基于在线智能优化PI控制器的电动加载系统控制方法

为抑制电动加载系统在加载的过程中受到承载系统主动运动带来的多余力矩的干扰,而经典的PI控制难以满足控制的准确性、快速性的需求,提出了基于Hopfield神经网络在线优化PI控制。研究了电动加载系统工作原理并采用机理法建立了电动加载系统的数学模型,利用Hopfield网络能够寻优的特性优化PI参数,通过选取合适的能量函数,网络从初始状态经过有限次网络反馈与迭代计算过程向着能量极小点演变,此时Hopfield神经网络趋于稳定,其输出即为最优的PI控制参数。仿真结果表明该控制方法能够更好地抑制多余力矩,加载精度高。

Online Optimization PI Control Based on Hopfield Neural Network on Electric Load Simulator

To inhibit the interference of the surplus torque caused by the active movement of the steering gear in the loading process,the online optimization PI controller based by Hopfield neural network is proposed in this paper while it＇s difficult to meet the＂ requirements of a high accuracy and rapidity with a classical PI controller.The working principle is studied and the mathematical model is built on the basis of its mechanism.PI parameters is optimized by using the optimization property of Hopfield neural network.By selecting the appropriate energy function,the network evolves from an initial state towards the minimum energy after finite feedbacks and iterations.