A method and system for evaluating the health status of a pumping well

Abstract

The present invention discloses a method for evaluating the health status of a pumping well, including: determining various influencing factors related to the operation efficiency, output and operation cost of the pumping well to form initial evaluation indicators; performing correlation analysis, principal component analysis and pump efficiency correlation degree analysis on various influencing parameters in the initial evaluation indicators according to the historical operation status and output data of multiple pumping wells, and selecting representative evaluation indicators for evaluating the health status of the pumping well from the initial evaluation indicators; setting a standard value range for each representative evaluation indicator; when the pumping well is running, regularly obtaining real-time data corresponding to the representative evaluation indicator, comparing various real-time data with the corresponding standard value range, and using the comparison results to monitor the operation status of the pumping well. The present invention correctly handles the contradictions between the production output, cost and safety of the pumping well, improves the production timeliness and controllability of the pumping well, and realizes the maximization of the economic benefits of the pumping well.

Description

Technical Field

[0001] The invention relates to the technical field of oil and gas exploitation, and in particular to a method and a system for evaluating the health status of a pumping well. Background Art

[0002] With the continuous deepening of oilfield exploitation and the continuous improvement of the requirements for economic efficiency of exploitation, my country's oilfield exploitation has gradually developed from extensive exploitation to refined exploitation. Moreover, with the continuous deepening of exploitation, the difficulty of oilfield exploitation has continued to increase, and the number of low-permeability and low-yield wells has continued to increase, and the economic efficiency of oilfield exploitation has become particularly significant.

[0003] The main exploitation area of ​​Tahe Oilfield contains the Ordovician Yijianfang Formation fracture-cave oil reservoirs, which have the characteristics of one ultra and five highs, namely, ultra-deep, high temperature, high pressure, high hydrogen sulfide content, high mineralization and high viscosity. The main block, Tuofutai area, also has the characteristics of poor liquid supply, high asphaltene content and high gas content.

[0004] Due to the particularity of the above-mentioned reservoir types, the working conditions of the pumping wells are complex and the abnormal risks are high. There are four main characteristics: ① The pump hanging is generally deep. Due to the lack of fluid supply, the rod pump deep pumping technology is widely promoted. At present, the average pump hanging has reached 2668m, the deepest pump hanging is 3599m, and there are 91 pumping wells with pump hanging exceeding 3000m; ② The working environment of the equipment is harsh. The pipes, rods, and pumps generally work in a fluid environment with high corrosion, high viscosity, and high impurities, and the risk of damage is high; ③ The gas-oil ratio of the thin oil wells is high. The original gas-oil ratio of the thin oil block is generally 100m3 / t~500m3 / t, and some wells are as high as 2000m3 / t, which greatly affects the pump efficiency and causes frequent gas locks in severe cases; ④ Some wells have a long production cycle, and the downhole pipes, rods, pumps and other equipment have been operating at high load for a long time, with high corrosion and aging, which further increases the risk.

[0005] In view of the above problems faced by the environment of the pumping wells, the existing technology lacks a health evaluation and monitoring system for the pumping wells suitable for the Tahe Oilfield area to conduct a comprehensive investigation and analysis of the pumping wells, so as to improve the lifting efficiency of the oilfield from the perspective of monitoring the operating status of the pumping wells. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a method for evaluating the health status of a pumping well, the method comprising: step one, determining various influencing factors related to the operating efficiency, output and operating cost of the pumping well to form initial evaluation indicators; step two, according to the historical operating status and output data of multiple pumping wells, respectively performing correlation analysis, principal component analysis and pump efficiency correlation degree analysis on various influencing parameters in the initial evaluation indicators, and selecting representative evaluation indicators for evaluating the health status of the pumping well from the initial evaluation indicators; step three, setting a corresponding standard value range for each representative evaluation indicator; step four, when the pumping well is in operation, regularly obtaining real-time data corresponding to the representative evaluation indicator, comparing each type of real-time data with the standard value range of the corresponding type, and thereby monitoring the operating status of the pumping well based on the comparison result.

[0007] Preferably, the representative evaluation indicators include: pump efficiency, oil production time rate, system efficiency, load utilization rate, balance rate, submergence, maintenance-free period, back pressure, alternating load and safe production days.

[0008] Preferably, in step 2, the first type of parameters to be analyzed within the initial evaluation parameters and the corresponding historical data are obtained, wherein the first type of parameters to be analyzed include pump diameter, stroke, number of strokes, pump depth, pump efficiency, submergence, oil production time rate, balance rate, back pressure, rated load, maximum load, minimum load, alternating load and load utilization rate; according to the historical data of the first type of parameters to be analyzed, the correlation coefficient and the corresponding degree of correlation between any two parameters to be analyzed are calculated, and based on this, the representative evaluation index based on correlation analysis is obtained.

[0009] Preferably, according to the historical data of the first type of parameters to be analyzed, the correlation coefficient and the corresponding degree of correlation between any two parameters to be analyzed are calculated, and based on this, the step of obtaining the representative evaluation index based on the correlation analysis includes: dividing the first type of parameters to be analyzed into oil pump type parameters, load type parameters and overall index type parameters, and determining the corresponding historical data; according to the historical data of the oil pump type parameters, grey correlation analysis is performed on the pump diameter, stroke, number of strokes, pump depth and sinking degree respectively, and the parameters with the most correlation and the highest degree of correlation with other oil pump type parameters are selected as the oil pump type parameters. Representatives of pump parameters; according to the historical data of the load parameters, grey correlation analysis is performed on the rated load, maximum load, minimum load, alternating load and load utilization rate respectively, and based on the degree of correlation between any two types of load parameters, parameters that are correlated with other load parameters are selected as representatives of load parameters; for the overall index parameters in the first type of parameters to be analyzed, grey correlation analysis is performed on pump efficiency, submergence, oil production time rate, balance rate, back pressure, alternating load and load utilization rate respectively, and parameters that are not correlated with other overall index parameters are selected as representatives of overall index parameters.

[0010] Preferably, in step 2, the N types of influencing factors in the initial evaluation index are taken as N-dimensional variables, and the historical data of each variable is determined; based on the sample space constituted by the historical data of each variable, the eigenvector corresponding to each variable is determined in the sample space, and the number P of target principal components is clarified; through the orthogonal transformation method, P target principal component variables are generated, and used as the representative evaluation index based on principal component analysis.

[0011] Preferably, in the step 2, a preset neural network model is constructed according to the initial evaluation parameters; based on the historical data corresponding to the initial evaluation parameters of multiple pumping wells, the correlation degree between various influencing parameters in the initial evaluation indicators and the pump efficiency is analyzed by training the neural network model to obtain multiple initial influencing parameters and pump efficiency relationship coefficients; the multiple initial influencing parameters and pump efficiency relationship coefficients are arranged from high to low, and the initial influencing parameters corresponding to a preset number of relationship coefficients are selected starting from the maximum relationship coefficient as the representative evaluation indicators based on the pump efficiency correlation degree analysis.

[0012] Preferably, the standard value range is constructed as a standard value range for different health levels.

[0013] Preferably, the method also includes: when the actual data of the current pumping well is in a sub-healthy or unhealthy state, adjusting the operating parameters of the current pumping well according to the actual operating state of the current pumping well and the geological conditions of the oil field area where the current pumping well is located, so that the real-time data corresponding to the representative evaluation index of the current pumping well returns to a healthy state.

[0014] On the other hand, the present invention also provides a system for evaluating the health status of a pumping well, the system comprising: an initial influencing factor determination module, configured to determine various influencing factors related to the operating efficiency, output and operating cost of the pumping well, and form initial evaluation indicators; a representative indicator determination module, configured to perform correlation analysis, principal component analysis and pump efficiency correlation degree analysis on various influencing parameters in the initial evaluation indicators according to the historical operating status and output data of multiple pumping wells, and screen out representative evaluation indicators for evaluating the health status of the pumping well from the initial evaluation indicators; a standard range determination module, configured to set a corresponding standard value range for each representative evaluation indicator; an evaluation module, configured to regularly obtain real-time data corresponding to the representative evaluation indicators when the pumping well is in operation, and compare each type of real-time data with the standard value range of the corresponding type, so as to monitor the operating status of the pumping well based on the comparison result.

[0015] Preferably, the representative evaluation indicators include: pump efficiency, oil production time rate, system efficiency, load utilization rate, balance rate, submergence, maintenance-free period, back pressure, alternating load and safe production days.

[0016] Compared with the prior art, one or more embodiments of the above scheme may have the following advantages or beneficial effects:

[0017] The present invention discloses a method and system for evaluating the health status of a pumping well. The method and system include: selecting a pumping well health evaluation system indicator according to an established pumping well management health evaluation principle; optimizing the health evaluation system indicator from aspects such as the correlation between the indicators, the importance of the indicators to the economic operation of the pumping well, the relationship between the indicators and the pump efficiency, the operation cost of the pumping well, and the operation safety of the pumping well; formulating a health threshold range for the preferred system indicator (representative evaluation indicator) in combination with the actual production situation of the oil well for different development units and different working parameters of different pumping wells; and evaluating the health status of the pumping well by regularly monitoring the actual data of the preferred system indicator during the production and operation of the pumping well. In addition, the present invention can also adjust the actual production situation of the pumping well in combination with the development unit and working parameters of the current pumping well for the actual data of the pumping well in a sub-healthy or unhealthy state, so that the above-mentioned sub-healthy or unhealthy state parameters return to the healthy range.

[0018] In this way, the present invention can utilize the health evaluation system of the pumping well to correctly handle the contradictions between output, cost and safety in the production process of the pumping well, improve the production timeliness and controllability of the pumping well, and apply the evaluation system to maximize the optimization of the operating parameters of the pumping well, optimize the output, minimize the operating costs, and maximize the economic benefits of the pumping well.

[0019] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0021] Figure 1 It is a step diagram of a method for evaluating the health status of a pumping well according to an embodiment of the present application.

[0022] Figure 2 It is a schematic diagram of the analysis results of the principal component analysis process in the method for evaluating the health status of a pumping well according to an embodiment of the present application.

[0023] Figure 3 It is a schematic diagram of a mathematical model of neurons in the pump efficiency correlation degree analysis process in the method for evaluating the health status of a pumping well in an embodiment of the present application.

[0024] Figure 4 It is a schematic diagram of the BP network structure of the pump efficiency correlation degree analysis process in the method for evaluating the health status of a pumping well in an embodiment of the present application.

[0025] Figure 5 It is a schematic diagram of the analysis results of the pump efficiency correlation degree analysis process in the method for evaluating the health status of a pumping well in an embodiment of the present application.

[0026] Figure 6 It is a module block diagram of a system for evaluating the health status of a pumping well according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] The following will describe the implementation methods of the present invention in detail with reference to the accompanying drawings and embodiments, so that the implementation process of how the present invention applies technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly. It should be noted that as long as there is no conflict, the various embodiments of the present invention and the various features in the embodiments can be combined with each other, and the technical solutions formed are all within the protection scope of the present invention.

[0028] In addition, the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions. Also, although a logical sequence is shown in the flowchart, in some cases, the steps shown or described can be performed in a sequence different from that here.

[0029] With the continuous deepening of oilfield exploitation and the continuous improvement of the requirements for economic efficiency of exploitation, my country's oilfield exploitation has gradually developed from extensive exploitation to refined exploitation. Moreover, with the continuous deepening of exploitation, the difficulty of oilfield exploitation has continued to increase, and the number of low-permeability and low-yield wells has continued to increase, and the economic efficiency of oilfield exploitation has become particularly significant.

[0030] The main exploitation area of ​​Tahe Oilfield contains the Ordovician Yijianfang Formation fracture-cave oil reservoirs, which have the characteristics of one ultra and five highs, namely, ultra-deep, high temperature, high pressure, high hydrogen sulfide content, high mineralization and high viscosity. The main block, Tuofutai area, also has the characteristics of poor liquid supply, high asphaltene content and high gas content.

[0031] Due to the particularity of the above-mentioned reservoir types, the working conditions of the pumping wells are complex and the abnormal risks are high. There are four main characteristics: ① The pump hanging is generally deep. Due to the lack of fluid supply, the rod pump deep pumping technology is widely promoted. At present, the average pump hanging has reached 2668m, the deepest pump hanging is 3599m, and there are 91 pumping wells with pump hanging exceeding 3000m; ② The working environment of the equipment is harsh. The pipes, rods, and pumps generally work in a fluid environment with high corrosion, high viscosity, and high impurities, and the risk of damage is high; ③ The gas-oil ratio of the thin oil wells is high. The original gas-oil ratio of the thin oil block is generally 100m3 / t~500m3 / t, and some wells are as high as 2000m3 / t, which greatly affects the pump efficiency and causes frequent gas locks in severe cases; ④ Some wells have a long production cycle, and the downhole pipes, rods, pumps and other equipment have been operating at high load for a long time, with high corrosion and aging, which further increases the risk.

[0032] In view of the above problems faced by the environment of the pumping wells, the existing technology lacks a health evaluation and monitoring system for the pumping wells suitable for the Tahe Oilfield area to conduct a comprehensive investigation and analysis of the pumping wells, so as to improve the lifting efficiency of the oilfield from the perspective of monitoring the operating status of the pumping wells.

[0033] Therefore, in order to solve the above technical problems, the present invention proposes a method and system for evaluating the health status of a pumping well. The method and system conduct a comprehensive investigation and analysis of the production parameters, various process effects, technical operation management and other aspects of the pumping well, and in combination with the operation efficiency, and / or output, and / or operation cost of the pumping well, 10 representative evaluation indicators are clarified for the outstanding problems of the current Tahe pumping well; in the actual operation process of the pumping well, the real-time operation data of the 10 representative evaluation indicators are classified and compared with the standard value range corresponding to each indicator, and the health of the indicator is used to measure the operation quality of the pumping well. In this way, the on-site staff can monitor the pumping well in real time according to the current operation quality of the pumping well, and adjust the operation parameters of the pumping well in combination with the actual production situation of the oil well, so that it returns to the healthy range, and then improve the pump inspection cycle and operation efficiency of the pumping well.

[0034] Figure 1 This is a step diagram of a method for evaluating the health status of a pumping well according to an embodiment of the present application. Figure 1 , the health status evaluation method of pumping wells in fracture-cavity reservoirs in the main exploitation area of ​​Tahe Oilfield is explained.

[0035] Step S110 determines various influencing factors related to the operation efficiency, and / or production, and / or operation cost of the pumping well to form initial evaluation indicators.

[0036] In step S110, with reference to information on the properties of oil products in different reservoirs, factors affecting optimization of actual production output in oil fields, factors affecting optimization of pumping unit operating efficiency, factors affecting minimization of pumping well operating costs, and factors affecting safe production of pumping wells (health assessment principles), a number of initial influencing factor parameters, i.e., initial evaluation indicators, are determined that can characterize the characteristics of maximizing the economic benefits of pumping wells.

[0037] Specifically, the present invention targets two different oil properties, thin oil and heavy oil, and combines the actual production situation of the oil field to select unit consumption, efficiency, benefit, quality, safety and other indicators that support the trinity of "output, cost, and safety", can reflect the management level of the oil production management area, and are operational and comparable as evaluation items. Based on the petroleum industry standard "SY / T6126-2017 Statistical Method for Production Indicators of Pumping Units, Electric Submersible Pumps and Screw Pump Oil Wells", and in combination with the relevant standards and specifications of the Tahe Oilfield, 25 main initial evaluation indicators that affect the production and operation of pumping wells are sorted out according to equipment indicators, economic and technical indicators and management indicators. Among them, equipment indicators (6 items): stroke and its utilization rate, stroke frequency and its utilization rate, torque and its utilization rate, pump balance rate, load utilization rate, alternating load economy (including maximum load and minimum load); technical indicators (8 items): pump efficiency, pump inspection cycle, maintenance-free period, system efficiency, power consumption per ton of liquid, power consumption per ton of oil, power consumption per ton of liquid per hundred meters, power consumption per ton of liquid per hundred meters; management indicators (11 items): well opening rate, oil production time rate, utilization rate, comprehensive utilization rate, lying well rate, average pump hanging, average dynamic liquid level, average submergence, maintenance frequency, back pressure compliance rate, safe production (time).

[0038] After the initial evaluation index is generated, the process proceeds to step S120. Step S120 performs correlation analysis, principal component analysis, and pump efficiency correlation degree analysis on various influencing parameters in the initial evaluation index obtained in step S110 according to the historical operating status of the plurality of pumping wells and the historical production data of the plurality of pumping wells, and selects representative evaluation indexes for evaluating the health status of the pumping wells from the initial evaluation indexes.

[0039] First, the principles for selecting health evaluation indicators for pumping wells are explained. When selecting representative evaluation indicators for pumping wells, it is necessary to combine the actual situation of the oil field and select unit consumption, efficiency, benefit, quality, safety and other indicators that support the trinity of "production, cost, and safety", can reflect the management level of the oil production area, and are operational and comparable as evaluation items. In addition, the following principles need to be followed: ① Comprehensiveness principle. The scope of digital evaluation design of oil wells is relatively wide and contains rich content. When evaluating, it is necessary to make a comprehensive and effective analysis of the effect, structure, level of oil field production and the relationship and interaction between the above factors. It is not possible to generalize. Comprehensive consideration of the indicators involved in economic and technical aspects can obtain accurate and reliable evaluation results. ② The principle of combining qualitative and quantitative. The nature and category of evaluation indicators are different. Some can only be analyzed qualitatively. Therefore, the principle of combining qualitative and quantitative should be adopted. ③ Incompatibility principle. The relationship between evaluation indicators is complicated. There can be correlation between indicators. However, compatible indicators should be excluded to ensure that the evaluation results are objective and consistent with the facts. ④ Comparability principle. When selecting indicators, we should try to adopt domestically and internationally recognized indicator concepts, or be able to be converted into comparable factors after dimensionless processing such as calculation.

[0040] In the method for evaluating the health status of a pumping well described in the embodiment of the present invention, since the object to be evaluated is a pumping well in the Tahe Oilfield with complex reservoir characteristics, when selecting representative evaluation indicators suitable for such reservoir characteristics, it is necessary to screen based on a large amount of historical operating status data and production data of multiple pumping wells in the Tahe Oilfield with complex reservoir characteristics. Preferably, the historical operating status data of the pumping well can be the historical data corresponding to the parameters of each initial influencing factor in the initial evaluation indicator of the pumping well.

[0041] Specifically, in step S120, first, it is necessary to conduct correlation analysis, principal component analysis and pump efficiency correlation analysis on the historical operation data and historical production data of multiple pumping wells in the target oil reservoir area, respectively, for each initial influencing factor parameter in the initial evaluation index, so as to obtain the representative evaluation index based on this type of analysis for each analysis.

[0042] Furthermore, correlation analysis is the process of analyzing the correlation and correlation degree between any two initial influencing factor parameters within the initial evaluation index. It is necessary to find the correlation between random factor data sequences and discover the main contradictions, find the main characteristics and the main influencing factors. If the correlation degree between the historical data sequences of several influencing factor parameters is large, it means that these influencing factor parameters have a high degree of similarity. Therefore, a column of data should be selected from these similar influencing factor parameters as a representative evaluation index based on correlation analysis.

[0043] Further, in an embodiment of the present invention, it is necessary to use a grey correlation analysis method to analyze the correlation degree of each initial influencing factor parameter in the initial evaluation index, and based on the analysis results of the correlation degree of each factor between different initial influencing factors, select representative parameters that can characterize other influencing factors in the same type of parameters from similar influencing factors, so that the representative parameters for different types of parameters are used as representative evaluation indicators based on correlation analysis. Specifically, first, obtain the first type of parameters to be analyzed and the corresponding historical data in the initial evaluation parameters. Among them, the first type of parameters to be analyzed include but are not limited to: pump diameter, stroke, stroke frequency, pump depth, pump efficiency, submergence, oil production time rate, balance rate, back pressure, rated load, maximum load, minimum load, alternating load and load utilization. Then, according to the historical data of the first type of parameters to be analyzed, calculate the correlation coefficient and the corresponding correlation degree between any two parameters to be analyzed, based on which, obtain the representative evaluation index based on correlation analysis.

[0044] The following describes in detail the principle and method of correlation analysis in the embodiment of the present invention.

[0045] The grey correlation analysis method can process the historical data series corresponding to the initial influencing factors to be analyzed and studied in incomplete information, find their correlation between the random initial influencing factor series, discover the main contradictions, find the main characteristics and the main influencing factors. When selecting the health indicators of the pumping well, the historical data corresponding to the initial influencing factors are used as reference series, such as pump diameter, stroke, stroke, pump depth, submergence, pump efficiency, oil production time rate, balance rate, back pressure, rated load, maximum load, minimum load, load utilization rate and other initial influencing factors. The historical data corresponding to the data series are used to calculate the correlation coefficient between the data series. According to the correlation coefficient, the correlation degree of each initial influencing factor can be obtained, so as to judge whether there is correlation and the degree of correlation between the indicators. If the correlation degree between certain columns of data is large, it means that the data has a high degree of similarity. Therefore, a column of data should be selected from these similar data as the evaluation indicator.

[0046] Furthermore, in order to facilitate the removal of the correlation between the various initial influencing factors, in the embodiment of the present invention, the above-mentioned correlation analysis process also needs to first perform parameter category classification processing, and then perform correlation analysis. Specifically, (the first step) the above-mentioned first category of parameters to be analyzed are divided into oil pump parameters, load parameters and overall indicator parameters, and the corresponding historical data are determined; (the second step) according to the historical data corresponding to each influencing factor in the oil pump parameters, the pump diameter, stroke, number of strokes, sinking degree and pump depth are respectively subjected to grey correlation analysis, and (from these five parameters) the parameters that have the most correlation and the highest degree of correlation with other oil pump parameters (except themselves) are selected as representatives of the oil pump parameters; (the third step) according to the historical data corresponding to each influencing factor in the load parameters, the rated load, maximum load, and the maximum load are respectively subjected to grey correlation analysis. Grey correlation analysis is performed on load, minimum load, alternating load and load utilization rate. Based on the correlation between any two types of load parameters, the parameters that are correlated with all other load parameters (except themselves) are selected as representatives of load parameters (from these five parameters); (Step 4) For the overall index parameters in the first type of parameters to be analyzed, grey correlation analysis is performed on pump efficiency, submergence, oil production time rate, balance rate, back pressure, alternating load and load utilization rate, and the parameters that are uncorrelated with other overall index parameters (except themselves) are selected as representatives of overall index parameters (from these seven parameters).

[0047] Further, the theoretical basis of grey relational analysis method:

[0048] ① Determination of parent sequence and subsequence and preprocessing of raw data

[0049] The purpose of standardizing the data series of the initial influencing factors to be analyzed is to overcome unreasonable factors and transform the variables with different units and dimensions into variables under a certain standard scale. Normalizing the variables requires transforming various variables into standardized variables under the same scale.

[0050] ② Find the correlation

[0051] Take each influencing factor data sequence to be analyzed as the corresponding parent sequence and subsequence (for example, if the correlation coefficient of sequence A with sequences B, C, and D is calculated, then A is taken as the parent sequence and sequences B, C, and D are the subsequences), and calculate the absolute value Δ of the difference between the parent sequence and each subsequence at each time point 0i (t j ),Right now:

[0052] △ 0i (t j )=X 0 (t j )-X i (t j )

[0053] Among them, Δ 0i (t j ) represents the relationship between the parent sequence and each subsequence j The absolute value of the difference at the time, X 0 (t j ) represents the t in the parent sequence j The data at the moment, X i (t j ) represents the number of t in subsequence i j The data at the time, i represents the sequence number of the subsequence, and j represents the time sequence number in the sequence.

[0054] Then, find the correlation coefficient at each time point, that is:

[0055]

[0056] Among them, Δmax and Δmin represent the above Δ 0i (t j ) The maximum and minimum values ​​in the calculation results, ρ represents the resolution coefficient, L 0i (t j ) represents the relationship between the parent sequence and each subsequence j The correlation coefficient at the time. Usually, ρ is (0,1). The smaller the value, the higher the resolution of the correlation degree. However, the key to the grey correlation method is to sort out the correlation order, which has nothing to do with the size of the difference. That is, the order of the correlation value represents the size of the correlation degree. In the embodiment of the present invention, ρ=1.

[0057] Next, calculate the correlation of each reference sequence, that is, calculate the average value of its correlation coefficient:

[0058]

[0059] Among them, γ 0i It represents the sequence correlation coefficient between the parent sequence and each subsequence, and n represents the total number of data in each sequence.

[0060] ③Remove the associated order

[0061] In order to accurately evaluate the degree of association between each subsequence and the parent sequence, the degree of association needs to be arranged in a row from large to small, which is the association sequence. Each subsequence needs to compare its size relationship with the same parent sequence, and then clarify its superiority and inferiority relationship with the parent sequence.

[0062] Further, according to the process described in ① to ③ above, the application in the selection of health indicators of pumping wells based on correlation analysis is as follows:

[0063] The pumping well data given on site (the first type of parameters to be analyzed) include: pump diameter, stroke, stroke times, pump depth, submergence, pump efficiency, oil production time rate, balance rate, back pressure, rated load, maximum load, minimum load, load utilization rate, etc. The grey correlation analysis method is used to process the relevant data separately to determine the key evaluation indicators.

[0064] A. Selection of oil well pump parameters:

[0065] After grey correlation analysis of pump diameter, stroke, stroke frequency, pump depth, submergence, etc., the following results are obtained, see Table 7. Table 7 shows the results of grey correlation analysis of oil well pump parameters.

[0066] Table 7 Correlation analysis results of oil well pump parameters

[0067] / stroke Rush Pump depth Submergence Pump diameter Not relevant Significant correlation Significant correlation Significant correlation stroke / Significant correlation Significant correlation Significant correlation Rush Significant correlation / Significant correlation Significant correlation Pump depth Significant correlation Significant correlation / Significant correlation

[0068] According to the above results, it is believed that pump diameter is significantly correlated with stroke frequency, pump depth, and submergence. Therefore, only one of these four parameters can be selected to represent the remaining parameters. In actual production management, submergence is usually selected. In addition, stroke is not correlated with pump diameter, but is significantly correlated with stroke frequency, pump depth, and submergence. Stroke can be selected as an independent variable and therefore can be omitted as an evaluation indicator.

[0069] B. Selection of load parameters:

[0070] After grey correlation analysis of rated load, maximum load, minimum load, load utilization, alternating load, etc., the following results are obtained, see Table 8. Table 8 shows the results of grey correlation analysis for load parameters.

[0071] Table 8 Correlation analysis of load class parameters

[0072] / Load utilization Rated load Alternating load Maximum load Minimum load Load utilization / Significant correlation Not relevant Significant correlation Not relevant Rated load Significant correlation / Significant correlation Significant correlation Not relevant Alternating load Not relevant Not relevant / Significant correlation Significant correlation

[0073] As can be seen from the table above, the load utilization is significantly correlated with the rated load and the maximum load. Therefore, these three parameters have a high degree of similarity, and the load utilization can be selected as an evaluation indicator. In addition, the alternating load is not correlated with the rated load, and the alternating load is the difference between the maximum load and the minimum load. Therefore, the alternating load is selected as another evaluation indicator.

[0074] C. Determination of overall indicator parameters:

[0075] After completing the selection of pump parameters and load parameters, further combined with pump efficiency, production cost and continuous production time, we can make a basic evaluation of the overall production situation of the pumping well. At the same time, using the grey correlation analysis method, we conducted correlation analysis on pump efficiency, balance rate, oil production time rate, back pressure, load utilization rate, alternating load, etc., and obtained the following results, see Table 9. Table 9 shows the results of the grey correlation analysis of the overall index parameters.

[0076] Table 9 Correlation analysis of overall indicator parameters

[0077]

[0078]

[0079] The above results show that there is no correlation between parameters such as pump efficiency, submergence, alternating load, load utilization, balance rate, back pressure, etc., which means that there is basically no similarity between the columns of data of these indicators, and these indicators are independent of each other. Therefore, they can be used for indicator evaluation and have good representativeness. Finally, the analysis results of the above-mentioned oil pump parameters, the analysis results of the pump load parameters, and the analysis results of the overall indicator parameters are all used as representative evaluation indicators based on correlation analysis.

[0080] In this way, through correlation analysis, representative parameters that can characterize the change status of other unselected parameters in the set are selected from the parameter set with a higher degree of correlation, so that representative parameters in different parameter sets with a lower degree of correlation or no correlation are used as representative evaluation indicators based on correlation analysis.

[0081] Furthermore, principal component analysis is a process of analyzing the main influencing factors in the parameters of each initial influencing factor in the initial evaluation index. It is a statistical analysis method that requires selecting the most important variable from multiple factor variables, and has the characteristics of moisture conservation, energy conservation, decorrelation, and energy redistribution and concentration. In an embodiment of the present invention, it is necessary to use the principal component analysis method to analyze the principal component factors of each initial influencing factor parameter in the initial evaluation index respectively, so as to determine multiple principal component factors as representative evaluation indicators based on principal component analysis.

[0082] The principle and method of principal component analysis in the embodiment of the present invention are described in detail below.

[0083] Principal component analysis is an optimal orthogonal transformation method under the mean square error criterion. It has the advantages of moisture preservation, energy preservation, decorrelation, and energy redistribution and concentration. It is a multivariate statistical analysis method that selects the most important variable target principal component variable (based on the representative evaluation index under principal component analysis) from multiple initial influencing factor variables through linear transformation.

[0084] First, the basic principle of the principal component analysis method in the embodiment of the present invention is described.

[0085] Let X be an N-dimensional random variable: X = (X 1 ,X 2 ,X 3 ,...X N ) T , then the random variable X i The mathematical expectation of (i=1,2,...N) is:

[0086]

[0087] The covariance matrix of X is denoted as:

[0088]

[0089] In the formula, Let λ 1 >λ 2 >...>λ N ≥0,λ 1 , 2 , ...λ N Respectively represent ∑ x The eigenvalues ​​of , and accordingly, the eigenvectors corresponding to each eigenvalue are: 1 ,φ 2 ,φ 3 ...,φ n , then i and φ i (i=1, 2, ..., N) satisfies the following formula, where i represents the sequence number of the eigenvalue or eigenvector:

[0090] ∑ x φ i =λ i φ i

[0091] Construct the following orthogonal transformation matrix T:

[0092]

[0093] Let {φ i} satisfies the following relationship:

[0094] ∑ x φ i =λ i φ i

[0095] This means that {φ i} is a canonical orthogonal transformation vector group, and the following transformation is performed:

[0096] Y=T T X

[0097] In this way, it is called the principal component transformation of Y to X. From the above formula and {φ i} is a normalized orthogonal vector group, it is easy to obtain:

[0098] Y=TX

[0099] For the N feature data X that have been obtained i ={x 1 ,x 2 ,x 3 ,...x n}, first do principal component analysis.

[0100]

[0101] Get a new set of data {y 1 ,y 2 ,...,y n}, where each y i are all the original N data {x 1 ,x 2 ,x 3 ,...x N}, and then, in {y 1 ,y 2 ,...,y n}, select the first K data to form a subset {y 1 ,y 2 ,...,y k} to characterize the characteristics of the processed object. Although the number of features is reduced from N to K, the K features all contain the influence of the original N features. It can be seen that feature compression is achieved through principal component analysis, and finally, the main feature parameters are formed for nonlinear mapping modeling of formation pressure.

[0102] Therefore, the result of principal component analysis is to find the principal axis of the sample point concentration in the multidimensional sample space geometrically, and to find the eigenvector of the covariance matrix of p variables algebraically. In practical applications, as few new variable parameters as possible are constructed through the analysis of multidimensional variable parameters. These new variable parameters not only contain the information of the original multidimensional variable parameters, but more importantly, have a greater correlation with the formation pressure than the original parameters, which can achieve the purpose of accurate monitoring of formation pressure.

[0103] The selection of the number of principal components is completed as follows:

[0104] Principal component analysis is to try to recombine the original numerous indicators with certain correlation (such as p indicators) into a new set of independent comprehensive indicators to replace the original indicators. Usually, the mathematical treatment is to make a linear combination of the original p indicators as a new comprehensive indicator. The most classic approach is to express it with the variance Var(PC1) of PC1 (the first linear combination selected, that is, the first comprehensive indicator), that is, the larger the Var(PC1), the more information PC1 contains. Therefore, PC1 selected from all linear combinations should have the largest variance, so PC1 is called the first principal component. If the first principal component is not enough to represent the information of the original p indicators, then consider selecting PC2, that is, the second linear combination. In order to effectively reflect the original information, the existing information of PC1 does not need to appear in PC2. In mathematical language, it requires Cov(PC1, PC2) = 0, then PC2 is called the second principal component, and the third, fourth, ..., Pth principal component can be constructed by analogy.

[0105] If there are p target random variables, there are p principal components. Since the total variance value does not increase or decrease, the variance of the first few comprehensive variables such as PC1 and PC2 is larger, while the variance of the latter few comprehensive variables such as PCp and PCp-1 is smaller. Strictly speaking, only the first few comprehensive variables can be called main (important) components, and the latter few comprehensive variables are actually secondary (important) components. In practice, the first few are always retained and the latter are ignored. The number of principal components retained depends on the percentage of the cumulative variance of the retained part in the total variance (that is, the cumulative contribution rate), which indicates how much information the first few principal components summarize. In practice, a roughly specified percentage can determine how many principal components to retain; if one more principal component is retained, the cumulative variance will not increase much, so no more will be retained.

[0106] In this way, the principle of the above-mentioned principal component analysis technology is applied to the principal component analysis process for each initial influencing factor in the initial evaluation index in the embodiment of the present invention. Specifically, step S21 takes the N types of influencing factors in the initial evaluation index as N-dimensional variables, and determines the historical data of each variable (taking the historical data of each variable as the sample data of the variable); step S22 determines the eigenvector corresponding to each variable in the sample space based on the sample space composed of the historical data of each variable, and clarifies the number P of target principal components; step S23 generates P target principal component variables through the orthogonal transformation method, and uses them as representative evaluation indicators based on principal component analysis.

[0107] Furthermore, in the above step S21, the parameters of different influencing factors such as pump diameter, stroke, stroke frequency, pump depth, submergence, rated load, maximum load, minimum load, load utilization rate, alternating load, etc. are processed as N-dimensional original variables in turn for the target number of principal components and orthogonal transformation, so as to obtain the principal component analysis result consistent with the target number of principal components, refer to Figure 2 . Figure 2 Schematic diagram of the analysis results of the principal component analysis process in the method for evaluating the health status of a pumping well in an embodiment of the present application. Figure 2 As shown in the figure, the representative evaluation indicators based on principal component analysis are arranged in order of importance: load utilization, submergence, alternating load, balance rate, pump efficiency and back pressure.

[0108] In this way, through principal component analysis, parameters with high energy and low correlation among the initial evaluation indicators are used as representative evaluation indicators based on principal component analysis.

[0109] Furthermore, the pump efficiency correlation degree analysis is the process of analyzing the correlation degree (influence degree) between each initial influencing factor parameter in the initial evaluation index and the pump efficiency. The data sequence of each factor is used as the data basis for finding the correlation between the factor and the pump efficiency. In the embodiment of the present invention, it is necessary to use an artificial neural network method (for example: BP neural network algorithm) to analyze the correlation degree between each initial influencing factor parameter in the initial evaluation index and the pump efficiency.

[0110] The principle and method of pump efficiency correlation degree analysis in the embodiment of the present invention are described in detail below.

[0111] As a mathematical method for nonlinear processing, artificial neural network is mainly a mathematical modeling method for processing the uncertainty (nonlinear) relationship between parameters. Artificial neural network is a widely parallel interconnected network composed of many simple neurons. Its organization can simulate the interactive response of real-world objects in the biological nervous system. It is an information system that simulates the information processing mechanism of the human brain in physical mechanism. It is a highly nonlinear ultra-large-scale continuous-time dynamic system with global network effects, large-scale parallel distributed processing and associative learning capabilities.

[0112] Figure 3 The figure is a schematic diagram of the mathematical model of the neuron in the pump efficiency correlation degree analysis process in the method for evaluating the health status of the pumping well in the embodiment of the present application. The artificial neuron is an information processing unit with multiple inputs and a single output. It receives information through other neurons connected to it. Its processing of information is nonlinear. The neuron can be abstracted into a simple mathematical model (refer to Figure 3). In the actual application process, it can be simply understood as taking the historical data of multiple initial influencing factor parameters as information input, and taking the historical data of pump efficiency as a single output. Through continuous training and calculation of the mathematical network, a model that can be used to directly predict the output results based on the input is obtained. Such processing not only maintains the integrity of the original information, but also does not destroy the physical meaning of any parameter. However, the risk is that when the number of irrelevant parameters or negatively correlated parameters involved is large, the network will not converge in calculation or the output result will be wrong.

[0113] exist Figure 3 In, x 1 , x 2 , …, x n is the input of the neuron, that is, the information of the axons of the previous n neurons; θ i is the threshold of neuron i; w 1i , w 2i ,…,w ni are the i-th neuron pair x 1 , x 2 …,x n The weight connection, that is, the transmission efficiency of the synapse; y i is the output of the i-th neuron; f is the transfer function, which determines the input x to the i-th neuron 1 , x 2 , …, x n How the output is taken when the combined effect reaches the threshold.

[0114] Furthermore, in the embodiment of the present invention, a BP neural network algorithm is used for analysis. Figure 4 This is a schematic diagram of the BP network structure of the pump efficiency correlation degree analysis process in the method for evaluating the health status of a pumping well in an embodiment of the present application. The BP (Back-Propagation) neural network is a relatively mature artificial neural network. About 80% of neural networks use the BP network. It is a feedback-type fully connected multi-layer neural network with the advantages of simple structure and stable working state. It also has strong association, memory and promotion capabilities, and can approximate any nonlinear connection function with arbitrary accuracy. The BP network is a multi-layer forward network with one-way propagation. The BP network is a neural network with three or more layers, and its structure is as follows: Figure 4As shown in the figure, it includes input layer, middle layer (hidden layer) and output layer, and the upper and lower layers are fully connected. However, there is no connection between neurons in each layer. When the learning sample is provided to the network, the activation value of the neuron propagates from the input layer to the output layer through the middle layer, and the connection weights are corrected layer by layer in the output layer through the middle layer, and finally return to the input layer. This algorithm is called "error back propagation algorithm", that is, BP algorithm. As this error back propagation correction continues, the accuracy of the network's response to the input pattern continues to increase.

[0115] In this way, the principle of the above-mentioned pump efficiency correlation degree analysis technology is applied to the pump efficiency correlation degree analysis process for each initial influencing factor in the initial evaluation index in the embodiment of the present invention. Specifically, step S31 constructs an initial (untrained) BP neural network model based on the number of various initial influencing factors in the initial evaluation index, the factor type, and the data volume of the data sequence. Then, step S32 analyzes the correlation degree between various influencing parameters in the initial evaluation index and the pump efficiency by training the neural network model according to the historical data corresponding to the initial evaluation parameters of multiple pumping wells, and obtains the relationship coefficients between multiple initial influencing factor parameters and pump efficiency. That is to say, step S32 uses the historical data corresponding to each initial influencing factor in the initial evaluation parameters of multiple pumping wells as the training input data, and uses the historical data corresponding to the pump efficiency of multiple pumping wells as the training output data. Based on this, by training the initial BP neural network model, the correlation degree between various influencing parameters in the initial evaluation index and the pump efficiency is analyzed respectively, and the relationship coefficients between multiple initial influencing factor parameters and pump efficiency are obtained. In one embodiment, in step S32, the historical data corresponding to the initial influencing factors in the initial evaluation indicators, including pump diameter, stroke, number of strokes, pump depth, submergence, rated load, maximum load, minimum load, load utilization, alternating load, etc., are used as the training input data of the initial BP neural network model, and the historical data corresponding to the pump efficiency is used as the training output data of the BP neural network model to train the initial BP neural network model, thereby obtaining a trained BP neural network model (i.e., a pump-efficiency relationship analysis model).

[0116] Next, step S33 uses each weight in the pump efficiency relationship analysis model as the correlation (coefficient) between the initial influencing factor parameter and the pump efficiency, thereby entering step S34. Step S34 arranges the multiple initial influencing parameters and the pump efficiency relationship coefficients from high to low, and selects the input item initial influencing factor parameters corresponding to a preset number of relationship coefficients starting from the maximum relationship coefficient as representative evaluation indicators based on the pump efficiency correlation degree analysis.

[0117] Figure 5FIG. 1 is a schematic diagram of the analysis results of the pump efficiency correlation degree analysis process in the method for evaluating the health status of a pumping well in an embodiment of the present application. Figure 5 As shown in the figure, the representative evaluation indicators based on the pump efficiency correlation degree analysis include: submergence, pumping unit balance rate, load utilization rate, back pressure, alternating load and pump depth.

[0118] In this way, through the pump efficiency correlation degree analysis, the parameters with a higher correlation degree with the pump efficiency of the well pump among the initial influencing factor parameters are used as representative evaluation indicators based on the pump efficiency correlation degree analysis.

[0119] Finally, according to the representative evaluation index based on correlation analysis, the representative evaluation index based on principal component analysis, and the representative evaluation index based on pump efficiency correlation degree analysis, combined with the initial evaluation index and the operating cost of the pumping well, operating safety and other factors, the final 10 representative evaluation indicators for evaluating the health status of the pumping well are determined. Therefore, the present invention selects 10 representative evaluation indicators of the pumping well with low correlation between each other, which can characterize the parameter change state of other unselected initial influencing factors, and can be quantitatively and qualitatively evaluated. Among them, the representative evaluation indicators include: pump efficiency, oil production time rate, system efficiency, load utilization rate, balance rate, submergence, maintenance-free period, back pressure, alternating load and safe production (days), see Table 1. Table 1 shows that after analyzing the initial evaluation parameters based on multiple influencing factors and a large amount of historical data, the representative evaluation indicators that can highly represent the health evaluation indicators of the pumping well are obtained.

[0120] Table 1 represents the evaluation index types and index parameters

[0121]

[0122] After the representative evaluation index of the pumping well is selected, the process proceeds to step S130. Step S130 sets a corresponding standard value range for each representative evaluation index.

[0123] In step S130, when formulating the optimal interval range of the indicator for each representative evaluation indicator of the health of the pumping well, it is first necessary to collect the historical data of each representative evaluation indicator for multiple pumping wells in the Tahe Oilfield, and count the distribution of each representative evaluation indicator. Based on this, combined with industry and enterprise standards, and the economic operation and management of oil wells, a corresponding reasonable or economic standard value range is determined for each representative evaluation indicator of the health of the pumping well as the standard value range of the evaluation indicator.

[0124] Furthermore, in order to more accurately and finely perform health evaluation monitoring and management on the representative evaluation indicators of the pumping well, in the embodiment of the present invention, the standard value range of each representative evaluation indicator is constructed as a standard value range of different health levels. Among them, the health level includes: healthy state, sub-healthy state and unhealthy state. Table 2 shows the standard value range formulated for each selected representative evaluation indicator of the pumping well in the embodiment of the present invention, and further shows the standard value range of different health levels corresponding to each representative evaluation indicator.

[0125] Table 2 represents the standard value range of evaluation indicators

[0126]

[0127] Taking the balance rate of an oil pumping unit as an example, the process of determining the standard value range corresponding to the balance rate of the oil pumping unit and the standard value ranges corresponding to different health levels will be explained below.

[0128] 1) Data distribution of historical data of pumping unit balance rate:

[0129] (1) The balance rate data of 582 pumping wells in Tahe Oilfield are statistically analyzed. Among them, the balance rate of pumping wells in the range of 80% to 110% accounts for 75.26%; the balance rate of pumping wells in the range of 75% to 80% accounts for 3.61%; the balance rate of pumping wells in the range of 110% to 115% accounts for 6.36%; the balance rate of pumping wells in the range of <75% accounts for 5.15%; the balance rate of pumping wells in the range of >115% accounts for 9.62%. The balance rate of the pumping wells statistically analyzed is mainly distributed in the range of 80% to 110%. Table 3 is the distribution characteristics of the balance rate of pumping wells in Tahe Oilfield.

[0130] Table 3 Balance rate distribution of pumping units in Tahe Oilfield

[0131]

[0132]

[0133] (2) Refer to the economic operation management of the oil pump: The current balance method is the simplest and most commonly used method to measure whether the oil pump is balanced. According to the "SY / T 6374-2016 Economic Operation Specifications for Mechanical Oil Production Systems", the balance of the oil pump system should reach 80% to 110%, which means that the current balance rate has the lowest energy consumption within this range.

[0134] 2) Refer to the recommended evaluation values ​​of industry standards:

[0135] Based on the above historical data statistical analysis and the economic operation management of the pumping unit, combined with the actual situation of Tahe Oilfield and referring to the general practice of the industry, the final standard value range of the pumping unit balance rate is determined. Table 4 shows the final standard value range of the pumping unit balance rate in Tahe Oilfield.

[0136] Table 4 Standard value range of balance rate of pumping units in Tahe Oilfield

[0137]

[0138] After the evaluation index and the optimal interval range of the index are formulated, the process proceeds to step S140, and the health status of the pumping wells in the Tahe Oilfield is regularly evaluated during the actual operation of the pumping wells. Step S140 During the actual operation of the pumping wells, real-time data corresponding to the evaluation indexes are regularly obtained, and various types of real-time data are compared with the standard value range of the corresponding type, thereby recording the comparison results to monitor the operation status of the pumping wells.

[0139] In step S140, during the actual operation of the pumping wells in the Tahe Oilfield, the real-time data corresponding to the representative evaluation index of each pumping well is obtained regularly (for example, once a month) and recorded. Then, each real-time representative evaluation index data of the pumping well is compared with the above-mentioned standard value range of the corresponding type (refer to Table 2), and the health level of the current pumping well for each real-time representative evaluation index data is determined, so as to evaluate the health status of the current pumping well by using the health level of each directional index in the representative evaluation index data.

[0140] In addition, when the real-time data corresponding to one or more representative evaluation indicators are in a sub-healthy or unhealthy state level, the operation control parameters and production parameters of the corresponding pumping wells can be adjusted based on the regularly recorded real-time representative evaluation indicator parameters of the pumping wells and the health status evaluation levels corresponding to the parameters, combined with the geological development conditions of the oil field area where the current pumping wells are located, the operation status of the pumping wells and the actual production conditions, so as to return the real-time data of the various representative evaluation indicators of the current pumping wells to the health status level. In this way, the embodiment of the present invention can adjust the parameters for different development units and different working parameters of different pumping wells, combined with the actual production conditions of the oil wells, return to the healthy range, and improve the production timeliness and controllability of the pumping wells.

[0141] According to the "one well, one policy" management model, a "health evaluation form" (see Table 5 and Table 6) is established for each pumping well. The actual values ​​of 10 evaluation index parameters are obtained according to the operating conditions of the pumping well. The health level (health degree) is determined based on the size of the actual value, and the health degree of the index (see Table 2) is used to measure the operation of the pumping well. Compare the actual value parameters of the admission with the health standard (see Table 2), monitor the current, back pressure and other parameters of each well, and after analysis and evaluation, prevent problems in the oil reservoir production process in advance to ensure the safe, efficient and more economical operation of the pumping well, so that the pumping well can be effectively monitored. The oil production management area uses the current health evaluation form to "diagnose" the problem pumping wells in a timely manner, formulate rectification countermeasures, evaluate the implementation effect, and continuously improve the controllability of the operation of the pumping wells.

[0142] Table 5 is the front content of the daily health evaluation record sheet of the pumping wells in the Tahe Oilfield area, showing the actual value records of the representative evaluation indicators of the pumping wells and the health level evaluation. Table 6 is the back content of the daily health evaluation record sheet of the pumping wells in the Tahe Oilfield area, showing the basic profile of the pumping wells, production status, equipment process status, production parameters, special downhole technical conditions, special downhole technical processes and daily management points (for example: well washing, dosing, water mixing, operation, etc.).

[0143] Table 5 Health evaluation table of pumping wells in Tahe Oilfield (front)

[0144]

[0145]

[0146] Table 6 Health evaluation table of pumping wells in Tahe Oilfield (back)

[0147]

[0148]

[0149] In this way, the present invention constructs a health evaluation method for pumping wells, including: pumping well health evaluation principles, pumping well health evaluation method indicators, mathematical basis for selecting pumping well health indicators, optimal interval range of pumping well health evaluation indicators, and pumping well health evaluation framework (Table 5, Table 6).

[0150] On the other hand, based on the above method for evaluating the health status of a pumping well, the present invention also proposes a system for evaluating the health status of a pumping well. Figure 6 is a block diagram of a system for evaluating the health status of a pumping well according to an embodiment of the present application. Figure 6As shown, the system for evaluating the health status of a pumping well according to the present invention includes: an initial influencing factor determination module 61 , a representative index determination module 62 , a standard range determination module 63 and an evaluation module 64 .

[0151] Specifically, the initial influencing factor determination module 61 is implemented according to the method described in step S110 above, and is configured to determine various influencing factors related to the operating efficiency, output and operating cost of the pumping well, and form the initial evaluation index. The representative index determination module 62 is implemented according to the method described in step S120 above, and is configured to perform correlation analysis, principal component analysis and pump efficiency correlation degree analysis on various influencing parameters in the initial evaluation index according to the historical operating status and output data of multiple pumping wells, and select representative evaluation indicators for evaluating the health status of the pumping well from the initial evaluation index. The standard range determination module 63 is implemented according to the method described in step S130 above, and is configured to set a corresponding standard value range for each representative evaluation index (wherein, the standard value range setting result refers to Table 2 above). The evaluation module 64 is implemented according to the method described in step S140 above, and is configured to regularly obtain real-time data corresponding to the representative evaluation index when the pumping well is running, and compare various real-time data with the standard value range of the corresponding type, so as to monitor the operating status of the pumping well based on the comparison result.

[0152] Further, in an embodiment of the present invention, the above-mentioned representative evaluation indicators include: pump efficiency, oil production time rate, system efficiency, load utilization rate, balance rate, submergence, maintenance-free period, back pressure, alternating load and safe production days.

[0153] The present invention discloses a method and system for evaluating the health status of a pumping well. The method and system include: selecting a pumping well health evaluation system indicator according to an established pumping well management health evaluation principle; optimizing the health evaluation system indicator from aspects such as the correlation between the indicators, the importance of the indicators to the economic operation of the pumping well, the relationship between the indicators and the pump efficiency, the operation cost of the pumping well, and the operation safety of the pumping well; formulating a health threshold range for the preferred system indicator (representative evaluation indicator) in combination with the actual production situation of the oil well for different development units and different working parameters of different pumping wells; and evaluating the health status of the pumping well by regularly monitoring the actual data of the preferred system indicator during the production and operation of the pumping well. In addition, the present invention can also adjust the actual production situation of the pumping well in combination with the development unit and working parameters of the current pumping well for the actual data of the pumping well in a sub-healthy or unhealthy state, so that the above-mentioned sub-healthy or unhealthy state parameters return to the healthy range.

[0154] In this way, the present invention can utilize the health evaluation system of the pumping well to correctly handle the contradictions between output, cost and safety in the production process of the pumping well, improve the production timeliness and controllability of the pumping well, and apply the evaluation system to maximize the optimization of the operating parameters of the pumping well, optimize the output, minimize the operating costs, and maximize the economic benefits of the pumping well.

[0155] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any person familiar with the technology within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0156] It should be understood that the embodiments disclosed in the present invention are not limited to the specific structures, processing steps or materials disclosed herein, but should be extended to equivalent substitutions of these features understood by ordinary technicians in the relevant field. It should also be understood that the terms used herein are only used for the purpose of describing specific embodiments and are not meant to be limiting.

[0157] The "one embodiment" or "embodiment" mentioned in the specification means that the specific features, structures or characteristics described in conjunction with the embodiment are included in at least one embodiment of the present invention. Therefore, the phrases "one embodiment" or "embodiment" appearing in various places throughout the specification do not necessarily refer to the same embodiment.

[0158] Although the embodiments disclosed in the present invention are as above, the above contents are only embodiments adopted for facilitating the understanding of the present invention and are not intended to limit the present invention. Any technician in the technical field to which the present invention belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present invention, but the patent protection scope of the present invention shall still be subject to the scope defined in the attached claims.

Claims

1. A method for evaluating the health status of a pumping well, characterized in that: The method comprises: Step 1: determine various influencing factors related to the operating efficiency, output and operating cost of the pumping well to form initial evaluation indicators; Step 2: According to the historical operation status and production data of multiple pumping wells, correlation analysis, principal component analysis and pump efficiency correlation degree analysis are respectively performed on various influencing parameters in the initial evaluation indicators, so as to screen out representative evaluation indicators corresponding to corresponding analysis angles from the initial evaluation indicators, and then, according to the representative evaluation indicators based on correlation analysis, the representative evaluation indicators based on principal component analysis and the representative evaluation indicators based on pump efficiency correlation degree analysis, combined with the initial evaluation indicators and the pumping well operation cost factors and operation safety factors, determine the representative evaluation indicators used to evaluate the health status of the pumping well; Step 3: setting a corresponding standard value range for each representative evaluation indicator; Step 4: When the pumping well is running, regularly obtain the real-time data corresponding to the representative evaluation index, compare each type of the real-time data with the standard value range of the corresponding type, and monitor the operating status of the pumping well based on the comparison result, wherein, in step 2, it includes: The correlation analysis is carried out through the following steps: Obtaining the first type of parameters to be analyzed in the initial evaluation index and corresponding historical data, wherein the first type of parameters to be analyzed include pump diameter, stroke, stroke frequency, pump depth, pump efficiency, submergence, oil production time rate, balance rate, back pressure, rated load, maximum load, minimum load, alternating load and load utilization rate; Dividing the first type of parameters to be analyzed into oil well pump parameters, load parameters and overall index parameters, and determining corresponding historical data; According to the historical data of the oil well pump parameters, grey correlation analysis is performed on the pump diameter, stroke, stroke frequency, pump depth and submergence, and the parameters with the most correlation and the highest degree of correlation with other oil well pump parameters are selected as representatives of the oil well pump parameters; According to the historical data of the load parameters, grey correlation analysis is performed on the rated load, maximum load, minimum load, alternating load and load utilization rate respectively, and based on the correlation degree of any two types of load parameters, parameters that are correlated with other load parameters are selected as load parameter representatives; According to the historical data of the overall index parameters, grey correlation analysis is performed on pump efficiency, submergence, oil production time rate, balance rate, back pressure, alternating load and load utilization rate, and parameters that are unrelated to other overall index parameters are selected as representatives of the overall index parameters, wherein the overall index parameters are obtained by combining pump efficiency, production cost and continuous production time factors after completing the selection of representative parameters of oil well pump parameters and representative parameters of load parameters; Taking representative parameters of oil well pump parameters, representative parameters of load parameters and representative parameters of overall index parameters as representative evaluation indicators based on correlation analysis; The principal component analysis is carried out through the following steps: Taking the N types of influencing factors in the initial evaluation index as N-dimensional variables, and determining the historical data of each variable; According to the sample space formed by the historical data of each variable, the eigenvector corresponding to each variable is determined in the sample space, and the number P of the target principal components is determined; Generate P target principal component variables through orthogonal transformation method, and use them as the representative evaluation index based on principal component analysis; The pump efficiency correlation analysis is carried out through the following steps: According to the initial evaluation index, construct a preset neural network model; According to the historical data corresponding to the initial evaluation index of multiple pumping wells, the correlation between various influencing parameters in the initial evaluation index and the pump efficiency is analyzed by training the neural network model to obtain the relationship coefficients between multiple initial influencing parameters and the pump efficiency; The multiple initial influencing parameters and pump efficiency relationship coefficients are arranged from high to low, and initial influencing parameters corresponding to a preset number of relationship coefficients are selected starting from the maximum relationship coefficient as the representative evaluation indicators based on the pump efficiency correlation degree analysis.

2. The method according to claim 1, characterized in that The representative evaluation indicators include: pump efficiency, oil production time rate, system efficiency, load utilization rate, balance rate, submergence, maintenance-free period, back pressure, alternating load and safe production days.

3. The method according to claim 1 or 2, characterized in that: The standard value range is constructed as a standard value range for different health levels.

4. The method according to claim 3, characterized in that The method further comprises: When the actual data of the current pumping well is in a sub-healthy or unhealthy state, the operating parameters of the current pumping well are adjusted according to the actual operating status of the current pumping well and the geological conditions of the oil field area where the current pumping well is located, so that the real-time data corresponding to the representative evaluation indicators of the current pumping well returns to a healthy state.

5. A system for evaluating the health status of a pumping well, characterized in that: The system implements the method according to any one of claims 1 to 4, and the system comprises: An initial influencing factor determination module is configured to determine various influencing factors related to the operating efficiency, output and operating cost of the pumping well to form initial evaluation indicators; A representative index determination module is configured to perform correlation analysis, principal component analysis and pump efficiency correlation degree analysis on various influencing parameters in the initial evaluation indexes according to the historical operation status and production data of multiple pumping wells, and select representative evaluation indexes for evaluating the health status of the pumping wells from the initial evaluation indexes; A standard range determination module, configured to set a corresponding standard value range for each representative evaluation indicator; The evaluation module is configured to periodically obtain real-time data corresponding to the representative evaluation index when the pumping well is in operation, compare each type of real-time data with the standard value range of the corresponding type, and monitor the operating status of the pumping well based on the comparison result.

6. The system according to claim 5, characterized in that The representative evaluation indicators include: pump efficiency, oil production time rate, system efficiency, load utilization rate, balance rate, submergence, maintenance-free period, back pressure, alternating load and safe production days.

Images