Study On Intelligent Control Method Of Electromagnetic Pumping Unit Load Simulation System

The beam pumping units are widely applied in the oil field.In order to improve the performance of the pumping unit,it needs to be tested for performance before leaving the factory.However,the actual working conditions of the pumping unit are more complicated,it is difficult to meet the requirements of conventional loading methods such as heavy loads.Therefore,a new type of electromagnetic pumping unit load simulation device was designed in this paper.By collecting the on-site dynamometer charts,the status of the pumping unit under actual operating conditions is simulated,so that the loading force on the pumping unit is consistent with the pumping unit under the actual working conditions.This article mainly did the following work:1.Through the previous data collection and analysis,the overall scheme of the electromagnetic pumping unit load simulation system is designed.The transfer function of each component is deduced in detail,and the overall mathematical model of the electromagnetic pumping unit load simulation system is established.The step response of the model is analyzed in MATLAB,and it is found that the traditional output stability of the PID controller is poor.2.Due to the inherent nonlinearity and hysteresis of the electromagnetic damper,and the traditional PID control method has poor control and control effect on the system,a fuzzy PID controller and fuzzy PID controller based on BP neural network is designed in this paper.Utilizing the powerful learning capabilities of these intelligent controllers and their adaptation to the characteristics of nonlinear systems can increase the effectiveness of control.3.We built the model of the system in the MATLAB software’s Simulink module and compared the control effects of the three controllers using the step response signal as an input signal.It is found that the use of traditional PID controller system in the early run there is a large shock,the adjustment time is 3s,the overshoot of up to 16 kN,three parameters of the PID controller are KP=0.5,KI=0.5,KD=0.05.The system adjustment time using the analog PID controller is 4s,the overshoot is 2kN,and the PID controller parameters are: KP=10,KI=1,KD=2.The adjustment time of the fuzzy PID controller system based on BP neural network is 2s.The PID controller parameters are: KP=6.2,KI=2.3,KD=1.5.Finally,the model of pumping unit in a laboratory was used to verify that the control effect of fuzzy PID controller based on BP neural network was superior to that of fuzzy PID controller and was superior to the traditional PID controller,which was consistent with the simulation results.