Optimal Sensor Placement Using Chaotic Monkey Search Algorithm

Abstract

The optimal sensor placement (OSP) problem is integral to modern large scale structures for their health monitoring. Evolutionary algorithms for the OSP problem are attractive as they can result in global optima without gradient information. In this paper, a modification of the Monkey Algorithm with a chaotic search strategy and adaptive parameters is proposed. It includes chaotic initialization, variable search step length, and adaptive watching time. The performance of the proposed chaotic Monkey Algorithm (cMA) is compared with the original Monkey Algorithm. Convergence property of cMA is established. The proposed method is applied to an optimal sensor placement problem for structural health monitoring. The OSP problem is solved for a mass-spring-damper system and then for a model of the I-40 bridge developed by the Los Alamos National Laboratory. Numerical results demonstrate that the proposed Chaotic Monkey Algorithm has capability of solving mixed-variable optimization problems and that it performs better than the originally proposed Monkey algorithm. Finally, nonparametric uncertainty modeling is used to evaluate variability in a model and its effect on the optimal sensor placement.