Study On The Stochastic Simulation Model And Application Of Forecast Errors In The Process Of Inbound Runoff

For a long time,hydrological forecast error has always been one of the key factors restricting the benefit of water conservancy projects.Through the establishment of a stochastic simulation model of forecast error in the process of inbound runoff,the quantitative description and analysis of its distribution and evolution law are carried out on this basis.Power generation dispatch risk analysis and application research work has important theoretical significance and practical application value for improving the efficiency of reservoir power generation dispatch under the condition of runoff uncertainty.Based on the Gaussian Mixture Random Simulation(GMM-Copula)method,this paper uses AIC,BIC criteria and K-means++algorithm to optimize the selection of Gaussian distribution mixture numbers,and establishes an improved Gaussian Mixture Random Simulation(IGMM-Copula)method.The stochastic simulation model of the forecast error of the inbound runoff process of the method(SMRPE);secondly,considering the characteristics of high and low changes in the annual runoff of Jinping I,it is divided into flood season,transition period and dry season.The models in this paper were used to simulate the forecast errors of the Jinping-1 inbound runoff process,and compared with the results of the GMM-Copula method,the simulation results closer to the actual inbound runoff process forecast error sequence were obtained;and finally;In order to reveal the impact of the forecast error of the inflow runoff process on the short-term power generation dispatching of the reservoir,the risk of insufficient power generation and the loss of abandonment opportunities are used as risk indicators to estimate and quantify the short-term power generation dispatch risk of the reservoir,which is a water conservancy project under the changing environment.The full use of benefits provides an important theoretical basis and decision-making basis.