Design Of Data Acquisition Experimental Platform And Identification Modeling For Photovoltaic Power Generation System

Solar energy, as a renewable, clean energy, has an extremely broad application prospect. With the continuous rise of the photovoltaic industry and support for PV projects with governments around the world, photovoltaic power generation has been increasing the proportion of power systems and new energy. Solar cells, as only energy source of whole photovoltaic generation system, the accurate PV model established based on system identification is the foundation, prerequisite and core of simulation, analysis and forecast for photovoltaic power generation system. However, due to the sensitive output characteristics of solar cells affected by temperature, it requires quantitative analysis of the output characteristics ofsolar cells affected by temperature. Yet solar cell temperature, which is the main temperature factor in pv system, can not be replaced by air temperature in the actual working condition. in summary, establishing the experimental platform of photovoltaic power generation system and then modeling solar cell temperature and solar cell/module/array based on the accuracy, reliability and stability of the pv system is the primary theoretical basis for the whole photovoltaic-storage system.Firstly, based on the requirements of system identification theory, it establishes photovoltaic array platform with a real-time tracker and experimental platform for Solar cell temperature and data acquisition system of photovoltaic power generation system respectively. Then it estabilishes steady state thermal model based on Differential Evolution Algorithm and precise prediction model based on Particle Swarm Optimization Algorithm optimized Support Vector Machine for solar cell, whose input variables are air temperature and total solar radiation intensity and input variable is solar cell temperature after its physical mechanism is certainly determined. Secondly, in order to make the best accuracy in modeling, it also establishes the physical and mathematical model for solar cell/module/array, which Differential Evolution Algorithm, due to its system identification theory and function optimization character, is used to identify unknown parameters and special parameters under ideal working condition aftet the experimental data is obtained. Thirdly, The reliability of the PV system are fully considered: reliability requirement for solar cells is proposed after quantitatively analyzing its reverse characteristic, it introduces bypass diode into model of photovoltaic module in consideration under the working condition of inhomogeneous solar radiation and extracts the unknown parameters of the model by Differential Evolution Algorithm again. Then the blocking diode is introduced in PV array performance model. Lastly, The stability of the PV system are fully considered: it establishes dynamic model by introducing the nonlinear junction capacitance and parasitic inductance and shows its corresponding parameter identification method. At the same time, it quantitatively analyzes effect of temperature on the output characteristics of solar cells in both ideal working condition and limit working condition.The researching results show that data acquisition platform bulit will meet the requirements of high-precision and multi-channel acquisition, and that both Solar cell/module/ array model and Solar cell temperature model established in different practical working conditions has obtained the ideal effect no matter what model selection, identification algorithm selection or modeling accuracy.