Rod Pumping Research Articles

Data Augmentation Method Based on Improved Generative Adversarial Network for the Sucker Rod Pump System

Data Augmentation Method Based on Improved Generative Adversarial Network for the Sucker Rod Pump System

Multi-well dynamic liquid-level prediction method of pumping well based on dynamic and static information feature fusion neural network

In a rod pumping system, accurate prediction of dynamic liquid level is a key to rational optimization of pumping system parameters. At present, it is difficult to use a uniform prediction model for dynamic liquid-level prediction in different rod pumping systems. To this end, a method for predicting the dynamic liquid level of multiple wells in a pumping well based on dynamic and static information feature fusion (DSIFF) neural network is proposed in this paper. According to the principle of dynamic liquid-level calculation, suspension displacement, suspension load, pumping rod parameters, formation crude oil density, surface crude oil density, oil pressure, and casing pressure are used as inputs. The subnetwork feature extraction method is proposed for the problem of large network structure caused by the high dimension of the important dynamic data of suspension displacement and suspension load. The Huber loss function is used as the loss function of the prediction model for the case of abnormal data in the large amount of data required for model training. Finally, the results of comparative analysis show that the proposed method solves the problem of unified establishment of dynamic liquid-level prediction model for multiple oil wells and has better robustness for abnormal data.

Implementation of automated selection of downhole pumping equipment in the ABAI information system

This article provides a description of the main approaches and principles for constructing individual algorithms used in the "Selection of downhole pumping equipment" module of the ABAI information system. The calculations were based on generally recognized methods for constructing an inflow curve, strength calculation of the rod string of a sucker rod pumping unit, etc.
 The main goals of creating a module for the selection of downhole pumping equipment, in particular, the subsystem "Selection of a sucker rod pump" are to increase the efficiency of the mechanized well stock, reduce premature failures of downhole pumping equipment due to the correct calculation of the rod string, pumping equipment mode, automation of the selection of downhole pumping equipment due to the mass use of the program by process engineers when working with a mining mechanized fund.
 The calculation algorithm, concise functionality, friendly interface in the module are built taking into account the work of field process engineers and shop foremen, in the conditions of making operational decisions on the selection of downhole pumping equipment in case of current failures.
 The authors compared the selection results in the process of implementing the module with generally recognized foreign software systems.

Application of Industrial Internet of Things (IIoT) in Crude Oil Production Optimization Using Pump Efficiency Control

The collapse of oil prices in mid-2014 and early 2016 was the biggest in modern history that witnessed more than a 70% drop in the price to around $40 barrel. It prompted companies to think seriously to maintain profitability. Most companies were able to survive partly by simplifying their operations. Price recovery in 2021 is only 70% of its peak value. Companies are focusing on reducing the cost of operations, increasing production simultaneously, and finding new and different strategies to survive. The current scenario is witnessing strong research focusing on the development of process control for oil and gas upstream and downstream to improve the control and preventive maintenance to reduce operating costs and increase production. This paper presents the Industrial Internet of Things (IIoT) practical solution that improves the oil production rate from the well and increases the average pump efficiency (fillage) to 90%. This paper proposes a mechanism for collecting, storing, and analyzing all required parameters to build valuable charts. These charts’ data help optimize the values and parameters for controller setpoints to prevent pump gas lock problems. An artificial lift is required to lift the oil from the well. In this paper, the sucker rod pump is driven by a gas engine fed by the well’s gas. At the same time, SCADAPack 535E remote terminal unit collects all pump and well parameters such as hydraulic pressure, casing pressure, tubing pressure, and pump speed in stroke per minute (SPM). The remote terminal unit sends the data through a wireless network using a 5 GHz antenna to the main control room. The IIoT platform is designed using a visual basic programming language. Microsoft DDE (Dynamic Data Exchange) and Kepware OPC server were used to work on the received data to monitor, generate charts, and apply the controllers.

Technology Focus: Artificial Lift (October 2022)

The oil and gas industry tends to have a poor reputation externally when it comes to innovation. Artificial lift is particularly notable in this regard, because the primary lifting techniques used today have existed in some form or format for more than a century. Artificial lift’s status as the “nursing home for wells” probably doesn’t help either. The medical and electronics industries often are praised for innovation, but some antiquated practices and dichotomies remain even in those fields. For example, although great advancements have been made recently in the medical realm in combating a certain virus, just last year I was required to fax something to my medical provider. (When was the last time you were asked to fax something?) And, although the electronics industry has consistently developed smaller, more-powerful computers with incredible storage capacities, the software residing on them has expanded to tax those processors and fill that space. How many megabytes do today’s operating systems require compared with the first version you ever used? It’s been widely reported that the electronics industry is reaching a turning point. Transistors can’t get much smaller because of heat-dissipation limitations, quantum interference effects, and other physics boundaries. Recent articles have lamented the end of Moore’s Law. New paths of thinking, designing, and engineering must be taken to sustain the march of computational progress. Serious attempts are now being made to advance new semiconductor materials, quantum computing, novel system architectures, optimized software, spintronics, and optical computing. Exciting, right? That brings us back to artificial lift, which has long been thriving in the exciting innovation zone that the electronics industry is only now approaching. Just when you think the artificial lift community knows everything, people find new ways to turn old maxims on their head. The selections this month, both the primary selections and recommended-reading selections, are true paradigm-changers. Some examples of hypotheses they challenge include the following: - SPE 204521 vs. “ESPs should never be deadheaded.” - SPE 205134 vs. “Gas-lifted well integrity is not possible to predict.” - SPE 209253 vs. “Rod pumps can’t be modeled using only pressure.” - And one of my favorites, SPE 207596 vs. “Brownfields can’t justify real-time surveillance.” I hope these selections pique your curiosity—or are you still trying to remember the last functional fax machine you encountered? Recommended additional reading at OnePetro: www.onepetro.org. SPE 209253 Rod Pump Status and Efficiency Monitoring With High-Frequency, High-Resolution Surface Pressure Data by Ryan David Gordon, Toku Systems, et al. SPE 204524 Overview of Opportunities and Challenges of Electrical Submersible Pumps in Geothermal Energy Production Systems by Pejman Shoeibi Omrani, TNO, et al. SPE 207596 The Benefits of Real-Time Monitoring Application Artificial Lifting Parameter for Surveillance, Optimization, and Cost Reduction in Mature Field: Case Study From Prabumulih Field by Taufik Fansuri, Pertamina Hulu Energi, et al.

Analysis of corrosion fatigue steel strength of pump rods for oil wells

Purpose is to perform analysis of corrosion durability (fatigue) of pump rod materials in terms of various chemically active simulation environments, and study influence of economically modified rare-earth impurity on corrosion fatigue strength of pump rod materials. Methods. 40 and 20N2M steel grades have been applied as well as experimental steel (ES). Steel of the conditinal ES grade has been melted within a pilot site of Institute of Electric Welding Named after E.O. Paton of the National Academy of Sciences of Ukraine. The steel was alloyed economically by means of a micro impurity of a rare-earth element (REE) being 0.03% of cerium; in addition, it contained comparatively low concentration of sulfur and phosphorus as well as minor concentration of dissolved hydrogen. The following has been used as simulation environments: 1) NACE environment (i.e. 5% NaCl solution which contained 0.5% СН3СООН, and saturated H2S; t = 22 ± 2°C; pH = 3.8-4.0); 2) 3% NaCl solution without hydrogen sulphide. Once every day, the environment was replaced to oxygenate it up to 8-10 mg/l concentration. Findings. Stability against sulfide stress-corrosion cracking (SSCC), hydrogen initiated cracking (HIC), and corrosion fatigue of steel of deep pump rods for oil industry has been studied. It has been defined that the experimental steel, modified economically by means of micro impurities of a REE, meets NACE MR0175-96 standard in terms of chemical composition as well as strength; in turn, 20N2M and 40 steel grades have high resistance neither to SSCC (threshold stresses are < 0.8 s) nor to corrosion fatigue attack; moreover, steel grade 40 has demonstrated low resistance to HIC (CLR > 6% and CTR > 3%). Originality. It has been identified that corrosion fatigue attack results from hydrogen penetration of steel initiating its cracking and hence destruction under the effect of alternating loads accelerated by the action of corrosive environment. Further, surface micro destructions, influenced by micro stresses, transform into large discontinuities and cracks with following macro destructions. Practical implications. It has been proved that high resistance to corrosion cracking can be achieved by means of refining of pump-rod steel of ferrite and perlite type using metallurgical methods, i.e. 0.01-0.03% REE microalloying.

Study on the Combination of Pump Rod Pipe in Complex Structure

The 3-D mechanical model of the inclined shaft pumping pump and the optimization model of anti-partial wear structure and security position are established in this paper. Based on the real data of well, 148-inclined 2 slope measurement, the ordinary pipe rod combination, inclined well pumping pump lining oil pipe, and oil pump laser oil pipe are optimized and calculated. In order to improve the computing power and reduce computational time consumption, this paper uses cloud computing technology in the computing process. The results show that the combination method of the lining oil pipe and the ordinary continuous pumping rod has less axial force and lateral force, and finally, the combination method of different well pipes is obtained.

Effect of fluid properties and boundary conditions on rod pump efficiency – Novel modeling

Among artificial lift systems, the most widely used method is sucker rod pumping. Since its inception in the oil industry, inefficiency in sucker rod pumping system due to gas interference is one of major concerns. The need for better understanding of gas interference during sucker rod pumping process is significantly required. Although, many models and techniques are available but there are many parameters, which need to be well studied and adequately addressed. This delineates the need to research all empirical parameters that could affect efficiency of conventional rod pump. The pump performance appears to be low in presence of gas which in extreme situation locks the pump and make the pump paralyzed completely. The models available in literature neglect mass transfer during suction process, which brings unreliable calculation of gas phase in pump. Models in literature on efficiency of pump and gas lock are studied to analyze the vital parameters and reasons for the inefficiency. The expedient analytical models to predict gas lock criteria and efficiency of conventional pump, based on liberation of gas during pumping, are established in this work. The models emphasize the importance of the fluids properties to be handled by pump, which affect the production rate while pumping with a gaseous phase. This new model considers free gas phase presence and mass transfer during working of pump. Sensitivity and comparative analysis of numerous parameters contributing to pump efficiency is established in this work. The study results provide some theoretical and analytical references for further advancements in system efficiency of conventional rod pump system and to improve economic benefits.

Comparison of a novel finite element method for sucker rod pump downhole dynamometer card determination based on real world dynamometer cards

• Evaluation of the downhole dynamometer card based on the finite elements method. • The novel advanced finite element method overcomes limitation of existing models and is based on physical parameters. • An excellent match with measured downhole dynamometer cards was reached. • The simulation results allow a the evaluation of the performance parameters of the sucker rod pumping system. An advanced dynamic finite element model is presented that diagnoses the downhole pump performance of sucker rod pumping systems, applicable for any pumping conditions and equipment used. The results are compared to downhole measurements and other state of the art evaluation techniques. Existing diagnostic tools exhibit specific limitations that reduce their applicability and output quality. This paper introduces a diagnostic tool, which can predict the rod string's stress field and its movement not only at the pump plunger but all along the rod string. Moreover, this tool can account for the interaction between rod guides and tubing as well as rod string and tubing. To this end, innovative tube-to-tube contact modeling is applied. The high precision results are accomplished by running a dynamic finite element simulation. The basic principle is to evaluate the plunger load incrementally by consecutively applying restarts of each time step, fully automated and computation time optimized. This publication shows that both the plunger load and the rod string's dynamic behavior can be determined for any given wellbore as long as the borehole trajectory and surface dynamometer measurements are known. The dynamic finite element model is evaluated for a deviated system and a vertical system equipped with two different downhole pump types. Comparing the simulation results with the available downhole measurements shows an excellent match. The the proposed solution provides a considerable amount of details about the overall system's behavior. The evaluation has shown that the performance of standard and novel downhole pump types can be successfully diagnosed in detail, which is just possible under limitations with commercial software solutions.. The novelty of the shown technique is the consideration of the full 3D trajectory, the implementation of only physical properties of the equipment used, and a realistic setup. The validation of the model output with measured downhole data indicates an excellent accuracy of the shown model.

Indicator diagram analysis based on deep learning

At present, more than 90% of China’s oil production equipment comprises rod pump production systems. Indicator diagram analysis of the pumping unit is not only an effective method for monitoring the current working condition of a rod pump production system but also the main way to prevent, detect, and rectify various faults in the oil production process. However, the identification of the pumping unit indicator diagram mainly involves manual effort, and the identification accuracy depends on the experience of the monitoring personnel. Automatic and accurate identification and classification of the pumping unit indicator diagram using new computer technology has long been the research focus of studies for monitoring the pumping unit working condition. In this paper, the indicator diagram is briefly introduced, and the AlexNet model is presented to distinguish the indicator diagram of abnormal wells. The influence of the step size, convolution kernel size, and batch normalization (BN) layer on the accuracy of the model is analyzed. Finally, the AlexNet model is improved. The improved model reduces the calculation cost and parameters, accelerates the convergence, and improves the accuracy and speed of the calculation. In the experimental analysis of abnormal well diagnosis, the data are preprocessed via data deduplication, binary filling, random line distortion, random scaling and stretching, and random vertical horizontal displacement. In addition, the image is expanded by transforming several well indicator diagrams. Finally, data sets of 10 types of indicator diagrams are created for better adaptability and application in the analysis and classification of indicator diagrams, and the ideal application effect is achieved in actual working conditions. In summary, this technology not only improves the recognition accuracy but also saves manpower. Thus, it has good application prospects in the field of oil production.

IMPROVEMENT OF ROD EQUIPMENT OF DEEP PUMPING UNITS IN THE CONDITIONS OF OIL FIELD DEVELOPMENT

Currently, it is relevant to reduce the cost of maintenance, selection and purchase of fishing equipment, primarily deep-water pumps, tubing, pumping rods. After all, they are most susceptible to wear and the supply of the rod pumping unit depends on their condition. However, despite the well-known improvement of production technology introduced by manufacturers, there is a tendency not only to increase the service life of pumps, but also to decrease. The reason for the decrease in the service life of deepwater pumps is the continuous deterioration of their operating conditions. The reason for the decrease in the service life of depth pumps is the continuous deterioration of their operating conditions. It is established that this is due to an increase in the depth of discharge of pump compressor pipes (SCS), progressive wetting of well production, an increase in the corrosive properties of the pumped liquid, etc. The cause mentioned above in the article was solved in the following way, studying the consequences. The task of the proposed modification of the stand is to improve the design of the unit for testing rod pumps by changing the position of the pump drive and ensuring accounting for leaks through the plunger pair. The mechanism of mutual action of the plunger in the pump cylinder is solved by the fact that the internal channel of the plunger of the rod pump testing unit with a system for measuring and recording pump parameters is blocked, a pressure gauge with a high-pressure inert gas is installed on the pressure line. Modeling of fluid flow in the stand position was carried out. A plan of the experiment Matrix was drawn up and the minimum number of repetitions of the experiment was determined.The obtained experimental data were processed statistically. The obtained regression models were tested for statistical significance. Detection rates were more than 90% in all cases. It was found that during turbulent fluid flow, the amount of fluid leakage in plunger pairs with a constant microregion in the form of a right triangle is less than in smooth plunger pairs. The Reynolds number for quantitative experiments for each hole was carried out by changing the value of the constant viscosity, a given flow of liquid entering the hole of the plunger pair.

A novel method for trajectory recognition and working condition diagnosis of sucker rod pumping systems based on high-resolution representation learning

The pumping unit is the most important lifting equipment in the rod pumping system, which transmits ground power to the downhole pump through its sucker rod. Trajectory identification of the rod pumping system can be used to study the motion law of the pumping unit, and it is the basis for the design of the pumping parameters and the analysis of the working conditions. Most of the studies on the dynamics of the pumping unit are to discuss the simple harmonic motion of the suspension point or the motion of the crank-slider in the ideal state, which is quite different from the actual situation. In this paper, a key point detection method for oil pumping units based on high-resolution representation learning is proposed. The model adopts a multi-resolution parallel network, which connects multiple network branches with different resolutions in parallel. Then it performs multi-stage processing on the input pumping unit images and outputs the predicted Gaussian heatmaps of key points to realize the trajectory recognition of the pumping unit. Different from the traditional series convolutional network structure, which needs to recover high resolution from low resolution, the parallel structure of this model is able to maintain the high resolution of the backbone network and thus the learned representations are more accurate in spatial location. This method is also experimentally verified and compared with the latest convolutional network structure. The results show that the model accuracy of the high-resolution representation learning method is more than 6% higher than that of the comparison methods. In addition, the results of this model can be used to judge the balance of the pumping unit, automatically calculate the maximum stroke and polished rod stroke, and have guiding significance for the working condition diagnosis of the rod pumping system.

A Novel Hybrid Method for Indirect Measurement Dynamometer Card Using Measured Motor Power in Sucker Rod Pumping System

The dynamometer card (DC) is crucial for the remote monitoring of the sucker rod pumping system (SRPS). However, the DC sensors have poor real-time performance and high maintenance costs leading to difficulties in obtaining datasets of the downhole working states from different oil wells. Moreover, the motor power curve is easy to obtain in real-time and can reflect the changes in pumping unit load. Thus, this paper proposed a novel hybrid model for indirect measurement DC based on measured motor power. Firstly, based on the mechanical model analysis of SRPS, the dataset is constructed by transforming the motor power and geometric parameters of the SRPS into the polished rod torque, the first-derivative of polished rod torque, the second-derivative polished rod torque, and the torque factor. These four parameters are merged as the inputs of the data-driven model. Subsequently, a data-driven model implements the particle swarm (PSO) algorithm to optimize the weights and thresholds of the Back Propagation (BP) neural network model, which was trained to predict DCs of three common working states. Finally, the proposed method is verified experimentally through the measured DC and motor power data. Both experimental and prediction results demonstrate the effectiveness of the proposed method for generating DC. The mean DC area relative error of the hybrid method proposed in this paper is 2.52%.

Research on mechanical characteristics and safety of working platform in rod pumping process

Rod pumping on offshore platform provides a solution to low efficient extraction of offshore heavy oil and the safety of working platform is the prerequisite for the implementation of this new technology. In this paper, a coupling dynamics model of rod pumping system and the jacket platform is established, and the dynamic characteristics of the pumping unit excitation, rod pumping system, and jacket platform is studied. Based on the analysis of the multi-body dynamics model, the three frequencies (pumping unit excitation, rod pumping system, jacket platform) are far apart, so the dynamic effect of rod pumping on the platform can be analyzed according to static load. Based on the limit working state and the cumulative damage theory of the structure, the safety evaluation standard for the increased load of the jacket platform is established. Under the extreme marine environment once in 50 years, the safety factor of the main structure of the platform is greater than 4, and the overall safety factor is 3% lower than that before the rod pumping operation. Offshore rod pumping has less influence on the jacket platform from analysis, and which lays a theoretical and safety foundation for offshore oil pumping.

Mathematical modeling of the deformation of a cylindrical ring under the influence of an external load distributed along the generatrix

The article is devoted to the mathematical modeling of the deformation of a cylindrical ring under the influence of an external load applied to it by pressing parallel plates. It is shown that at the last stage of oil field operation, the main method of artificial lift is the use of sucker rod pumping units (SRPU), on whose failure-free operation the profitability of oil production largely depends. It is also shown that the technical condition of the SRPU is controlled by the dynamometry method, which is the dependence of the rod string force on the suspension on the stroke of the polished rod. In this case, the main source of information is the force transducer. In the article, we propose using a hollow thick-walled cylinder, which is compressed between traverses consisting of parallel flat plates, as the primary sensitive element of the force transducer. The question of refinement and determination of the non-Hertzian contact deformation during compression of a thick-walled hollow cylinder by two parallel flat plates within the elasticity range is considered. It is shown that the well-known formula for the convergence of the axes of the cylinders being compressed according to the Hertzian method does not take into account the real deformation and convergence of the points of the cylinder due to compression and contact deformation between the generatrices of the cylinders. By solving the differential equation, formulas are derived to determine the value of the elastic deformation of a cylinder depending on its geometric dimensions under the influence of an external load applied to it by pressing plates. An example of calculating elastic deformation for different values of cylinder thicknesses depending on the load is given and its graph is plotted.

Express-diagnostics of rods’ parting-twist off at wells equipped with sucker rod pump

The development of fields that are at a late or final stage of operation is carried out mainly with the use of downhole sucker rod pumps. The most characteristic failures for these installations are rods parting-twist off and malfunctions of pump valves. The methods of eliminating these accidents vary significantly the parting or twist off of the rods involves lifting of the downhole equipment, and the "sticking" of the shut-off valve assembly of the pump is eliminated by "resuscitation" of the downhole equipment (flushing with water, hot oil or solvent). An error in fault identification leads to incorrect planning of work to restore the operability of the downhole equipment and, as a result, to economic losses.
 The cause of malfunctions of the downhole equipment of downhole rod pumping units, as a rule, is determined by analyzing dynamograms. However, in many cases dynamograms do not allow to distinguish the lower twist off of the rods from the malfunction of the valves of the rod pump. In the presented work, a method for the operational determination of the parting or twist off of rods in the well is considered, which consists in creating an electrical circuit "rod string tubing string" and monitoring its integrity. To determine the type of malfunction, the synchronization unit measures the resistance of the system on the dielectric insert via the electromagnetic channel. In case of parting or twist off of the rods, the electrical resistance of the "tubing-pump-rod string" circuit will be much higher (more than 2 ohms) than in the absence of this failure (0...2 ohms).

Multi-Weighted Partial Domain Adaptation for Sucker Rod Pump Fault Diagnosis Using Motor Power Data

Motor power curves (MPCs) have received great attention for use in diagnosing the working conditions of sucker rod pumping systems (SRPSs) because of their advantages in accessibility and real-time performance. However, existing MPC-based approaches mostly need a rigorous assumption that the MPC instances of different working conditions are sufficient, which does not hold in industrial scenarios. To this end, this paper proposes an unsupervised fault diagnosis methodology to leverage readily available dynamometer cards (DCs) to diagnose collected unlabeled MPCs. Firstly, a mathematical model of the SRPS is presented to convert actual DCs to MPCs. Secondly, a novel diagnostic methodology based on adversarial domain adaptation is proposed for the problem of data distribution discrepancy across the collected and converted MPCs. Specifically, the collected unlabeled MPCs may only cover a subset of the working conditions of the abundant DCs, which will easily cause negative transfer and lead to dramatic performance degradation. This proposed methodology employs class-level and distribution-level weighting strategies so as to guide the network to focus on the instances from shared categories and down-weight the outlier ones. Validation experiments are performed to evaluate the mathematical model and the diagnostic methodology with a set of actual MPCs collected by a self-developed device. The experimental result indicates that the accuracy of the proposed algorithm can reach 99.3% in diagnosing actual MPCs when only labeled DCs and unlabeled actual MPCs are used.

Unsupervised Fault Diagnosis of Sucker Rod Pump Using Domain Adaptation with Generated Motor Power Curves

The poor real-time performance and high maintenance costs of the dynamometer card (DC) sensors have been significant obstacles to the timely fault diagnosis in the sucker rod pumping system (SRPS). In contrast to the DCs, the motor power curves (MPCs), which are accessible easily and highly associated with the entire system, have been attempted to predict the working conditions of the SRPS in recent years. However, the lack of labeled MPCs limits the successful applications in the industrial scenario. Thereby, this paper presents an unsupervised fault diagnosis methodology to leverage the generated MPCs of different working conditions to diagnose the actual unlabeled MPCs. Firstly, the MPCs of six working conditions are generated with an integrated dynamics mathematical model. Secondly, a framework named mechanism-assisted domain adaptation network (MADAN) is proposed to minimize the distribution discrepancy between the generated and actual MPCs. Specifically, benefiting from introducing the mechanism analysis to label the collected MPCs preliminarily, a conditional distribution discrepancy metric is defined to guarantee a more accurate distribution matching with respect to different working conditions. Eventually, validation experiments are performed to evaluate the mathematical model and the diagnosis method with a set of actual MPCs collected by a self-developed device. The experimental result demonstrates that the proposed method offers a promising approach for the unsupervised diagnosis of the SRPS.

Adaptive fault diagnosis of sucker rod pump systems based on optimal perceptron and simulation data

A highly precise and timely diagnosis technology can help effectively monitor and adjust the sucker rod production system (SRPS) used in oil wells to ensure a safe and efficient production. The current diagnosis method is pattern recognition of a dynamometer card (DC) based on feature extraction and perceptron. The premise of this method is that the training and target data have the same distribution. However, the training data are collected from a field SRPS with different system parameters designed to adapt to production conditions, which may significantly affect the diagnostic accuracy. To address this issue, in this study, an improved model of the sucker rod string (SRS) is derived by adding fault-parameter dimensions, with which DCs under 16 working conditions could be generated. Subsequently an adaptive diagnosis method is proposed by taking simulated DCs generated near the working point of the target SRPS as training data. Meanwhile, to further improve the accuracy of the proposed method, the DC features are improved by relative normalization and using additional features of the DC position to increase the distance between different types of samples. The parameters of the perceptron are optimized to promote its discriminability. Finally, the accuracy and real-time performance of the proposed adaptive diagnosis method are validated using field data.

Statistical analysis of the relative position of the rod hanger and the wellhead

One of the leading methods of exploitation of oil fields is oil production with the help of downhole rod pumping units (DRPU). Over 80 % of the operating well stock of Azneft PA is equipped with deep well pumps and about 30 % of oil is produced in the country with their help.
 The widespread use of DRPU is associated with a fairly high maturity of installations, simplicity of its design and maintenance, repair in field conditions, ease of adjustment, the possibility of servicing the installation by unskilled workers, a small effect on the operation of DRPU of the physical and chemical properties of the pumped liquid, as well as high efficiency.
 However, along with the high efficiency of the applied DRPU, there are also complaints regarding the need to increase the reliability and resource of wellhead equipment, including in order to improve the environmental situation in the oil fields.
 One of the conditions for ensuring high reliability of the ground equipment of the DRPU is to ensure the tightness of the wellhead rod-wellhead stuffing box assembly, the violation of which is not only a failure of the installation, but also leads to environmental pollution.
 This is facilitated by inaccuracies in the assembly and installation of DRPU at the wellhead. When mounting the pumping unit, for many reasons, the tolerance of the wellhead rod with the suspension point of the rod string to the balancer head is not ensured.
 In this regard, in the requirements for the accuracy of mounting the pumping unit at the point of application, a certain mismatch of the axes within the circular coordinates is allowed. So, for widely used pumping units of the CK8 type, the permissible mismatch between the axis of the wellhead rod and the suspension point of the rods is determined by the conditions under which the projection of the suspension point of the rods onto the plane of the base of the pumping unit at any position of the balancer is allowed within a circle with a diameter of 25 mm.