摘要
该文将自动分类技术应用于手机螺丝锁附的结果分类中,由此提出了一种改进的最小二乘支持向量机算法(least squares support vector machine,LS-SVM)。一方面,通过在径向基函数上进行泰勒展开,并选择前3项改进目标函数减少计算量;另一方面,在参数选取时考虑计算速度因子,以提高计算速度。仿真结果表明,改进后的LS-SVM算法与传统的LS-SVM算法具有相同的准确率,但运算速度更快,具有更强的实用性。
The article aims to apply the automatic classification technology to the classification of the results attached to the mobile phone screw locking,so an improved least square support vector machine(LS-SVM)algorithm is proposed.On the one hand,Taylor expansion is performed on the radial basis function,and take the first three items to reduce the calculation amount.On the other hand,the calculation speed factor is considered in the parameter selection to increase the calculation speed.The simulation results show that the improved LS-SVM algorithm has the same accuracy as the traditional LS-SVM algorithm,but the operation speed is faster,so it’s more practical.
作者
刘金燕
王冬青
崔建伟
LIU Jinyan;WANG Dongqing;CUI Jianwei(School of Automation Engineering,Qingdao University,Shandong Qingdao 266071,China;School of Electrical Engineering,Qingdao University,Shandong Qingdao 266071,China)
出处
《工业仪表与自动化装置》
2020年第4期12-15,24,共5页
Industrial Instrumentation & Automation
基金
国家自然科学基金项目(61873138,1573205)。
关键词
螺丝锁附
LS-SVM
分类
泰勒展开
参数选取
screw locking
LS-SVM
classification
Taylor expansion
parameter selection
