Optimal Control Of Shape Of The Piezoelectric Shell Structure Based On Genetic Algorithm

With the rapid growth of science and technology, people want to control mechanicalstructure more precisely under various working environment. However, traditional materialis difficult or impossible to be monitored in use after being made to product. Moreover,some key components of precise instrument are not being able to maintain the desiredshape under special working environment, leading to them not being able to work.Therefore, the active control methods have to be used to make up these deficiencies.Because piezoelectric material is light and easy to control, it is widely used in modernengineering. One aspect is using piezoelectric actuator to efficiently and accurately controlstructure deformation to meet the system requirements. In this paper, piezoelectricactuators are used to control the deflection of piezoelectric shell effectively with the leastand most reasonable layout and multipoint voltage.Due to the features of shell structure, calculating curved shell is more complex thancalculating straight beam or plate structure, so the shell structure is mostly analysised byfinite element method. In order to overcome shear locking, trapezoidal locking, thicknesslocking and improve the accuracy,8nodes solid shell element is used to buildelectromechanical coupling solid shell element of piezoelectric structure. Hybrid stress andassumed natural strain (HS-ANS) method are also invoked. Number and position ofpiezoelectric actuators are transformed to a string, designed with Genetic Algorithm (GA)combinatorial optimization. Based on optimized piezoelectric layout, actuator voltage isdesigned with GA numerical optimization. The validity and high precision of the shellelement and optimization design method constructed in this paper are verified by severalnumerical examples. The main innovation of this article is the solid shell element and GAoptimization method that is brief effective and precise having successfully controlled theshape of shell structure. Using this method can overcome self-locking phenomenon, solvenon-continuous variable optimization and multipoint input problem. This method can also suitably control shape for arbitrary piezoelectric curve shell structure and is able to begeneralized to optimal design of active vibration control.