论文部分内容阅读
天然气作为四川省的一种主要能源,正确合理地对天然气消费量进行预测有着重要的意义。在历史数据有限的情况下,为解决四川省天然气消费量预测的问题,提出一种基于最小二乘支持向量机与马尔可夫链的组合预测模型。首先,利用经过优化后的最小二乘支持向量机建立四川省天然气消费量的时间序列预测模型;然后借助马尔可夫链进行系统状态划分,缩小预测区间以提高预测精度;最后运用该模型对四川省2007～2011年天然气消费量做出预测。计算结果表明,该模型具有一定的精度和可靠性。
Natural gas, as a major energy source in Sichuan Province, is of great significance in correctly and reasonably predicting natural gas consumption. In the case of limited historical data, in order to solve the problem of gas consumption forecast in Sichuan Province, a combined forecasting model based on least square support vector machine and Markov chain is proposed. Firstly, the optimized time series prediction model of natural gas consumption in Sichuan Province is established by using the optimized least squares support vector machine. Then the system state is divided by Markov chain to narrow the prediction interval to improve the prediction accuracy. Finally, the model is applied to Sichuan Provincial 2007-2011 natural gas consumption forecast. The calculation results show that the model has certain accuracy and reliability.