Reasear on Application of Combination Forecasting Model Based on LS-SVM and Markov Chain——Take Consumption of Natural Gas in Sichuan Province as Example
Guo Zhi-gang · Coal Technology · 2010
As one of primary energy in Sichuan Province,it is great significance to forecast natural gas consumption.On the condition of limited historical data,for resolving Sichuan nature gas consumption,puts forward forecasting model based on LS-SVM and Markov chain.First,build Sichuan natural gas consumption time-series forecasting model by optimized use of LS-SVM.Then,divide system state into five parts by Markov chain in order to improve the prediction accuracy.Finally,make a prediction for 2007～2011 Sichuan nature gas consumption.The results show that the model has a certain accuracy and reliability.