电子器件2025，Vol.48Issue(1)：31-37,7.DOI:10.3969/j.issn.1005-9490.2025.01.006
阵列传感器气体浓度检测的改进型海鸥算法研究
Research on Improved Seagull Algorithm for Gas Concentration Detection of Array Sensor
摘要
Abstract
Targeting at the problem of low detection accuracy due to cross-sensitivity when the array sensors are used to detect binary mixed gas quantitatively,an improved seagull algorithm is proposed to optimize the kernel extreme learning machine algorithm.Firstly,the kernel principal component analysis(KPCA)is used to reduce the dimension of the data and extract the features.The principal com-ponent with large contribution rate is selected as the input vector.Secondly,the nonlinear convergence factor(B)and two spiral shape coefficients(u and v)in the seagull algorithm(SOA)are improved.Then,the improved seagull algorithm is used to optimize the key pa-rameters of the kernel extreme learning machine(KELM),namely the regularizition coefficient(C)and the kernel parameter(σ),to form the SOA-KELM quantitative detection model.Finally,the output results are calculated and analyzed.The experimental results show that the improved seagull algorithm has better search ability and higher accuracy.The improved seagull algorithm optimizes the kernel extreme learning machine(SOA-KELM)with stronger regression ability.The correlation coefficient detection is above 0.991 6,which provides a new method for gas concentration detection by using array sensors.关键词
浓度检测/核主成分分析/核极限学习机/改进型海鸥算法Key words
concentration detection/kernel principal component analysis/kernel extreme learning machine/improved seagull algorithm分类
信息技术与安全科学引用本文复制引用
李鹏,纵彪,林事力,张立豪,刘轩宇..阵列传感器气体浓度检测的改进型海鸥算法研究[J].电子器件,2025,48(1):31-37,7.基金项目
国家自然科学基金项目(41075115) （41075115）
江苏省重点研发计划社会发展项目(BE201569) （BE201569）
无锡市社会发展科技示范工程项目(N20191008) （N20191008）
