Abstract: Aiming at the problem of unknown parameters, dead-time input and non-linearity in traditional DC motor control strategy, a dead-time control strategy of DC motor based on Lyapunov closed-loop stability Convolutional Neural Network (CNN) is proposed. Firstly, the dynamic system model of DC motor and the control objective of adaptive convolution neural network are given. The unknown parameters of DC motor system are approximated by convolution neural network, and the Lyapunov function is used to design the state feedback adaptive controller. Then, the signal definition in closed-loop control system is given. Its stability is analyzed theoretically. Finally, the proposed control strategy is modeled and simulated by using MATLAB platform. The results show that the proposed algorithm has good control characteristics.

Key words:

DC motor, dead-time control; Lyapunov function, convolutional neural network, state feedback

摘要： 针对传统直流电机控制策略中，未考虑参数未知、死区输入以及非线性特性问题，提出一种基于Lyapunov闭环稳定的卷积神经网络(Convolutional Neural Network， CNN)直流电机死区控制策略。首先给出直流电动机的动力学系统模型，以及自适应卷积神经网络的控制目标，利用卷积神经网络逼近直流电机系统中的未知参数，并利用Lyapunov函数进行状态反馈自适应控制器设计，然后给出闭环控制系统中的信号有界定理，从理论上对其稳定性进行分析。最后利用MATLAB平台对所提控制策略进行模型建立和实验仿真，结果显示所提算法具有良好的控制特性。

关键词:

"> 直流电机；死区控制；Lyapunov函数；卷积神经网络；状态反馈