城市天然气短期日需求量预测新模型

A new model for forecasting the short-term daily demand of urban natural gas

准确预测短期城市天然气需求量,对于城市天然气的合理调峰调压、安全供应、管网优化等都具有重要的现实意义。目前,城市天然气短期需求量预测模型主要包括时间序列、回归分析、支持向量机、灰色关联等,但其精度均不很理想,神经网络的精度虽然较高,但却容易陷入局部最小值,降低了泛化性。较之于上述诸多模型,最小二乘支持向量机基于结构风险最小化的原则,对于非线性问题也能得到较高的精度和泛化性,并且不容易出现过拟合现象。为此,基于对城市天然气短期日需求量的各种影响因素的全面分析和讨论,最终将气象、日期、政策确定为影响因素,并采用模糊综合评价法、经验打分法及专家评分法处理因素中的定性数据,采用极差变换法处理其他定量数据,最终利用最小二乘支持向量机建立了城市天然气短期日需求量预测新模型。仿真实验以四川省成都市为例,新模型预测结果平均绝对百分比误差为1.423%,较之于ARIMA、灰色关联、BP神经网络以及非线性回归等模型,新模型的预测精度有了很大的提高。结论认为,新模型的预测结果可作为确定城市天然气短期日需求量的依据和参考。

To accurately predict the short-term urban natural gas demand is of great significance to urban gas peak-shaving and adjustment, stable supply, pipeline network optimization, etc. The present prediction models mainly include time series, regression analysis, least squares support vector machine（LS-SVM）, grey relational analysis, etc. but their accuracies are not satisfactory except that of the BP neural network analysis, the generalization of which is reduced by its frequent occupancy by local minimums. Comparatively, the LSSVM prediction method based on the minimizing structural risk principle, is proved to be of higher accuracy and generalization, the most important of all, with an overfitting phenomenon rarely appeared. In view of this, we thoroughly analyzed and discussed all involved factors that affect the short-term daily urban gas demand, and finally determined the three major dimensions of weather, date and policy. Then, we adopted the fuzzy comprehensive evaluation method, the experience scoring method and the expert scoring method to deal with the qualitative data in the above three factors, processed other quantitative data by the method of extreme difference transformation. Finally, we established a new LS-SVM-based model to predict the short-term daily demand of urban natural gas. Furthermore, taking Chengdu as an example, we made pilot tests and demonstrated that the average absolute percentage error of prediction results was only 1.423% with this new model, the accuracy of which, compared with the ARIMA, gray correlation, BP neural network and the nonlinear regression models, is significantly improved. Therefore, this new prediction model can be used as reference for forecasting short-term daily urban gas demand.