Simultaneous optimization of orientations and locations of actuators and sensors for morphing structural shapes

Abstract

Optimal design of the orientations and locations of collocated piezoelectric actuators/sensors pairs for a plate-like structure under bending load uncertainty are determined with the objective of minimizing the deformation and electrical input under any sort of loading. The bending moments generated by the piezoelectric actuator actuators are used for deformation control, i.e., to minimize the deformation. The plate-like structure is subjected to an arbitrary load which lies in an uncertainty domain with regard to its magnitude and direction. The uncertain loading studied in the present paper involves a load of unknown magnitude and direction, which should be determined to produce the arbitrary deformation. Two optimization variables are considered for each piezoelectric actuator/sensor device: the location of its center and its orientation. An optimal control algorithm and three types of artificial intelligence algorithms (PSOOOL algorithm—particle swarm optimization algorithm for optimization of orientations and locations of actuators; SAOOL algorithm—Simulated Annealing algorithm for optimization of orientations and locations of actuators; EMOOL algorithm—Electromagnetism-like Method for optimization of orientations and locations of actuators; optimal control algorithm) are presented for the determination of the orientation and location of piezoelectric actuators/sensors in the application to shape control of plate-like structures. Numerical results show that simultaneous optimization of both orientations and locations can lead to optimum configurations that consume less electrical energy and minimizing the deflection. SAOOL algorithm can handle the optimization of orientations and locations of actuators/sensors better than PSOOOL algorithm and EMOOL algorithm. The different algorithms exhibit similar performance. However, exhaustive EMOOL algorithm and PSOOOL algorithm require significantly higher computational effort.