Research Of The Short Term Prediction Of Photovoltaic Power Generation In Microgrid

With the development of economy,the demand for energy is increasing,and fossil energy is facing the danger of being exhausted.The pollution problems such as haze and acid rain caused by large amount of energy consumption are also more prominent.In this case,distributed generation,microgrid and other technical applications are widely used in various industries.As an important part of microgrid,power generation prediction of photovoltaic power generation system with intermittent strong features has a certain degree of difficulty.In order to realize the optimal allocation of micro-grid energy and the economic operation of micro-grid,this paper studies the short-term power prediction of micro-grid PV power and makes the following work:(1)Based on the basic principle of photovoltaic cells,the correlation analysis of the photovoltaic cell model is carried out.Based on the analysis of microgrid photovoltaic power generation system structure,the related factors that influence the power generation of microgrid PV system are studied in-depth.(2)This paper proposes a similar sample combination screening method for the traditional sample selection method.First,the samples are selected based on the temporal characteristics of the samples.Secondly,the similarity samples are selected according to the factor of time distance and the Euclidean distance.Finally,the gray correlation analysis is used to select the similar samples.The above-mentioned screening method is applied to the BP neural network,and the short-term prediction model of the power of photovoltaic power generation is carried out.Simulation results show that the proposed method can filter out effective sample data.(3)The least squares support vector machine has the advantages of high modeling precision,strong generalization ability,less computation and less adjustable parameters.Based on the improvement of the traditional particle swarm optimization algorithm,the improved PSO-LSSVM is applied to power generation short-term forecast model.The model is applied to the case calculation.The simulation results show that the proposed method has a better performance than traditional methods and can meet the basic requirements of short-term power prediction of microgrid PV power,which can lay a solid foundation for energy optimization and economic operation of microgrid.