ANewMPPTControlMethodofPhotovoltaicGrid-connectedInverterSystem

1 INTRODUCTIONIn the photovoltaic power generation systems, theoutput characteristics of PV arrays have nonlinearcharacteristics, and its output is affected by sunlightintensity, ambient temperature and load conditions. Undercertain sunlight intensity and ambient temperature, thephotovoltaic array may operateat different output voltages.But only at a certain output voltage, the output power ofthe PV array is able to reach the maximum value, then theoperating point of the PV array can reach the highest pointof output power voltage curve, which is called themaximum power point MPP.In order to take fulladvantage of the energy generated by the photovoltaicsystem, work efficiency of solar cells is improved, theoperating point of the PV arraysneed to be adjustedtimely,so that it is always working in the vicinity of the maximumpower point, this processis called maximum power pointtracking MPPT.Currently, aiming at the control technology ofmaximum power point tracking of the PV array, it hasacertain degree of research at home and abroad, and avariety of new control methods are proposed. Ref.[1] usessingle-stage photovoltaic grid-connected inverter systemas an example, a variable step improved MPPT algorithmis proposed, adopts different step length value whenincreasing or decreasing the PREF, and the purpose oftracking solar maximum working point is achieved byincreasing or decreasing the output active power of theinverter circuit. In Ref.[2], the fuzzy PID control isintroduced into theMPPT control, under certain conditionsof bus voltage stability, the power of the feedback to theThis work is supported by National Natural Science Foundation ofChinaNo 51077046, Hunan Provincial Natural Science Foundation ofChinaNo 14JJ2116, the Construct Program of the Key Discipline inHunan ProvinceNo 201176electrical grid is controlled through adjusting the outputcurrent to ensure the system can steadily and maximumtrack power point at the case of that the illuminationchange is quite dramatic. Due to adopt the fuzzy controlstrategy, control precision of the system is not high, thedesign processis relatively complicated and this methodisa lack of practicality.Taking single-phase photovoltaic grid- connectedpower generation system as an example, this paperproposes a artificial fish swarm algorithm AFSA, thecontrol algorithm has the advantages of fast dynamicresponseand strong optimization capability, and can avoidthe situation of maximum power point falling into a localoptimum value point, thus this control method caneffectively improve the output efficiency of thephotovoltaic array and has good engineering practicalvalue.2 Photovoltaic grid-connected system topologyand control principleThe biggest advantage of photovoltaic grid-connectedpower generation system is the high efficiency, there isonly one link of energy transformation, and the topologystructure is simple without energy storagesectors[3-6] .Thesystem structure is shown in Figure 1, it is mainlycomposed of the PV array, DC-AC link, LCL filter linkand the like. Considering the output voltage of thephotovoltaic cell is low, therefore, the voltage of theinverter bridge also needsto add boost links to connect tothe load terminal.A New MPPT Control Method of Photovoltaic Grid-connected Inverter SystemLI Sheng-qing, ZHANG Bin, XU Tian-jun, YANG JunCollege of Electrical and Information Engineering, Hunan University of Technology, Zhuzhou 412007E-mail zhangb27163.comAbstract Photovoltaic grid-connected inverter system has the advantages of simple topology and low cost. Since theoutput power of photovoltaic devices is a nonlinear function of the external environment load, to maximize theperformance of photovoltaic devices, Maximum Power Point Tracking MPPT control should adjust the duty cycledisturbance based on the work of photovoltaic devices. Although MPPT control with using conventional perturbationand observation method has the advantages of simple in structure and easy to implement the hardware circuit, but thecontrol is imprecise and general poor. Therefore, aiming at the potential inadequacies of using the traditional MPPTcontrol algorithm in a single-stage photovoltaic inverter system, in this paper, the traditional MPPT algorithm isproposed based on a new control algorithm, the artificial fish swarm algorithm is applied to MPPT control of asingle-stage photovoltaic grid-connected system. Parametersettings of AFSA are changed through simulation to seek theoptimal power value, and this control method is compared with the traditional MPPT control method P theschematic diagram is shown in Figure 3, assuming that theinitial point power of the photovoltaic battery working isPa, as the voltage increases,the operating point moves toPb at the next time, at this time HPPb-Pa0, it showsvoltages “disturbance“ is in the right direction, and cancontinue “disturbance“ according to the original direction;if the initial point of power is Pc, use the disturbanceHPPd-Pc Y ithis is seeking maximum, X i is moved a step in thedirection of X j, the formula 2 is performed; Conversely, ifthe condition is not satisfied after repeating try_numbertimes, the formula 3 is performed.ijijinexti| X-XX-XSteprandXX2SteprandXX inexti| 33 AF-Swarm Behavior It refers to an optimizationbehavior of that every fish moves as far as possible to thecenter of the adjacent partnersin the processof swimmingand avoids overcrowding. Artificial fish X i explore thenumber of partners nf and the center position Xc in thecurrent field, If meet Y cnf iY , then move one steptoward the center position, the formula 4 is performed;Otherwise perform foraging behavior.icicinexti| X-XX-XSteprandXX44 AF-Follow Behavior It refers to an act that the fishmove to the optimal direction of the visual area. Artificialfish Xi explore the number of partnersnf in the current fieldand that Y j is the largest partner X j in partners, If meet Y cnfiY , then move one steptoward the center position, theformula 5 is performed; Otherwise perform foragingbehavior.ijijinexti| X-XX-XSteprandXX55 Simulation and Experimental Verification5.1 SimulationIn order to verify the above MPPT algorithm this paperbuild the simulation circuit based on matlab/simulinksimulation software platform and write M-files of artificialfish swarm algorithm, schematic diagram of the systemsimulation is shown in Figure 4. The system mainlyconsists of photovoltaic array and the DC-DC converter,the load side is replacedby a resistor R0. Where C1200uF,L350mH, C2600uF, R010P, the rated voltage andcurrent of the diode is 500V/15A, light intensity is800W/m 2, the temperature is 25 .Figure .4 Schematic diagram of the simulation circuitFor AFSA, its convergence is related to the maximumnumber of iterations, the fish scaleand the maximum steplength, in order to find out the valuesof the parametersthatmaking the algorithm achieve optimal, the A, B, C threekind of scheme are designed, parameter settings are asshown in Table 1.Table1. Parameters of AFSAparameters SchemeASchemeBSchemeCThe maximumnumber ofiterations100 150 150Fish scale 50 50 100Maximumstep2-4 2-4 2-4Crowdingfactor0.618 0.618 0.618MaximumforagingTentativenumber100 100 100Perceiveddistance1 1 1Safety factor 0.8 0.8 0.8Through simulation, the changesof the system optimalvalue can be got in different scenarios; as the number ofiterations increases,the changesof power-optimized valueare as shown in Figure 5SchemeA2014 26th Chinese Control and Decision Conference CCDC 2755Scheme BScheme CFigure .5 Simulation comparison of fish algorithm iterative process ofdifferent schemesFor scheme A, after the start of the simulation, theoptimum value of the maximum power PmaxA 20.501 hasbeenfound when the iteration to the eighth generation withusing 1.823s; scheme B increasesthe number of iterations,power-optimized value can reach the maximum valuePmaxB 20.503 after three jumps when iterating to 62generations, although the optimal value compared toscheme A has slightly improved, but little practicalsignificance, and because the iteration number increaseand lead to increase the amount of time, scheme B uses2.314s.Thus, the optimal effect of simulation is not obviousonly by changing the number of iterations. Scheme Csimultaneously changesthe size of the fish and the numberof iterations, maximum power optimization valuePmaxC20.510 can be got when iteration number is 6 withusing 1.023s, the optimal value and the computing speedhave been improved significantly with comparing withschemeA andB.5.2 ExperimentIn order to verify whether the artificial fish swarmalgorithm in practical condition also ensures that thesystem work stably and optimize to the maximum powerpoint, the settings of simulation parameters arethe sameasthat of section 5.1, the hardware circuit is designed anddeveloped and these experiments were carried out withcomparing with the P there are slight regular fluctuations whenreaching steadystate,while the output waveforms of usingAFSA haven’t overshoot. The output voltage has kept in aconstant state,the output voltage amplitude of using thesetwo algorithms are 16.2V and 18.5V respectively.Obviously, the control precision of PO algorithm ispoorer, and the error difference of the simulation results islarge, it indicates that AFSA has better dynamic responsecharacteristics.6 ConclusionsOn the basis the study of existing control algorithm ofsingle-stage photovoltaic grid-connected inverter system,this paper proposes a new kind of MPPT algorithm,namely, artificial fish swarm algorithm. Through theestablishment of the system simulation model, theoperation of the system is simulated and it is comparedwith the control effect of PO algorithm. Simulation andexperimental results show that single-stage typegrid-connected system can be stable and effectively trackmaximum power point of photovoltaic array trackingunder the control of AFSA, it shows a good dynamiccharacteristic.REFERENCES[1] GU Junyin, CHEN Guocheng. A Current-sensor-lessMPPTAlgorithm for Single-stage Grid-connected PV Inverters[J].Proceedings of the CSEE, 2012, 3227149 -153 inChinese.[2] WU Libo, ZHAO Zhengming, LIU Jianzheng, et al.Research on the Stability of MPPT Strategy Applied inSingle-stage Grid-connected Photovoltaic System[J].Proceedings of the CSEE, 2006,26673-77.inChinese.2756 2014 26th Chinese Control and Decision Conference CCDC[3] Zheng Biwei, Cai Fenghuang, Wang Wu. Analysis andResearch of MPPT Algorithm for a Single-Stage PVGrid-Connected Power Generation System [J].Transactionsof China Electrotechnical Society, 2011, 26790-96.[4] ZHU Xianglin, LIAO Zhiling, LIU Guohai. Researchon theInitial Values of MPPT Algorithm for Solar Cell[J].PowerElectronics,2010,4427-9.[5] ZHANG Chao, HE Xiangning, ZHAO Dean. Research onVariable Perturb Step MPPT Control of PhotovoltaicSystem[J].Power Electronics, 2009,43 1047-49.[6] LIU Fei, DUAN Shanxu, YIN Jinjun, ZHOU Yan. TheMPPT Realization and Stability Study of the Single-stagePhotovoltaic Power System[J].Power Electronics,2008,42328-30.[7] Altas I H, Sharaf A M. A novel maximum power fuzzy logiccontroller for photovoltaic solar energy systems[J].Renewable Energy, 2008, 333 388- 399.[8] Chao Zhang, Dean Zhao. MPPT with asymmetric fuzzycontrol for photovoltaic system[C]. 2009 4th IEEEConference on Industrial Electronics and Applications,Xi ’ an,2009 2180-2183.[9] Li Jing, Dou Wei, Xu Zhengguo, et al. Researchon MPPTMethods of Photovoltaic Power Generation System[J].ActaEnergiae Solaris Sinica, 2007, 283268-273.[10] ZHOU Wenyuan, YUAN Yue, FU Zhixin, et al. ConstantVoltage Tracking Combined with Newton Method MPPTControl for Photovoltaic System[J].Proceedings of theCSU-EPSA, 2012, 24 66-13.[11] LI Xiaolei, SHAO Zhijiang, QIAN Jixin. An Optimizing MethodBased on Autonomous AnimatsFish-swarm Algorithm [J].SystemsEngineering-theory Practice,2002,1132-38.2014 26th Chinese Control and Decision Conference CCDC 2757