In order to improve the reliability and accuracy of photovoltaic power generation power prediction

a short-term photovoltaic power generation power prediction model based on improved sparrow search algorithm(t SSA) optimization support vector machine(SVM) was proposed. Firstly

the self-adaptive t-distribution was used to enhance the population diversity of the sparrow search algorithm

and then t SSA was used to optimize the penalty parameters and kernel function parameters in the SVM

the photovoltaic power generation power was predicted when the optimal penalty parameters and kernel function parameters were obtained. An analysis on the data of a photovoltaic power station in Australia within one year found that

the photovoltaic power in each season was significantly different. For this reason

the influence of external environmental parameters such as irradiance

humidity

temperature

wind speed and wind direction was fully considered

and the photovoltaic power generation power on a certain day in different seasons was predicted based on the t SSA-SVM model. The simulation results show that

the above model has high prediction accuracy

especially for the prediction of photovoltaic power generation in autumn and winter

so it has high practical value.