Parameter identification of contact model for tank saddle ring system based on genetic algorithm and BP neural network

Abstract

This paper is devoted to present an experimental identification method of tank saddle ring system parameters based on BP neural network algorithm and genetic algorithm. The contact parameters in mechanical system cannot be obtained directly through measurement, and it is inevitable that there will be errors between the parameters estimated by engineering experience and the real value. In this paper, a 6-degree-of-freedom swing experimental device of tank turret is built, the excitation source is 1Hz sine function, the contact stiffness and contact damping coefficient are identified by constructing error functions based on simulation results and data of four laser displacement sensors distributed around the saddle ring, by the combination of BP neural network and genetic algorithm. Then, the sensor data in the test results and simulation results are compared to further verify the accuracy of the turret model and parameter identification method.