Short-term Photovoltaic Output Prediction Model For ESN Networks Based On Improved Grey Wolf Algorithm Optimization

As the global energy crisis continues to deepen,countries have increased the use and research of new energy,and photovoltaic power generation technology,with its unique advantages,has received attention and research from countries around the world.However,the meteorological environment can have an important impact on the application of photovoltaic power generation,and the shortcomings of uncertainty and volatility in photovoltaic output still exist.With the continuous improvement of photovoltaic installed capacity,the shortcomings of uncertainty in photovoltaic output need to be addressed.Otherwise,the upgrading of photovoltaic power generation systems will be a costly project,and it will affect the safe and stable operation of the system itself Economic dispatch and power management pose enormous challenges and threats.Especially in a short period of time,the uncertain output and intermittent fluctuations of photovoltaic power generation systems can have a significant impact on power dispatching operations.The research on photovoltaic output prediction of photovoltaic power generation systems can effectively solve this problem.Based on this background,this paper conducts research on photovoltaic power generation power prediction.This article briefly introduces the basic principles of photovoltaic power generation and the composition of photovoltaic power generation systems,analyzes the characteristic factors that affect photovoltaic output,and performs correlation analysis.Based on the analysis structure,the input variables required for this article are determined.BP and ESN network models are established for short-term prediction of photovoltaic power generation.Aiming at the local optimization problems under a single model,genetic optimization algorithm(GA)and gray wolf optimization algorithm(GWO)are used to optimize the BP network model and ESN network model with super parameters respectively,so as to establish a combined prediction model.In addition,the convergence factor in the Grey Wolf algorithm(GWO)was improved,and an improved IGWO-ESN prediction model based on the Grey Wolf algorithm was proposed to improve the limitations of the Grey Wolf algorithm(GWO).Finally,the principal component analysis(PCA)was introduced to reduce the dimension of input data in photovoltaic power generation power prediction,establishing a PCA-IGWO-ESN prediction model,reducing the impact of data on photovoltaic power generation prediction.The accuracy of the proposed model is verified by analyzing and comparing the experimental structures of various prediction models.