Fault diagnosis model of the variable torque pumping unit well based on the power-displacement diagram

3rd International Symposium on Resource Exploration and Environmental Science

生态环境科学

Chen, Dechun^1 ; Zhou, Ruiqi^1 ; Meng, Hongxia^1 ; Peng, Yuandong^2 ; Chang, Feng^3 ; Jiang, Dong^3 ; Wei, Bin^3

China University of Petroleum, Qingdao, Shandong

266580, China^1

SINOPEC Henan Oilfield Branch, Zhengzhou, Henan

473000, China^2

SINOPEC Shengli Oilfield Branch, Dongying, Shandong

257000, China^3

关键词: Average relative error;    Fault diagnosis model;    Fitting coefficient;    Grey correlation analysis;    Intelligent diagnosis;    Method of diagnosing;    Production optimization;    Production system;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/300/2/022030/pdf
DOI  :  10.1088/1755-1315/300/2/022030

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

This work proposes a new method of diagnosing the variable torque pumping unit well, by analysing the eigenvalue of power-displacement diagram. Considering the actual angular velocity, the unbalanced weight of structure and the efficiency of motor and gearbox, the relationship between the polished rod load and the input power of motor are derived by analysing the internal relation between the polished rod load, the output shaft torque of the gearbox, the output shaft power of motor and the input power of motor. Then, the conversion model of the power-displacement diagram based on the dynamometer card is derived. By transforming the dynamometer card in typical working condition into the corresponding power-displacement diagram, 11 power-displacement diagrams in typical working conditions are established. By extracting the grayscale statistical eigenvalues from typical power-displacement diagrams, a working condition diagnosis model of oil well is established based on the grey correlation analysis. The test of 5 wells in the oilfield shows the average relative error of average normalized power of up and down stroke between the measured power-displacement diagram and the calculated power-displacement diagram are 6.31% and 5.83%, and the average fitting coefficient is 0.91. This shows that this model has a good accuracy, and it also proves the reliability of the typical power-displacement diagram. According to the test of 80 wells in the oilfield, the accuracy of diagnosis results based on the measured power-displacement diagram is 93.3%. This shows that the model has high accuracy and practicability, and can provide technical support for intelligent diagnosis and production optimization decision in production system of variable torque pumping unit well.