Grey-related least squares support vector machine optimization model and its application in predicting natural gas consumption demand

作者：

Highlights：

•

• In order to eliminate redundant information on LS-SVM modeling, the GRA-LSSVM model is proposed.

• Weight adaptive SecPSO algorithm is proposed to optimize the parameters of GRA-LSSVM model.

• Experiments with China’s natural gas consumption data prove that WASecPSO-GRA-LSSVM is superior to other models.

摘要

•In order to eliminate redundant information on LS-SVM modeling, the GRA-LSSVM model is proposed.•Weight adaptive SecPSO algorithm is proposed to optimize the parameters of GRA-LSSVM model.•Experiments with China’s natural gas consumption data prove that WASecPSO-GRA-LSSVM is superior to other models.

论文关键词：Grey related analysis,Least squares support vector machine,Particle swarm optimization,Natural gas demand

论文评审过程：Received 17 July 2017, Available online 9 February 2018, Version of Record 23 February 2018.

论文官网地址：https://doi.org/10.1016/j.cam.2018.01.033