Experimental Research On The Control System Of Excavator’s Trajectory Based On Neural Network PID

Excavator due to its high performance and high efficiency in the fie ld of architecture has been widely used, but in the process of excavator is working, the vast majority rely on manual operation of a driver, and should not be in dangerous environment or work for a long time, to achieve automation and intelligent of excavator is particularly important.The research content of this paper mainly includes the modeling of the track control system of excavator, the simulation and experimental research of the controller. In the modeling process of excavator control system, the co ntrol system of boom and bucket is modeled separately, and the transfer function of each link is calculated. The dynamic mechanism of the hydraulic system is modeled by the reverse movement of the piston in the hydraulic cylinder, and the transfer function is obtained by the least square method. In order to realize the digital computer control, the continuous transfer function is discretized, and the pulse transfer function and difference equation are obtained.Based on the establishment of hydraulic loading trajectory control system of discrete mathematics model and the incremental PID control algorithm and design of ordinary PID controller variable integral digital PID controller and by the neural network implementation and regulate parameters of controller of single neuron adaptive PID controller, BP neural network PID controller based on and in the MATLAB software written controller algorithm program, complete the controller parameter tuning and simulation, dynamic arm control system and the bucket rod control system for test signal response curve and PID controller parameter curve simulation results under the role of each controller.Based on the laboratory experiment platform for the excavator, write the control program of the control algorithm, experimental study, will common PID controller, gearshift integral PID controller, single neuron adaptive PID controller and BP neural network PID controller of the experiment results are analyzed and compared. Experimental results show that based on neural networ k of single neuron adaptive PID controller and BP neural network PID controller control effect is better than the conventional PID controller control, adaptable, and both BP neural network PID control effect is better, with a good application prospect.