Method and system for identifying steam channeling pathways in shallow heavy oil reservoirs

Abstract

The present application provides a kind of for shallow thick oil reservoir vapor channeling passage discrimination method and system, this method includes: step 1, establish observation data matrix, standardization is carried out to observation data matrix;Step 2, establish vapor channeling passage classification grade, to evaluation matrix normalization processing;Step 3, initial evaluation matrix is calculated with observation data matrix, and classification center is determined;Step 4, the Euclidean distance matrix of each observation well distance classification center is calculated, and evaluation matrix is revised using distance matrix;Step 5, using distance matrix and evaluation matrix to calculate objective function;Step 6, according to maximum evaluation coefficient, the grade of the well vapor channeling passage is determined, and the classification evaluation result of each observation well vapor channeling passage is given.The method and system for shallow thick oil reservoir vapor channeling passage discrimination provide important theoretical basis for improving vapor drive development efficiency, preventing steam from prematurely vapor channeling, and developing reasonable steam drive development scheme, and have important application value in heavy oil thermal recovery development field.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil and gas field development, and particularly relates to a method and system for discriminating steam channeling channels of a shallow and thin heavy oil reservoir. BACKGROUND

[0002] The shallow and thin heavy oil reservoir generally has the characteristics of shallow buried depth, thin oil layer thickness, high viscosity of crude oil, high porosity and permeability of the reservoir. The buried depth of the reservoir is generally 100-400m, the oil layer thickness is generally 5-10m, the viscosity of the crude oil is generally greater than 20000mPa·s, the porosity is generally greater than 30%, and the permeability is generally greater than 2000mD. The steam injection development of the reservoir has the following problems: 1. The oil layer pressure is low, resulting in low single well production; 2. Steam channeling occurs frequently, and the steam profile sweep range is limited; 3. The reservoir pore type changes, and the “hot wormhole” is easily formed; and 4. The reservoir heterogeneity has obvious control effect on the steam channeling channel. Therefore, the discrimination of the steam channeling channel is a technical problem in the development of the reservoir.

[0003] In the long-term steam injection development process of the heavy oil reservoir, due to the reservoir heterogeneity and unreasonable working system, a low-resistance seepage flow channel, i.e. a steam channeling channel, is formed in the injection-production direction. After the formation of the steam channeling channel, a large amount of injected steam circulates along the channel in a low-efficiency or even invalid manner, so that other parts of the reservoir are difficult to be swept, and the oil displacement efficiency is seriously reduced. At present, the identification methods of the steam channeling channel mainly include well logging interpretation, coring observation, tracer monitoring, pressure drop test, and reservoir engineering method. However, the above methods use observation information in a certain aspect, and cannot completely reflect the actual control factors of the steam channeling channel. Meanwhile, in the process of qualitative interpretation, there are a large number of subjective factors, resulting in different qualitative interpretation results for different people.

[0004] In the Chinese patent application with the application number CN201610958641.7, a quantitative description method of steam channeling channel in the process of steam injection in a heavy oil reservoir is disclosed, which includes the following steps: identifying the steam channeling degree of each production well in a steam injection well group, and determining the steam channeling type of each production well; and calculating the steam channeling volume in the steam injection well group according to the steam channeling type of each production well in the steam injection well group and different characteristic parameters of the production well. The quantitative description method of the steam channeling channel in the process of steam injection in the heavy oil reservoir defines and classifies the channeling degree of all production wells in the steam injection well group, and finally obtains the volume of the multi-directional channeling steam channeling channel in the injection-production well group. The method provides a quantitative and operable technical method and implementation steps to calculate the volume of the steam channeling channel, and can be widely applied to the research field of steam injection development of onshore and offshore heavy oil reservoirs.

[0005] In the Chinese patent application No. CN202011623829.9, a method for identifying interwell connectivity of steam flooding reservoir of heavy oil reservoir is disclosed, which belongs to the technical field of oil and gas field development. The method comprises: obtaining basic parameters and production dynamic data of an interwell block of the reservoir, wherein the basic parameters are parameter data related to geological information in the interwell block of the reservoir, and the production dynamic data are variable parameter data in the production process; and identifying the interwell connectivity of the reservoir according to the basic parameters and the production dynamic data by using an established interwell dynamic connectivity inversion model of steam flooding reservoir of heavy oil reservoir. The embodiment of the invention obtains the basic parameters and the production dynamic data of the target reservoir block, establishes the interwell dynamic connectivity inversion model of steam flooding reservoir of heavy oil reservoir, and obtains the interwell dynamic connectivity coefficient by solving the interwell dynamic connectivity inversion model of steam flooding reservoir of heavy oil reservoir, so as to quickly identify the interwell connectivity of the reservoir.

[0006] In the Chinese patent application No. CN202010645663.4, a method for determining the degree of steam channeling between steam stimulation wells is disclosed, wherein the method comprises calculating a steam channeling degree evaluation index by using dimensionless steam channeling time and dimensionless temperature increment, and evaluating the degree of steam channeling between steam stimulation wells by using the evaluation index. The invention quantifies the degree of steam channeling between steam stimulation wells by theoretical calculation, and provides a theoretical basis for determining a synchronous injection-production well group on site by comparing and analyzing the quantified indexes.

[0007] The above prior art has great differences from the present invention and cannot solve the technical problems we want to solve. Therefore, we have invented a new method and system for identifying steam channeling channels in shallow and thin heavy oil reservoirs. SUMMARY

[0008] The purpose of the present invention is to provide a method and system for identifying steam channeling channels in shallow and thin heavy oil reservoirs, which quantitatively classifies and evaluates the steam channeling channels by using geological static parameters collected from each observation well.

[0009] The purpose of the present invention can be achieved by the following technical measures: a method for identifying steam channeling channels in shallow and thin heavy oil reservoirs, which comprises:

[0010] Step 1: establishing an observation data matrix and normalizing the observation data matrix;

[0011] Step 2: determining the classification level of steam channeling channels and normalizing the evaluation matrix;

[0012] Step 3: calculating the initial evaluation matrix and the observation data matrix to determine the classification center;

[0013] Step 4, calculating the Euclidean distance matrix of each observation well from the center of the classification, and correcting the evaluation matrix by using the distance matrix;

[0014] Step 5, calculating the objective function by using the distance matrix and the evaluation matrix;

[0015] Step 6, determining the grade of the steam channeling channel of the well according to the maximum evaluation coefficient, and giving the classification evaluation result of the steam channeling channel of each observation well.

[0016] The object of the application can also be achieved by the following technical measures:

[0017] In step 1, a plurality of observation wells are selected, and a observation data matrix is established by taking permeability, sedimentary microfacies, oil layer thickness, heterogeneity and interlayer as characteristic parameters, and the observation data matrix is standardized by using Z-score standardization method.

[0018] Step 1 includes:

[0019] 1) Establishing an observation matrix X:

[0020]

[0021] Wherein, each row is the characteristic parameter value of the permeability, sedimentary microfacies, oil layer thickness, heterogeneity and interlayer of each well, and each column is the n observation values of a characteristic parameter, that is, X is an observation data matrix composed of the observation values of p characteristic variables of n observation wells (x1, x2,..., xn). n

[0022] 2) Standardizing the observation data matrix by using Z-score standardization method:

[0023]

[0024] Wherein, x′ j is the column vector of each characteristic parameter after standardization, x j is the column vector of each characteristic parameter before standardization; is the average value of each characteristic parameter, and σ j is the variance of each characteristic parameter;

[0025] After standardization, the influence of different characteristic parameters on the dimension is eliminated, and the observation data matrix X′ after Z-score standardization is obtained.

[0026] In step 2, the classification grade of the steam channeling channel is determined, that is, the number of steam channeling channel classifications is determined, an initial random evaluation matrix is established by using a random generation function, the evaluation matrix coefficients are uniformly distributed between [0, 1], and the evaluation matrix is normalized.

[0027] ​Step 2 includes:

[0028] 1) According to the actual production needs, the classification level of the gas channeling passage is determined, and then the classification number r of the gas channeling passage is determined, and the initial random evaluation matrix A is established by using the pseudo-random generating function of the computer:

[0029]

[0030] Wherein, A is the initial random evaluation matrix; a rn is a random number generated by the pseudo-random generating function, which represents the evaluation coefficient of the rth classification and the nth observation well.

[0031] 2) Normalization processing of the initial evaluation matrix: each evaluation coefficient in an arbitrary row is divided by the sum of the evaluation coefficients in the row, so that each evaluation coefficient is uniformly distributed between [0, 1], and it is ensured that:

[0032]

[0033] In step 3, according to the set weighted index factor of the evaluation matrix, the initial evaluation matrix and the observation data matrix are calculated to determine the classification center.

[0034] In step 3, the determined classification center is:

[0035]

[0036] Wherein, m is the weighted index factor of the evaluation matrix, c i is the ith clustering center, x' k is the standardized column vector of the kth feature parameter, a ik is the evaluation coefficient of the ith classification and the kth observation well; r is the classification number; n is the total number of observation wells.

[0037] Step 4 includes:

[0038] 1) The formula for calculating the Euclidean distance matrix of each observation well from the classification center is:

[0039]

[0040] Wherein, d ik is the Euclidean distance from the kth observation well to the ith classification center; p is the number of feature variables of each observation well; x′ kj is the standardized column vector of the kth observation well to the jth feature variable; c ij is the ith classification center, the jth feature variable; r is the classification number; n is the total number of observation wells.

[0041] 2) The formula for correcting the evaluation matrix by using the distance matrix is:

[0042]

[0043] wherein a ik is the evaluation coefficient of the ith class for the kth observation well; d ik is the Euclidean distance from the kth observation well to the ith class center; d jk is the Euclidean distance from the kth observation well to the jth class center.

[0044] In step 5, the objective function is calculated using the distance matrix and the evaluation matrix, and if the error value of the objective function is greater than the termination precision of the objective function, the iterative evaluation matrix and the class center are constantly updated until the value of the objective function reaches the minimum.

[0045] Step 5 includes:

[0046] 1) The formula for calculating the objective function using the distance matrix and the evaluation matrix is:

[0047]

[0048] wherein I is the objective function; A is the evaluation matrix; C is the class center matrix; a ik is the evaluation coefficient of the rth class for the kth observation well; d ik is the Euclidean distance from the kth observation well to the ith class center; m is the evaluation matrix weighting index factor; r is the number of classes; and n is the total number of observation wells.

[0049] 2) The iterative evaluation matrix and the class center are constantly updated until the error value of the objective function after any two iterations is less than the termination precision, or the number of iterations is greater than the maximum given number of times:

[0050] |I (l) -I (l-1) |<ε, or l>L max

[0051] wherein ε is the termination precision; l is the number of iterations; and L max is the maximum number of iterations;

[0052] At this time, the value of the objective function reaches the minimum, and the class center reaches the optimum.

[0053] In step 6, the maximum value of the evaluation matrix coefficient corresponding to each observation well is obtained, and the grade of the gas channeling channel of the well is determined according to the maximum evaluation coefficient, thereby giving the classification evaluation result of the gas channeling channel of each observation well.

[0054] In step 6, the grade of the gas channeling channel of the observation well is determined according to the value of the coefficient in the evaluation matrix A, and it is assumed that:

[0055]

[0056] wherein, is the maximum value of the kth well r evaluation matrix coefficient, according to the position of the maximum value, the observation well can be classified into the ith steam channel grade, the same work is completed for all wells, and then the steam channel classification evaluation result of each observation well is obtained.

[0057] The purpose of the present application can also be achieved by the following technical measures: a steam channel discrimination system for shallow and thin heavy oil reservoirs, the steam channel discrimination system for shallow and thin heavy oil reservoirs comprises:

[0058] An observation data establishment module establishes an observation data matrix and normalizes the observation data matrix;

[0059] An initial random evaluation matrix generation module determines steam channel classification grades and normalizes the evaluation matrix;

[0060] A classification center calculation module calculates the classification center by using the initial evaluation matrix and the observation data matrix;

[0061] An evaluation matrix correction module calculates the Euclidean distance matrix of each observation well from the classification center and corrects the evaluation matrix by using the distance matrix;

[0062] A target function calculation and optimization iteration module calculates the target function by using the distance matrix and the evaluation matrix;

[0063] A classification result discrimination module determines the grade of the steam channel of the well according to the maximum evaluation coefficient and gives the classification evaluation result of the steam channel of each observation well.

[0064] The steam channel discrimination method and system for shallow and thin heavy oil reservoirs in the present application introduce an iterative optimization method, which tries to comprehensively consider the static information related to the steam channel in the geological aspect and give an objective and quantitative classification evaluation of the steam channel development. By collecting the characteristic parameters of each observation well, the optimal classification center and evaluation matrix are found by using the optimization iteration method, and then the classification of the steam channel is determined by the size of the evaluation matrix coefficient, which ensures that the classification of the steam channel is more objective and accurate, which provides an important theoretical basis for improving the steam flooding development efficiency, preventing steam channeling from occurring too early, and formulating a reasonable steam flooding development plan, and has important application value in the field of heavy oil thermal recovery development. BRIEF DESCRIPTION OF DRAWINGS

[0065] Figure 1 The flowchart of a specific embodiment of the steam channel discrimination method for shallow and thin heavy oil reservoirs of the present application;

[0066] Figure 2 The structural diagram of a specific embodiment of the steam channel discrimination system for shallow and thin heavy oil reservoirs of the present application. DETAILED DESCRIPTION

[0067] It should be noted that the following detailed description is exemplary in nature and is intended to provide further description of the application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0068] It is also important to note that the terms "or" and "and" as used herein, unless otherwise indicated, are used to mean either phonetic "or", that is, any or all possible combinations of one or more items, or phonetic "and" that is, any combination of all of the items.

[0069] The method and system for identifying steam channeling in shallow and thin heavy oil reservoirs of the present application uses the idea of optimization iteration to constantly update the classification center and evaluation matrix, so that the objective function is minimized, and the classification center and evaluation matrix are both optimized. The classification category of each observation well steam channeling is determined by the maximum value of the evaluation matrix coefficient. This method can be applied to the effective prediction and objective evaluation of steam channeling in shallow and thin heavy oil reservoirs, and has important application value for the rational design of gas drive development plan for shallow and thin heavy oil reservoirs.

[0070] The following are several specific embodiments of the application

[0071] Embodiment 1

[0072] In a specific embodiment 1 of the application, as shown in Figure 1 , the flow chart of the method for identifying steam channeling in shallow and thin heavy oil reservoirs of the present application is shown. The method for identifying steam channeling in shallow and thin heavy oil reservoirs includes the following steps: Figure 1 S1: Select multiple observation wells, take permeability, sedimentary microfacies, oil layer thickness, heterogeneity, and interlayer as characteristic parameters, establish an observation data matrix, and use the Z-score standardization method to standardize the observation data matrix; specifically including:

[0073] 1) The established observation matrix X:

[0074]

[0075]

[0076] Wherein, each row is the permeability, sedimentary microfacies, oil layer thickness, heterogeneity, and interlayer characteristic parameter value of each well, and each column is n observation values of a characteristic parameter, that is, X is an observation data matrix composed of p characteristic variables of n observation wells (x1, x2,..., xn). n ) of the p characteristic variables of the n observation wells.​

[0077] 2) Using Z-score standardization method, the observation data matrix is standardized:

[0078]

[0079] wherein x' is the column vector of each characteristic parameter after standardization, x is the column vector of each characteristic parameter before standardization; j is the average value of each characteristic parameter, and σ is the variance of each characteristic parameter; j j

[0080] After standardization, the influence of different characteristic parameter dimensions is eliminated, and the observation data matrix X' after Z-score standardization is obtained.

[0081] S2: Establishing the classification level of the gas channeling passage, i.e. determining the classification number of the gas channeling passage, using a random generation function to establish an initial random evaluation matrix, making the evaluation matrix coefficients uniformly distributed between [0, 1], and normalizing the evaluation matrix; specifically including:

[0082] 1) According to the actual production needs, the classification level of the gas channeling passage is determined, and then the classification number r of the gas channeling passage is determined, and a pseudo-random generation function of a computer is used to establish an initial random evaluation matrix A:

[0083]

[0084] 2) Normalization processing of the initial evaluation matrix: dividing each evaluation coefficient of an arbitrary row by the sum of the evaluation coefficients of the row, so that each evaluation coefficient is uniformly distributed between [0, 1], and ensuring that:

[0085]

[0086] S3: According to the set weighted index factor of the evaluation matrix, the initial evaluation matrix and the observation data matrix are calculated to determine the classification center;

[0087]

[0088] wherein m is the weighted index factor of the evaluation matrix, and is generally set to 2.

[0089] S4: Calculating the Euclidean distance matrix of each observation well from the classification center, and using the distance matrix to correct the evaluation matrix; specifically including:

[0090] 1) The formula for calculating the Euclidean distance matrix of each observation well from the classification center is:

[0091] ​​​

[0092] 2) The formula for correcting the evaluation matrix using the distance matrix is:

[0093]

[0094] S5: Calculate the objective function using the distance matrix and the evaluation matrix. If the error value of the objective function is greater than the termination precision of the objective function, continuously update the iterative evaluation matrix and the clustering center until the objective function value reaches the minimum. Specifically, it includes:

[0095] 1) The formula for calculating the objective function using the distance matrix and the evaluation matrix is:

[0096]

[0097] where I is the objective function; A is the evaluation matrix; C is the clustering center matrix;

[0098] 2) Continuously update the iterative evaluation matrix and the clustering center until the error value of the objective function after any two iterations is less than the termination precision, or the number of iterations is greater than the maximum given number:

[0099] |I (l) -I (l-1) |<ε, or l>L max

[0100] where ε is the termination precision; l is the number of iterations; L max is the maximum number of iterations;

[0101] At this time, the objective function value reaches the minimum, and the clustering center reaches the optimum.

[0102] S6: Find the maximum value of the evaluation matrix coefficient corresponding to each observation well, and determine the grade of the gas channeling channel of the well according to the maximum evaluation coefficient, thereby giving the classification evaluation result of the gas channeling channel of each observation well.

[0103] According to the value of the coefficient in the evaluation matrix A, the grade of the gas channeling channel of the observation well is determined. Let:

[0104]

[0105] where is the maximum value of the rth evaluation matrix coefficient of the kth well. According to the position of the maximum value, the observation well can be classified as the ith grade of gas channeling channel. The same work is completed for all wells, and thus the classification evaluation result of the gas channeling channel of each observation well is obtained.

[0106] Example 2

[0107] In a specific embodiment 2 of the application, taking the actual data of a shallow thin heavy oil reservoir in a block of Shengli Oilfield as an example, the effectiveness and practicability of the steam channeling passage discrimination method for shallow thin heavy oil reservoirs of the embodiment of the application are illustrated. The method comprises the following steps:

[0108] (1) Taking 26 wells of 3 well groups in the research area as the research objects, the obtained permeability, sedimentary microfacies, oil layer thickness, heterogeneity and interlayer of the target layer of each observation well are taken as characteristic parameters, and an observation data matrix is established as shown in the following table:

[0109] Table 1 Characteristic parameter table of different observation wells

[0110]

[0111]

[0112] Wherein, 1 represents a channel sand body, and 0 represents a non-channel sand body; the Z-score standardization method is used to standardize the observation data matrix:

[0113]

[0114] Wherein, x′ j is a column vector of each characteristic parameter after standardization, x j is a column vector of each characteristic parameter before standardization; is the average value of each characteristic parameter, and σ j is the variance of each characteristic parameter; after standardization, the influence of the dimension of different characteristic parameters is eliminated, and the observation data matrix X′ after Z-score standardization is shown in the following table:

[0115] Table 2 Standardized characteristic parameter table

[0116]

[0117]

[0118] (2) The classification level of the steam channeling passage in the research area is determined to be three, that is, the classification number of the steam channeling passage is 3, the initial random evaluation matrix is established by using a random generation function, so that the evaluation matrix coefficients are uniformly distributed between 0 and 1, the evaluation matrix is normalized, and the initial evaluation matrix obtained is shown in the following table:

[0119] Table 3 Initial evaluation matrix data table

[0120]

[0121]

[0122] (3) According to the set evaluation matrix weighted index factor, the initial evaluation matrix and the observation data matrix are calculated to determine the classification center;

[0123]

[0124] wherein m is the evaluation matrix weighted index factor, generally set to 2.

[0125] (4) The Euclidean distance matrix of each observation well from the classification center is calculated, and the distance matrix is used to correct the evaluation matrix;

[0126] 1) The formula for calculating the Euclidean distance matrix of each observation well from the classification center is:

[0127]

[0128] 2) The formula for correcting the evaluation matrix using the distance matrix is:

[0129]

[0130] (5) The target function is calculated using the distance matrix and the evaluation matrix, and if the error value of the target function is greater than the target function termination precision, the iterative evaluation matrix and the clustering center are constantly updated until the target function value reaches the minimum.

[0131] 1) The formula for calculating the target function using the distance matrix and the evaluation matrix is:

[0132]

[0133] wherein I is the target function; A is the evaluation matrix; C is the classification center matrix;

[0134] 2) The iterative evaluation matrix and the clustering center are constantly updated until the target function error value after any 2 iterations is less than the termination precision, or the iteration number is greater than the maximum given number:

[0135] |I (l) -I (l-1) |<ε, or l>L max

[0136] wherein ε is the termination precision; l is the iteration number; L max is the maximum iteration number;

[0137] At this time, the target function value reaches the minimum, the classification center reaches the optimum, and the finally obtained classification center is shown in the following table:

[0138] Table 4 Clustering center characteristic parameter table

[0139] Class Feature parameter 1 Feature parameter 2 Feature parameter 3 Feature parameter 4 Feature parameter 5 1 0.406 -0.349 1.107 -0.566 0.615 2 0.769 1.015 -0.28 -0.672 0.847 3 -0.852 -0.635 -0.355 1.085 -1.051

[0140] (6) finding the maximum value of each observation well corresponding to the evaluation matrix coefficient, determining the grade of the steam channeling channel of the well according to the maximum evaluation coefficient, thereby giving the classification evaluation result of the steam channeling channel of each observation well.

[0141] According to the value of the coefficient in the evaluation matrix A, the grade of the observation well steam channeling channel is determined, and the maximum value of the rth evaluation matrix coefficient of the kth well is set as:

[0142]

[0143] wherein, The maximum value of the rth evaluation matrix coefficient of the kth well is set as: according to the position of the maximum value, the observation well can be classified into the ith steam channeling channel grade, and the same work is completed for all wells, and then the classification evaluation result of the steam channeling channel of each observation well is obtained, and the evaluation matrix and the classification result are shown in the following table:

[0144] Table 5 Classification table of different observation well steam channeling channels

[0145] Hash Classification 1 Classification 2 Classification 3 Final judgment class W1 0.796 0.164 0.039 1 W2 0.016 0.019 0.965 3 W3 0.864 0.098 0.038 1 W4 0.251 0.411 0.339 2 W5 0.821 0.132 0.048 1 W6 0.193 0.679 0.128 2 W7 0.563 0.363 0.074 1 W8 0.148 0.386 0.466 3 W9 0.736 0.165 0.099 1 W10 0.133 0.171 0.696 3 W11 0.053 0.082 0.864 3 W12 0.165 0.776 0.059 2 W13 0.085 0.09 0.825 3 W14 0.234 0.681 0.085 2 W15 0.263 0.264 0.473 3 W16 0.554 0.314 0.132 1 W17 0.599 0.341 0.06 1 W18 0.051 0.057 0.892 3 W19 0.788 0.15 0.062 1 W20 0.14 0.758 0.102 2 W21 0.023 0.033 0.945 3 W22 0.853 0.097 0.051 1 W23 0.595 0.273 0.132 1 W24 0.129 0.13 0.74 3 W25 0.14 0.177 0.683 3 W26 0.011 0.015 0.974 3

[0146] Through the analysis of the above table, it can be seen that the well of the first classification is mostly homogeneous reservoir, and the steam channeling channel is not developed; the well of the second classification is relatively heterogeneous reservoir, and the steam channeling channel is likely to be formed after a long time of steam flooding; the well of the third classification is a strongly heterogeneous reservoir, and the steam channeling channel is developed, which is easy to steam channel, and the water cut of the production well will rise sharply. The classification results are compared with the tracer monitoring results, and the coincidence rate reaches 83.7%, which fully proves that the steam channeling channel discrimination method for shallow and thin heavy oil reservoirs established by the present application is correct, effective and practical.

[0147] Embodiment 3

[0148] In a specific embodiment 3 of the application, as Figure 2 It is a structure schematic diagram of a system for predicting the productivity of low-saturation oil reservoirs after water-oil two-phase pressure provided by the third embodiment of the present application. Since the device embodiment is basically similar to the method embodiment, it is described more simply, and the related parts refer to the part of the method embodiment. The device embodiment described below is only illustrative.

[0149] The system for discriminating steam channeling channels of shallow and thin heavy oil reservoirs provided by the embodiment of the present application comprises: an observation data establishing module 10, an initial random evaluation matrix generating module 20, a classification center calculating module 30, an evaluation matrix correcting module 40, a target function calculating and optimization iteration module 50, and a classification result discriminating module 60.

[0150] The observation data establishment module 10 is used for selecting multiple observation wells, taking permeability, sedimentary microfacies, oil layer thickness, heterogeneity and interlayer as characteristic parameters, establishing an observation data matrix, and adopting a Z-s core standardization method to standardize the observation data matrix;

[0151] The initial random evaluation matrix generation module 20 is used for determining a classification level of the gas channeling channel, determining the number of classifications of the gas channeling channel, adopting a random generation function to establish an initial random evaluation matrix, making the evaluation matrix coefficients uniformly distributed between [0, 1], and performing normalization processing on the evaluation matrix.

[0152] The classification center calculation module 30 is used for calculating the classification center according to a set weighting index factor of the evaluation matrix by using the initial evaluation matrix and the observation data matrix.

[0153] The evaluation matrix correction module 40 is used for calculating an Euclidean distance matrix of each observation well from the classification center, and correcting the evaluation matrix by using the distance matrix.

[0154] The target function calculation and optimization iteration module 50 is used for calculating a target function by using the distance matrix and the evaluation matrix, and constantly updating the iterative evaluation matrix and the clustering center until the target function value reaches a minimum if the error value of the target function is greater than a target function termination precision.

[0155] The classification result discrimination module 60 is used for obtaining a maximum value of the evaluation matrix coefficients corresponding to each observation well, determining the level of the gas channeling channel of the well according to the maximum evaluation coefficient, and thus giving a classification evaluation result of the gas channeling channel of each observation well.

[0156] The shallow and thin heavy oil reservoir gas channeling channel discrimination system provided by the present application has the same beneficial effects as the above-mentioned method for discriminating the gas channeling channel of the shallow and thin heavy oil reservoir, and thus will not be described here.

[0157] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the above embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and the description of the present application.

[0158] In addition to the technical features described in the specification, all are known technologies of professional technicians.

Claims

1. A method for identifying steam channeling pathways in a shallow heavy oil reservoir, characterized in that, The method for identifying steam channeling paths in a shallow and thin heavy oil reservoir comprises the following steps: Step 1: establishing an observation data matrix and normalizing the observation data matrix; Step 2: determining the classification level of the steam channeling path, i.e., determining the number of classifications of the steam channeling path, establishing an initial random evaluation matrix by using a random generation function, and making the coefficients of the initial random evaluation matrix uniformly distributed between 0 and 1, and normalizing the initial random evaluation matrix; Step 3: calculating the initial random evaluation matrix and the observation data matrix to determine the classification center; Step 4: calculating the Euclidean distance matrix of each observation well from the classification center, and correcting the initial random evaluation matrix by using the distance matrix; Step 5: calculating the objective function by using the distance matrix and the corrected initial random evaluation matrix; Step 6: determining the level of the steam channeling path according to the maximum evaluation coefficient, and giving the classification evaluation result of the steam channeling path of each observation well.

2. The method according to claim 1, characterized in that, In step 1, a plurality of observation wells are selected, and a feature parameter including permeability, sedimentary microfacies, oil layer thickness, heterogeneity and interlayer is used to establish an observation data matrix, and the observation data matrix is normalized by using a Z-score normalization method.

3. The method according to claim 2, characterized in that, Step 1 comprises the following steps: 1) Establish observation matrix : ； wherein each row is a parameter value of permeability, sedimentary microfacies, oil layer thickness, heterogeneity, interlayer characteristics of each well, and each column is n observation values of a characteristic parameter, that is, X is an observation data matrix composed of observation values of p characteristic variables of n observation wells ​ 2) the observation data matrix is normalized by using a Z-score normalization method; ； wherein, is a column vector of each feature parameter after standardization, is a column vector of each feature parameter before standardization; is a mean value of each feature parameter, is a variance of each feature parameter; After standardization, the influence of different characteristic parameter dimensions is eliminated, and a Z-score standardized observation data matrix is obtained .

4. The method for discriminating steam channeling pathways in a shallow and thin heavy oil reservoir according to claim 1, characterized in that, Step 2 comprises the following steps: 1) According to the actual production needs, determine the classification level of the gas channeling passage, and then determine the classification number of the gas channeling passage , using the pseudo-random generation function of the computer to establish the initial random evaluation matrix : ； wherein: is an initial random assessment matrix; is a random number generated by a pseudo-random generating function, representing the th category, the th observation well assessment coefficient; 2) the normalization of the initial random evaluation matrix: each evaluation coefficient in an arbitrary row is divided by the sum of the evaluation coefficients in the row, so that each evaluation coefficient is uniformly distributed between 0 and 1, and the following conditions are ensured: 。 5. The method for discriminating steam channeling pathways in a shallow and thin heavy oil reservoir according to claim 1, characterized in that, In step 3, the classification center is determined by calculating the initial random evaluation matrix and the observation data matrix according to the set weighting index factor of the initial random evaluation matrix.

6. The method for discriminating steam channeling pathways in a shallow and thin heavy oil reservoir according to claim 5, characterized in that, In step 3, the determined classification center is as follows: ； in, The initial random evaluation matrix is ​​weighted by an exponential factor. For the first Cluster centers, For the standardized number The column vector of feature parameters, For the first Category 1, No. Evaluation coefficients for observation wells; Number of categories; This represents the total number of observation wells.

7. The method for discriminating steam channeling pathways in a shallow and thin heavy oil reservoir according to claim 1, characterized in that, Step 4 comprises the following steps: 1) the formula for calculating the Euclidean distance matrix of each observation well from the classification center is as follows: ； wherein, is the number of classes, is the Euclidean distance from the i-th observation well to the j-th cluster center; is the number of feature variables for each observation well; is the i-th feature variable of the j-th observation well after standardization; is the i-th feature variable of the j-th observation well after standardization; is the i-th feature variable of the j-th observation well after standardization; is the i-th cluster center, the j-th feature variable; is the i-th cluster center, the j-th feature variable; is the i-th cluster center, the j-th feature variable; is the i-th cluster center, the j-th feature variable; is the number of classes; is the total number of observation wells; 2) the formula for correcting the initial random evaluation matrix by using the distance matrix is as follows: ； in, For the first Category 1, No. Evaluation coefficients for observation wells; For the first Observation well to the first Euclidean distance between the classification centers; For the first Observation well to the first Euclidean distance of each classification center.

8. The method for discriminating steam channeling pathways in a shallow and thin heavy oil reservoir according to claim 1, characterized in that, In step 5, the objective function is calculated by using the distance matrix and the corrected initial random evaluation matrix, and if the error value of the objective function is greater than the termination precision of the objective function, the initial random evaluation matrix and the clustering center are updated iteratively until the value of the objective function reaches the minimum.

9. The method for discriminating steam channeling pathways in a shallow and thin heavy oil reservoir according to claim 8, characterized in that, Step 5 comprises the following steps: 1) the formula for calculating the objective function by using the distance matrix and the corrected initial random evaluation matrix is as follows: ； wherein, is the objective function; is the evaluation matrix; is the classification center matrix; is the first classification, the first evaluation coefficient of the observation well; is the first Euclidean distance from the observation well to the first classification center; is the initial random evaluation matrix weighting exponential factor; is the number of classifications; is the total number of observation wells; 2) the initial random evaluation matrix and the clustering center are updated iteratively until the error value of the objective function after any two iterations is less than the termination precision or the number of iterations is greater than the maximum given number: ； wherein: is the termination accuracy; is the number of iterations; is the maximum number of iterations; At this time, the value of the objective function reaches the minimum, and the classification center reaches the optimum.

10. The method for discriminating steam channeling pathways in a shallow and thin heavy oil reservoir according to claim 1, characterized in that, In step 6, the maximum value of the evaluation matrix coefficient corresponding to each observation well is obtained, the level of the steam channeling path is determined according to the maximum evaluation coefficient, and the classification evaluation result of the steam channeling path of each observation well is given.

11. The method for discriminating steam channeling pathways in a shallow and thin heavy oil reservoir according to claim 10, characterized in that, In step 6, the steam channeling passage grade of the observation well is determined according to the evaluation matrix The coefficient value condition, let: ； wherein, is the well maximum value of the evaluation matrix coefficient, according to the position of the maximum value, the observation well can be classified into the class of steam channeling channel grade, the same work is completed for all wells, and then the steam channeling channel classification evaluation result of each observation well is obtained.

12. A system for discriminating steam channeling pathways in a shallow heavy oil reservoir, characterized in that, The system for identifying steam channeling paths in a shallow and thin heavy oil reservoir comprises the following modules: An observation data establishing module for establishing an observation data matrix and normalizing the observation data matrix; An initial random evaluation matrix generating module for determining the classification level of the steam channeling path, i.e., determining the number of classifications of the steam channeling path, establishing an initial random evaluation matrix by using a random generation function, making the initial random evaluation matrix uniformly distributed between 0 and 1, and normalizing the evaluation matrix; The classification center calculation module calculates using the initial random evaluation matrix and the observation data matrix to determine the classification center; The evaluation matrix correction module calculates the Euclidean distance matrix of each observation well from the classification center, and corrects the initial random evaluation matrix using the distance matrix; The target function calculation and optimization iteration module calculates the target function using the distance matrix and the corrected initial random evaluation matrix; The classification result discrimination module determines the grade of the gas channeling channel according to the maximum evaluation coefficient, and gives the classification evaluation result of the gas channeling channel of each observation well.

Images