基于IPSO-SVR的水泥分解炉温度预测模型研究  

Research on cement decomposing furnace temperature prediction model based on IPSO-SVR

作　　者：金星[1] 徐婷[1] 冷淼[1] JIN Xing;XU Ting;LENG Miao(College of Electrical and Electronic Engineering,Changchun University of Technology,Changchun 130012,China)

机构地区：[1]长春工业大学电气与电子工程学院,吉林长春130012

出　　处：《现代电子技术》2017年第9期148-151,共4页Modern Electronics Technique

基　　金：吉林省科学技术厅计划项目(20150203003SF)

摘　　要：为建立稳定可靠的分解炉温度预测模型,结合与分解炉温度密切相关的几个主要运行参数,提出一种粒子群参数优化的支持向量回归机算法(PSO-SVR),并在粒子群算法中引入自适应惯性权重的思想,构建出分解炉温度预测模型。与未改进的模型进行仿真对比实验,实验结果表明,该IPSO-SVR模型具有较佳的预测能力,预测相关系数达到0.707 5,温度预测误差绝对值不超过7℃,误差率在0.8%以内。In order to establish a stable and reliable temperature prediction model for the decomposing furnace,in combination with several main operating parameters closely related to the decomposing furnace temperature,a particle swarm optimization based support vector regression(PSO?SVR)machine algorithm is proposed.The thought of adaptive inertia weight is introduced into the particle swarm optimization algorithm to construct the decomposing furnace temperature prediction model.The model is compared with the unimproved one by means of simulation experiment.The experimental results show that the IPSOSVR model has better forecasting ability,the correlation coefficient reached to0.7075,the temperature prediction error absolute value is less than7℃,and the error rate is within0.8%.

关 键 词：分解炉温度 粒子群算法 惯性权重 支持向量回归机 预测模型 

分 类 号：TN911.1-34[电子电信—通信与信息系统] TP273[电子电信—信息与通信工程]