Research On Energy Demand Forecast Analysis And Development Countermeasure Of Beijing,Tianjin And Hebei

Energy is an important material basis for economic development and social progress,and it is the guarantee of all kinds of human activities.Beijing Tianjin Hebei coordinated development is a major national development strategy,but also an important core area of national economic development.In the rapid development of the economy at the same time,energy is essential to ensure that.According to the analysis,due to the economic losses caused by the lack of energy,about 20 of the value of the energy itself-60 times.Energy demand forecast is the basis of energy planning and policy formulation,analysis of the regional energy demand in Beijing,Tianjin and help the government to accurately formulate energy policies to ensure sustainable and healthy development of Beijing,Tianjin and the economy.Therefore,it has important theoretical and practical significance to predict the energy demand of Beijing,Tianjin and Hebei.A lot of energy demand forecasting methods used ARIMA model,multiple regression,grey prediction,exponential smoothing,trend extrapolation method and the neural network model prediction method.These methods have their advantages and disadvantages,but they are not mutually exclusive,but they are mutually compatible and complementary.Therefore,Bates and Grange put forward the idea of combination forecasting.The combination forecasting method can effectively keep the valuable data information of the single forecasting method,and has higher prediction precision and better stability than the single forecasting method.However,the current energy demand forecasting literature,how to choose a single prediction method,the number of single forecasting methods without the basis,and the combination of the form of most remain in the linear combination or nonlinear combination of simple form,and weight coefficient of nonlinear combination determined the difficult,complex calculation.Aiming at these problems,the following research work is done in this paper:First,this paper studies the factors affecting energy demand,analyzes the impact of economic growth,industrial structure,population and city,energy consumption structure,technological progress,the level of consumption,energy prices and environmental policy related factors and energy demand.Second,this paper puts forward the basic principle of choosing a single prediction method,the number of subjective and objective combination,linear model and nonlinear model of single forecasting method to moderate(3 to 5 more suitable principles).This paper selects the above principles based on the ARIMA model(objective linear model),grey prediction method(objective nonlinear model)and multiple regression(subjective linear model),two nonlinear energy demand forecast model(subjective nonlinear model)and grey neural network model(subjective nonlinear model)and other 5 kinds of prediction methods.Third,the nonlinear combined model is more accurate than the linear combination model,but it is difficult to determine the weight of the combined model,and the calculation is complex.Therefore,this paper proposes a combination forecasting model based on BP neural network.BP neural network is a nonlinear mapping model,weights can be determined in the network training,to avoid the problem of calculating the weight of the problem.Fourth,for the slow convergence of the neural network,easy to converge to the local extremum problems,the use of chaos genetic algorithm to optimize the performance of neural networks.Chaos genetic algorithm has the characteristics of fast convergence,strong global search ability and suitable parameter optimization,which can greatly improve the network performance and improve the prediction accuracy by optimizing the neural network with chaotic genetic algorithm.Finally,the use of neural network model for the forecast and analysis of the energy demand of Beijing,Tianjin and 2020 to 2015,and gives policy recommendations.