基于BP-GSA优化的某随动平台滑模控制  

Sliding Mode Control of Certain Type Servo Platform Based on BP-GSA Optimization

作　　者：戴宇辰 陈机林[1] 刘政 李玉腾 Dai Yuchen;Chen Jilin;Liu Zheng;Li Yuteng(School of Mechanical Engineering,Nanjing University of Science and Technology,Nanjing 210094,China)

机构地区：[1]南京理工大学机械工程学院,南京210094

出　　处：《兵工自动化》2023年第4期43-47,53,共6页Ordnance Industry Automation

摘　　要：为实现某随动平台负载模拟器响应的快速性和系统的鲁棒性,提出一种基于遗传模拟退火算法(genetic simulated annealing,GSA)优化的BP神经网络(BP-GSA)滑模控制方法。根据负载模拟器各环节硬件组成,建立系统等效数学模型;采取非奇异终端滑模实现对系统的控制,并采用BP神经网络对状态方程中未定项进行逼近,利用GSA算法调整网络节点权值。实验仿真结果表明:相比于传统滑模控制和PID控制,该方法在具有扰动输入的情况下,具有最小的稳态误差和最快的跟踪速度,能够有效提升系统的响应速度和力矩跟踪精度。In order to achieve the rapid response and robustness of a servo platform load simulator,a BP neural network(BP-GSA)sliding mode control method based on genetic simulated annealing algorithm(GSA)is proposed.According to the hardware composition of the load simulator,the equivalent mathematical model of the system is established.Non-singular terminal sliding mode is used to control the system,and BP neural network is used to approximate the undetermined terms in the state equation,and GSA algorithm is used to adjust the network node weights.The simulation results show that compared with the traditional sliding mode control and PID control,the proposed method has the smallest steady-state error and the fastest tracking speed in the case of disturbance input.The method can effectively improve the response speed and the torque tracking accuracy of the system.

关 键 词：负载模拟器 BP神经网络 遗传模拟退火算法 非奇异终端滑模控制 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]