改进的蝗虫优化算法及其应用  

Improved grasshopper optimization algorithm and its application

作　　者：乔亚茹 吴怀宇[1,2] 陈志环 陈晨 QIAO Ya-ru;WU Huai-yu;CHEN Zhi-huan;CHEN Chen(Engineering Research Center for Metallurgical Automation and Detecting Technology of Ministry of Education,Wuhan University of Science and Technology,Wuhan 430081,China;Institute of Robotics and Intelligent Systems,Wuhan University of Science and Technology,Wuhan 430081,China;Planning Department,China Renewable Energy Engineering Institute,Beijing 100120,China)

机构地区：[1]武汉科技大学冶金自动化与检测技术教育部工程研究中心,湖北武汉430081 [2]武汉科技大学机器人与智能系统研究院,湖北武汉430081 [3]水电水利规划设计总院规划处,北京100120

出　　处：《计算机工程与设计》2023年第12期3621-3627,共7页Computer Engineering and Design

基　　金：国家自然科学基金项目(62073250、62003249、62173262);湖北省重点研发计划基金项目(2020BAB021)。

摘　　要：针对蝗虫优化算法(GOA)收敛精度不高、易陷入局部最优等问题,提出一种非线性自适应模拟退火的蝗虫优化算法(SA-NGOA)。利用非线性自适应更新GOA的关键参数,更好地协调算法的全局探索和局部开发;利用模拟退火算法,以一定概率接收GOA的劣势解,提高算法的收敛精度。实验结果表明,SA-NGOA算法具有良好的优化性能。同时,将SA-NGOA算法用于扩展同步参数观测器(ESPO)参数优化,其结果表明,该算法优化后的ESPO具有更快的辨识速度和更稳定的精度。The nonlinear adaptive and simulation annealing grasshopper optimization algorithm(SA-NGOA)was proposed to avoid the disadvantages that the locust optimization algorithm is easy to fall into local optimum and has low convergence accuracy.The key parameters of GOA were updated by nonlinear adaptation to better coordinate the global exploration and local development of the algorithm.The simulated annealing algorithm was used to receive the inferior solution of GOA with a certain probability to improve the convergence accuracy of the algorithm.The results show that the SA-NGOA algorithm has good optimization performance.At the same time,the SA-NGOA algorithm was used to optimize the parameters of the extended synchronous parameter observer(ESPO).The results show that the ESPO optimized using the algorithm has higher identification speed and more stable accuracy.

关 键 词：蝗虫优化算法 非线性自适应 模拟退火算法 混合算法 混沌系统 观测器 参数优化 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]