig. 20. Experimental results. (a) I V characteristic curve of a real PV system. (b) Proposed MPPT under real PV conditions.

VIII. C ONCLUSION
Fig. 19. Experimental results of the proposed MPPT under partial shading. TABLE III PV M ODULE DATA

For further verication of the proposed MPPT, a roofinstalled PV array at our university building is used to collect data. The PV array system consists of eight PV modules connected in series with the specication, at standard test conditions, as shown in Table III. The PV array system is articially shaded to generate an I V characteristic with three local MPPs as shown in Fig. 20(a). After programming the I V characteristic in the PV simulator, the proposed MPPT tracks the global MPP by applying a large perturbation step to the system duty cycle, to scan and store the global MPP. Once the global maximum is found, the fuzzy perturbation and observation take over to move the operating point to the global MPP, with a small oscillation around it. The power, voltage, and current curves of the PV system with the proposed MPPT are shown in Fig. 20(b).

In this paper, a fuzzy-logic-based MPPT has been proposed to extract the global MPP under partially shaded PV system conditions. The proposed MPPT has been implemented by combining fuzzy-logic-based MPPT with a scanning and storing system. Three scanning techniques have been proposed to scan the PV power characteristic curve and store the maximum power value, during initial and varying weather conditions. A new mathematical model has been proposed to represent the behavior of the P V characteristic under partial shadowing conditions. Matlab/Simulink simulations and practical experiments of a partially shaded PV system have been carried out to validate the proposed MPPT. The results show that the proposed MPPT is able to reach the global MPP under any partial shading conditions. Moreover, the controller exhibits a fast converging speed, with small oscillation around the MPP during steady state. R EFERENCES
[1] I. J. Balaguer, L. Qin, Y. Shuitao, U. Supatti, and P. F. Zheng, Control for grid-connected and intentional islanding operations of distributed power generation, IEEE Trans. Ind. Electron., vol. 58, no. 1, pp. 147157, Jan. 2011. [2] J. C. Vasquez, R. A. Mastromauro, J. M. Guerrero, and M. Liserre, Voltage support provided by a droop-controlled multifunctional inverter, IEEE Trans. Ind. Electron., vol. 56, no. 11, pp. 45104519, Nov. 2009.

ALAJMI et al.: MPP TRACKING TECHNIQUE FOR PARTIALLY SHADED PHOTOVOLTAIC SYSTEMS IN MICROGRIDS

1605

[3] J. M. Guerrero, J. C. Vasquez, J. Matas, M. Castilla, and L. G. de Vicuna, Control strategy for exible microgrid based on parallel line-interactive UPS systems, IEEE Trans. Ind. Electron., vol. 56, no. 3, pp. 726736, Mar. 2009. [4] J. C. Vasquez, J. M. Guerrero, A. Luna, P. Rodriguez, and R. Teodorescu, Adaptive droop control applied to voltage-source inverters operating in grid-connected and islanded modes, IEEE Trans. Ind. Electron., vol. 56, no. 10, pp. 40884096, Oct 2009. [5] J. M. Guerrero, J. C. Vasquez, J. Matas, L. G. de Vicuna, and M. Castilla, Hierarchical control of droop-controlled AC and DC microgridsA general approach toward standardization, IEEE Trans. Ind. Electron., vol. 58, no. 1, pp. 158172, Jan. 2011. [6] J. M. Guerrero, F. Blaabjerg, T. Zhelev, K. Hemmes, E. Monmasson, S. Jemei, M. P. Comech, R. Granadino, and J. I. Frau, Distributed generation: Toward a new energy paradigm, IEEE Ind. Electron. Mag., vol. 4, no. 1, pp. 5264, Mar. 2010. [7] K. H. Ahmed, A. M. Massoud, S. J. Finney, and B. W. Williams, A modied stationary reference frame-based predictive current control with zero steady-state error for LCL coupled inverter-based distributed generation systems, IEEE Trans. Ind. Electron., vol. 58, no. 4, pp. 13591370, Apr. 2011. [8] M. Liserre, T. Sauter, and J. Y. Hung, Future energy systems: Integrating renewable energy sources into the Smart power grid through industrial electronics, IEEE Ind. Electron. Mag., vol. 4, no. 1, pp. 1837, Mar. 2010. [9] J. M. Carrasco, L. G. Franquelo, J. T. Bialasiewicz, E. Galvan, R. C. P. Guisado, M. A. M. Prats, J. I. Leon, and N. Moreno-Alfonso, Powerelectronic systems for the grid integration of renewable energy sources: A survey, IEEE Trans. Ind. Electron., vol. 53, no. 4, pp. 10021016, Jun. 2006. [10] C. Yuen, A. Oudalov, and A. Timbus, The provision of frequency control reserves from multiple microgrids, IEEE Trans. Ind. Electron., vol. 58, no. 1, pp. 173183, Jan. 2011. [11] J. T. Bialasiewicz, Renewable energy systems with photovoltaic power generators: Operation and modeling, IEEE Trans. Ind. Electron., vol. 55, no. 7, pp. 27522758, Jul. 2008. [12] R. Lanzafame, S. Nachtmann, M. Rosa-Clot, P. Rosa-Clot, P. F. Scandura, S. Taddei, and G. M. Tina, Field experience with performances evaluation of a single-crystalline photovoltaic panel in an underwater environment, IEEE Trans. Ind. Electron., vol. 57, no. 7, pp. 24922498, Jul. 2010. [13] C. Rodriguez and J. D. K. Bishop, Organic architecture for smallto large-scale photovoltaic power stations, IEEE Trans. Ind. Electron., vol. 56, no. 11, pp. 43324343, Nov. 2009. [14] L. Wuhua and H. Xiangning, Review of nonisolated high-step-up DC/DC converters in photovoltaic grid-connected applications, IEEE Trans. Ind. Electron., vol. 58, no. 4, pp. 12391250, Apr. 2011. [15] G. Lijun, R. A. Dougal, L. Shengyi, and A. P. Iotova, Parallel-connected solar PV system to address partial and rapidly uctuating shadow conditions, IEEE Trans. Ind. Electron., vol. 56, no. 5, pp. 15481556, May 2009. [16] T. Esram and P. L. Chapman, Comparison of photovoltaic array maximum power point tracking techniques, IEEE Trans. Energy Convers., vol. 22, no. 2, pp. 439449, Jul. 2007. [17] A. Safari and S. Mekhilef, Simulation and hardware implementation of incremental conductance MPPT with direct control method using Cuk converter, IEEE Trans. Ind. Electron., vol. 58, no. 4, pp. 11541161, Apr. 2011. [18] L. Fangrui, K. Yong, Z. Yu, and D. Shanxu, Comparison of P&O and hill climbing MPPT methods for grid-connected PV converter, in Proc. 3rd ICIEA, 2008, pp. 804807. [19] J. W. Kimball and P. T. Krein, Discrete-time ripple correlation control for maximum power point tracking, IEEE Trans. Power Electron., vol. 23, no. 5, pp. 23532362, Sep. 2008. [20] H. Patel and V. Agarwal, Matlab-based modeling to study the effects of partial shading on PV array characteristics, IEEE Trans. Energy Convers., vol. 23, no. 1, pp. 302310, Mar. 2008. [21] T. Noguchi, S. Togashi, and R. Nakamoto, Short-current pulse-based maximum-power-point tracking method for multiple photovoltaic-andconverter module system, IEEE Trans. Ind. Electron., vol. 49, no. 1, pp. 217223, Feb. 2002. [22] M. A. S. Masoum, H. Dehbonei, and E. F. Fuchs, Theoretical and experimental analyses of photovoltaic systems with voltage and current-based maximum power-point tracking, IEEE Trans. Energy Convers., vol. 17, no. 4, pp. 514522, Dec. 2002. [23] S. Kazmi, H. Goto, O. Ichinokura, and H.-J. Guo, An improved and very efcient MPPT controller for PV systems subjected to rapidly varying

[24] [25] [26]

[27] [28]

[29]

[30] [31]

[32]

atmospheric conditions and partial shading, in Proc. AUPEC, 2009, pp. 16. H. Patel and V. Agarwal, Maximum power point tracking scheme for PV systems operating under partially shaded conditions, IEEE Trans. Ind. Electron., vol. 55, no. 4, pp. 16891698, Apr. 2008. Y. Liu, Advance control of photovoltaic converters, Ph.D. dissertation, Dept. Eng., Univ. Leicester, Leicester, U.K., Apr., 2009. M. Miyatake, T. Inada, I. Hiratsuka, H. Zhao, H. Otsuka, and M. Nakano, Control characteristics of a Fibonacci-search-based maximum power point tracker when a photovoltaic array is partially shaded, in Proc. 4th IPEMC, 2004, vol. 2, pp. 816821. T. L. Nguyen and K.-S. Low, A global maximum power point tracking scheme employing DIRECT search algorithm for photovoltaic systems, IEEE Trans. Ind. Electron., vol. 57, no. 10, pp. 34563467, Oct. 2010. M. Miyatake, M. Veerachary, F. Toriumi, N. Fujii, and H. Ko, Maximum power point tracking of multiple photovoltaic arrays: A PSO approach, IEEE Trans. Aerosp. Electron. Syst., vol. 47, no. 1, pp. 367380, Jan. 2011. K. Kobayashi, I. Takano, and Y. Sawada, A study on a two stage maximum power point tracking control of a photovoltaic system under partially shaded insolation conditions, in Proc. IEEE Power Eng. Soc. Gen. Meeting, 2003, vol. 4, p. 2617. D. Nguyen and B. Lehman, An adaptive solar photovoltaic array using model-based reconguration algorithm, IEEE Trans. Ind. Electron., vol. 55, no. 7, pp. 26442654, Jul. 2008. G. Velasco-Quesada, F. Guinjoan-Gispert, R. Pique-Lopez, M. RomanLumbreras, and A. Conesa-Roca, Electrical PV array reconguration strategy for energy extraction improvement in grid-connected PV systems, IEEE Trans. Ind. Electron., vol. 56, no. 11, pp. 43194331, Nov. 2009. B. Alajmi, K. Ahmed, S. Finney, and B. Williams, Fuzzy logic controlled approach of a modied hill climbing method for maximum power point in microgrid stand-alone photovoltaic system, IEEE Trans. Power Electron., vol. 26, no. 4, pp. 10221030, Apr. 2011.

Bader N. Alajmi received the B.Sc. and M.Sc. degrees from California State University, Fresno, in 2001 and 2006, respectively. He is currently working toward the Ph.D. degree in electrical engineering in the Department of Electronic and Electrical Engineering, University of Strathclyde, Glasgow, U.K. His research interests are digital control of power electronic systems, microgrids, and distributed generation, photovoltaic inverters, and dcdc converters.

Khaled H. Ahmed (S08M10) received the B.Sc. (rst-class honor) and M.Sc. degrees from the Faculty of Engineering, Alexandria University, Alexandria, Egypt, in 2002 and 2004, respectively, and the Ph.D. degree in electrical engineering from the Department of Electronic and Electrical Engineering, University of Strathclyde, Glasgow, U.K., in 2008. In 2009, he was appointed Lecturer at Alexandria University. His research interests are digital control of power electronic systems, power quality, microgrids, and distributed generation. He has published over 21 technical papers in refereed journals and conferences in the last three years. Dr. Ahmed is a Reviewer for various IEEE T RANSACTIONS and several conferences.

1606

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 60, NO. 4, APRIL 2013

Stephen J. Finney received the M.Eng. degree from Loughborough University of Technology, Loughborough, U.K., in 1988 and the Ph.D. degree from Heriot-Watt University, Edinburgh, U.K., in 1995. For two years, he was with the Electricity Council Research Centre laboratories near Chester, U.K. He is currently a Professor with the University of Strathclyde, Glasgow, U.K. His areas of research interest are soft-switching techniques, power semiconductor protection, energy recovery snubber circuits, and low-distortion rectier topologies.

Barry W. Williams received the M.Eng.Sc. degree from The University of Adelaide, Adelaide, Australia, in 1978 and the Ph.D. degree from the University of Cambridge, Cambridge, U.K., in 1980. After seven years as a Lecturer at Imperial College, University of London, London, U.K., he was appointed Chair of electrical engineering at Heriot-Watt University, Edinburgh, U.K., in 1986. He is currently a Professor with the University of Strathclyde, Glasgow, U.K. His teaching covers power electronics (in which he has a free Internet text) and drive systems. His research activities include power semiconductor modeling and protection, converter topologies, soft-switching techniques, and application of application-specic integrated circuits and microprocessors to industrial electronics.