Research On Maximum Power Point Tracking And Fault Diagnosis Of Photovoltaic System

With the development of world, how to improve the efficiency of energy, optimizing the structure and developing the new energy have become the key to the sustainable development of the every country. The efficiency is the main problem in the development process need to be solved. How to improve the output efficiency of photovoltaic cell based on the existing conversion rate has become a hot spot. The measures to improve the efficiency of a photovoltaic system are maximum power point tracking(MPPT) and fault diagnosis. However the output of the photovoltaic array is easily influenced by the irradiance, temperature and the shading condition, and then output characteristics would present nonlinear. Especially under the partial shading condition the P-V curve of photovoltaic would appear multiple peak, while conventional MPPT algorithm(such as hill climbing method,perturbation and observation, etc.) is easy to fall into local optimum instead of maximum power point, which will reduce the system efficiency greatly. The conventional fault diagnosis methods either need a lot of information leads to high cost or can not judge the fault of system online and real-time. Therefore it is greatly significant to put forward a more efficient and accurate MPPT algorithm and online fault diagnosis with low cost to improve the efficiency and safety of photovoltaic system.In order to solve MPPT problem, this paper presents an improved Global scanning& variable step PO method(GS&VSPO). At first this algorithm adopts variable step to search the peak and storage at the high speed to track the local optimum, and then compare all the peak through the global scanning method to obtain the real maximum power point to make the photovoltaic system work at the real maximum power point.In order to solve fault diagnosis problem, this paper presents a novel method to optimize the number of sensor. This method measures the current and voltage online to compare the current and voltage of branch to detect the fault type and location effectively.This paper took use of MATLAB/SIMULINK to compare the three MPPT algorithms through simulation experiment. The results show the improved variable step PO &Global scanning method can effectively track the maximum power point of photovoltaic array under different shading condition with higher accuracy, less power oscillation and less energy loss. In addition, this paper also made simulation of the fault diagnosis method proposed which combine the improved MPPT, and the results show the feasibility and effectiveness in reducing the cost and detecting the fault type and location.