摘要
针对悬浮间隙传感器输出特性非线性严重的问题,提出一种结合径向基核函数和多项式核函数优点的混核最小二乘支持向量机(HKLSSVM)作为电容悬浮间隙传感器的非线性校正模型并采用浣熊优化算法(COA)对HKLSSVM的惩罚因子和核函数参数进行优化。为验证模型的有效性,分别采用径向基神经网络模型、传统LSSVM模型、粒子群优化(PSO)算法-HKLSSVM模型以及COA-HKLSSVM模型进行非线性校正仿真分析。结果表明,COA-HKLSSVM模型在电容悬浮间隙传感器非线性校正的应用中表现出最佳的校正精度与稳定性,校正后的电容悬浮间隙传感器线性度为0.43%,均方根误差为0.022 mm,最大误差为0.068 mm,满足悬浮控制系统对悬浮间隙传感器的线性要求。
To address the issue of severe nonlinearity in the output characteristics of the capacitive suspension gap sensor,a hybrid kernel least squares support vector machine(HKLSSVM)model is proposed as the nonlinearity correction model.This model combines the advantages of the radial basis function and the polynomial kernel function.Furthermore,the coati optimization algorithm(COA)is employed to optimize the penalty factor and kernel function parameters of the HKLSSVM model.To validate the effectiveness of the model,the radial basis neural network model,the traditional LSSVM model,the particle swarm optimization(PSO)algorithmHKLSSVM model and the COAHKLSSVM model are used for nonlinear correction simulation analysis.The results show that the COAHKLSSVM model shows the best correction precision and stability in the application of nonlinear correction of capacitive suspension gap sensor,and the linearity of the corrected capacitive suspension gap sensor is 0.43%,the root mean square error is 0.022 mm,and the maximum error is 0.068 mm,which meets the linearity requirements of the suspension control system for the suspension gap sensor.
作者
郑洋阳
王滢
陈康
李贵
陈友豪
ZHENG Yangyang;WANG Ying;CHEN Kang;LI Gui;CHEN Youhao(School of Electrical Engineering,Southwest Jiaotong University,Chengdu 611756,China;Key Laboratory of Maglev Technology and Maglev Train,Ministry of Education,Chengdu 611756,China)
出处
《传感器与微系统》
2024年第12期16-20,共5页
Transducer and Microsystem Technologies
基金
四川省面上基金资助项目(2022NSFSC0473)。
关键词
电容悬浮间隙传感器
非线性校正
浣熊智能优化算法
capacitive suspension gap sensor
nonlinear correction
coati intelligent optimization algorithm
