Method for analyzing influence factors of rod-pumped well in maintenance-free period of medium-high permeability full-loading oil reservoir

Abstract

The invention provides a method for analyzing influence factors in a maintenance-free period of a rod-pumped well in a medium-high permeability full-package oil reservoir, which comprises the following steps of: 1, carrying out standardization processing on related data of the rod-pumped well in an oil field to obtain a rod-pumped well data set; 2, analyzing the distribution condition of the repair-free period of the oil well in each block, and screening to obtain the data of the abnormal well in the repair-free period of each block; 3, searching specific parameters of the abnormal well sample library in the repair-free period; and 4, analyzing the specific parameters of the abnormal oil well in the maintenance-free period to obtain a frequent occurrence mode between the specific parameters and the maintenance-free period abnormity, so as to obtain factors influencing the maintenance-free period and an influence rule. According to the method for analyzing the influence factors in the maintenance-free period of the rod-pumped well in the medium-high permeability full-loaded oil reservoir, the specific parameters of the abnormal oil well in the maintenance-free period can be found out, the mode frequently occurring between the specific parameters and the maintenance-free period abnormity can be found out, and the influence factors and the influence rule of the maintenance-free period are clarified.

Description

Technical Field

[0001] The present invention relates to the technical fields of data mining algorithms and oil production technologies, and particularly to a method for analyzing influencing factors of the trouble-free period of pumping wells in medium-high permeability integral oil reservoirs. Background Art

[0002] The supporting level of oil production process technology and management level are directly related to the level of oilfield development. The trouble-free period of oil wells is a comprehensive reflection of the working level of oil production engineering. It is of great significance to do a good job in extending the trouble-free period of oil wells, which can reduce the maintenance workload, save material consumption and human resources; reduce oil layer pollution and provide conditions for the normal production capacity of oil wells; improve the normal production rate of oil wells and lay a foundation for realizing a virtuous cycle of development; and reduce the operation cost of the oil production system.

[0003] In the medium-high permeability integral oil reservoir entering the middle and late stages of development, the comprehensive water cut of oil wells continuously rises, and the phenomena of oil well scaling and formation sand production become increasingly serious. During the development process, the production of oil wells is affected by factors such as scaling and sand production, resulting in failures such as scale jamming, sand jamming, and breakage and disconnection, causing the oil wells to lie idle, reducing the trouble-free period of oil wells, and affecting the development efficiency of oil wells. By analyzing the frequently operated wells, finding out the reasons affecting the normal production of oil wells, and taking targeted solutions, it is of great significance to do a good job in extending the trouble-free period of oil wells.

[0004] In the patent "A Method for Evaluating the Construction Effect of Measure Operations in Downhole Workover Operations", by using the effect evaluation method, it is possible to timely understand and master the workover operation effect and its influencing factors, improve the quality of measure operations, and thus avoid unnecessary major repairs and complex operations as much as possible. However, only 6 evaluation factors are selected to construct the evaluation of measure operations that meet various blocks. It fails to explore which specific production parameters can affect the length of the trouble-free period.

[0005] In the paper "Analysis of the Causes and Countermeasures Affecting the Trouble-Free Period of Rod Pump Wells in the High Water-Cut Period", through the investigation of the operation reasons for 1045 well-times in the Shengli Oilfield and on the basis of seriously analyzing various operation reasons, the main reasons for the short trouble-free period of rod pump oil wells are found, and the application of special oil pumps and supporting process new technologies is proposed to extend the service life of the oil pump and improve the pump efficiency. However, only the statistical analysis of various operation reasons is carried out, and data is not selected from multiple perspectives of geology, fluid, production, and lifting system to explore the factors affecting the trouble-free period.

[0006] In the papers "Technical Measures for Extending the Trouble-Free Period of Oil Wells in Oilfields" and "Discussion on the Causes and Countermeasures Affecting the Pump Inspection Period of Oil Wells", the inductive method is adopted to list and analyze the common influencing factors, but the quantitative analysis of specific data is lacking, and it is impossible to obtain the influence degree of each index on the trouble-free period of oil wells, lacking a deeper understanding.

[0007] The above existing technologies are quite different from the present invention and cannot solve the technical problems we want to address. Therefore, we have invented a new method for analyzing the influencing factors of the trouble-free period of pumping wells in medium-high permeability whole-set oil reservoirs. Summary of the Invention

[0008] The objective of the present invention is to provide a method for analyzing the influencing factors of the trouble-free period of pumping wells in medium-high permeability whole-set oil reservoirs, which has high reliability and good accuracy, clarifies the internal influencing factors and influencing rules of the trouble-free period of oil wells, provides a decision-making recommendation for supporting measures for current oil wells, and is of great significance for reducing the operation workload and extending the trouble-free period.

[0009] The objective of the present invention can be achieved through the following technical measures: A method for analyzing the influencing factors of the trouble-free period of pumping wells in medium-high permeability whole-set oil reservoirs, which includes:

[0010] Step 1: Normalize the relevant data of the oilfield pumping wells to obtain a dataset of pumping wells;

[0011] Step 2: Analyze the distribution of the trouble-free periods of oil wells in each block, and screen out the data of abnormal wells with trouble-free periods in each block;

[0012] Step 3: Find the specific parameters of the abnormal well sample library with trouble-free periods;

[0013] Step 4: Analyze the specific parameters of the abnormal oil wells with trouble-free periods to obtain the frequently occurring patterns between the specific parameters and the abnormalities of the trouble-free periods, so as to obtain the influencing factors and influencing rules of the trouble-free period.

[0014] The objective of the present invention can also be achieved through the following technical measures:

[0015] Step 1 includes:

[0016] Step 1.1, perform data cleaning, including missing value processing, outlier processing, and digital processing of categorical data;

[0017] Step 1.2, perform feature selection, using the principal component analysis method to improve the sample quality and reduce the data processing dimension.

[0018] In Step 1.1, when performing missing value processing, use isnull() to determine whether it is null, and correct the null values by deletion or filling.

[0019] In Step 1.1, when performing outlier processing, use the 3σ principle to monitor data outliers, and check and correct the data.

[0020] In Step 1.1, when performing digital processing of categorical data, use One-hot encoding to digitize the categorical data in the index system.

[0021] In step 2, screening of wells with abnormal overhaul-free periods in each block is carried out, the distribution of the overhaul-free periods of oil wells in each block is analyzed, and an outlier analysis algorithm is used to screen out the data of wells with abnormal overhaul-free periods in each block.

[0022] Step 2 includes:

[0023] Step 2.1: Calculate the local outlier factor. The larger the factor value, the higher the degree of being an outlier; the smaller the factor value, the lower the degree of being an outlier.

[0024] Step 2.2: Perform elliptical model fitting to find an ellipse that is as close as possible to the overhaul-free period sample points.

[0025] Step 2.3: Use the isolation forest algorithm to characterize the degree of alienation. Split the data using a binary tree. The sample selection, feature selection, and split point selection are all randomized. If a certain sample is an outlier, it may be split out with fewer times.

[0026] Step 2.4: Integrate three different outlier value algorithms to identify the oil wells with abnormal overhaul-free periods and include them in the sample library of wells with abnormal overhaul-free periods.

[0027] In step 2.1, when calculating the local outlier factor, based on the density-based local outlier detection algorithm, according to the density of the overhaul-free periods of oil wells, first calculate the local reachability density of each overhaul-free period, and then further calculate the outlier factor of each overhaul-free period through the local reachability density. This outlier factor indicates the degree of being an outlier of an overhaul-free period. The larger the factor value, the higher the degree of being an outlier; the smaller the factor value, the lower the degree of being an outlier, so as to obtain the wells with abnormal overhaul-free periods in the block.

[0028] In step 2.2, when performing elliptical model fitting, for a set of sample points on a given plane, find an ellipse that is as close as possible to the overhaul-free period sample points. The points not within the elliptical model are the wells with abnormal overhaul-free periods.

[0029] In step 2.3, use the isolation forest algorithm to characterize the degree of alienation. Split the data using a binary tree. Randomly select features, and then randomly select the split value of the features to recursively generate partitions of the overhaul-free periods. Compared with the normal points in the dataset, fewer random partitions are required for the outliers to be isolated. Therefore, the outliers of the overhaul-free periods are the points with shorter paths in the tree. The path length is the number of edges passed from the root node.

[0030] In step 3, from multiple perspectives of geology, fluid, production, lifting system, and supporting technology, find the specific parameters of the sample library of wells with abnormal overhaul-free periods.

[0031] Step 3 includes:

[0032] Step 3.1: Standardize each index to eliminate the dimensional differences between different indexes so that they can be directly compared;

[0033] Step 3.2: Adopt the box plot data analysis method to compare the differences between the wells with abnormal repair-free periods and conventional oil wells in each block in terms of indexes such as geology, fluid, production, lifting system, and supporting technology;

[0034] Step 3.3: Quantitatively analyze and determine the influence degree of each index on the abnormal repair-free period, obtain the parameters with significant differences in the wells with abnormal repair-free periods, and use them as specific parameters.

[0035] In Step 4, analyze the specific parameters of the wells with abnormal repair-free periods, analyze the data of the sample library of the wells with abnormal repair-free periods, and analyze from evaluation indexes such as the standard deviation and frequency of the specific parameter indexes to obtain the frequently occurring patterns between the specific parameters and the abnormal repair-free period, so as to obtain the influencing factors and influence rules of the repair-free period.

[0036] The object of the present invention can also be achieved by the following technical measures: an analysis system for influencing factors of the repair-free period of pumping wells in medium-high permeability integral reservoirs, which uses the analysis method for influencing factors of the repair-free period of pumping wells in medium-high permeability integral reservoirs to analyze and clarify the internal influencing factors and influence rules of the repair-free period of oil wells.

[0037] The analysis method for influencing factors of the repair-free period of pumping wells in medium-high permeability integral reservoirs in the present invention involves the excavation of internal influencing factors of wells with abnormal repair-free periods, especially the analysis of influencing factors of the repair-free period of pumping wells in medium-high permeability integral reservoirs. The analysis method for influencing factors of the repair-free period of pumping wells in medium-high permeability integral reservoirs is based on the global data of the oilfield, divides the abnormal wells according to the number of days of the repair-free period, comprehensively uses outlier analysis algorithms such as local outlier factor, elliptical model fitting, and isolation forest algorithm, classifies by block, and finds the oil wells with abnormally low or high repair-free periods in different blocks. For the found oil wells with abnormally low and high repair-free periods, conduct parameter analysis of multiple indexes from geology, fluid, production, and lifting system, compare the differences in each parameter between the wells with abnormal repair-free periods and conventional oil wells, find the specific parameters of the wells with abnormal repair-free periods, find the frequently occurring patterns between the specific parameters and the abnormal repair-free period, and clarify the influencing factors and influence rules of the repair-free period. Description of the Drawings

[0038] Figure 1 It is a schematic diagram for the preparation of a sample library of wells with abnormal repair-free periods in a specific embodiment of the present invention;

[0039] Figure 2 It is a schematic diagram for summarizing the influence rules of abnormal repair-free periods in a specific embodiment of the present invention;

[0040] Figure 3 Schematic diagram of statistics data with extremely low repair-free period in a specific embodiment of the present invention;

[0041] Figure 4 Schematic diagram of standard deviation of each index of wells with extremely low repair-free period in a specific embodiment of the present invention;

[0042] Figure 5 Schematic diagram of statistics data with extremely high repair-free period in a specific embodiment of the present invention;

[0043] Figure 6 Schematic diagram of standard deviation of each index of wells with extremely high repair-free period in a specific embodiment of the present invention. Detailed implementation manners

[0044] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0045] It should be noted that the terms used herein are only for describing specific implementation manners and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular forms are also intended to include the plural forms. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they specify the presence of features, steps, operations, and / or combinations thereof.

[0046] The method for analyzing influencing factors of the repair-free period of pumping wells in medium-high permeability integrated oil reservoirs of the present invention includes:

[0047] Step 1: Normalize the relevant data of the pumping wells in a certain oilfield to obtain a dataset of pumping wells.

[0048] Step 1.1: Data cleaning:

[0049] Missing value processing: Use isnull() to judge whether it is null, and correct the null values by deletion or filling methods;

[0050] Outlier processing: Adopt data outlier monitoring including but not limited to the 3σ principle, and check and correct the data.

[0051] Digital processing of categorical data: Adopt digitalization of categorical data in the index system including but not limited to One-hot encoding.

[0052] Step 1.2: Feature selection: Adopt principal component analysis method including but not limited to improve the sample quality and reduce the data processing dimension.

[0053] Step 2: Screening of wells with abnormal overhaul-free periods in each block. Analyze the distribution of the overhaul-free periods of the oil wells in each block, and use the outlier analysis algorithm to screen out the data of wells with abnormal overhaul-free periods in each block.

[0054] Step 2.1: Local Outlier Factor. Based on the density-based local outlier detection algorithm, according to the density of the overhaul-free periods of the oil wells, first calculate the local reachability density of each overhaul-free period, and then further calculate an outlier factor for each overhaul-free period through the local reachability density. This outlier factor indicates the degree of outlier of an overhaul-free period. The larger the factor value, the higher the degree of outlier; the smaller the factor value, the lower the degree of outlier, so as to obtain the wells with abnormal overhaul-free periods in the block.

[0055] Step 2.2: Ellipse model fitting. For a given set of sample points on a plane, find an ellipse that is as close as possible to the sample points of the overhaul-free periods. The points not within the ellipse model are the wells with abnormal overhaul-free periods.

[0056] Step 2.3: Isolation Forest algorithm. By randomly selecting features and then randomly selecting the splitting values of the features, recursively generate partitions of the overhaul-free periods. Compared with the normal points in the dataset, the outliers to be isolated require fewer random partitions. Therefore, the outliers of the overhaul-free periods are the points with shorter paths in the tree, and the path length is the number of edges passed from the root node.

[0057] Step 2.4: Integrate the three different outlier algorithms to identify the oil wells with abnormal overhaul-free periods and include them in the sample library of wells with abnormal overhaul-free periods.

[0058] Step 3: From multiple perspectives such as geology, fluid, production, lifting system, and supporting technology, find the specific parameters of the sample library of wells with abnormal overhaul-free periods.

[0059] Step 3.1: Standardize each index to eliminate the dimensional differences between different indexes so that they can be directly compared.

[0060] Step 3.2: Use data analysis methods including but not limited to box plots to compare the differences between the wells with abnormal overhaul-free periods and the conventional oil wells in each block in terms of indexes such as geology, fluid, production, lifting system, and supporting technology.

[0061] Step 3.3: Quantitatively analyze and determine the influence degree of each index on the abnormal overhaul-free period, obtain the parameters with significant differences in the wells with abnormal overhaul-free periods, and use them as specific parameters.

[0062] Step 4: Analyze the specific parameters of the oil wells with abnormal overhaul-free periods, analyze the data of the sample library of oil wells with abnormal overhaul-free periods, and analyze from evaluation indexes such as the standard deviation and frequency of the specific parameter indexes to obtain the frequently occurring patterns between the specific parameters and the abnormal overhaul-free periods, so as to obtain the influencing factors and influencing rules of the overhaul-free period.

[0063] Step 4.1: Analyze the frequencies of the specific parameter indicators being abnormally high, normal, and abnormally low, and calculate the standard deviations in these three cases to determine the influencing factors of the overhaul-free period.

[0064] Step 4.2: Quantitatively analyze the abnormal conditions of the specific parameter indicators and their impact on the abnormally low or high overhaul-free period.

[0065] Step 4.3: Analyze the frequencies of the abnormal occurrences of the specific parameter indicators to determine the indicators that have the greatest impact on the abnormal overhaul-free period.

[0066] The following are several specific embodiments of applying the present invention

[0067] Embodiment 1

[0068] In a specific Embodiment 1 of applying the present invention, as Figure 1 、 2 shown in the flowchart of the method, the method for analyzing the influencing factors of the overhaul-free period of pumping wells in medium-high permeability integral reservoirs includes the following steps:

[0069] Step 1: Normalize the relevant data collected from the pumping wells in a certain oilfield to obtain a dataset of pumping wells.

[0070] Step 1.1: Data cleaning:

[0071] Missing value processing: Use isnull() to determine whether it is null, and correct the null values by deletion or filling methods;

[0072] Outlier processing: Adopt data outlier monitoring including but not limited to the 3σ principle, and check and correct the data.

[0073] Digital processing of categorical data: Adopt digitalization of categorical data in the index system including but not limited to One-hot encoding.

[0074] Step 1.2: Feature selection: Adopt principal component analysis method including but not limited to improve the sample quality and reduce the data processing dimension.

[0075] Step 2: Screen the abnormal wells in the overhaul-free period of each block, analyze the distribution of the overhaul-free periods of the oil wells in each block, calculate indicators such as the number of wells included in the statistics, the change range of the overhaul-free period, the average value of the overhaul-free period, and the standard deviation of the overhaul-free period in each block, and use the outlier analysis algorithm to screen and obtain the data of the abnormal wells in the overhaul-free period of each block.

[0076] Step 2.1: Local Outlier Factor, Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm. Based on the density of the trouble-free periods of oil wells, first calculate the local reachability density of each trouble-free period, and then further calculate the outlier factor of each trouble-free period through the local reachability density. This outlier factor indicates the degree of outlier of a trouble-free period. The larger the factor value, the higher the degree of outlier; the smaller the factor value, the lower the degree of outlier, thereby obtaining the abnormal wells in terms of trouble-free periods in the block.

[0077] Step 2.2: Ellipse model fitting. For a given set of sample points on a plane, find an ellipse that is as close as possible to the trouble-free period sample points. The points not within the ellipse model are the abnormal wells in terms of trouble-free periods.

[0078] Step 2.3: Isolation Forest algorithm. By randomly selecting features and then randomly selecting the splitting values of the features, recursively generate partitions of the trouble-free periods. Compared with the normal points in the dataset, the outliers to be isolated require fewer random partitions. Therefore, the abnormal values in terms of trouble-free periods are the points with shorter paths in the tree, and the path length is the number of edges passed from the root node.

[0079] Step 2.4: Integrate the three different outlier algorithms to identify the oil wells with abnormal trouble-free periods and include them in the sample library of abnormal wells in terms of trouble-free periods. The screening of abnormal wells using a combination of multiple methods enhances the comprehensiveness and reliability of the screening of abnormal wells in terms of trouble-free periods compared to the method of only judging whether it is abnormal based on the number of days of the trouble-free period.

[0080] Step 3: From multiple perspectives such as geology, fluid, production, lifting system, and supporting technology, find the specific parameters of the abnormal wells in terms of trouble-free periods.

[0081] Step 3.1: Standardize each index to eliminate the dimensional differences between different indexes so that they can be directly compared. Scale the values to near 0, and the distribution of the data becomes a standard normal distribution with a mean of 0 and a standard deviation of 1, reducing the huge differences between various evaluation indexes and ensuring the reliability of the results.

[0082] Step 3.2: Use data analysis methods including but not limited to box plots to compare the differences between the abnormal wells in terms of trouble-free periods and the conventional oil wells in each block in terms of indexes such as geology, fluid, production, lifting system, and supporting technology.

[0083] Taking the box plot as an example: Define the box plots drawn for each index. The data greater than the upper quartile is regarded as an abnormally high value, the data less than the lower quartile is regarded as an abnormally low value, and the data between the upper and lower quartiles is regarded as a normal value. Count the number of abnormal values and normal values of each index, compare the differences in each parameter between the abnormal oil wells in terms of trouble-free periods and the conventional oil wells, and find the specific parameters of the abnormal oil wells in terms of trouble-free periods.

[0084] For example, from the perspective of fluid, compare the differences in water cut indicators between abnormal wells and conventional wells, standardize the water cut, and draw box plots of abnormal wells and conventional oil wells in terms of water cut. It is found that the water cut of abnormal wells is generally higher than that of conventional oil wells. The water cut can be regarded as a specific parameter for abnormal oil wells during the trouble-free period, and it has a greater impact on the abnormality during the trouble-free period.

[0085] Step 3.3: Quantitatively analyze the influence degrees of indicators from multiple perspectives such as geology, fluid, production, lifting system, and supporting technology on the abnormality during the trouble-free period, and obtain the parameters with significant differences in abnormal wells during the trouble-free period.

[0086] Step 4: Analyze the specific parameters of abnormal oil wells during the trouble-free period, analyze the abnormal data of the trouble-free period of oil wells, and analyze from evaluation indicators such as the standard deviation and frequency of the specific parameter indicators to obtain the frequently occurring patterns between the specific parameters and the abnormality during the trouble-free period, so as to obtain the influencing factors and influencing rules of the trouble-free period.

[0087] Step 4.1: Analyze the frequencies of the specific parameter indicators showing extremely high, normal, and extremely low values, and calculate the standard deviations of the three situations to determine the influencing factors of the trouble-free period.

[0088] Step 4.2: Quantitatively analyze the abnormal conditions of the specific parameter indicators and their influence on the extremely low or extremely high values during the trouble-free period.

[0089] Step 4.3: Analyze the frequencies of the abnormal occurrences of the specific parameter indicators and determine the indicators with the greatest influence on the abnormality during the trouble-free period.

[0090] Embodiment 2

[0091] In a specific Embodiment 2 of applying the present invention, the method for analyzing the influencing factors of the trouble-free period of pumping wells includes the following steps:

[0092] Step 1: Normalize the relevant data of the pumping wells.

[0093] Step 2: Integrate three different outlier algorithms such as the local outlier factor, the elliptical model fitting isolation forest algorithm, etc. to identify the oil wells with abnormalities during the trouble-free period.

[0094] Taking Block 1 as an example, where 1 represents the normal value and -1 represents the abnormal value.

[0095] Table 1 Screening of Abnormal Wells in Block 1

[0096]

[0097]

[0098] Table 2 Abnormal Wells in Block 1

[0099]

[0100] The screening of abnormal wells by combining three methods is shown in Table 2. A total of 5 abnormal wells were screened in Block 1, enhancing the comprehensiveness and reliability of the screening of abnormal wells during the workover-free period.

[0101] Step 3: From multiple perspectives such as geology, fluid, production, lifting system, and supporting technology, find the specific parameters of the abnormal well sample library during the workover-free period. For example, select 22 indicators such as stroke frequency, casing pressure, annual oil production, annual water production, and annual gas production to analyze the relationship between the specific parameters and the abnormally low workover-free period, as Figure 3 shown.

[0102] Step 4: Analyze the specific parameters of the abnormal oil wells during the workover-free period to obtain the frequently occurring patterns between the specific parameters and the abnormalities during the workover-free period, so as to obtain the influencing factors and influencing laws during the workover-free period. Analyze the frequencies of the 22 indicators showing abnormally high, normal, and abnormally low values, and calculate the standard deviations of the three situations to determine the main control factors, as Figure 4 shown.

[0103] 1. The standard deviation is less than 6, and there are 5 indicators: maximum upward load, daily oil production capacity, number of days without moving the tubing string, system efficiency, and pump efficiency; the number of abnormal values and normal values of each indicator has little difference and has little impact on the workover-free period.

[0104] 2. 10 indicators such as stroke frequency, casing pressure, annual oil production, annual water production, annual gas production, daily oil production capacity, monthly oil production, displacement, pump diameter, and monthly water production are abnormally low and have an impact on the abnormally low workover-free period.

[0105] 3. 5 indicators such as minimum downward load, crude oil density, liquid level, crude oil viscosity, and pump depth are abnormally high and have an impact on the abnormally low workover-free period.

[0106] 4. 2 indicators such as stroke and water cut are not within the normal range and have an impact on the abnormally low workover-free period.

[0107] Table 3 Frequencies of each indicator for abnormally low workover-free period

[0108]

[0109]

[0110] 5. The pump diameter appears 92 times, which is the highest among the 10 abnormally low indicators statistically affecting the abnormally low workover-free period. The lower the pump diameter, the greater the impact on the abnormally low workover-free period.

[0111] 6. The pump depth appears 83 times, which is the highest among the 5 abnormally high indicators statistically affecting the abnormally low workover-free period. The higher the pump depth, the greater the impact on the abnormally low workover-free period.

[0112] Example 3

[0113] In a specific embodiment 3 of applying the present invention, the analysis method for influencing factors of the trouble-free period of pumping wells in medium-high permeability whole-reservoir includes:

[0114] Step 1: Normalize the relevant data of the pumping well.

[0115] Step 2: Combine three different outlier algorithms, such as the local outlier factor and the elliptical model fitting isolation forest algorithm, to identify the wells with abnormal trouble-free periods.

[0116] Taking Block 2 as an example, where 1 represents the normal value and -1 represents the abnormal value.

[0117] Table 4 Screening of Abnormal Wells in Block 2

[0118]

[0119] Table 5 Abnormal Wells in Block 2

[0120]

[0121]

[0122] Using the combination of three methods to screen abnormal wells, as shown in Table 5, a total of 18 abnormal wells are screened out in Block 2, enhancing the comprehensiveness and reliability of the screening of abnormal wells with abnormal trouble-free periods.

[0123] Step 3: From multiple perspectives such as geology, fluid, production, lifting system, and supporting technology, find the specific parameters of the abnormal well sample library for the trouble-free period. For example, select 22 indicators such as the pumping speed, casing pressure, annual oil production, annual water production, and annual gas production to analyze the relationship between the specific parameters and the abnormally high trouble-free period, as Figure 5 shown.

[0124] Step 4: Analyze the specific parameters of the abnormal wells with abnormal trouble-free periods to obtain the frequently occurring patterns between the specific parameters and the abnormal trouble-free period, so as to obtain the influencing factors and influencing laws of the trouble-free period. Analyze the frequencies of the 22 indicators showing abnormally high, normal, and abnormally low values, and calculate the standard deviations of the three situations to determine the main control factors, as Figure 6 shown.

[0125] 1. The standard deviation is less than 6, and there are 10 indicators: the minimum downward load, casing pressure, annual oil production, annual water production, crude oil density, system efficiency, daily liquid capacity, displacement, monthly water production, and pump depth; the number of abnormal values and normal values of each indicator has no significant difference and has little impact on the trouble-free period.

[0126] 2. Four indicators, namely the maximum upward load, stroke, number of days without moving the pipe string, and crude oil viscosity, are abnormally low and have an impact on the abnormally high trouble-free period.

[0127] 3. Six indicators, namely pumping frequency, daily oil production capacity, annual gas production, water cut, monthly oil production, and pump efficiency, are extremely high, which have an impact on the extremely high maintenance-free period.

[0128] 4. One indicator such as pump diameter not being within the normal range has an impact on the extremely high maintenance-free period.

[0129] Table 6 Frequency of occurrence of each indicator with an extremely high maintenance-free period

[0130] Abnormally low index Frequency Abnormally high index Frequency Maximum uplink load 48 Stroke frequency 61 Stroke 91 Daily oil production capacity 32 Number of days with no tubing movement 65 Annual gas production 21 Crude oil viscosity 66 Water cut 72 Monthly oil production 66 Pump efficiency 67

[0131] 5. The stroke appears 91 times, which is the highest among the four abnormally low indicators statistically affecting the extremely high maintenance-free period. The lower this indicator of the stroke, the greater the impact on the extremely high maintenance-free period.

[0132] 6. The water cut appears 72 times, which is the highest among the six abnormally high indicators statistically affecting the extremely high maintenance-free period. The higher this indicator of the water cut, the greater the impact on the extremely high maintenance-free period.

[0133] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

[0134] Except for the technical features described in the specification, they are all known technologies to those skilled in the art.

Claims

1. Analysis Method for Influencing Factors of Trouble - free Period of Beam - Pumping Wells in Medium - high Permeability Homogeneous Reservoirs Characterized in that the analysis method for influencing factors of trouble - free period of beam - pumping wells in medium - high permeability homogeneous reservoirs includes: Step 1: Normalize the relevant data of the beam - pumping wells in the oilfield to obtain a dataset of beam - pumping wells; Step 2: Analyze the distribution of the trouble - free periods of the oil wells in each block, and screen out the data of the wells with abnormal trouble - free periods in each block; Step 3: Search for the specific parameters of the sample library of wells with abnormal trouble - free periods; Step 4: Analyze the specific parameters of the wells with abnormal trouble - free periods to obtain the frequently occurring patterns between the specific parameters and the abnormal trouble - free periods, so as to obtain the influencing factors and influencing rules of the trouble - free period.

2. The analysis method for influencing factors of trouble - free period of beam - pumping wells in medium - high permeability homogeneous reservoirs according to claim 1 Characterized in that Step 1 includes: Step 1.1: Conduct data cleaning, including missing value processing, outlier processing, and digital processing of categorical data; Step 1.2: Conduct feature selection, using the principal component analysis method to improve the sample quality and reduce the data processing dimension.

3. The analysis method for influencing factors of trouble - free period of beam - pumping wells in medium - high permeability homogeneous reservoirs according to claim 2 Characterized in that In Step 1.1, when conducting missing value processing, use isnull() to judge whether it is empty, and correct the null values by deletion or filling methods.

4. The analysis method for influencing factors of trouble - free period of beam - pumping wells in medium - high permeability homogeneous reservoirs according to claim 2 Characterized in that In Step 1.1, when conducting outlier processing, adopt the 3σ principle to monitor data outliers, and check and correct the data.

5. The analysis method for influencing factors of trouble - free period of beam - pumping wells in medium - high permeability homogeneous reservoirs according to claim 2 Characterized in that In Step 1.1, when conducting digital processing of categorical data, adopt One - hot encoding to digitalize the categorical data in the index system.

6. The analysis method for influencing factors of trouble - free period of beam - pumping wells in medium - high permeability homogeneous reservoirs according to claim 1 Characterized in that In Step 2, when screening the wells with abnormal trouble - free periods in each block, analyze the distribution of the trouble - free periods of the oil wells in each block, and adopt the outlier analysis algorithm to screen out the data of the wells with abnormal trouble - free periods in each block.

7. The analysis method for influencing factors of trouble - free period of beam - pumping wells in medium - high permeability homogeneous reservoirs according to claim 6 Characterized in that Step 2 includes: Step 2.1: Calculate the local outlier factor. The larger the factor value, the higher the degree of outlier; the smaller the factor value, the lower the degree of outlier; Step 2.2: Conduct elliptical model fitting to find an ellipse that is as close as possible to the sample points of the trouble - free period; Step 2.3: Adopt the isolation forest algorithm to depict the degree of isolation. The algorithm uses a binary tree to split the data, and the sample selection, feature selection, and split point selection are all randomized; if a certain sample is an outlier, it may be split out with very few times; Step 2.4: Integrate 3 different outlier value algorithms to identify the oil wells with abnormal trouble - free periods and include them in the sample library of wells with abnormal trouble - free periods.

8. The method for analyzing the influencing factors of the trouble-free period of pumping wells in medium-high permeability integrated oil reservoirs according to claim 7, characterized in that, in step 2.1, when calculating the local outlier factor, based on the density-based local outlier detection algorithm, according to the density of the trouble-free periods of oil wells, first calculate the local reachability density of each trouble-free period, and then further calculate an outlier factor for each trouble-free period through the local reachability density. This outlier factor identifies the degree of outlier of a trouble-free period. The larger the factor value, the higher the degree of outlier; the smaller the factor value, the lower the degree of outlier, so as to obtain the abnormal wells of the trouble-free period in the block.

9. The method for analyzing the influencing factors of the trouble-free period of pumping wells in medium-high permeability integrated oil reservoirs according to claim 7, characterized in that, in step 2.2, when performing ellipse model fitting, for a set of sample points on a given plane, find an ellipse to make it as close as possible to the trouble-free period sample points. The points not within the ellipse model are the abnormal wells of the trouble-free period.

10. The method for analyzing the influencing factors of the trouble-free period of pumping wells in medium-high permeability integrated oil reservoirs according to claim 7, characterized in that, in step 2.3, use the isolation forest algorithm to characterize the degree of alienation. Use a binary tree to split the data. Randomly select features, and then randomly select the split values of the features to recursively generate partitions of the trouble-free period; compared with the normal points in the dataset, the abnormal values to be isolated require fewer random partitions. Therefore, the abnormal values of the trouble-free period are the points with shorter paths in the tree, and the path length is the number of edges passed from the root node.

11. The method for analyzing the influencing factors of the trouble-free period of pumping wells in medium-high permeability integrated oil reservoirs according to claim 1, characterized in that, in step 3, from multiple perspectives of geology, fluid, production, lifting system, and supporting technology, find the specific parameters of the abnormal well sample library of the trouble-free period.

12. The method for analyzing the influencing factors of the trouble-free period of pumping wells in medium-high permeability integrated oil reservoirs according to claim 11, characterized in that, step 3 includes: Step 3.1: Standardize each index to eliminate the dimensional differences between different indexes so that they can be directly compared; Step 3.2: Use the box plot data analysis method to compare the differences between the abnormal wells of the trouble-free period and the conventional oil wells in each block in terms of these indexes of geology, fluid, production, lifting system, and supporting technology; Step 3.3: Quantitatively analyze and determine the influence degree of each index on the abnormality of the trouble-free period, obtain the parameters with significant differences in the abnormal wells of the trouble-free period, and use them as specific parameters.

13. The method for analyzing the influencing factors of the trouble-free period of pumping wells in medium-high permeability integrated oil reservoirs according to claim 1, characterized in that, in step 4, analyze the specific parameters of the abnormal oil wells of the trouble-free period, analyze the data of the abnormal well sample library of the trouble-free period of the oil wells, and analyze from the evaluation indexes such as the standard deviation and frequency of the specific parameter indexes to obtain the frequently occurring patterns between the specific parameters and the abnormality of the trouble-free period, so as to obtain the influencing factors and influencing rules of the trouble-free period.

14. The system for analyzing the influencing factors of the trouble-free period of pumping wells in medium-high permeability integrated oil reservoirs, characterized in that, The analysis system for influencing factors of the trouble-free period of pumping wells in medium-high permeability homogeneous reservoirs uses the method for analyzing influencing factors of the trouble-free period of pumping wells in medium-high permeability homogeneous reservoirs described in any one of claims 1-13 to analyze and clarify the internal influencing factors and influencing laws of the trouble-free period of oil wells.

Images