基于主成分分析和支持向量机相结合的天然气消费量预测

Consumption Prediction of Natural Gas Based on Principal Component Analysis and Support Vector Machine

为了提高了天然气消费量的预测精度,提出一种主成分分析和支持向量机相结合的天然气消费量预测模型(PCA-SVM)。首先采用主成分分析选择影响天然气消费量的影响因素,然后输入到非线性预测能力强的支持向量机进行训练,并采用遗传算法优化支持向量机参数,最后建立天然气消费量预测模型。仿真实验结果表明,PCA-SVM加快了天然气预测型的学习速度,提高了天然气消费量的预测精度。

In order to improve the accuracy of the prediction of natural gas consumption, a model is presented to forecast natural gas consumption and support vector machine combined with a principal component analysis （PCA-SVM）. The first principal component analysis was used to select factors that influence the consumption of natural gas, and then input to a nonlinear prediction ability of training support vector machine, genetic algorithm is used to optimize the parameters of SVM, the last natural gas consumption prediction model. The simulation results show that, PCA-SVM accelerate the learning speed prediction of natural gas, to improve the accuracy of the prediction of natural gas consumption.