Analysis And Prediction Of Photovoltaic Power Generation With Influence Factors Of Haze

With the global pollution and energy shortage becoming more and more serious,as an efficient and clean power generation method,photovoltaic(PV)power generation has broad development prospects.However,due to the continuous large-scale haze has caused a significant impact on PV power generation,when the haze reaches a certain concentration,it is the main issue that affects irradiance,which critically decreases PV power generation,affecting grid operation and optimizing dispatching areas。Haze has different influence on the intensity of irradiance,leading to the loss of PV power generation.Making the quantitative analysis for the loss of PV power generation resulted by haze conditions,which plays a significant role in the design of PV power generation systems and the optimal scheduling of regional energy.In this thesis,a kind of improved grey slope correlation degree approach is used to analyze the connection between different concentrations of haze and PV power generation.An exponential linear model is proposed to establish the coupling relationship between haze and irradiance.Combined with PV power generation system model,the loss of PV power generation caused by haze is quantitatively calculated.Through the modeling and analysis of data samples of PV power generation system in Hangzhou,it is calculated that the loss of PV power generation caused by haze in 2017 and 2018 is5.25 ± 1.19% and 6 ± 1.16%,respectively.Secondly,increase the haze influence factors to improve the PV power generation prediction system,combined with the analysis of the characteristics of haze weather,propose a short-term PV power generation prediction model that takes into account the haze factors,the approach of particle swarm optimization(PSO)combining with Cross-validation(CV)is applied to optimize parameters,and support vector regression(SVR)is used to improve the shortterm PV power generation prediction model subject to the influence of haze.Through analyzing the actual data experiment,the prediction effect of the forecast model with haze factor and without haze element under different weather types is compared.It is found that the accuracy of PV power prediction can be significantly improved when considering the influence factors of haze.