Fault diagnosis of sucker rod pumping systems based on Curvelet Transform and sparse multi-graph regularized extreme learning machine

Abstract

A novel approach is proposed to complete the fault diagnosis of pumping systems automatically. Fast Discrete Curvelet Transform is firstly adopted to extract features of dynamometer cards that sampled from sucker rod pumping systems, then a sparse multi-graph regularized extreme learning machine algorithm (SMELM) is proposed and applied as a classifier. SMELM constructs two graphs to explore the inherent structure of the dynamometer cards: the intra-class graph expresses the relationship among data from the same class and the inter-class graph expresses the relationship among data from different classes. By incorporating the information of the two graphs into the objective function of extreme learning machine (ELM), SMELM can force the outputs of data from the same class to be as same as possible and simultaneously force results from different classes to be as separate as possible. Different from previous ELM models utilizing the structure of data, our graphs are constructed through sparse representation instead of K-nearest Neighbor algorithm. Hence, there is no parameter to be decided when constructing graphs and the graphs can reflect the relationship among data more exactly. Experiments are conducted on dynamometer cards acquired on the spot. Results demonstrate the efficacy of the proposed approach for faults diagnosis in sucker rod pumping systems.

Prediction and Analysis of Polished Rod Dynamometer Card in Sucker Rod Pumping System with Wear

“…According to the parameter data of the pump unit in Table 2, the polished rod displacement curve can be achieved by solving Equation (11), as shown in Figure 10. By determining the surface boundary conditions of the polished rod displacement, the wave equation can be solved, and the polished rod load of sucker rod pumping system can be obtained.…”

“…It not only reflects the downhole working conditions of the beam pumping unit [2] but also calculates the balance of the pumping unit, reducer output shaft torque, and motor power [3,4]. At the same time, the actual polished rod dynamometer card can be matched with the fault cards, which are often used to diagnose the fault of the oil well [5][6][7][8][9][10][11]. us, it is very meaningful to study the polished rod load.…”

Diagnostic of Operation Conditions and Sensor Faults Using Machine Learning in Sucker-Rod Pumping Wells

“…Among the methods that use pumps, sucker-rod pumping is the oldest and the most used in the world. It is estimated that 90% of wells with artificial lift use sucker-rod pumping systems [ 1 ].…”

“…He also compared the results with the random forest and K-nearest neighbors (KNN) algorithms. In 2018, Zheng and Gao [ 23 ] diagnosed downhole cards via decomposition and hidden Markov model; Zhang and Gao [ 1 ] used the fast discrete curvelet transform as dynamometer cards descriptors and sparse multi-graph regularized extreme learning machine (SMELM) as the algorithm; Zhou et al [ 24 ] proposed a classification model based on Hessian-regularized weighted multi-view canonical correlation analysis and cosine nearest neighbor multi-classification for pattern detection; finally, Ren et al [ 25 ] highlighted successful results when proposing root-mean-square error (RMSE) for card classification.…”

Catenary image enhancement method based on Curvelet transform with adaptive enhancement function

“…This makes it more effective in processing image edges, texture and other geometric features than conventional algorithms. As for now, Curvelet transform has been used in many fields such as denoising [1,2], detection [3,4], recognition [5,6], diagnosis [7,8], enhancement [9], compression [10], watermarking [11], estimation [12], steganography [13], segmentation [9,14], classification [9, 15,1 6], etc. Moreover, many combination methods based on Curvelet transform and other methods have been proposed in many areas.…”