基于多目标优化策略的最佳无线传感器布置  

Optimal Wireless Sensor Placement Based on Multi-objective Optimization Strategy

作　　者：安葳鹏[1] 刘镕飞 AN Weipeng;LIU Rongfei(School of Software,Henan Polytechnic University,Jiaozuo 454003,China;School of Computer Science and Technology,Henan Polytechnic University,Jiaozuo 454003,China)

机构地区：[1]河南理工大学软件学院,河南焦作454003 [2]河南理工大学计算机科学与技术学院,河南焦作454003

出　　处：《小型微型计算机系统》2025年第3期751-758,共8页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(61872126)资助.

摘　　要：无线传感网络消除了物理电缆,降低了结构健康监测系统收集信息的难度,但通常具有较低的带宽和有限的能量.为了使结构健康监测系统获得最有效的信息和最高的网络性能,将最佳无线传感器布置问题表述为一个多目标优化问题,其中信息有效性被指定为模态保证准则,网络性能通过能量效率与网络连接性相结合来衡量.针对此问题,提出了一种多种群自动学习的鲸鱼优化算法,用整数编码代替二进制编码对解进行编码;采用多种群策略搜索最优解;最后通过自动学习机制加快寻找帕累托最优解集.数值实验表明,优化准则可以在信息有效性和网络性能之间进行权衡,算法能够有效解决最佳无线传感器布置问题,并且优于常用的非支配排序遗传算法.Wireless sensor networks eliminate physical cables,making it easier for structural health monitoring systems to gather information,but typically have lower bandwidth and limited energy.In order to obtain the most effective information and the highest network performance for the structural health monitoring system,the optimal wireless sensor placement problem is formulated as a multi-objective optimization problem in which information effectiveness is specified as a modal assurance criterion and network performance is measured by combining energy efficiency with network connectivity.To solve this problem,a whale optimization algorithm based on multi-population automatic learning is proposed,which uses integer coding instead of binary coding to encode the solution.The multi-population strategy was used to search the optimal solution.Finally,an automatic learning mechanism is used to accelerate the search for Pareto optimal solution set.Numerical experiments show that the optimization criteria can be balanced between information validity and network performance,and the algorithm can effectively solve the problem of optimal wireless sensor placement,and is better than the common non-dominated sorting genetic algorithm.

关 键 词：无线传感网络 结构健康监测 鲸鱼优化算法 最佳传感器布置 多目标优化 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]