【摘要】 针对球杆系统的位置控制难题,利用ADAMS软件建立球杆系统的三维虚拟样机模型,由此确定球杆系统的动力学模型,提出一种基于神经网络补偿控制的PID双闭环球杆位置控制方案,基于ADAMS/Controls与Matlab/Simulink进行了球杆位置控制的联合仿真。仿真结果表明:位置定值控制与方波信号跟踪控制时,PID双闭环-神经网络补偿控制较PID双闭环控制的稳态精度分别由0.06 m、0.08 m提高到0.002 m、0.003 m,正弦信号跟踪时,跟踪误差由0.15 m减小到0.05 m,PID双闭环-神经网络补偿控制具有更好的动态性能和较高的稳态控制精度。更多还原

【Abstract】 In allusion to the position control problem of the ball-and-beam system, a 3D virtual prototype model of a ball-and-beam system was established with ADAMS to determine the kinetic model, a double closed-loop PID position control scheme based on neural network compensation control was proposed, and the co-simulations of the ball-and-beam position control based on ADAMS/Controls and Matlab/Simulink were conducted. Simulation results indicate that compared with the double closed-loop PID control, the double closed-loop PID-neural network compensation control improves steady state accuracy from 0.06 m to 0.002 m, and 0.08 m to 0.003 m in value-fixed position control and square-wave signal tracking control, respectively. The double closed-loop PID-neural network compensation control also reduced the tracking error from 0.15 m to 0.05 m in sine-wave signal tracking control. The double closed-loop PID-neural network compensation control is proven to have excellent dynamic performance and high steady state control accuracy.更多还原