Development Of Daily Runoff Prediction System For Jinping Hydropower Station Based On Neural Network

As an important energy source for economic and social development,electricity is an indispensable part of people's production and life.It plays a decisive role in the national economic development.Hydropower is highly regarded as a clean energy source.Jinping Hydropower Station is an important project of the “West-to-East Power Transmission Project” of the country.It has made tremendous efforts to solve the shortage of power in East China.contribution.Runoff forecasting is a powerful means and an important link for fully utilizing water resources,realizing the optimal operation of reservoirs and giving full play to the economic benefits of power plants.With an accurate and reliable runoff prediction system,deterministic forecast values and their estimated errors can be used to quickly and efficiently describe future runoff.Finally,a practical approach to the operation of the power plant is effectively planned.It will be the formulation of the shortterm reservoir operation mode of the hydropower station and the daily generation of electricity.The preparation of the power plan provides the basis for decision-making,thus providing a basis for the hydropower stations to participate in the preparation of the preplan for futures power market development and improving the efficiency of power generation.Daily runoff forecasting is an important part of the runoff forecast for hydropower stations.Artificial neural networks have become a research hotspot in the field of artificial intelligence.Because of this,this topic proposes to use the artificial neural network model as a foundation to establish a combined forecasting model,and through the establishment,verification,and improvement of the research path,it is finally applied to the daily runoff prediction system of the Jinping Hydropower Station.This thesis studies the daily runoff prediction system of the Jinping hydropower station from two aspects of prediction algorithm and program design.The article first elaborates the main background of the research,points out the important value and impact brought by the research,and discusses the relevant research at home and abroad.clear the subject research ideas.Then,the general situation of the Yalong River Basin is summarized from the aspects of physical geography,climate characteristics,runoff characteristics,rainstorm and flood characteristics.The water situation forecasting scheme was designed.The combination forecasting model based on BPNN,GRNN and RBFNN neural network was mainly discussed.The daily runoff of Jinping hydropower station was forecasted.The prediction model was constructed.Finally,the business requirements of the daily runoff forecasting system of the Jinping hydropower station are analyzed,and the system design principles and functional components are pointed out.The overall architecture of the system is designed from logical structure,physical structure and technical architecture,and the system development tools are introduced.The neural network model is introduced.The background program of the combined forecasting model based on neural network and SSH foreground program are designed and implemented.The daily runoff forecasting system of Jinping Hydropower Station Based on neural network is used to improve the accuracy of daily runoff forecast in the river basin,to effectively help the power station to effectively draw up the practical operation strategy and scheduling plan,to improve the efficiency of power generation,and to achieve the purpose of safety service for flood control in the lower reaches of the river.The realization of the optimal scheduling of water energy resources system is a powerful means and important link for the hydropower station to play economic benefits.It has a very important significance to the social and economic development.