Research And Development Of Surface-driven Cavity Pump Fault Diagnosis System Based On PLC

The progressing cavity pump oil production system has the advantages of strong adaptability and high system efficiency,and has been widely used in oil production operations in recent years,especially for the exploitation of oil wells with high sand content,high gas content or high viscosity.Due to the complex downhole operation environment of the progressing cavity pump,it is difficult to carry out efficient and accurate fault diagnosis of the progressing cavity pump after abnormal working conditions occur in the production process,and the commonly used diagnostic analysis system has low compatibility with the software framework of the oilfield site management system.In order to improve the efficiency and economic benefits of the oil production system of the progressing cavity pump,take into account the accuracy of fault detection and system compatibility,and realize the "efficiency" of the oilfield,this paper researches the fault diagnosis method of the progressing cavity pump and the development of PLC fault diagnosis system as follows:Surface-driven progressing cavity pump as one of the most widely used methods of oil and gas extraction,this paper takes the Shengli oilfield surface-driven progressing cavity pump well as an example,firstly,to study its organizational structure and working principle,to provide a theoretical basis for the analysis of its failure causes and parameters.Secondly,the functional requirements and design framework of the fault diagnosis system are combined to derive the overall PLC system design scheme.The S7-1200 of Siemens is selected as the CPU controller,and each PLC module is connected to each other using industrial Ethernet for communication.The Portal V15 software developed by Siemens is used as the configuration programming software,and the statement table and ladder diagram with strict logic and clear organization are selected as the programming language in combination,and the modular programming of the overall program is carried out through modules such as OB module and DB module,and then Using missing value filling and outlier processing to improve the data accuracy of the collected progressing cavity pump data to ensure the accuracy of the fault diagnosis model construction,while using Pearson correlation coefficients combined with principal component analysis methods to determine the characteristic parameters used for fault modeling.Then,through the analysis of common faults and diagnosis methods of progressing cavity pumps,a fault diagnosis method combining wavelet packet analysis and Fisher discriminant analysis to construct a fault model using the selected feature parameters is proposed,and a corresponding PLC progressing cavity pump fault diagnosis system is developed based on this method.The accuracy test of this paper shows that the correct rate of progressing cavity pump fault diagnosis using the fault modeling method is basically the same as and higher than that of the traditional fault diagnosis method using the probabilistic neural network diagnosis method,and the PLC fault diagnosis system developed based on the fault model is more compatible with the control management system of the oilfield site.The feasibility and practicality of the fault modeling method can be proved through the testing of the progressing cavity pump fault detection examples in the Shengli oilfield.