基于ABC算法优化支持向量机的温度测点恢复方法研究  

Improved Artificial Bee Colony Optimization Based on Support Vector Machine for Temperature Measurement Points Restoration

作　　者：戴红霞[1] 唐经纶 赵力[2] DAI Hongxia;TANG Jinglun;ZHAO Li(Department of Electronics Communications Engineering,Jiangsu Information Occupation Technical College,Wuxi Jiangsu 214061,China;School of Information Science and Engineering,Southeast University,Nanjing 210096,China)

机构地区：[1]江苏江苏信息职业技术学院电子信息工程系,江苏无锡214153 [2]东南大学信息科学与工程学院,南京210096

出　　处：《电子器件》2020年第6期1364-1367,共4页Chinese Journal of Electron Devices

摘　　要：为了改善当前参数优化算法易陷入局部最优的问题,研究一种基于人工蜂群算法优化SVR的温度测点恢复方法。以机器空转温度测点为对象,根据支持向量机的实现原理建立支持向量回归机模型,引入人工蜂群算法对SVR模型的参数惩罚因子和核函数进行优化,并用优化后的算法对缺失的温度测点进行恢复。经人工蜂群优化后的模型(ABC-SVR)的平均相对误差为6.57%,优于其他优化算法,因而是一种实用可行的优化算法。In order to solve the problem that the current parameter optimization algorithm is prone to fall into local optimization,a temperature measurement point restoration method based on the artificial bee colony(ABC)algorithm to optimize SVR was investigated in this study.Taking the idling temperature measuring point of the machine as the object,the support vector regression(SVR)model was established according to the principle of support vector machine,and the artificial swarm algorithm was introduced to optimize the parameter penalty factor and kernel function of the SVR model,and the optimized algorithm was then used to recover the missing temperature measuring point.The average relative error of the model(ABC-SVR)optimized by artificial bee colony algorithm(ABC-SVR)is 6.57%,which is better than other optimization algorithms.So it is a practical and feasible optimization algorithm.

关 键 词：支持向量回归机 人工蜂群算法 参数优化 测点恢复 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]