Rod Pumping System Research Articles

An evolutional SVM method based on incremental algorithm and simulated indicator diagrams for fault diagnosis in sucker rod pumping systems

Fault diagnosis of the sucker rod pumping system (SRPS) is a challenging problem in the oil industry. Recently, the computer-aided indicator diagram (ID) recognition techniques are becoming useful measurements to help engineers monitor the wells. However, the slow variation of SRPS state makes it difficult for a well to experience all the fault conditions. Using the IDs collected from other oil wells as training data will lead to a large difference between the distribution of training data and target data because of the interference information including noise and system operation characteristics in IDs. Moreover, the different occurrence probability of each fault leads to the imbalance of training data, therefore the fault tolerance of the diagnosis models will result in lower accuracy in identifying the fault IDs with less training samples. In addition, due to one-off training, the parameters of diagnostic model are always fixed resulting in lack of adaptive ability. To address this issue, an evolutional support vector machine (SVM) method for SRPS diagnosis is proposed. First, to obtain balanced training data with similar distribution to the target data, a novel model describing the working process of SRPS under various fault conditions is established. Subsequently, the static apparent stiffness feature and its extraction algorithm is proposed to fully retain the fault information of SRPS in IDs. Then, the incremental SVM is used to construct the diagnosis model based on generated IDs. At last, the proposed method is verified experimentally through the system parameters and IDs of many wells collected from oilfields, and then some conventional techniques are employed in the comparison studies. The obtained results show that the accuracy of the diagnosis model trained by simulated IDs is 11.2% higher than that trained by collected IDs. Besides, the parameters of proposed diagnosis model can be continuously evolved to improve the diagnosis accuracy and generalization ability. Furthermore, the proposed diagnosis model also has advantages in training efficiency and memory consumption.

Simulation for determining the rod pump system parameter values using finite difference method

<p>Lifting <span>petroleum to the surface requires a mechanism called an artificial lift. This mechanism is useful for increasing the flow of fluids from a well. One that uses this method is the rod pump. The rod pump system needed the appropriate design (through the determination of the value of the parameters that affect) so that the pump runs optimally. But changing the appropriate design in real terms is not economically and time-efficient. The ideal rod pump system analysis procedure is to take downhole data (in the well) which is commonly called a pump card. A pump card calculation simulation is needed to efficiently analyze the rod pump system. The wave equation can describe the model of the rod pump system. A simulation of pump data calculation is performed using the finite difference method for the wave equation solution. Through the process of tuning parameter values using trial and error methods, the steps of tuning the parameter value of the rod pump system are proposed in this paper. For further research, a pump card calculation method based on a surface card can be developed to efficiently analyze the rod pump.</span></p>

Full Reproduction of Surface Dynamometer Card Based on Periodic Electric Current Data

Summary The surface dynamometer card is composed of ground load and ground displacement, which is of great significance to reflect the operation of rod pumping and the exploitation of crude oil. However, the current method of obtaining the surface dynamometer by sensors is a huge financial investment on the sensor installations and maintenance. In this paper, we propose an innovative method based on deep learning to reproduce the surface dynamometer card directly from electrical parameters. In our method, the convolution neural network is used as the basic layer to automatically extract the spatial characteristics of input data. A long short-term memory (LSTM) network as the core component is used for the output layer to consider the time dependence of the dynamometer card. Finally, the experimental shows that the proposed method achieves the mean relative error (MRE) of 4.00% on the real oil well data in A-oilfield, and the dynamometer card calculated by our model is basically consistent with the field data. In addition, the method has been tested in new wells with a rod pumping system, and the results show that the accuracy of the model is close to 90%, which has already greatly outperformed the previous methods.

Sucker Rod Antibuckling System: Development and Field Application

SummaryWhen the oil price is low, cost optimization is vital, especially in mature oil fields. Reducing lifting costs by increasing the mean time between failure and the overall system efficiency helps to keep wells economical and increase the final recovery factor. A significant portion of artificially lifted wells currently use sucker rod pumping systems. Although its efficiency is in the upper range, there is still room for improvement compared with other artificial-lift systems.This paper presents the field-tested sucker rod antibuckling system (SRABS), which prevents buckling of the entire sucker rod string, achieved by a redesign of the standing valve, the advantageous use of the dynamic liquid level, and, on a case-by-case basis, application of a tension element. The system allows full buckling prevention and a reduction of the overall stresses in the sucker rod string.The resulting reduction in the number of well interventions combined with the higher system efficiency prolongs economic production in mature oil fields, even in times of low oil prices. The analysis of SRABS, using finite-element simulations, showed a significant increase in system efficiency. The SRABS performance and wear tests under large-scale conditions were performed at Montanuniversität Leoben’s Pump Test Facility and in the oil field. The results of intensive laboratory testing were used to optimize the pump-body geometry and improve the wear resistance by selecting optimal materials for the individual pump components. The ongoing field-test evaluation confirmed the theoretical approach and showed the benefits achieved by using SRABS. SRABS itself can be applied within every sucker rod pumping system; the installation is as convenient as a standard pump, and manufacturing costs are comparable with those of a standard pump.This paper shows improved performance of the SRABS pumping system compared with a standard sucker rod pump. SRABS is one of the first systems that prevents the sucker rod string from buckling without any additional equipment, such as sinker bars. Testing of SRABS has identified significant benefits compared with standard sucker rod pumps.

Downhole Dynamometer Sensors for Sucker Rod Pumps

This publication introduces the downhole dynamometer sensor technology for digitalizing sucker rod pumps’ downhole behavior. The presented field application enables insight into the sensor principles, data evaluation processes and highlights the value chain. Sucker rod pumps (SRP) are used worldwide since the last century. Nevertheless, the system is still prone to fail in certain situations. The lack of measured downhole data is a severe problem that significantly slows down the innovation process. The presented downhole dynamometer sensor measurement system can successfully fill this gap and provides valuable, requested information. Each downhole dynamometer sensor records the three-dimensional stress field, the movement characteristics, and the produced fluid temperature. The recorded data allow a critical evaluation of the rod string behavior. Rod string loading and movement, friction forces between rod guides and tubing, and volumetric parameters can be evaluated. The recordings of several sensors are linked together to provide a comprehensive picture of the sucker rod string condition. The field case data analysis has indicated severe wear at specific positions of the sucker rod string and shown a significant optimization potential of the system. This publication shows the significant optimization potential for the tested sucker rod pumping system to postpone the economic limit and increase the recovery factor. The novelty is the wealth of information obtained and the associated detailed evaluation possibilities.

Nonlinear Finite Element Analysis of Rod String in Rod Pumping System

Rod pumping is the main oil recovery method at home and abroad. The corresponding displacement and load of the sucker rod section are important parameters in the system. This paper uses the advantages of the finite element method to solve the three nonlinear problems in the rod string model, and establishes its finite element analysis model. According to the upper and lower boundary conditions and initial conditions, the software is used to program and calculate. The surface dynamometer card of multi-level steel rod columns and glass fiber reinforced plastics-steel hybrid rod string is predicted, and the calculation results with and without considering mass nonlinearity are compared and analyzed. The feasibility of predicting the surface dynamometer card by the finite element method is verified. It is proved that the coupling has a great influence on the mechanical properties of the glass fiber reinforced plastics-steel hybrid rod string.

A novel method of output metering with dynamometer card for SRPS under fault conditions

Remote monitoring of sucker rod pumping system (SRPS) based on polished rod dynamometer cards (DC) is an effective means to reduce the operating costs and realize the real-time control. Nevertheless, how to obtain the working process and the production rate of SRPS from polished rod DC is always a puzzle because of complex down-hole motions. The typical solution for this problem is deriving the pump DC via the wave equation to achieve the effective stroke of plunger. But some key parameters involved in this approach such as valve leakage and the fill factor are difficult to determine accurately, especially under fault conditions, which may influence the accuracy of output metering.In order to address this tough issue, in this paper, a novel method of output metering with DC is proposed based on quantitative analysis of fault. The method consists of two parts, the first part is an approach to simulate the working process of SRPS under fault conditions, specifically it's an improved simulation model characterized by a set of fault parameters, which is presented through analyzing the mechanisms of sucker rod pumps at normal and several faulty scenarios. And the second part is a method implemented by genetic optimization algorithm to determine the fault parameters set of the actual SRPS.Using parameters of typical well, the analysis of effects of faults on SRPS is carried out. The simulation results show that valves leakage reduces the flow rate of liquid significantly in the period of effective stroke, which leads to the distortion of output metering results of effective stroke method. Compared with single faults, coupling faults aggravate the production reduction of SRPS. Furthermore, when the well fluid contains gas, the impact of traveling valve leakage on the reduction of production is greater than that of standing valve leakage, but the consequence is opposite when formation energy is insufficient. Besides, the carbon fiber rod and fiberglass rod SRPS are more sensitive to valve leakage, while the steel rod and wire rope SRPS are more sensitive to the condition of insufficient liquid supply.Eventually, the proposed method is verified experimentally through the productive parameters and measured DC of actual wells collected from an oilfield. The obtained results demonstrate the effectiveness of the proposed method for output metering and fault diagnosis as well.

Identification of Downhole Conditions in Sucker Rod Pumped Wells Using Deep Neural Networks and Genetic Algorithms (includes associated discussion)

Summary In this paper, deep learning artificial neural networks (ANNs) are used to analyze the features of downhole dynamometer cards and identify the sucker rod pumping system conditions. A description model for the dynamometer cards, using Fourier descriptors, was established for card feature extraction. Then, neural networks were trained to generate failure prediction models to recognize downhole faults of the rod pumping systems. The failure prediction models were validated and tested with a large database of previously interpreted cards. The proposed model is trained by using 4,467 dynamometer cards—29.2% of these cards represent sucker rod pumping systems of normal conditions, while the rest (70.8%) represent faulty sucker rod pumping systems. Genetic algorithms (GAs) were used to search for the best deep ANN structure that gives highest accuracy for the testing data. Accuracy of the proposed ANN model was measured with 1,915 cards that were not used in developing the ANN. The proposed model identified the sucker rod system failure successfully with very high accuracy (99.69%).

Dynamometer-Card Classification Uses Machine Learning

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 194949, “Beam-Pump Dynamometer Card Classification Using Machine Learning,” by Sayed Ali Sharaf, Tatweer Petroleum; Patrick Bangert, SPE, Algorithmica Technologies; and Mohamed Fardan, Tatweer Petroleum, et al., prepared for the 2019 SPE Middle East Oil and Gas Show and Conference, Manama, Bahrain, 18–21 March. The paper has not been peer reviewed. In reciprocating rod pumping systems, analysis of dynamometer-card data can deliver valuable insight into the status of the pump, and can indicate if future action is required. The complete paper explains the steps taken to improve surveillance of beam pumps using dynamometer-card data and machine-learning techniques, and reviews lessons learned from executing the operator’s first artificial intelligence (AI) project. Background The oldest and most widely used method of artificial lift is called beam pumping, or the sucker-rod lift method. A dynamometer is a device used on beam pumps that measures load on the polished rod (top) and plots the load in relation to the rod position as the pumping unit moves through a stroke cycle. This plot is known as the surface card. A pump card is a plot of load vs. position on the pump’s plunger. The pump card is more useful for surveillance purposes because it filters effects of anything above the plunger and provides standard pump card shapes for interpreting pump operating conditions. Identification and diagnosis of beam pumps using the pump card is an expensive human visual-interpretation process, not only requiring significant labor time but also deep expertise in the domain. Use of machine-learning techniques for pattern recognition can help automate the visual interpretation process, increasing efficiency and reducing maintenance activities resulting from missed early diagnosis. Data Collection Sucker-rod pumps are widely deployed in the Bahrain Field. There are two different types of communication layers used across the field, radio and optical fiber. Almost 300 optical-fiber- connected beam-pump units (BPUs) have been selected for surveillance and advanced analysis. Fiber is used to avoid any communication issues during data collection for the AI project. The sampling rate of collection for each pump is one reading per 20 seconds, thus covering three pump cards per minute. All BPUs are connected to a central server that converts hardware-communication protocol used by a programmable logic controller into the open-platform-communication (OPC) protocol. Fig. 1 shows the details of the 209 values representing a single pump card being read from a BPU and stored in the operator database. During the data-collection period, a total of 5,380,163 pump cards from 297 BPUs were collected and stored in the database. Data Labeling and Preparation In machine learning, an important step after data collection is data labeling, which is usually performed manually by experts. The labeled data set is used to feed machine-learning algorithms that detect patterns specific to each class. A total of 35,292 pump cards has been labeled (more than 1,000 cards per day on average). A useful feature of the labeling software is that it allows labeling an entire time period of cards in one shot, although such labeling should be performed with care.

Improved method on hydraulic power calculations for conventional sucker rod pumping system

The annual energy consumption of suck rod pumping units in China is more than 10 billion kWh, but the average efficiency is only about 25-31%. Limited information is known, about the efficiency of various components of the pumping unit. The current methods for calculating the hydraulic power neglect the effect of friction and free gas. This research introduces an improved method for calculating hydraulic power which takes into account the effect of friction and free gas. The model was obtained from the product of two parameters which were, predicted pressure increased by the downhole pump and predicted flow rate. The pressure increased by the pump was obtained by the nodal analysis approach in which the concepts of multiphase flow were included. The Computer program was developed and all calculations in this paper were computed by this program developed from python programming language version 3.5. This model was compared with the previous models and showed that, the previous models underestimated the hydraulic power since the influence of gas and friction were not included.

Energy dissipation analysis of oil wells sucker rod string

Sucker rod pumping systems are well-known artificial lift methods, thereby improving the efficiency of oil production. However, the periodic movement performed by the string makes the system susceptible to mechanical failures, promoting operational shutdowns. This paper presents a new approach to the vibrational analysis of the sucker rod string through a study of energy decay. The damped wave equation representing the oscillatory motion of the sucker rod lift system was solved analytically, and the energy of the solution was characterized by analyzing its behavior in space and time. The results show that the energy has a dissipative nature, promoting the vibrational attenuation in three different exponential decay forms. The results allow to characterize the damping limits of each exponential decay type. The approach proposed in this paper, by energy decay analysis, can serve as base to the development of new methodologies capable to detect anomalies and optimum operational conditions in sucker rod pump system.

Development and Efficiency Testing of Sucker Rod Pump Downhole Desanders

Summary Tense economic situations push the demand for low-cost oil production, which is especially challenging for production in mature oil fields. Therefore, an increase in the meantime between failure and the limitation of equipment damage is essential. A significant number of wells in mature fields are suffering under sand by-production. The objective of this paper is to show the development process and the testing procedure of an in-house-built, effective downhole desander for sucker rod pumps on the basis of a sophisticated analytical design model. In weak reservoir zones, often the strategy to prevent equipment damage due to sand by-production is the sand exclusion method using a gravel pack. Nevertheless, a certain amount of small sand grains still enter the wellbore and may damage the sucker rod pumping system over time. In early 2018, various types and sizes of downhole desander configurations were tested at the pump testing facility (PTF) at the University of Leoben (Montanuniversitaet Leoben). In a period of about 4 months, testing took place under near field conditions to find the optimum and most efficient design. The design optimization was focused on the geometry of the swirl vanes and the sand separation distance at the sucker rod pump intake. An analytical model provided the basis for geometric optimization. Concurrently, field tests of the in-house downhole desander were performed in the Vienna Basin that confirmed the findings of the tests at the PTF. The test results have shown that the downhole desander design and the pumping speed are the most influencing parameters on sand separation efficiency. Poor design in combination with a wrongly selected pumping speed can reduce the sand separation efficiency to lower than 50%, while if all parameters are chosen correctly, the sand separation efficiency can be 95% or higher. The grain size distribution is the additional parameter that enables a decision and ranks the performance. The sensitivity analysis, performed for several downhole desander types, has shown the high dependency of the sand separation efficiency on the major desander design parameters. Proper selection of the components and operating parameters will contribute to an increase in the meantime between failures. This paper will present the testing configurations, the development of the high-efficiency in-house downhole desander, and the sensitivity analysis performed on the design.

Application of CNN-LSTM in Gradual Changing Fault Diagnosis of Rod Pumping System

Owing to the importance of rod pumping system fault detection using an indicator diagram, indicator diagram identification has been a challenging task in the computer-vision field. The gradual changing fault is a special type of fault because it is not clearly indicated in the indicator diagram at the onset of its occurrence and can only be identified when an irreversible damage in the well has been caused. In this paper, we proposed a new method that combines the convolutional neural network (CNN) and long short-term memory (LSTM) network to perform a gradual changing fault classification. In particular, we employed CNN to extract the indicator diagram multilevel abstraction features based on its hierarchical structure. We considered the change in the time series of indicator diagrams as a sequence and employed LSTM to perform recognition. Compared with traditional mathematical model diagnosis methods, CNN-LSTM overcame the limitations of the traditional mathematical model theoretical analysis such as unclear assumption conditions and improved the diagnosis accuracy. Finally, 1.3 million sets of well production were set as a training dataset and used to evaluate CNN-LSTM. The results demonstrated the effectiveness of utilizing CNN and LSTM to recognize a gradual changing fault using the indicator diagram and characteristic parameters. The accuracy reached 98.4%, and the loss was less than 0.9%.

Simulation Analysis of Nonlinear Friction of Rod String in Sucker Rod Pumping System

The nonlinear friction of rod string is an important foundation of dynamic simulation and efficiency optimization in sucker rod pumping system (SRPS). As we know, there are geometrical nonlinearity and contact nonlinearity of rod string in special oil well. Considering the earlier practical problems, a novel computational method of nonlinear friction of rod string is given. First, the simulation model of nonlinear friction is built. In the detail, the nonlinear friction of sucker rod string is made up of four parts: the viscous damping friction of rod surface, the viscous damping friction of rod coupling, liquid friction of centralizer, and the nonlinear Coulomb friction between rod and tube. Second, considering the rheology property of multiphase fluid in annular space, a simulation method of fluid pressure is proposed with software ansys. Based on earlier models, a numerical simulation model of nonlinear friction of rod string is built with visualbasic 6.0. Comparing experiment results with simulated results, good agreement is found, which shows the simulation model is feasible. The characteristics of fluid pressure and pressure gradient are analyzed. The instantaneous friction of rod string is simulated, and the sensitive factors of rod friction are analyzed. The novel simulation method solves the problem of environmental load calculation of large deformation rod. The analysis results have an important significance to guidance of optimization and design of SRPS.

Optimization and Application of Reciprocating Direct-Drive Electric Submersible Plunger Pump Lifting System in the Xinjiang Oilfield

Background: A reciprocating direct-drive electric submersible plunger pump (RDD-ESPP) lifting system is an innovative rodless lifting method, which uses a downhole linear motor to directly drive the plunger pump to lift the fluid. This method effectively overcomes the problems encountered in a conventional sucker rod pumping system, such as severe eccentric wear between the sucker rod and the tubing, higher energy consumption and frequent pump maintenance. Objective: In this manuscript, the objective is to propose a methodology on optimizing the working parameters of RDD-ESPP lifting system and validate the theory with field case. Methods and Results: Mathematical models for the wellbore temperature distribution, lifting load, pump efficiency, system efficiency and tubing strength validation are established. A selection process for the optimal working parameters is proposed, and a parameter sensitivity analysis is implemented. The results show that pump efficiency is largely affected by gas and leakage, and there are reasonable pump submergence depths in oil wells for different produced gas-to-oil ratios. A field application test in a low-production well resulted in a 33.5% power savings rate and 41.6% improvement in the pump efficiency. Conclusion: RDD-ESPP lifting system has a promosing potential on power savings and consumption reduction compared with conventional sucker rod pumping system.

On Leakage Issues of Sucker Rod Pumping Systems

The issues of leakage with respect to the clearance between the pump plunger outer diameter and the pump barrel inner diameter and other operation conditions have been revisited in this paper. Both Poiseuille flow rate due to the pressure difference and Couette flow rate due to the plunger motion have been considered. The purpose of this study is to better understand the nature of the leakage with respect to pressure difference, eccentricity, and motion related to the plunger of typical sucker rod pump systems. More specifically, based on the newly derived relaxation time scales for transient solutions of the governing Navier–Stokes equations, the quasi-static nature of relevant measurement techniques is confirmed for the current production systems. This key observation is also demonstrated with a computational model using the experimentally measured pressure difference and the plunger movement.

Fluid characteristic of multiphase fluid in annular space between pump barrel and plunger

The pump performance parameters, such as pump pressure, plunger friction and pump valve resistance, are fundamental parameters of optimal design of pump efficiency and sucker rod pumping system (SRPS). In this paper, considering the characteristic of geometrical nonlinear and rheology property of multiphase fluid, the pump performance parameters are studied. Firstly, a dynamics model of annular fluid flow is built. In the detail, a partial differential equation of annular fluid is established and a computing model of fluid pressure gradient is built. Secondly, the simulation models of plunger friction and hydraulic resistance of pump valve are built. Finally, a novel simulation method of fluid pressure in annular space is proposed with software ANSYS. In order to check up the correction of models proposed in this paper, the comparison curves of experiment and simulation results are given. Based on above model, the whole simulation model of plunger pump is simulated with Visual Basic 6.0. The results show that the fluid friction of pump plunger and instantaneous resistance of pump valve are nonlinear. The impact factors of pump performance parameters are analyzed, and their characteristic curves are given, which can help to optimize the pump motion parameters and pump structural.

Simulation Analysis of Critical Contact Load of Buckling Rod in Tubing String

Sucker rod pumping system is the main equipment of domestic mechanical oil production. The eccentric wear of the rod is the main factor that affects the optimal design of the rod pump system and the reasonable evaluation of pump detection period. Therefore, considering the effect of rod string buckling in tubing, a simulation method for the critical contact load of buckling rod string in tubing is established. In the detail, with double coordinate system method, and the finite element analysis dynamic model of sucker rod string is established. Using ANSYS software, a simulation method for critical contact load of rod string buckling in tubing is established. The simulation results show that the pipe contact stress is affected by the rod string diameter. The maximum stress increases with decreasing the diameter of rod string. The maximum stress of rod string increases with increasing the polished rod load. The effect of polished rod load and rod diameter on buckling stress is considered comprehensively, and the optimum design area of rod string is given. The simulation results have important theoretical and practical for optimal design of sucker rod pumping system and reasonable evaluation of oil well pump testing period.

Study Evaluates Ability of Tailpipe Systems To Optimize Artificial Lift in Horizontal Wells

This article, written by JPT Technology Editor Judy Feder, contains highlights of paper SPE 190938, “Evaluating Tailpipe Systems Designed To Optimize Artificial Lift Performance in Horizontal Wells,” by C.R. Humphreys, SPE, B.N. Vangolen, SPE, A.P. Allison, SPE, D. Yin, SPE, and C. Yicon, SPE, Occidental Petroleum, prepared for the 2018 SPE Artificial Lift Conference and Exhibition–Americas held in The Woodlands, Texas, USA, 28–30 August. The paper has not been peer reviewed. The performance of artificial-lift systems on horizontal wells is greatly influenced by both the volume of gas produced and the tendency for gas slugging. With a sucker rod pump (SRP) system, gas slugging leads to gas interference at the pump, reducing system efficiency and equipment run life. With an electrical submersible pump (ESP), gas slugs can cause cycling of the ESP, which may shorten run life significantly. A trial project was launched to evaluate the performance of two tailpipe systems that could be applied to both forms of artificial lift. Introduction The two tailpipe systems were tested in a number of wells using SRPs, and one well was tested using an ESP. The goal was to Reduce the frequency and magnitude of slugging behavior seen at the pump Reduce the flowing bottomhole pressure without having to land pumps past the kickoff point (KOP) Improve separation of free gas from the produced fluid before it reaches the pump intake The two systems differ in both separator design and packer location. The first uses a conventional packer-style gas separator with a reduced-inner-diameter (ID) tailpipe extending below the separator and past the KOP. The second uses a special cyclonic separator with a reduced-ID tailpipe, and the packer is located at the lower end of the tailpipe. The authors examined the differences between the two systems regarding the isolation location, whether at the top or bottom of the tailpipe, to aid in designing future systems. A comparison of the two separators was attempted, and various operational challenges are discussed in the complete paper to improve the design, installation, startup, and operation of these systems. Some installations were outfitted with downhole gauges (DHGs) measuring pressure and temperature at several points along the tailpipe. The DHGs recorded pressure at the tailpipe inlet and outlet, pump intake pressure, and pump discharge pressure. This surveillance package allowed for real-time monitoring of the performance of both the tailpipe and the artificial-lift system while also providing data for modeling the flow regime through the tailpipe. The modeling results were used to forecast long-term performance of the system as well production declines over time. Results from the field trial show the performance of each system from a variety of standpoints: changes in flowing bottomhole pressure, flowing behavior through the tailpipe, separation effectiveness, and changes in production.

Supervised dictionary-based transfer subspace learning and applications for fault diagnosis of sucker rod pumping systems

Sucker rod pumping wells are systems that their operation states varies slowly. For a relatively new well, it is hard to collect all kinds of fault samples for training. Moreover, samples from different wells are not always have similar distributions, so directly using samples from other wells as the training data may hardly get good results. In this paper, a novel framework is proposed to solve the aforementioned problems. For the source data from one well and the target data from another well, a transform matrix is calculated to transfer these data into a common low dimensional subspace. In this subspace, the source data that contain all kinds of fault samples and the target data that lack some kinds of fault samples can be represented by a shared dictionary matrix. By introducing two idea regularization terms, the structure information of source data and target data are included into the dictionary learning process. So the obtained dictionary has discriminative ability. Extensive experiments are conducted to evaluate the effectiveness of the proposed method.