论文信息 - Application of Fourier Descriptors and Pearson Correlation for fault detection in Sucker Rod Pumping System

Application of Fourier Descriptors and Pearson Correlation for fault detection in Sucker Rod Pumping System

This paper presents a study of a new proposal for down-hole dynamometer cards patterns for the automatic fault diagnosis of Sucker Rod Pumping System. The accuracy and quick identification of down-hole problems is essential to the risk decrease and production improvement of the petroleum industry. The main idea is a card recognition through a digital image processing (Fourier Descriptors) and statistical (Pearson Correlation) tools. Successful results were reached when this proposal was applied in simulated cards and PETROBRAS real cards.

[1] Rafael C. González,et al. Digital image processing using MATLAB , 2006 .

[2] H. A. Tripp. A Review: Analyzing Beam-Pumped Wells , 1989 .

[3] Athanasios Papoulis,et al. Probability, random variables, and stochastic processes , 2002 .

[4] Celso Kazuyuki Morooka,et al. Automatic Downhole Card Generation And Classification , 1996 .

[5] Lideniro Alegre,et al. Intelligent Diagnosis of Rod Pumping Problems , 1993 .

[6] Roderick R. Dickinson,et al. Use of Pattern-Recognition Techniques in Analyzing Downhole Dynamometer Cards , 1990 .

[7] J. D. Rogers,et al. Artificial Neural Networks for Identification of Beam Pump Dynamometer Load Cards , 1990 .

[8] Kaveh Ashenayi,et al. Application of Artificial Neural Network to Pump Card Diagnosis , 1994 .

[9] P. Schirmer,et al. Use of Advanced Pattern-Recognition and Knowledge-Based System in Analyzing Dynamometer Cards , 1991 .

[10] Ari Visa,et al. Efficient Fourier shape descriptor for industrial defect images using wavelets , 2005 .