摘要
针对非线性摩擦对直流伺服系统性能的影响,提出一种基于模拟退火布谷鸟算法提高摩擦参数辨识精度的方法.采用LuGre摩擦模型建立伺服系统静摩擦参数与力矩的关系表达式,并分析其稳态特性,以此构建参数辨识的目标函数.引入模拟退火中的Boltzmann选择机制,结合迭代全局最优值对布谷鸟算法发现概率的自适应性进行了增强,以改善算法的寻优效果,通过分析比较得知改进算法的时间复杂度与原算法一致.在相同摩擦参数设置下,改进布谷鸟算法的辨识精度优于传统方法,利用其辨识结果补偿伺服系统运行的非线性摩擦力矩,得到了良好的控制效果.实验结果表明,该方法可以使摩擦参数的辨识适应度值提高2~3个数量级,有效改善了系统的控制性能.
Aiming at the influence of nonlinear friction on the performance of DC servo system a method using cuckoo search algorithm based on simulated annealing to improve the accuracy of friction parameters identification is proposed.LuGre friction model is used to establish the relationship between the static friction parameters and the torque of the servo system and the objective function for the parameters identifying is available.The Boltzmann selection mechanism in simula ted annealing is introduced and the iterative global optimization value is applied to enhance the self-adaptation of discover y probability for cuckoo search algorithm to strengthen the optimization ability of the strategy.The time complexity of the im proved method is consistent with that of the original way through analysis.The identification accuracy of the presented cuck oo search algorithm is better than that of the traditional method in the same friction parameter setting and perfect controlled effect is obtained by using the identified results to compensate the nonlinear friction torque of the system.Experimental re sults show that the method can increase the identification fitness value of friction parameters by 2~3 orders of magnitude and enhance the control performance of the servo system.
作者
李伟
张长胜
LI Wei;ZHANG Changsheng(Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,P.R.China)
出处
《重庆邮电大学学报（自然科学版）》
CSCD
北大核心
2019年第6期892-897,共6页
Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)
基金
国家自然科学基金(51665025)~~
关键词
直流伺服系统
摩擦参数辨识
布谷鸟搜索算法
模拟退火
DC servo system
friction parameters identification
cuckoo search algorithm
simulated annealing
