IEEE INDICON 2015 1570176549 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 60 61 62 63 64 65
Asymmetrical Fuzzy logic control to PV Module Connected micro-grid Nikita Gupta1, Rachana Garg2
Parmod Kumar3 MAIT3 Delhi, India [email protected]
Delhi Technological University (DTU) Delhi, India [email protected]
, [email protected]
factors involved in input and requirements[4-6]. Further, the selected fuzzy functions must be asymmetrical to meet dynamic uncertainties. Asymmetric fuzzy functions generates more efficient trigging pulses especially during transient state and complex situation [7-9].
Abstract: This paper proposes asymmetrical fuzzy logic control (AFLC) for PV module connected micro grid. PV module connected micro grid is considered as two stages control, (i) maximum power point tracking control, and (ii) three phase PWM voltage source control. Voltage source converter (VSC) is used for interfacing solar energy with the micro grid and the dc link voltage is regulated using conventional and intelligent controller. During load change transients, there is considerable variation in dc link voltage in case of conventional controller whereas with asymmetrical fuzzy logic control, transient performance is improved considerably. The system has been modelled and implemented in MATLAB Simpower Simulink. Simulation results are presented and analyzed to validate the proposed simulation model.
II. PROBLEM DEFINATION Variation in input parameter (photon energy), variation of demand (loads), and intermediate parameter (temperature) need asymmetrical functions based fuzzy logic control, called asymmetrical fuzzy logic control. In the present study, the authors have simulated a 10kW proposed system as shown in fig.1. PV array is followed by DC-DC boost converter to raise the voltage to desired voltage level and perform maximum power tracking function using perturb and observe algorithm. DC link is connected to micro grid tied DC-AC converter. The converter is controlled using synchronous reference frame theory to control the DC link voltage and for harmonics reduction.
Index Terms- Asymmetrical fuzzy logic control, photovoltaic system, maximum power point tracking, Voltage source converter.
Energy is an important ingredient for any process operation and sustainability. Electrical energy being clean form of energy and easy to transport, utilize, process, and control, it is preferred in all walks of life, industries, communication, transportation, domestic appliances such as light, fan, air conditioning, fridge, or, washing machine, etc. Today, due to various factors including environment pollution, renewable energy sources (Photo Voltaic Cell, Wind Turbine-generator, storage battery, micro hydro-power) for electrical generation have taken the central stage. PV module converters solar light energy to electrical power. Converter plays a vital role for processing, control and synchronization of generated power with micro grid. There can be single or multiple power converter stages for optimization, and processing PV module power output, and for synchronization with grid. Light energy input to PV module, and environment temperature may vary continuously and possess the factor of uncertainty [2, 3]. Also the type and quantum of loads connected to microgrid depend upon the customers need and requirement. Thus, it is preferred to select the fuzzy logic controller based triggering of converter devices, to take into account the uncertainty
III. MODELLING A. PV Panel Model Solar PV array is designed for 10kW. When exposed to sunlight, a solar cell produces an open circuit voltage of about 0.4 to 0.6 volts DC, with the current flow being proportional to the light energy (photons) . So, to obtain the desired power, solar cells are connected in series and parallel and hermetically sealed to constitute a photovoltaic module panel. PV module can be mathematically described by equation (1, 2). I
is the photon current, k is the Boltzmann constant Where (1.38 x 10-23 J/K), q is the charge (1.602 x 10-19 C), Tcell is the cell temperature (K); Rse the series resistance (Ω), Rpar is the shunt resistance (Ω), is module reference temperature of 250C and Tcell is the module operating temperature of 25 to
[978-1-4673-6540-6/15/$31.00 ©2015 IEEE]
Fig. 1. Block diagram of the grid connected solar PV system with synchronous reference frame based controller for Voltage source converter.
700C, Nse and Npar, number of cells in series and parallel respectively and K is the short-circuit current temperature coefficient at ISC = 0.01 /oC. The crystals of PV cell are sensitive to amount and temperature of sunlight falling on it. Fig. 2 depicts the effect of irradiation level and temperature on an IV and P-V curve of illuminated PV array, simulation study carried out by the authors, considering the solar panel of 10 kW peak power capacity. The panel used in proposed simulation work has an open circuit voltage (Voc) of 452 V and short circuit current (Isc) of 30.5 A.
Fig.2. 10 kW PV array I-V and P-V Characteristics for varying irradiation and temperature levels.
B. Boost converter Model In this paper, dc-dc converter boosts the voltage value of PV array to 800 V and perform the maximum power tracking function by applying perturb and observe algorithm. The design parameter of the converter includes passive elementsInductor (L), and Capacitor (C)  and their values can be computed using the equation (7), and (7) ∆
where D is the Duty cycle given by equation (8), 1
and Vi is the input voltage, Vo is the output voltage, ∆ is output current peak to peak ripple, taken as 10% of input current, f is the switching frequency, ∆ is peak to peak ripple in output voltage, taken as 3% of the output voltage and Io is the output current.
fundamental and harmonic components given by equations (10, 11): iLd= iddc + idac (10) iLq= iqdc + iqac (11) They are then passed through LPF to extract dc components iLd, iLq i.e. harmonics are separated from reference signal. For proper operation of Voltage source converter the DC link voltage should be greater than amplitude of grid voltage. The sensed DC link voltage is passed through a low pass filter to suppress the switching noise and harmonic ripples. The output of low pass filter is designated as Vdc. PI and Fuzzy logic controllers are used to maintain the DC link voltage to reference DC link voltage. The output of controller is considered as current for meeting the losses. The reference daxis source current then becomes (12): (12) id*= iddc + iloss The resultant currents in the dq0 reference frame are reverse transformed into reference source current (isa*, isb*, isc*) in abc frame using reverse Park’s transformation given by equation (13), cos sin 1 sin 1 cos (13)
IV. CONTROL ALGORITHMS There are two control algorithms: (i) MPPT algorithm, and (ii) Voltage source converter control algorithm. A. MPPT algorithm Tracking of output of PV module panel is necessary in order to track the maximum power point (MPP) under varying meteorological parameters. These MPPT techniques are based on the reference voltage or reference current signal of the PV system which is adjusted in order to achieve maximum power point . In this paper, perturb and observe MPPT algorithm has been used. This method can be implemented by applying perturbation to the reference voltage or reference current signal of the solar PV system. After the application of perturbation the output power is compared with the previous perturbation cycle power output. If the power increases then the increment in voltage or current remains continuous in same direction. If power decreases then the variation in voltage or current in reverse direction. Fig. 3 shows the schematics of implementation of MPPT control algorithm with a boost converter.
cos sin 1 These reference source currents are fed to hysteresis current controller along with sensed source currents, thus generating switching signal for Voltage source converter. The hysteresis band current controller, controls the source currents (isa, isb, isc) close to three phase reference currents (isa*, isb*, isc*). Indirect current control scheme is employed which compares these reference currents with sensed currents. When the source current is exceeding the upper hysteresis limit, it turns on a negative voltage switching function and causes the source current to decrease. And if the current violates the lower hysteresis limit, then it turns on a positive voltage switching function to increase the source current.
Fig. 3. Block diagram of MPPT algorithm
B. Voltage source converter control In this work, Synchronous reference frame theory is used as the control algorithm for controlling the gating pulse of Voltage source converter . The basic block diagram for synchronous reference frame theory control algorithm is shown in Fig. 1. Feedback signal for the algorithm are taken as sensed values of Load current, PCC voltage and dc bus voltage. In the synchronous reference frame, real and reactive power of the inverter can be controlled separately by d- axis current (Id) and q- axis current (Iq) of inverter. SRF isolator extracts the fundamental component of source currents by transformation of current signals into d-q reference frame using park’s transformation given by equation (9): cos
1)Proportional Integral Controller Proportional-Integral control is the most common control algorithm used in industry. PI algorithm consists of two basic coefficients; proportional and integral, which are varied to get optimal response. The proportional plus integral controller produces an output value which is proportional to voltage error value and also proportional to integral of error value. The output of the PI controller is given by equations (14): (14)
where, Kp = proportional gain, Ki = integral gain, e(t) is the (9) voltage error signal. The advantages of Proportional controller sin sin sin and Integral controller are combined in this P-I controller. In 1/2 1/2 1/2 the present study, PI algorithm is used to maintain the DC bus The phase locked loop (PLL) block generates a signal voltage since this algorithm eliminates the offset of voltage synchronized in the phase to the grid side voltage to provide source converter output. the reference phase angle ‘wt’ used for the ‘abc’ to ‘d-q’ transformation. The d-axis and q- axis currents consist of
2)Fuzzy logic Controller Conventional PI controller are not suitabble for nonlinear variations as they increases the overshoot, settling time and the control action performed is not so approppriate. Fuzzy logic controller is much efficient in dealing with w nonlinearities. Fuzzy Logic unlike Boolean logic, deals with w problems that have fuzziness or vagueness. The generall methodology of reasoning in FL is by “IF…THEN…”statem ments or rules. In general one introduces for each variabble functions of memberships. In present work, seven sets s of triangular memberships are considered as- -Hi: Neggative High. -Me: Negative medium -Lo: Negative low. Ze: Zero. Z +Lo: positive low +Me: positive medium. +Hi: Positive High. H The rule base is shown in Table-I. Fuzzy logic membershipp functions can be of two types- symmetrical and asymmetriccal. In the present studies ALFC is used. Fig. [4-6] shows the asymmetrical membership function used. de
TABLE I. Fuzzy RULE BASE FOR COMPUTTING ILOSS -Hi -Me -Lo Ze +L Lo +Me
The MAX-MIN method is used for inferencing of fuzzy logic controller. The output membeership function of each rule is given by minimum operator, whereas w collective fuzzy output is given by maximum operattor [4, 5]. DC link voltage is regulated with FLC. Sensed DC D voltage value is compared with reference value which generate iloss factor, used for regulating d-axis current. Figg. 7 shows the dynamics of optimized fuzzy logic controlller. From the surface view the control rules of fuzzy logic system s are changed and so the control action varies with respeect to rules.
e -Hi -Me -Lo Ze +Lo +Me +Hi
-Hi -Hi -Hi -Me -Me -Lo Ze
-Hi -Hi -Me -Me -Lo Ze +Lo
-Hi -Me -Me -Lo Ze +Lo +Me
-Me -Me -Lo Ze +Lo +Me +Me
-M Me -L Lo Zee +L Lo Me +M +M Me +H Hi
-Lo Ze +Lo +Me +Me +Hi +Hi
Fig. 7. Surface view of rulee base computing ‘iloss’ component
Ze +Lo +Me +Me +Hi +Hi +Hi
V. SIMULLATION RESULTS The solar PV system connectted to the grid is modeled and simulated using MATLAB Simulink S platform. The system data is given in appendix-I. The T simulation results of PCC voltage (Vpcc), dc link volttage (Vdc), grid side voltage (Vgrid), grid side current (Iggrid), load current (Iload), PV voltage (Vpv), PV current (Ipvv) and PV power (Ppv) for linear load (RL load with 0.9 pf) are presented. p A. Performance of PI controlller The performance for PI conntroller based Voltage source converter is shown in Fig. 8 with time as y- axis, for variation in load. At 0.3 sec additional liinear load with 0.9 lagging pf is added to the system. Its effectt on the various parameters can be seen. As seen from the grapphs, Vdc remains constant, PCC voltage is regulated, current from f the grid side increases to meet the load. DC link voltaage and PCC voltage are well maintained and load currents are a balanced and THD level of grid current is 2.92% and PCC current is 3.55%, which is well within IEEE limits.
Fig. 4. Membership function for input vaariable ‘error’
B. Performance of Fuzzy Loggic controller The performance Fuzzy Logic controller based Voltage source converter is shown in Fig. 10 with time as y- axis, for variation in load. At 0.3 sec additional linear load with 0.9 lagging pf is added to the syystem. Its effect on the various parameters can be seen. As seeen from the graphs, Vdc remains constant, PCC voltage is regullated, current from the grid side increases to meet the load. DC C link voltage and PCC voltage are well maintained and load currents c are balanced and THD level of grid current is 1.53% % and PCC current is 2.40%, which is well within IEEE limiits.
Fig. 5. Membership function for input variablle ‘change in error’
0Fig. 6. Membership function for output vaariable ‘iloss’
Fig. 8: Performance of PI controlleer
Fig. 11. Performancce of Fuzzy logic controller
Fig. 9. Waveform and harmonic analysis for Grid Current (Igrid) for PI controller
Fig. 12. Waveform and harmoniic analysis for Grid Current (Is) for FLC coontroller
Fig. 13. Waveform and harmonic analysis for PCC Current (IPCC) for FLC coontroller
C Current (IPCC) for Fig. 10. Waveform and harmonic analysis for PCC PI controller
 Fig. 14. DC link voltage under transient response of system for PI and FLC
As seen from Fig. 14, the performance of Asymmetrical FLC is better under transient conditions i.e. when the load is varied. Table-II describes the graph. During the load change, although settling time of both the controllers are same, but the overshoot/ undershoot in ALFC controlled system is much less. Performance evaluation of the PI and fuzzy controller are shown in Table-II.
TABLE II. PERFORMANCE EVALUATION OF PI AND FUZZY CONTROLLER Load connected
Undue shoot more less
Over shoot present absent
Under shoot present absent
Over shoot more less
THD Grid Current 2.92 1.53
PCC Current 3.55 2.40
VI. CONCLUSIONS Asymmetrical fuzzy logic control for three-phase micro-gridconnected PV system has been studied. The model has been implemented in MATLAB Simulink. The obtained simulation results show the proposed controller achieved better performance in comparison to PI controller in terms of dc link voltage dynamic responses. THD level of grid side and PCC currents are also improved in case of proposed asymmetrical fuzzy logic control.
APPENDIX-I Data for Sunpower SPR Solar PV-module system Peak power of array=10kW, Short circuit current of module= 8.7 A, open circuit voltage of module = 36.5 V, Cells per module=60, series resistance=0.032Ω, parallel resistance=500.02Ω, Module current at MPP=7.74 A, Module voltage at MPP=29.7. Data for DC-DC boost converter D=0.531, L=1.45mH, C=350uF, fs=10 kHz
Haitham Abu Rub, Mariusz Malinowski, Kamal Al-Hadad, “Power electronics of renewable energy systems, transportation, and industrial application.”, First edition, IEEE press and john wiley and sons publication, 2014 Augustin McEvoy, Tom Markvart and Luis Castaner, “Practical Handbook of Photovoltaics-Fundamentals and Applications”, Second Edition, Elsevier, Wyman Street, USA, 2012 Erickson, Robert W, “Fundamentals of Power Electronics”. Second Edition. Secaucus, NJ, USA, Kluwer Academic Publishers, 2000. C. C Lee, “Fuzzy logic in control systems: Fuzzy logic controller – Part I,” IEEE Transactions on Systems, Man. and Cybernetics, vol. 20, pp. 404-418, April 1990. Devendra K. Chaturvedi, “Soft Computing”, First edition, SpringerVerlag Berlin Heidelberg, 2008 F. Cao and Y. Wang, “Design of a single-phase grid-connected photovoltaic systems based on fuzzy-PID controller” ICIC 2009, LNAI 5755, pp. 912–919, 2009. M. Yatak, O. Bay, “Fuzzy control of a grid connected three phase two stage photovoltaic system”, International conference on power engineering, Spain May 2011. A. Bouafia, F. Krim and J. P. Gaubert, “Design and implementation of high performance direct power control of three-phase PWM rectifier, via fuzzy and PI controller for output voltage regulation”, Energy Conversion and Management, vol. 50, pp. 6-13, 2009. A. Sakhare,A. Davari and A. Feliachi, “Fuzzy logic control of fuel cell for stand-alone and grid connection, ” Journal of Power Sources, vol. 135, pp. 165–176, 2004. TrishanEshram and Patrick L. Chapman, “Comparison of Photovoltaic Array Maximum Power Point Techniques”, IEEE Transaction on Energy Conversion, vol.22, no.2, pp. 439-449,June 2007. Akagi Hirofumi, Watanabe Edson H, Aredes M., “Instantaneous power theory and applications to power conditioning”. IEEE Press-Wiley Press Pvt. Ltd.; 2007. Chinmay Jain, Bhim Singh, “A single- phase two-stage grid interfaced SPV system with adjustable DC link voltage for VSC under non ideal grid conditions”, in proc. Of the IEEE International conference on Power Electronics, Drives, and energy systems, 2014. IEEE Recommended Practices and Requirements for Harmonics Control in Electric Power Systems, IEEE Standard 519, 1992.