改进的粒子群算法在圆度误差评价中的应用—崔星星.docxVIP

ADDIN CNKISM.UserStyle改进的粒子群算法在圆度误差评价中的应用 崔星星,陈岳坪来稿日期： 基金项目：广西自然科学基金资助项目（2014GXNSFAA118337）；广西教育厅科研重点资助项目（2013ZD048，ZD2014074）广西科技大学科学研究基金资助项目（院科博12Z15）；广西重点实验室建设项目（13-051-38） 作者简介:崔星星，（1989-），男，广东深圳人，硕士研究生，主要研究方向:精密测量； 陈岳坪，（1973-），男，湖南岳阳人，博士，副教授，主要研究方向：精密检测技术 ，卢海燕 （广西科技大学 机械工程学院，广西 柳州 545006） 摘 要：由于用最小区域法（MZC）评价圆度的目标函数是非线性函数，用传统的优化方法难以进行评价，而且还容易陷入局部最优解。本文基于最小区域法(MZC)准则的基础上，运用一种改进的粒子群算法—协同粒子群优化（CPSO）算法来评价圆度误差，此方法将原来的粒子群分成若干个子种群，每个子种群中粒子各自寻求自己的最优值，各种群粒子信息共享，共同进化,直到达到指定的进化代数，最后比较得出最优值。相比遗传算法（GA）和标准的粒子群优化(PSO)算法具有全局有哪些信誉好的足球投注网站能力强，收敛速度快，精度高的优点。最后在MATLAB软件编程环境下，用实例比较验证了CPSO算法的有效性。对其他的几何量评价具有指导意义。 关键词：协同粒子群优化（CPSO）；粒子群优化(PSO)；圆度；优化 中图分类号：TH16 文献标识码：A 文章编号： The Application of Improved Particle Swarm Optimization Algorithm in the Evaluation of Roundness Error CUI Xing-xing, CHEN Yue-ping, LU Hai-yan (School of Mechanical Engineering, Guangxi University of Science and Technology, Guangxi Liuzhou 545006,China) Abstract: As the objective function is a nonlinear function using the minimum zone method (MZC) to evaluate the roundness, the traditional optimization method is difficult to evaluate and easy to fall into local optimal solution. Based on the minimum zone method (MZC) criterion, it utilizes an improved particle swarm optimization (PSO) algorithm—cooperative particle swarm optimization (CPSO) algorithm, to evaluate the roundness error. This method divides the original particle swarm into several sub populations, each of which searches its own optimum respectively. All particle groups share their information and evolve together until the certain evolution number is reached. In the end, the optimal value is obtained through comparison. Compared to the genetic algorithm (GA) and PSO algorithm, CPSO has stronger global search ability , faster convergence rate and higher precision. Finally, the validation of CPSO algorithm is tested with an example coding in MATLAB, which is important in guiding other geometrical quantity. Key Words: Cooperative Pa