具有随机系数的自激振动问题的人工神经网络求解方法

Solution on Self-exited Vibration with Random Coefficient in Mechanical Systems by an Artificial Neural Network

利用人工神经网络技术 (BP网络 )研究具有随机系数的初轧机自激振动问题 ,提出了一种Runge-Kutta法和人工神经网络相结合的求解新方法 .即利用数值计算和 BP网络建立随机系数和稳态振幅之间的关系 ,从而可直接利用数值积分计算出稳态振幅的统计特性 .计算结果表明 ,所提方法通用性强且可提高计算精度 ,同时说明摩擦系数的随机特性是轧钢机发生重大事故的重要原因之一 .

In this paper an artificial neural network (BP network) is applied to solve the problem of the self-exited vibration with random coefficients in a rolling mill and a new method has been presented in which Runge-Kutta method and a BP network are applied successively. First the relations between random coefficients and the stable amplitude of the self-exited vibration are built up by numerical calculation and a BP network, then the statistical property of the stable amplitude can be calculated directly by numerical integral. It has been shown that the method presented in the paper is widely available in the engineering, the accurateness of calculation is improved greatly and one of the main reasons that bring on great accidents in a rolling mill is still random property of the friction factor.