Table of Contents Author Guidelines Submit a Manuscript
Journal of Control Science and Engineering
Volume 2017, Article ID 5710408, 7 pages
Research Article

A Fault Diagnosis Method for Oil Well Pump Using Radial Basis Function Neural Network Combined with Modified Genetic Algorithm

1Harbin University of Science and Technology, Harbin 150001, China
2Research Institute of Oil Production Engineering, Daqing Oilfield Company, Daqing 163000, China
3Tongji University, Shanghai 200092, China
4Harbin Institute of Technology, Harbin 150001, China

Correspondence should be addressed to Deliang Yu; moc.361@2301gnaileduy

Received 26 April 2017; Revised 16 August 2017; Accepted 29 August 2017; Published 8 October 2017

Academic Editor: Manuel Pineda-Sanchez

Copyright © 2017 Deliang Yu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


This paper presents a new method to diagnose oil well pump faults using a modified radial basis function neural network. With the development of submersible linear motor technology, rodless pumping units have been widely used in oil exploration. However, the ground indicator diagram method cannot be used to diagnose the working conditions of rodless pumping units because it is based on the load change of the polished rod suspension point and its displacement. To solve this problem, this paper presents a new method that is applicable to rodless oil pumps. The advantage of this new method is its use of a simple feature extraction method and advanced genetic algorithm to optimize the threshold and weight of the RBF neural network. In this paper, we extract the characteristic value from the operation parameters of the submersible linear motor and oil wellhead as the input vector of the fault diagnosis model. Through experimental analysis, the proposed method is proven to have good convergence performance, high accuracy, and high reliability.