中国计量大学学报2023，Vol.34Issue(4)：519-526,8.DOI:10.3969/j.issn.2096-2835.2023.04.004
基于WOA的SVM在近红外甲烷测量系统中的应用
Application of SVM based on whale optimization algorithms for NIR methane measurement systems
摘要
Abstract
Aims:To improve the prediction accuracy of the support vector machine(SVM)regression in NIR methane measurement systems,the whale optimization algorithm(WOA)was proposed to optimize the parameters C and g of SVM regression that were compared with conventional regression models.Methods:The whale optimization algorithm was used to search the global optimal solution adaptively;and the optimal penalty coefficient C and gamma parameter were searched after quick iteration.A near-infrared methane measurement system based on laboratory setup was used to measure methane standard gas within a concentration range of 0～1％.SVM was combined with GridSerach,particle swarm optimization algorithm(PSO),Gray Wolf Optimization algorithm(GWO),and WOA separately to establish four regression models:GridSerach-SVM,PSO-SVM,GWO-SVM and WOA-SVM,which were applied to experimental data processing for comparison.Results:Among the four models,the WOA-SVM regression model had the highest prediction accuracy and the smallest root mean square error with a reduction of more than 2 times in average relative errors and a reduction of more than 10 times in average absolute errors.Conclusions:The SVM model based on WOA has high prediction accuracy in near-infrared methane detection systems.关键词
支持向量机回归/鲸鱼优化算法/甲烷检测Key words
SVM/WOA/methane detection分类
信息技术与安全科学引用本文复制引用
李炜楠,徐鹏飞,叶有祥,陈红岩..基于WOA的SVM在近红外甲烷测量系统中的应用[J].中国计量大学学报,2023,34(4):519-526,8.基金项目
浙江省基础公益研究计划(No.LGF21E040005) （No.LGF21E040005）
