基于LS-SVM的天然气管网负荷组合预测

Combination Method of Natural Gas Pipeline Network Load Forecasting Based on Least Squares Support Vector Machines

针对现有组合预测模型,基于经验风险最小化原则,克服预测精度受组合模型限制的缺点,提出一种基于最小二乘支持向量机(LS-SVM)的天然气管网负荷组合预测模型,并与AR模型、BP神经网络模型、GM(1,1)模型以及最优权重组合模型进行了比较,得出基于最小二乘支持向量机的天然气管网负荷组合预测模型能够得到更高的预测精确度,可为天然气管网的安全运行以及优化调度提供决策支持的结论。

In light of the existing combined prediction model based on the experience of risk minimization and of the forecast accuracy of the model by the combination of restrictions,a natural gas pipe network load forecasting model based on the least squares support vector machines（LS-SVM）is proposed and compared with the AR model,BP neural network model,GM（1,1）model as well as the top priority recombination model.Least squares support vector machines based on the natural gas pipeline network load forecasting model portfolio will provide a higher forecast accuracy for the safe operation of the pipeline network optimization as well as credible support for the theory.