Short Term Power Forecasting Of Distributed Photovoltaic Power Plant Based On Neural Network

With the rapid development of world economy,more and more of our energy needs,the fossil energy shortage and environmental protection gradually under the dual pressure,repeated use of clean energy in the power system occupies an increasingly large proportion.Solar energy has attracted great attention from society for its many advantages,and photovoltaic power generation has become a kind of renewable power generation.Because,the photovoltaic power generation with uncertainty and intermittent faults,bring a high degree of difficulty to the operation of large-scale grid connected photovoltaic power generation,the influence of the safety and reliability of power system,economic operation.Therefore,the accurate prediction of photovoltaic power generation has important practical significance and guiding significance.This is the short term prediction of photovoltaic power generation as the research object,through analyzing the affecting factors of short-term photovoltaic power,establish an accurate forecast model,the short-term power of photovoltaic power generation system of prediction.Most of the research results are completed under certain conditions.This paper first analyzes the status of photovoltaic power generation at home and abroad,followed by the introduction of some theoretical basis,composition,structure and classification of the neural network.On the basis of a large number of domestic and foreign literatures,some typical forecasting methods and forecasting models are summarized.This paper analyzes the advantages and disadvantages of the ANN prediction model,and proposes a short-term power prediction method based on the improved BP neural network for distributed PV generation.On the basis of a large amount of historical data,the training of neural network model is carried out.The prediction results of the model prove that the prediction method of short-term prediction power of photovoltaic power generation proposed in this paper is correct and feasible.This study provides a theoretical basis for a large number of photovoltaic power systems,can improve the prediction accuracy of photovoltaic power generation system,power generation scheduling and help formulate detailed plan,in order to improve the operation stability of power generation.