A method for screening low-yield and low-efficiency wells based on PCA and IF-THEN rule base

Abstract

The present invention provides a low-yield and low-efficiency well screening method based on PCA and IF-THEN rule base, collects and organizes the production data of the target oil field, uses PCA principal component analysis and other methods to pre-analyze the data, actively shut down wells, passively shut down wells, less than economic limit wells, greater than economic limit wells, and uses clustering and other methods to classify low-yield and low-efficiency wells. The method can quickly and accurately screen out low-yield and low-efficiency wells, not only can determine the cause type, but also can sort the priority, save labor costs, and determine the priority of implementing production increase measures.

Description

Technical Field

[0001] The invention relates to the technical field of oilfield measure guidance, and more specifically to a method for screening low-yield and low-efficiency wells based on PCA and IF-THEN rule base. Background Art

[0002] At present, there is no unified standard for defining low-yield and low-efficiency wells in domestic offshore oil fields, and experience is often used as the main method for screening low-yield and low-efficiency wells. Conventional screening methods for low-yield and low-efficiency wells are either based on the input-output ratio, or simply use two factors such as daily liquid production and water content to screen out low-yield and low-efficiency wells in oil fields. There are few factors to consider, and the screening process is limited by human experience. It is difficult to form a process-based and theoretical screening and discrimination process, and it is impossible to effectively establish the priority of governance, nor to know the main cause type. Summary of the invention

[0003] The present invention overcomes the shortcomings of the prior art. The existing definition standards for low-yield and low-efficiency wells are often based on empirical judgment, but do not have a theoretical screening and discrimination method. A low-yield and low-efficiency well screening method based on PCA and IF-THEN rule library is provided. The method can quickly and accurately screen out low-yield and low-efficiency wells, not only determine the cause type, but also sort the priority, save labor costs, and determine the priority of implementing production increase measures.

[0004] The purpose of the present invention is achieved through the following technical solutions.

[0005] A method for screening low-yield and low-efficiency wells based on PCA and IF-THEN rule base is carried out in the following steps:

[0006] Step 1, collect and organize the production data of the target oilfield: first collect the oilfield data, i.e. the first-hand data of oilfield development, including the maximum number of wells that can be treated this year, daily oil production, economic limit production, shutdown notes, operation notes, reservoir thickness, oil production intensity, water content, liquid production, water analysis, dynamic liquid level, physical property data, connectivity and other characteristic data that can represent the above parameters, but not limited to;

[0007] Step 2: Pre-analyze the data using methods such as PCA principal component analysis: If some low-yield and low-efficiency wells have been treated, you can choose to conduct a pre-analysis first. The collected data include daily oil production, oil production intensity, water content, liquid production, flow pressure, reservoir physical properties, formation pressure, and daily production data of a single well. The PCA principal component analysis is used to determine the priority weights for the subsequent adjustment of the priority order of logical nodes, as shown in Table 1.

[0008] Table 1 is a pre-analysis data table of the present invention

[0009]

[0010]

[0011] Step 3, active well shut-in judgment logic: Based on the daily oil production of the oil field, first judge the shut-in well. If it is an active well shut-in, the reason for low production and low efficiency is known according to the oil field management regulations, and directly go to step 7 for judgment;

[0012] Step 4, passive shut-in well judgment logic: According to the daily oil production of the oil field, the shut-in wells are judged. If they are passive shut-in wells, first determine whether there are on-site inspection problems. After finding out if it is a problem with underground equipment, determine whether the oil production is greater than the daily economic limit. If it is greater, they will not be screened or placed at level 1. If it is less, these wells will be processed in step 5. If it is due to ground equipment, they will not be screened or placed at level 1.

[0013] Step 5, judgment logic for wells less than the economic limit: The economic limit daily production is provided by the financial department. If the daily oil production is less than the economic limit production, it is necessary to make node judgments on oil production intensity, water content, flow pressure, physical properties, static pressure, water analysis, and operating condition analysis. The IF-THEN rule is a semi-fuzzy rule, which is expressed as a network structure composed of If node conditions then=>. The rules are as follows:

[0014] 1) IF (oil production intensity is low) and (water content is low) and (flow pressure is low) and (reservoir physical properties is high) and (formation pressure is high), then=>downhole reasons, analyze the working conditions;

[0015] 2) IF (oil recovery intensity is low) and (water content is low) and (flow pressure is low) and (reservoir properties is high) and (formation pressure is low), then => insufficient energy, analyze connectivity;

[0016] 3) IF (oil recovery intensity is low) and (water content is low) and (flowing pressure is low) and (reservoir properties is low) and (recovery degree is low) then=>maintain;

[0017] 4) IF (oil recovery intensity is low) and (water content is low) and (flowing pressure is low) and (reservoir physical properties is low) and (recovery degree is high) then => reservoir reasons, sidetracking, high investment and long cycle;

[0018] 5) IF (oil production intensity is low) and (water content is low) and (flow pressure is high) and (pump efficiency is low) then => underground reasons, pump replacement and inspection, low investment and short cycle;

[0019] 6) IF (oil production intensity is low) and (water content is low) and (flow pressure is high) and (pump efficiency is high) then => downhole reasons, replacement of tubing, high investment and long cycle;

[0020] 7) IF (oil production intensity is low) and (water content is high) and (liquid production is high) and (water analysis is injected water) then => production reasons, optimized water injection, low investment and short cycle;

[0021] 8) IF (oil production intensity is low) and (water content is high) and (liquid production is high) and (water analysis is formation water) then => reservoir reasons, water blocking, low investment and short cycle;

[0022] 9) IF (oil production intensity is high) and (recovery degree is high) then => reservoir reasons, sidetracking, high investment and long cycle;

[0023] 10) IF (oil recovery intensity is high) and (recovery degree is low) then => maintain;

[0024] According to the above 10 IF-THEN rules, the cause of the well is finally judged logically, where the order of the judgment nodes of each rule can be arranged in the order of weight in the analysis;

[0025] Step 6, judgment logic for wells greater than the economic limit: If the daily oil production is greater than the economic limit, first determine the oil production intensity and well number nodes, then classify different blocks, different water content, and different well types, and then determine the cause type and recommend measures for each category based on the 10 IF-THEN rules in step 5 above;

[0026] Step 7, use clustering and other methods to classify low-yield and low-efficiency wells: control the membership degree of different nodes through a set of membership functions or clustering algorithms, take the number of wells that can be treated in the oil field each year as the constraint variable, and use the data accuracy of the oil field and the number of wells that have been treated as the test set to fit the best parameters. According to the production situation of the oil field, the causes inferred by the system are used to establish a mapping relationship of the classification level, as shown in Table 2.

[0027] Table 2 is a description of the classification characteristics of low-yield and low-efficiency wells of the present invention.

[0028]

[0029] The beneficial effects of the present invention are as follows: the method of the present invention overcomes the limitations of strong spatiality and small data bodies of random data in oil fields, and achieves adaptation to various types of oil fields and production stages. At the same time, it draws on the experience of experts, conforms to the logic of the human brain, and complies with production reality. The results are controllable and highly scalable. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a tree-shaped judgment logic diagram of a long well stop in the present invention;

[0031] Figure 2 It is a tree-shaped judgment logic diagram of the output less than the economic limit of the present invention;

[0032] Figure 3 It is a tree-shaped judgment logic diagram of the output greater than the economic limit of the present invention; DETAILED DESCRIPTION

[0033] The technical solution of the present invention is further described below through specific embodiments.

[0034] Example

[0035] Screening of low-yield and low-efficiency wells in Q oil field:

[0036] The Q oilfield is a heavy oil reservoir with fluid viscosity between 200-300cp.

[0037] Step 1, first collect various data of the oil field, taking Well A2 as an example;

[0038] Step 2: Since the design idea is shown in Table 2, the weights of the influencing factors are the same as the design structure according to PCA analysis, so the order is not adjusted;

[0039] Step 3: This well is a producing well, so skip this step;

[0040] Step 4: This well is a producing well, so skip this step;

[0041] Step 5, the well production is 8m 3 / d, less than the economic limit output 10m 3 / d, so the well enters Figure 3 The framework shown, by comparison:

[0042] The oil production intensity of this well is only 0.13m 3 / (dm), the membership is judged as low, and the next node is entered; the water content of the well is 58%, and the clustering results of the two indicators of water content rise rate and cumulative oil production are used, and the membership is judged as low, and the next node is entered; the well is converted to a flowing pressure of 4.5MPa after the dynamic liquid surface, and the membership is judged as low, and the next node is entered; the reservoir permeability of the well is 600mD, and the membership is judged as high, and the next node is entered; the formation pressure of the well is 8.5MPa, and the membership is judged as low, and the formation is due to insufficient energy, and the next node is entered; the well is connected with the surrounding wells, and the membership is judged as high using the connectivity interpretation data, and the water injection can be strengthened and optimized;

[0043] Step 6, the well production is 8m 3 / d, less than the economic limit output 10m 3 / d, so skip this step;

[0044] Step 7: The domain value of each node is set by the field engineer according to the number of wells that can be arranged each year; the program is directly determined according to the description and data - enhanced water injection is high-yield, highly necessary, and easy to implement, and is implemented in accordance with the level I sequence.

[0045] The above results provide guidance for the identification of low-yield and low-efficiency wells and measures for the A2 well in Q Oilfield. After the well was re-injected, the daily oil production of the well group increased by 5m 3 / d, achieving good results.

[0046] The present invention is described above by way of example. It should be noted that, without departing from the core of the present invention, any simple deformation, modification or other equivalent replacement that can be made by those skilled in the art without inventive effort falls within the protection scope of the present invention.

Claims

1. A method for screening low-yield and low-efficiency wells based on PCA and IF-THEN rule base, characterized by: Follow the steps below: Step 1: Collect and organize production data of target oil fields; Step 2: Pre-analyze the data using the PCA principal component analysis method; Step 3, active well shut-in judgment logic; Among them, the technical route is: according to the analysis of the daily oil production of the oil field, first judge the shut-in wells. If the wells are shut down actively, the reasons for low production and low efficiency are known according to the oil field management regulations, and directly go to step 7 for judgment; Step 4, passive well shut-in judgment logic; The technical route is: according to the analysis of the daily oil production of the oil field, the shut-in wells are judged. If it is a passive shut-in well, first determine whether there is a problem operation in the field inspection. After finding out if it is a problem with the underground equipment, determine whether the oil production is greater than the daily economic limit. If it is greater, it will not be screened or placed in level 1. If it is less than, these wells will be put into step 5 for processing. If it is due to the ground equipment, it will not be screened or placed in level 1 directly. Step 5, judgment logic of wells less than the economic limit; The technical route is as follows: the economic limit daily production is provided by the financial department. If the daily oil production is less than the economic limit production, node judgments of oil production intensity, water content, flow pressure, reservoir physical properties, water analysis, and operating condition analysis are required. The IF-THEN rule is a semi-fuzzy rule, which is expressed as a network structure composed of If node conditions then=>. The rules are as follows: 1) IF (oil production intensity is low) and (water content is low) and (flow pressure is low) and (reservoir physical properties is high) and (formation pressure is high), then=>downhole reasons, analyze the working conditions; 2) IF (oil recovery intensity is low) and (water content is low) and (flow pressure is low) and (reservoir properties is high) and (formation pressure is low), then => insufficient energy, analyze connectivity; 3) IF (oil recovery intensity is low) and (water content is low) and (flowing pressure is low) and (reservoir properties is low) and (recovery degree is low) then=>maintain; 4) IF (oil recovery intensity is low) and (water content is low) and (flowing pressure is low) and (reservoir physical properties is low) and (recovery degree is high) then => reservoir reasons, sidetracking, high investment and long cycle; 5) IF (oil production intensity is low) and (water content is low) and (flow pressure is high) and (pump efficiency is low) then => underground reasons, pump replacement and inspection, low investment and short cycle; 6) IF (oil production intensity is low) and (water content is low) and (flow pressure is high) and (pump efficiency is high) then => downhole reasons, replacement of tubing, high investment and long cycle; 7) IF (oil production intensity is low) and (water content is high) and (liquid production is high) and (water analysis is injected water) then => production reasons, optimized water injection, low investment and short cycle; 8) IF (oil production intensity is low) and (water content is high) and (liquid production is high) and (water analysis is formation water) then => reservoir reasons, water blocking, low investment and short cycle; 9) IF (oil production intensity is high) and (recovery degree is high) then => reservoir reasons, sidetracking, high investment and long cycle; 10) IF (oil recovery intensity is high) and (recovery degree is low) then => maintain; According to the above 10 IF-THEN rules, the cause of the well is finally judged logically, where the order of the judgment nodes of each rule can be arranged in the order of weight in the analysis; Step 6, judgment logic of wells greater than the economic limit; The technical route is: if the daily oil production is greater than the economic limit, first determine the oil production intensity and well number nodes, then classify different blocks, different water content, and different well types, and then determine the cause type and recommend measures for each category based on the 10 IF-THEN rules in step 5 above; Step 7: Use clustering method to classify low-yield and low-efficiency wells.

2. The method for screening low-yield and low-efficiency wells based on PCA and IF-THEN rule base according to claim 1, characterized in that: The technical route of step 1 is to first collect oilfield data, that is, first-hand data on oilfield development, including the maximum number of wells that can be treated this year, daily oil production, economic limit production, shutdown notes, operation notes, reservoir thickness, oil production intensity, water content, liquid production, water analysis, dynamic liquid level, connectivity and other characteristic data including but not limited to those that can represent the above parameters.

3. The method for screening low-yield and low-efficiency wells based on PCA and IF-THEN rule base according to claim 1, characterized in that: The technical route of step 2 is: if some low-yield and low-efficiency wells have been managed, choose to conduct preliminary analysis first. The collected data include daily oil production, oil production intensity, water content, liquid production, flow pressure, reservoir properties, formation pressure, and daily production data of a single well. The priority weights are determined through PCA principal component analysis, which is used to adjust the priority order of logical nodes in the later stage.

4. The method for screening low-yield and low-efficiency wells based on PCA and IF-THEN rule base according to claim 1, characterized in that: The technical route of step 7 is: to control the membership degree of different nodes through a set of membership functions or clustering algorithms, to use the number of wells that can be treated in the oil field each year as the constraint variable, to fit the optimal parameters based on the data accuracy of the oil field and the number of wells that have been treated as the test set, and to establish a hierarchical mapping relationship for the causes inferred by the system according to the production situation of the oil field.

5. Application of the low-yield and low-efficiency well screening method based on PCA and IF-THEN rule base as described in any one of claims 1 to 4 in the screening of low-yield and low-efficiency wells.

Images