Intelligent Diagnosis Model of Working Conditions in Variable Torque Pumping Unit Wells Based on an Electric Power Diagram
Open Access
Abstract
Because of the problems, such as the lack of an electric power diagram atlas under different working conditions and the difficulty in intelligent diagnosis of variable torqued pumping unit wells, this paper proposes a diagnosis model of working conditions based on feature recognition. The mathematical relationship model between the polished rod load and motor output power is derived based on the analysis of geometric structure, motion law, and process of energy transformation and transfer of the variable torque pumping unit. It can calculate the electric power diagram based on a dynamometer card. On this basis, the electric power diagram atlas is created, and the feature analysis and eigenvalue extraction of the electric power diagrams under different working conditions are carried out to realize the direct diagnosis of the working conditions in the variable torque pumping unit wells. The application and analysis of examples show that the electric power diagram atlas created in this paper has good practicability, and the working condition diagnosis model accuracy is high. It can provide a theoretical basis and technical support for the intelligent diagnosis of oil production working conditions and improve the intellectual management level of the oilfield, which is conducive to reducing production management costs and improving the oilfield’s production efficiency and benefits.Keywords
Funding Information
- Dongying Science Development Fund (DJ2021003)
This publication has 17 references indexed in Scilit:
- Transformation of academic teaching and research: Development of a highly automated experimental sucker rod pumping unitJournal of Petroleum Science and Engineering, 2020
- Diagnosis of Sucker Rod Pump based on generating dynamometer cardsJournal of Process Control, 2019
- A novel prediction method for down-hole working conditions of the beam pumping unit based on 8-directions chain codes and online sequential extreme learning machineJournal of Petroleum Science and Engineering, 2018
- Sucker rod pumping diagnosis using valve working position and parameter optimal continuous hidden Markov modelJournal of Process Control, 2017
- A new model for predicting liquid loading in deviated gas wellsJournal of Natural Gas Science and Engineering, 2016
- A recognition and novelty detection approach based on Curvelet transform, nonlinear PCA and SVM with application to indicator diagram diagnosisExpert Systems with Applications, 2011
- Accurate extraction of valve opening and closing points based on the physical meaning of surface dynamometer cardPetroleum Exploration and Development, 2011
- On concepts, strategies and techniques to the secondary development of China's high water-cut oilfieldsPetroleum Exploration and Development, 2010
- Application of self-organizing competitive neural network in fault diagnosis of suck rod pumping systemJournal of Petroleum Science and Engineering, 2007
- Dynamic Measurements of Beam-Pump ParametersSPE Production Engineering, 1992