例证多维二阶近似式法的适用性  

On the Applicability of Multidimensional Two-order Approximation Method

作　　者：李春明[1] 陈峙村 许玟慧 李煦 朱明烨 LI Chun-ming;CHEN Zhi-cun;XU Wen-hui;LI Xu;ZHU Ming-ye(Department of Mechanical Engineering, School of Mechanical and Control Engineering,Shengli College, China University of Petroleum(East China),Dongying Shandong 257061, China;Shengli No.3 Middle School of Dongying City,Dongying Shandong 257051, China;Shengli No.2 Middle School of Dongying City,Dongying Shandong 257051, China)

机构地区：[1]中国石油大学胜利学院机械与控制工程学院机械工程系,山东东营257061 [2]东营市胜利第三中学,山东东营257055 [3]东营市胜利第二中学,山东东营257051

出　　处：《德州学院学报》2017年第6期12-14,共3页Journal of Dezhou University

基　　金：山东省自然科学基金资助项目(Q2006A08);中国石油大学胜利学院科技计划项目(KY2015025)

摘　　要：根据迭代点的形成特点,牛顿迭代优化方法更名为多维二阶近似式法较好.在文献[8]发现该方法的寻优效果与目标函数的幂有关之前,一直公认为理论性强的较好方法.在研究文献[8]的基础上采用数值计算的方法获得了寻优步长与目标函数幂的关系式,并进行了算例验证.According to the forming characteristics of the iteration points,the Newton iterative optimiza-tion method is better to be changed to the multidimensional two-order approximation method.Before ref-erences[8]finds that the effect of this method is related to the power of objective function,it has been recognized as a better theoretical method.In this paper,the relationship between the optimal step size and the power of the objective function is obtained by numerical calculation.And a numerical example is also carried out.

关 键 词：共位群内捕食模型 全局稳定性 LYAPUNOV函数 数值模拟 

分 类 号：O224[理学—运筹学与控制论] TP202[理学—数学]