Improved memetic algorithm for nonlinear identification of a three-dimensional elliptical vibration cutting system

Abstract

This article focuses on the nonlinear Wiener system identification of three-dimensional elliptical vibration cutting. The developed three-dimensional elliptical vibration cutting device is actuated by four piezoelectric stacks, which have hysteresis nonlinear characteristics. Our research proposes an improved memetic algorithm in order to identify the nonlinear elliptical vibration cutting model. The improved memetic algorithm displays the advantages of both particle swarm optimization and genetic algorithm, while simultaneously overcoming these programs’ shortcomings. Moreover, it can quickly and efficiently search for the global optimization. The improved memetic algorithm’s identification performance may be compared with the conventional memetic algorithm, particle swarm optimization, and genetic algorithm using two well-known test functions. Test results demonstrate that the improved memetic algorithm can search the global optimal solution more efficiently than available state-of-the-art algorithms. Based on the input–output data collected from experiments, the accuracy of the identification model can potentially reach 97.55%. This relatively small margin of error verifies the efficiency of the proposed improved memetic algorithm for system identification.