Optimization of piezoelectric wafer placement for structural health-monitoring applications

Abstract

Optimization of the number and location of piezoelectric wafers used in sensor networks for continuous monitoring of automotive and aerospace structures is not yet fully developed. This article presents a novel approach for the optimization of piezoelectric wafer networks for application in the field of structural health monitoring. A mixed integer nonlinear program is formulated and validated for different geometric shapes. The proposed objective function is to maximize the coverage of the area under study, described by a set of control points. In the optimal solution, each control point should be covered by a user-defined number of sensing paths, where each sensing path is the line joining any two piezoelectric transducers within the monitored plate. During the optimization process, any place on the plate is a potential location of a piezoelectric wafer. A mathematical programming language is used to code the algorithm, and selected simulation cases are executed to demonstrate the efficiency of the proposed optimization algorithm. The algorithm introduces several optimization parameters, such as coverage Levels 2 and 3, number of paths passing through a control point, and areas of concentrated coverage. Furthermore, the algorithm features the flexibility to change a wide range of problem parameters, such as the number of piezoelectric wafers, their coverage range, and the number of control points. The tractability of the model proposed is improved by feeding the solver an initial solution that makes the branch-and-bound technique less extensive. Data fusion is conducted to determine the quality of the coverage provided by the optimized piezoelectric wafer locations and to scrutinize the efficiency of the proposed approach. Finally, two sensor network configurations are selected and validated experimentally. The validation demonstrates a high level of coverage within the network, as well as evaluating the accuracy of damage localization within the optimized sensor networks.