Study On Medium And Long-term Load Forecasting Model Of Urban Natural Gas

With the country’s vigorous development of the natural gas industry,the proportion of natural gas in energy consumption will further increase.Accurate prediction of natural gas load has important practical significance.The forecast results can be used to guide gas companies to improve the dispatching capacity of the natural gas pipeline network.It also can maintain a balance between supply and demand in the natural gas market.Therefore,it is particularly important to establish a natural gas load forecasting model with high accuracy.Natural gas quarterly load forecasting can provide a basis for natural gas supply planning to ensure a balance between supply and demand in the natural gas market.To accurately predict the seasonal fluctuations of the natural gas load,we propose a seasonal grey model(SGM(1,1)model)based on the accumulation operators generated by seasonal factors.We use the proposed model to carry out an empirical analysis based on the seasonal electricity consumption data of the natural gas load in Chongqing from 2010 to 2016.The results from the SGM(1,1)model are compared with those obtained using the grey model(GM(1,1)),and the particle swarm optimization algorithm combines with the grey model(PSO-GM(1,1)model)The results of the comparison show that the SGM(1,1)model can effectively identify seasonal fluctuations in the natural gas load and its prediction accuracy is significantly higher than those of the GM(1,1)and PSO-GM(1,1)models.The forecast results from 2017 to 2020obtained using the SGM(1,1)model suggest that the natural gas load is expected to increase slightly,but obvious seasonal fluctuations will still be present.It is forecasted that the annual natural gas load in 2020 will be118.75×10~8m~3 with an annual growth rate of 9.41%.The annual natural gas load forecast can provide a basis for the laying of national gas pipeline networks and the formulation of energy policies.In order to accurately predict the annual natural gas load,the improved annual gray-Markov natural gas annual load model with the highest accuracy is obtained.Taking the natural gas load in Beijing from 2005 to 2016 as the original data,the data volatility can be reduced after the weakening process,and three improved grey prediction models are constructed.The three improved gray GM(1,1)models are analyzed and compared in turn.The best improved model is combined with the Markov error correction model,and the Beijing 2017-2020natural gas load forecasting model is constructed.The results show that the combined model can reduce the interference of the original data to the prediction system,the prediction accuracy meets the actual requirements,and the accuracy is much higher than the traditional gray model.It can truly reflect the development trend of annual natural gas load,and the prediction results are reliable and practical.