| With development of society and economic, the demand for energy becomemore and more great. At present, the research and application of new energy hasdrawn worldwide concern. Especially, after the Copenhagen conference, someconcepts like low-carbon cities, low-carbon transportation and low-carbon life havebeen put forward. It sketched out a future energy supply system and it also designateda clearly direction for future energy development.Solar energy resource has its unique advantages and occupies a large proportionin future energy. Solar photovoltaic generation technology is an ideal renewableenergy generation technology.Photovoltaic module is an important part of photovoltaic power generationsystem and important energy source. In order to ensure photovoltaic power generationsystem working durable and effective, must understand physical and electricalcharacteristics of PV module. PV module is a highly non-linear DC power, its outputcharacteristic is mainly affect by intensity of solar radiation, ambient temperature andso on. When some regions of PV module are obscured, battery cells in that regionlikely to occur hot spot effect. Through accessing bypass diode can avoid hot spoteffect, but under condition of partial shadow light, the extern electric character of PVarrays will present multi-peak. But traditional maximum power point tracking methodcan’t solve it, also it can’t distinguish local peak value and global peak value.For these reasons, this paper set up a PV power generation system model undercondition of partial shadow, which is according to nature of power consumption andactual usage. Secondly, this paper puts forward that some sensors like light sensor,temperature sensor in PV power generation system effect from Gaussian white noise.According to this effect, this paper designs a Kalman Filter to filter PV powergeneration system, and effectively proved that the Kalman Filter can improve MPPT.Finally, introduce particle swarm algorithm to optimize MPP under dynamic inputcondition, which is based on previous experiments.Therefor, this paper researches some important problem existing in PV power generation system, the main contents as follows.(1) According to basic operation principle to set up a correspondingmathematical model, then uses Matlab/simulink to simulate extern electric characterof PV module, and verifies the correctness and feasibility of simulation result. Build asimulation model of PV array under condition of partial shadow, also explains thepossibility and conditions of hot spot phenomenon occurring.(2) Compare with different tracking methods of maximum power point, andbased on the intensity of solar radiation, temperature, and the shadow of the solar cellelements electrically connected in parallel on the diode array of the outputcharacteristics such as the number of photovoltaic elements; and then emphaticallyexplains advantages and disadvantages of incremental conductance algorithm used attracking MPP, which is widely used. Improve incremental conductance method fromstep adjustment, power oscillation, possibility of misjudgment, model structure andcombined incremental conductance algorithm, to future improve the accuracy ofMPPT algorithm.(3) According to test data which is acquired under Nanchang city’s climatecondition and effect from Gaussian white noise, particle filter is used to filtermeasured sensor’s data, then apply self-programming function module to simulatesystem in Matlab/Simulink environment. Finally, analysis results from experiment,which proves that Kalman Filter can improve PV power generation system’sperformance.(4) According to test data which is acquired under Nanchang city’s climatecondition, effect from Gaussian white noise and Kalman particle filter, applyimproved particle swarm algorithm to track maximum power point. Compareexperiment result with3’s result, this method can achieve further completeness ofconsideration under condition of simulation tracking and diversification of trackingalgorithm. |
PDF Full Text Request