Study On Working Condition Identification Of Belt Pumping Unit Based On Electric Power Diagram

Belt pumping unit is widely used in oil field because of its simple structure and high production efficiency.Due to the pumping unit itself or the complex environmental problems underground,there will be many kinds of mechanical failures,motor failures and other conditions during the production process,which will affect the productive efficiency of the oil field.Aiming at the diagnosis of the pumping station's working condition,the condition diagnosis technology based on the data source of the electric power diagram of belt pumping unit is mainly studied.It can monitor the running condition of pumping unit in real time and improve the intelligence and information level of oil field production.First,the mechanical structure and working principle of the belt pumping unit are introduced.Based on kinematics and dynamics analysis,a mapping model of pumping electric and mechanical diagrams and indicator diagrams is constructed.On the basis of typical working diagram,the sample of electric power diagram with different working conditions is obtained.Then the characteristics of the typical electric power diagram under various working conditions are analyzed and defined.Based on the data,curves and frequency characteristics of the electric power graph,the extraction technology of multiple eigenvectors is studied.Because of the difference of data dimension,order of magnitude and data structure of multiple feature vectors,data fusion algorithm of multi feature data is studied.A multi feature vector clustering normalization method based on SOFM neural network is proposed.Reusing the normalized eigenvector of the electric power diagram under the typical working condition to train the Probabilistic Neural Network,The test samples is used to test the accuracy of the Probability Neural Network,and the effectiveness of the algorithm is proved by the MATLAB simulation experiment.Finally,the method is applied to practical application.The experiment proves the effectiveness of the diagnosis method based on the electric power diagram,which is of great significance for improving the intelligent management and monitoring level of the oil field.