基于自适应猴群算法的传感器优化布置方法研究

Optimal sensor placement based on adaptive monkey algorithm

针对猴群算法中爬过程和望过程的搜索方式较为机械,以及跳过程的方式较为单一的问题,提出一种用于传感器优化布置的自适应猴群算法。采用双重编码的方式,克服了原猴群算法只能解决连续变量优化问题的缺陷;对爬过程和望过程进行了改进,使其能够自适应选择这两个搜索方式以提高算法的局部搜索能力和效率;提出了两种全新的跳过程,即反射跳和变异跳,来增强算法的全局搜索能力。文末以大连国贸大厦为例,进行了参数敏感性分析以及传感器优化布置方案的选择,结果表明自适应猴群算法的搜索效率较原猴群算法有了大幅提高,能较好地解决传感器优化布置问题。

An adaptive monkey algorithm( AMA) used for optimal sensor placement( OSP) was proposed to solve the problem that the searching methods of the climb process and watch-jump process are mechanical and the pattern of the somersault process is single. Firstly,the dual-structure coding method was utilized to overcome the defect that the original monkey algorithm could only perform the optimization for continuous variables. Then,the climb process and watch-jump process were updated in order to adaptively select the two searching methods to improve the local searching ability and efficiency of the whole algorithm. In addition,the two new somersault processes,i. e.,reflection somersault process and variation somersault process,were introduced to strengthen the global searching ability of the algorithm. Finally,taking the Dalian international trade mansion as an example,the parametric sensitivity analysis and the selection of OSP schemes were performed. The results showed that AMA can better solve the OSP problem and its searching efficiency is greatly improved compared to the original algorithm.