Simulation-Based Optimal Sensor/Actuator Positioning on a Fin-Like Structure

Abstract

Aircraft structures are subjected to high amplitude dynamic loads under service conditions; hence, it is necessary to determine their dynamic properties. Dynamic properties of a structure can be determined using simulation-based methods (e.g., finite element method) or using experimental modal analysis (EMA) methods. An EMA system should be lightweight and accurate so that it does not have a negative impact on the dynamic performance of the structure. For this purpose, the transducers (sensor and actuators) of the EMA system should be positioned in an optimal way. Optimal sensor positioning studies in current literature are mostly focused on positioning of a pre-specifcase 1ied sensor type. In this study, multi-objective optimization of types, numbers and locations of sensors is conducted for modal analysis of a fin-like structure. Pareto optimal designs that minimize the mass loading error in natural frequency prediction, the mode shape observability error and the total cost are sought by using a simulation based approach, where the dynamic properties (e.g., natural frequencies, mode shapes) are determined by using NASTRAN, and the multi-objective optimization is performed by using multi-objective genetic algorithm (MOGA) in MATLAB. Minimization of the mass loading error is achieved by locating the sensors near with minimum modal constant in all modes of interest (near clamp and rightmost region). Minimization of mode shape observability error is obtained by locating the sensors to the points with large displacements and avoiding nodal lines. © 2015 American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.