Research On Intelligent Frequency Modulation Control Of Rod Pumping Well

During the operation of the beam pumping unit,due to its inherent structure and motion characteristics,the rod pumping system is subjected to severe periodic alternating loads,resulting in large fluctuations in the input power of the motor,low system efficiency,and energy consumption serious loss and other problems,so frequency modulation control is often used in the field for field application.Although good results have been achieved,there are also some problems.On the one hand,such control focuses on the configuration of hardware devices,including frequency conversion cabinets,a large number of sensors,and rarely conducts theoretical analysis of frequency modulation directly from the motor input power curve.On the other hand,in terms of frequency modulation control,the redistribution of the system’s pumping speed is mostly achieved by changing the stroke times,but it is rarely refined to adjust the distribution profile within the period of the stroke times,and the effect of the distribution of the up and down stroke time on the rod pump well is ignored.This article focuses on the motor input power curve,from establishing the power curve,to learning the power curve,to implementing frequency modulation control to change the motor power curve,and finally achieving a smooth output of the power curve.First of all,according to the kinematic and dynamic analysis of beam pumping unit is respectively establishes the hanging point of the crank rotating at a constant speed and variable speed rotation motion,polished rod load reducer,the mathematical model of the torque and motor power,and by using the computer simulation system working parameters under two kinds of driving mode,for frequency modulation control change motor power curve to lay the theoretical foundation.Secondly,in this paper,the corresponding relationship between motor power and crank Angle is obtained without relying on sensors such as inclination angle and displacement,and frequency modulation control is only carried out through motor power curve at different moments,which requires in-depth study of the power curve.This paper is based on the convolutional neural network to learn a large number of power curves under the operating parameters of different pump wells with rods.Due to the influence of the equilibrium condition,the fluctuation of the power curves in the upper and lower strokes is different,and the rule of which can be found and then the equilibrium condition of the pumping unit can be reversed to realize the intelligent learning of the power curve.Then,based on the influence of power supply frequency on power curve,the power supply frequency in the form of trapezoidal wave is adopted on the premise that the impulse times remain unchanged,and the influence of up and down stroke time distribution on the pump well with rod is fully considered,and the intelligent frequency modulation control combining power curve intelligent learning and frequency modulation control technology is implemented.The operation data of test wells show that the frequency modulation control can achieve the stable output of the motor input power and the working conditions of the rod pump well can be improved.Finally,Matlab and C# of the hybrid compilation of intelligent FM control monitoring software interface design,convenient for user monitoring and management.In this paper,the intelligent learning based on the motor input power curve and the frequency modulation control provide important theoretical significance and application value for the intelligent control of the pumping unit and the improvement of the operating conditions of the rod pumping well pumping system.