Power Load Forecasting Of Large Industrial Users Based On Correlation Analysis And Application

Power load forecasting is the basis of power grid operation management,dispatching and planning.With the adjustment of industrial structure and the implementation of sustainable development policy,large industrial users have become the main and important service objects of power supply enterprises,accurate load forecasting of large industrial users can optimize the structure of power supply and distribution,and improve the operation efficiency of power grid.The development and application of data analysis technology provides a new idea and method for the establishment of load forecasting model.Therefore,this thesis uses correlation analysis method to analyze the relationship between temperature and office load as well as temperature and industrial load of large industrial customers,so as to establish a load forecasting model of large industrial customers based on BP neural network.Firstly,this thesis describes the research background and research status at home and abroad of power load forecasting,and summarizes the application of data analysis technology in load forecasting.Secondly,this thesis introduces the basic concept,general process and analysis index of power load forecasting model,and expounds the neural network algorithm and fuzzy theory.Then,the relationship between temperature and office load as well as industrial load of large industrial customers is analyzed qualitatively and quantitatively,thus,the load forecasting model of large industrial customers is determined,that is,the power consumption of large industrial customers is divided into two parts:office load and industrial load.The fuzzy neural algorithm is introduced in the office power forecasting,and the monthly average temperature is fuzzified into membership function and input into BP neural network for forecasting.The industrial power is forecasted by BP neural network.Then the load forecasting model is programmed and simulated by MATLAB language,and the forecasting error is analyzed.The model is verified and analyzed by inputting different data of large industrial customers.The monthly load forecasting error of large industrial customers is controlled within 3%.and good forecasting effect is achieved.Finally,the short-term and long-term load forecasting of the industrial park is carried out.and the load forecasting results are applied to the 10kV distribution network planning of the industrial park.In order to realize the automation and intellectualization of the distribution network of the industrial park,the distribution network communication of the industrial park is planned,and calculate the economy of the power grid planning.