Research On Diesel Price Forecasting Based On BP Neural Network

As a kind of petroleum products,diesel oil is an indispensable energy in the industrialized society.Due to the wide use of diesel oil,the price fluctuation of diesel oil affects all aspects of society,especially the economic development of various countries and even the world,so countries pay close attention to the price fluctuation of diesel oil.In order to reduce the negative impact of diesel price fluctuation,it is very important to forecast the diesel price for both government departments and diesel trading enterprises.In this case,this paper uses a combination of qualitative and quantitative methods to identify the influencing factors of diesel price,and then to forecast and analyze the diesel price.In this paper,literature analysis method is used to explain the theoretical basis,and the domestic and foreign research on diesel price and its influencing factors and BP neural network are reviewed.Secondly,the Delphi method is used to conduct one round of open-ended research and three rounds of evaluation research.Eight main factors affecting diesel price,such as domestic market supply and demand,refining capacity,diesel import and export,economic policy factors,production cost,new energy substitution,diesel inventory and futures investment,are identified.The quantitative analysis shows that the indicators of each factor include diesel output,diesel consumption and diesel production Energy,export quantity,import quantity,GDP,diesel production cost,LNG supply,diesel inventory,ice diesel price,etc.On this basis,using these 10 indicators and diesel price data,the prediction model is built based on BP neural network,and the parameters of the model are determined by experiments.Finally,the model is used to verify the diesel price,and the fitting degree is 0.99998.Compared with Multiple Regression Model and SVM Model,the results show that the prediction effect of BP neural network is better than the other two algorithms.The algorithm constructed in this paper is used to predict the diesel price from January 2019 to December 2020.The research results of this paper will provide more scientific decision-making reference for government departments to formulate policies and systems,as well as the production and operation of diesel trading enterprises,so that the research has both theoretical value and practical significance.