A method and system for optimizing injection and production parameters based on connectivity analysis

Abstract

The present invention relates to an injection-production parameter optimization method and system based on connectivity analysis, comprising: obtaining water well stratified water injection data and oil well stratified production data of an oil field to be tested; processing the water well stratified water injection data and oil well stratified production data of the oil field to be tested and inputting them into a pre-established water injection volume-oil production volume relationship model to obtain a connectivity analysis result of the oil field to be tested; constructing a relationship between liquid production, water content and water injection volume of the oil field to be tested according to the water well stratified water injection data, oil well stratified production data and connectivity analysis result, and obtaining an injection-production parameter optimization scheme for the oil field to be tested. The present invention can be widely applied to the field of oil field development.

Description

Technical Field

[0001] The invention relates to the field of oilfield development, and in particular to an injection and production parameter optimization method and system based on connectivity analysis. Background Art

[0002] As oilfield development continues to deepen, the challenges faced in oilfield development are becoming greater and greater. In the actual daily production of oilfields, the injection-production connectivity is a difficult but very important issue to determine. The injection-production connectivity refers to the connectivity between water injection wells and oil production wells in water-driven oil reservoirs. The accurate identification of the injection-production connectivity has a certain guiding role in the production adjustment, oil stabilization and water control of the oilfield. There are many types of data involved in the injection-production connectivity analysis process. Methods such as tracers and interference well tests are costly and have poor timeliness. In addition, as the water drive time increases, the contradictions in the vertical and horizontal development of the reservoir increase, and manual analysis is difficult and inefficient.

[0003] At present, the methods for studying the connectivity between injection and production mainly include traditional analysis and inversion analysis. Traditional analysis methods include tracer testing, multi-well test analysis and geochemical methods. Such analysis methods often require additional construction work, are complex to operate, costly, and labor-intensive, and will affect the normal production and operation of the oil field during implementation. For the inversion analysis method, with the improvement of the degree of digitalization of oil fields, a large amount of data has been accumulated in the oil fields. These data can reflect the dynamic characteristics of the reservoir. Therefore, a large amount of data generated during the development of the oil field can be used to study the connectivity between injection and production. Such analysis methods do not require additional construction work, are simple to operate, and have low costs. They will not affect the normal production activities of the oil field. They mainly include Spearman correlation analysis model, gray correlation analysis model, multivariate linear regression model, CM model, IC-NS model and system analysis model. In the analysis methods of these injection-production connectivity situations, scholars have considered more and more factors and tried to establish models using more extensive data. However, due to the complexity of the injection-production system, some models require assumptions, some do not consider the time lag and attenuation of water, some have too low judgment accuracy, or require a large number of parameters to be solved. Each model has its shortcomings, which prevents it from being fully applied.

[0004] On the other hand, the traditional production and injection design method is to formulate several injection and production plans based on the actual geology and development conditions of the reservoir, and then compare and analyze the various plans through reservoir engineering methods or reservoir numerical simulation methods to select the best plan. This kind of traditional injection and production design method is still the main means of formulating injection and production plans. However, this method only compares and analyzes a limited number of plans, which are greatly affected by human subjectivity and cannot guarantee the optimal injection and production parameter plan. Summary of the invention

[0005] In view of the above problems, the purpose of the present invention is to provide an injection and production parameter optimization method and system based on connectivity analysis, which can obtain the optimal solution for injection and production parameters and further improve the efficiency of oil and gas development.

[0006] To achieve the above objectives, the present invention adopts the following technical solutions: In a first aspect, a method for optimizing injection and production parameters based on connectivity analysis is provided, comprising:

[0007] Obtain the water well stratified water injection data and oil well stratified production data of the oil field to be tested;

[0008] After data processing, the water well stratified water injection data and oil well stratified production data of the oil field to be tested are input into a pre-established water injection volume-oil production volume relationship model to obtain the connectivity analysis results of the oil field to be tested;

[0009] According to the water well stratified injection data, oil well stratified production data and connectivity analysis results of the oil field to be tested, the relationship between the liquid production, water content and injection volume of the oil field to be tested is constructed, and the injection and production parameter optimization plan of the oil field to be tested is obtained.

[0010] Furthermore, the process of constructing the water injection volume-oil production volume relationship model includes:

[0011] Conduct static injection-production connectivity analysis based on application block input data;

[0012] Based on the results of static injection-production connectivity analysis, a water injection volume-oil production relationship model was established using regression algorithm.

[0013] Furthermore, the static injection-production connectivity analysis based on the application block input data includes:

[0014] Based on the input data of the application block, with the oil well as the center and the well spacing as the constraint, the water injection wells that have a static corresponding relationship with the central oil well within the well spacing constraint range are searched, and a static injection-production corresponding relationship statistics table is obtained;

[0015] According to the static injection-production correspondence statistics table, determine the water injection wells and their corresponding layers that have a static correspondence with the central oil well;

[0016] The static injection-production correspondence statistical table was corrected to obtain the static injection-production connectivity analysis results.

[0017] Furthermore, based on the static injection-production connectivity analysis results, a regression algorithm is used to establish a water injection volume-oil production volume relationship model, including:

[0018] Process the water injection data and oil well production data in the static injection-production connectivity analysis results;

[0019] Select the regression algorithm to be used, set the hyperparameters of the regression algorithm, and establish a water injection volume-oil production relationship model;

[0020] The water injection data and oil production data after data processing are divided into training set, validation set and test set, and the established water injection volume-oil production volume relationship model is trained, verified and tested to obtain a trained water injection volume-oil production volume relationship model.

[0021] Furthermore, the data processing of the water injection volume data and the oil well production data in the static injection-production connectivity analysis results includes:

[0022] Select the water injection volume data and oil well production data during the period of stable injection-production relationship;

[0023] Eliminate production outliers in water injection data and oil well production data;

[0024] Remove water well outliers from water injection data and oil well production data;

[0025] Perform N-day average processing on the water injection data and the oil well production in the oil well production data;

[0026] Extract the oil well operation measures data from the water injection volume data and the oil well production data, and remove the production data for N days after the measures are removed.

[0027] Furthermore, the selected regression algorithm and setting of hyper parameters of the regression algorithm to establish a water injection volume-oil production volume relationship model include:

[0028] Select the regression algorithm to be used;

[0029] Set hyperparameters of the regression algorithm;

[0030] The global sensitivity analysis method is used to perform parameter sensitivity analysis on the hyperparameters of the regression algorithm and obtain the dynamic connectivity coefficient;

[0031] The selected regression algorithm is used to establish a water injection volume-oil production volume relationship model based on the dynamic connectivity coefficient, wherein the input of the water injection volume-oil production volume relationship model is the water injection volume injection data and the oil well production data, and the output is the connectivity analysis result.

[0032] Furthermore, the relationship between the production of liquid, water content and water injection volume of the oil field to be tested is constructed based on the water well stratified water injection data, oil well stratified production data and connectivity analysis results of the oil field to be tested, and the injection and production parameter optimization scheme of the oil field to be tested is obtained, including:

[0033] Using a regression algorithm, based on the obtained connectivity analysis results of the oil field to be tested, and according to the water well stratified water injection data and oil well stratified production data of the oil field to be tested, the relationship between the liquid production, water content and water injection volume of the oil field to be tested is constructed;

[0034] The Bayesian optimization algorithm is used to optimize the injection and production parameters according to the relationship between the liquid production, water content and injection volume of the oil field to be tested, and the injection and production parameter optimization scheme of the oil field to be tested is obtained.

[0035] In a second aspect, a system for optimizing injection and production parameters based on connectivity analysis is provided, comprising:

[0036] A data acquisition module, used to acquire the water well stratified water injection data and oil well stratified production data of the oil field to be tested;

[0037] The connectivity analysis result determination module is used to process the water well stratified water injection data and oil well stratified production data of the oil field to be tested and input them into a pre-established water injection volume-oil production volume relationship model to obtain the connectivity analysis result of the oil field to be tested;

[0038] The injection and production parameter optimization scheme determination module is used to construct the relationship between the liquid production, water content and water injection volume of the oil field to be tested based on the water well stratified water injection data, oil well stratified production data and connectivity analysis results of the oil field to be tested, and obtain the injection and production parameter optimization scheme of the oil field to be tested.

[0039] In a third aspect, a processing device is provided, comprising computer program instructions, wherein the computer program instructions, when executed by the processing device, are used to implement the steps corresponding to the above-mentioned method for optimizing injection and production parameters based on connectivity analysis.

[0040] In a fourth aspect, a computer-readable storage medium is provided, on which computer program instructions are stored, wherein the computer program instructions, when executed by a processor, are used to implement the steps corresponding to the above-mentioned method for optimizing injection and production parameters based on connectivity analysis.

[0041] The present invention adopts the above technical solution, which has the following advantages:

[0042] 1. The present invention establishes a water injection volume-oil production volume relationship model driven by the fusion of geological and injection-production data. The connectivity analysis results are consistent with the comparison results of the digital streamline model. The accuracy of the injection-production parameter prediction model is greater than 85%, and the prediction accuracy has reached the level of manual analysis.

[0043] 2. The present invention can be applied to injection-production correspondence analysis and daily injection adjustment work in oil field development. The application of the present invention can effectively improve work efficiency, liberate the labor of technical personnel, and enable them to devote more managers to in-depth analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Throughout the accompanying drawings, the same reference numerals are used to represent the same components. In the accompanying drawings:

[0045] Figure 1 It is a schematic diagram of a method flow provided by an embodiment of the present invention;

[0046] Figure 2 It is a schematic diagram of a static injection-production correspondence analysis idea provided by an embodiment of the present invention;

[0047] Figure 3 It is a schematic diagram of the basic flow of a recurrent neural network provided by an embodiment of the present invention;

[0048] Figure 4 It is a schematic diagram of the Bayesian search algorithm flow provided by an embodiment of the present invention;

[0049] Figure 5 is a schematic diagram comparing the model calculation results and the actual daily fluid production provided by an embodiment of the present invention;

[0050] Figure 6 It is a schematic diagram comparing the model calculation results provided by an embodiment of the present invention with the actual daily fluid production. DETAILED DESCRIPTION

[0051] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0052] It should be understood that the terms used herein are only for the purpose of describing specific example embodiments and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "include", "comprise", "contain", and "have" are inclusive, and therefore specify the existence of stated features, steps, operations, elements and / or parts, but do not exclude the existence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not interpreted as necessarily requiring them to be performed in the specific order described or illustrated, unless the execution order is clearly indicated. It should also be understood that additional or alternative steps may be used.

[0053] Although the terms first, second, third, etc. can be used in the text to describe multiple elements, components, regions, layers and / or sections, these elements, components, regions, layers and / or sections should not be limited by these terms. These terms can only be used to distinguish an element, component, region, layer or section from another region, layer or section. Unless the context clearly indicates, terms such as "first", "second" and other numerical terms do not imply order or sequence when used in the text. Therefore, the first element, component, region, layer or section discussed below can be referred to as the second element, component, region, layer or section without departing from the teaching of the example embodiments.

[0054] In recent years, with the rapid development of electronic computer technology, especially the progress of artificial intelligence technology represented by machine learning and deep learning, new solutions have been brought to the problems of artificial identification, prediction, screening and other problems that rely heavily on expert experience in traditional industrial production, and at the same time, the industrial informatization and intelligent transformation of traditional industrial enterprises have been further promoted. In the field of oil and gas exploration, many works have begun to combine big data and artificial intelligence means to assist and guide exploration, production and other links. The injection and production parameter optimization method and system based on connectivity analysis provided by the embodiment of the present invention comprehensively utilizes static data of reservoir geology and dynamic development data to establish a static connectivity analysis model. Starting from static connectivity, based on dynamic development data, using big data algorithms, a relationship model between water injection volume and liquid production volume is constructed, the model is solved and verified, and the potential relationship between injection and production data is deeply explored to realize the quantitative characterization of the injection and production connectivity relationship; based on the results of connectivity analysis, a big data regression algorithm is used to construct an injection and production parameter prediction model, and an optimization algorithm is used to realize automatic optimization of injection and production parameters, assisting further injection and production adjustments, improving water injection effects, and increasing recoverable reserves, so as to achieve the purpose of tapping the production potential of oil reservoirs.

[0055] Example 1

[0056] This embodiment provides an injection-production parameter optimization method based on connectivity analysis, comprising the following steps:

[0057] 1) If Figure 1 As shown in the figure, a static injection-production connectivity analysis is performed, specifically:

[0058] 1.1) Based on the input data of the application block, with the oil well as the center and the well spacing as the constraint, find the water injection wells that have a static corresponding relationship with the central oil well within the well spacing constraint range, and obtain the static injection-production correspondence statistics table.

[0059] Specifically, the application block input data includes dynamic and static data such as well location data, well deviation data, geological stratification data, sand body data, production dynamic data, and production well history of the application block.

[0060] Specifically, the static injection-production correspondence statistics table includes oil well numbers, water well numbers and corresponding strata.

[0061] 1.2) According to the static injection-production correspondence statistics table, determine the water injection wells and their corresponding strata that have a static correspondence with the central oil well.

[0062] 1.3) Using geological maps (structural maps, sub-layer plan maps, reservoir physical property contour maps, etc.), business experts calibrate the static injection-production correspondence statistical table to obtain the static injection-production connectivity analysis results.

[0063] Specifically, the results of the static injection-production connectivity analysis include the oil well number, water well number, corresponding strata, water injection volume data, and oil well production data.

[0064] 2) If Figure 2 As shown in the figure, based on the results of static injection-production connectivity analysis, a regression algorithm is used to establish a water injection volume-oil production volume relationship model, which is as follows:

[0065] 2.1) Data processing is performed on the water injection data and oil well production data in the static injection-production connectivity analysis results.

[0066] Specifically, data processing includes:

[0067] 2.1.1) The change of production layer of a single well will inevitably affect the connectivity between the oil wells / water injection wells related to it. Therefore, it is necessary to select the water injection volume injection data and oil production data of the time period when the injection-production relationship is stable, that is, there is no change in the production layer of the oil and water wells in the selected time period.

[0068] 2.1.2) Consider the impact of production time and eliminate abnormal production values ​​in water injection data and oil well production data caused by well closure, etc.

[0069] 2.1.3) Water injection outlier processing: Remove water well outliers in water injection volume data and oil well production data.

[0070] 2.1.4) Consider the impact of the injection-production cycle: perform N-day averaging on the water injection data and the oil well production data.

[0071] 2.1.5) Consider the impact of oil well measures on production: extract the oil well operation measures data from the water injection data and the oil well production data, and remove the production data for N days after the measures are taken.

[0072] 2.2) Select the regression algorithm to be used, set the hyperparameters of the regression algorithm, and establish the water injection volume-oil production relationship model.

[0073] 2.2.1) Based on actual needs, choose the regression algorithm to be used.

[0074] Specifically, the regression algorithm can be selected from support vector machine, neural network, random forest and recurrent neural network.

[0075] Specifically, the support vector machine algorithm finds the optimal regression hyperplane by minimizing the distance between the support vector and the hyperplane, and realizes nonlinear regression by combining the kernel function method. The training process includes: setting the optimization problem according to the definition and constructing the corresponding dual problem; using SVM (support vector machine) optimization algorithms such as SMO (sequential minimum optimization algorithm) to solve the optimization problem constructed in the previous step; determining the support vector based on the solution obtained in the second step, and further obtaining the regression hyperplane.

[0076] Specifically, multi-layer neural networks can theoretically fit any nonlinear relationship by stacking single-layer neurons and using linear combination + nonlinear activation. The training process includes: inputting training data into the network for prediction, and using the predicted value and the true value to calculate the loss function; calculating the partial derivative of the loss function with respect to the model parameters, and using the gradient descent method to update the parameters; iterating the second step until the loss function and accuracy of the model on the validation set converge.

[0077] Specifically, random forest is an integrated learning method that uses multiple weak models to average and combine into a strong model, in which decision trees are selected as weak models. The training process includes: based on the Bagging method, replacement sampling is used for the training data set, and a subset of the data is selected each time; split features and split points are selected according to criteria such as MSE (mean square error) / MAE (mean absolute error); iterative calculation is performed in the second step to obtain a binary tree for regression; a binary tree is trained for each data subset, and finally the trees are combined by averaging.

[0078] Specifically, a recurrent neural network is a neural network structure used to process time series data. When calculating the output, it uses both the current input value and the previous input history, thereby modeling the sequence data. The basic structure and calculation formula of the recurrent neural network are:

[0079]

[0080]

[0081] in, for The hidden state of the moment; for Input of time; for The hidden state of the moment; is the activation function; , , All are trainable hidden layer model parameters; for Output at the moment; , is the trainable output layer model parameter. The basic process of the recurrent neural network is as follows Figure 3 shown.

[0082] 2.2.2) Set the hyperparameters of the regression algorithm.

[0083] Specifically, the hyperparameters in the regression algorithm (such as learning rate, number of network layers, support vector machine kernel function) can be set according to actual conditions or determined using a parameter search algorithm.

[0084] Specifically, the parameter search algorithm iteratively evaluates the validation set accuracy of the model under certain hyperparameters, and selects the hyperparameters for the next iteration based on this. After multiple model trainings, the optimal solution is obtained. Commonly used parameter search algorithms include grid search algorithm, random search algorithm and Bayesian optimization algorithm.

[0085] Specifically, the grid search algorithm exhaustively enumerates all possible values ​​according to the numerical range of each hyperparameter. Assuming that each hyperparameter has The number of searches required is The random search algorithm randomly selects a set of hyperparameter configurations within the parameter range of the grid search each time until the preset number of searches is reached. The Bayesian optimization algorithm builds a probability distribution model for the optimal parameters, updates the probability model based on the verification accuracy of each search, and selects the search point for the next iteration until the preset number of iterations is reached.

[0086] 2.2.3) The global sensitivity analysis method (Sobol method) is used to perform parameter sensitivity analysis on the hyperparameters of the regression algorithm and obtain the dynamic connectivity coefficient, where the dynamic connectivity coefficient refers to the strength of the underground connectivity between oil wells and water wells.

[0087] Specifically, the Sobol algorithm uses the main sensitivity coefficient and the total sensitivity coefficient to identify the influence of the input parameters on the output results. The main sensitivity coefficient represents the sensitivity caused by a single parameter, and the total sensitivity coefficient represents the sensitivity caused by the coupling of a single parameter and other parameters. The main sensitivity coefficient and the total sensitivity coefficient are also called Sobol coefficients. The specific process of the algorithm is as follows: Assume that the model can be expressed as y=f(x1, x2, xn), 1 oil well corresponds to n water wells, representing n input variables (injection volume) in the well group; let each input variable change within the possible range of values, calculate the influence of the changes in these input variables on the model output value, and the influence The degree is called the connectivity coefficient of the input variable; the larger the connectivity coefficient, the greater the influence of the input variable (injection volume) on the model output. Considering the nonlinear characteristics of the established model, the variance-based Monte Carlo method Sobol method is used to perform sensitivity analysis of the model; the basic idea is: decompose the total variance of the model output into the sum of the variance of each input variable and the variance of the interaction of each input variable, and then grade the connectivity according to the contribution ratio of the input variable to the total output variance; obtain the influence and importance of the injection volume of the surrounding injection wells on the liquid production of the central oil production well, thereby quantifying the connectivity between the injection wells and the production wells, that is, the quantitative representation of the connectivity relationship.

[0088] 2.2.4) Using the selected regression algorithm, a water injection volume-oil production relationship model is established based on the dynamic connectivity coefficient.

[0089] 2.3) The water injection data and oil well production data in the static injection-production connectivity analysis results after data processing are divided into three parts: training set, validation set and test set. The established water injection volume-oil production relationship model is trained and parameter tuned through the training set and validation set. The model accuracy of the established water injection volume-oil production relationship model is evaluated through the test set to obtain the trained water injection volume-oil production relationship model.

[0090] Specifically, the input of the water injection volume-oil production volume relationship model is the injection data of the water injection well and the production data of the oil production well, and the output is the connectivity analysis result, including the oil well number, the water well number, the corresponding layer and the connectivity coefficient.

[0091] Specifically, the ratios of the training set, validation set, and test set can be set to 80%, 10%, and 10%.

[0092] 3) Obtain the water well stratified injection data and oil well stratified production data of the oil field to be tested, process the data and input them into the established water injection volume-oil production volume relationship model to obtain the connectivity analysis results of the oil field to be tested.

[0093] 4) According to the water well stratified injection data, oil well stratified production data and connectivity analysis results of the oil field to be tested, the relationship between the liquid production, water content and water injection volume of the oil field to be tested is constructed, and the injection and production parameter optimization scheme of the oil field to be tested is obtained, specifically:

[0094] 4.1) Using a regression algorithm, based on the obtained connectivity analysis results of the oil field to be tested, and according to the water well stratified water injection data and oil well stratified production data of the oil field to be tested, the relationship between the production of liquid, water content and water injection volume of the oil field to be tested is constructed. Specifically, the process of constructing the relationship between the production of liquid, water content and water injection volume of the oil field to be tested is similar to constructing a water injection volume-oil production volume relationship model, and the specific process will not be described in detail here.

[0095] 4.2) Using the Bayesian optimization algorithm, the injection and production parameters are optimized according to the relationship between the liquid production, water content and injection volume of the oil field to be tested, and the injection and production parameter optimization plan of the oil field to be tested is obtained.

[0096] Specifically, the specific process of the Bayesian optimization algorithm is as follows Figure 4 As shown in the figure, it includes: ① initializing parameters and selecting a surrogate model to describe the function f(x) to be optimized; ② updating the acquisition function a(x) according to the test results and the prior model; ③ determining the parameter x=argmaxxa(x) used in the next test according to the surrogate function; ④ iteratively repeating steps ② and ③ until the final search result is obtained. The difference between the Bayesian search algorithm and the grid search algorithm and the random search algorithm is that each time a search is performed, a probability model of the objective function is established based on the current and historical search results to select the next search value, thereby reducing the number of experiments. Its advantages include: high search efficiency, especially for high-dimensional input; not easy to fall into the local optimal solution; can solve black box optimization problems, only need to know the input-output correspondence, without the need for a specific output solution process; a certain amount of noise is allowed in the data, and this algorithm is more in line with the business characteristics of injection and production parameter optimization.

[0097] Assume that you need to optimize The liquid production of the oil well is , corresponding to The parameters to be allocated for the injection wells are The liquid production prediction model is , then the injection optimization problem can be described by the following formula:

[0098]

[0099] in, For the Liquid production of oil wells; For the Parameters of water injection wells to be allocated; For the Liquid production prediction model for oil wells.

[0100] Specifically, the injection-production parameter optimization scheme includes the adjusted stratified water injection volume of the water injection well, the predicted stratified liquid production and stratified oil production of the target oil well, and the stratified liquid production and stratified oil production of the associated oil wells.

[0101] The following is a detailed description of the injection and production parameter optimization method based on connectivity analysis of the present invention through specific embodiments:

[0102] Data preparation: Obtain the required input data, including well location data (horizontal and vertical coordinates), well inclination data, small layer stratification data, production well history data, operation well history data, water well daily data and oil well daily data, etc. Among them, the well location data includes the well number, the oil field to which it belongs, the block to which it belongs, the horizontal coordinate and the vertical coordinate; the well inclination data includes the well number, sounding, well inclination angle and azimuth; the small layer stratification data includes the well number, layer name, top depth and bottom depth; the production well history data includes the well number, top depth, bottom depth, start time and end time; the operation well history data includes the well number, measure category, start date, completion date and construction purpose; the water well daily data includes the well number, date, layer, water injection time, daily water injection volume, daily allocated water injection volume, pump pressure and oil pressure; the oil well daily data includes the well number, date, layer, production time, daily liquid production, daily liquid production, water content, displacement, nozzle, oil pressure and casing pressure.

[0103] 1) Conduct static injection-production connectivity analysis:

[0104] The input data include well location data, well deviation data, small layer stratification data, production well history data, water well daily data and oil well daily data.

[0105] Based on the daily data of water wells and oil wells, the classification of water and oil wells is determined in combination with the selected date. The well numbers in the daily data of water wells on the selected date are water injection wells, and the well numbers in the daily data of oil wells are oil production wells.

[0106] Based on the well location data, well inclination data and sub-layer stratification data, a target well location map of each sub-layer is generated. Based on the well classification judgment results, according to the production well history data and sub-layer stratification data, it is judged whether the oil wells and water injection wells are in production and injection in each sub-layer. The judgment rule is: there is an intersection between the top depth and bottom depth of the stratification and the top depth and bottom depth in the production well history, and the start time in the production well history data is before the selected date, and the end time is after the selected date; according to the well classification judgment results, the oil wells and water injection wells are marked in each map, indicating that the marked ones are producing wells and injection wells in a certain layer.

[0107] With the target points in each layer of the producing wells as the center and the well spacing of 500m as the radius (the well spacing can be customized), search for the water injection wells within the search range layer by layer, generate a static injection-production correspondence statistics table, and display the oil wells, water wells, and corresponding layers.

[0108] A target well location map of a certain layer is selected, and the static injection-production correspondence statistical table is corrected using geological maps to obtain the static injection-production connectivity analysis results.

[0109] 2) Based on the results of static injection-production connectivity analysis, a regression algorithm is used to establish a water injection volume-oil production volume relationship model:

[0110] 2.1) Data processing of static injection-production connectivity analysis results:

[0111] Specifically, daily data of oil wells, daily data of water wells, and operation well history data are obtained; sample data selection, based on the selected time, logical judgment is performed forward and backward, and a stable injection and production period without perforation and plugging operations is selected as sample data, and the sample data is further processed:

[0112] Processing of outliers in water well injection data: remove data with daily injection volume / daily allocated injection volume < 80%; eliminate the impact of oil well production time rate, convert daily liquid production into daily liquid production capacity, the conversion method is daily liquid production capacity = daily liquid production / production time * 24; eliminate the impact of injection-production effective cycle, consider the impact of injection-production effective cycle, and average the daily liquid production capacity of oil wells for 15 consecutive days (customizable according to the actual situation of the block); eliminate the impact of oil well measures on production, extract the operation category in the operation well history data, and remove the production data 15 days after the well opening date after the completion of the measures.

[0113] 2.2) Select the regression algorithm to be used, set the hyperparameters of the regression algorithm, and establish the water injection volume-oil production relationship model:

[0114] The regression algorithm selected is the RNN algorithm.

[0115] Set the hyperparameters of the RNN algorithm.

[0116] Set the global sensitivity analysis method parameters, set the number of random samples to 1024, and perform model simulation operations.

[0117] The selected regression algorithm is used to establish the water injection volume-oil production relationship model.

[0118] 2.3) The water injection data and oil well production data in the static injection-production connectivity analysis results after data processing are divided into three parts: training set, validation set and test set. The established water injection volume-oil production relationship model is trained and parameter tuned through the training set and validation set. The model accuracy of the established water injection volume-oil production relationship model is evaluated through the test set to obtain the trained water injection volume-oil production relationship model.

[0119] 3) Obtain the water well stratified injection data and oil well stratified production data of the oil field to be tested, process the data and input them into the established water injection volume-oil production volume relationship model to obtain the connectivity analysis results of the oil field to be tested:

[0120] Get input data, extract outcome data, set prediction parameters, daily fluid production / daily fluid production.

[0121] Data preprocessing: Obtain daily data of oil wells, daily data of water wells, and operation well history data; sample data selection, based on the selected time, perform logical judgment forward and backward, select the stable injection and production time period without perforation and plugging operations as sample data, select the data of the well group 1 from March 2005 to March 2013 for the past 8 years as samples, during which no perforation or plugging operations were performed on the oil and water wells, and further process the sample data, as follows:

[0122] Processing of outliers in water well injection data: remove data with daily injection volume / daily allocated injection volume < 80%; eliminate the impact of oil well production time rate, convert daily liquid production into daily liquid production capacity, the conversion method is daily liquid production capacity = daily liquid production / production time * 24; eliminate the impact of injection-production effective cycle, consider the impact of injection-production effective cycle, and average the daily liquid production capacity of oil wells for 15 consecutive days (customizable according to the actual situation of the block); eliminate the impact of oil well measures on production, extract the operation category in the operation well history data, and remove the production data 15 days after the well opening date after the completion of the measures.

[0123] like Figure 5 , Figure 6 The figure shows the comparison between the model calculation results and the actual data.

[0124] 4) According to the water well stratified injection data, oil well stratified production data and connectivity analysis results of the oil field to be tested, the relationship between the liquid production, water content and water injection volume of the oil field to be tested is constructed, and the injection and production parameter optimization scheme of the oil field to be tested is obtained:

[0125] With the goal of maximizing the daily liquid production of the target oil well (P-3), injection of associated water wells is carried out, while ensuring that the production of associated oil wells does not decrease.

[0126] The blending results are shown in Table 1 below:

[0127] Table 1 P-3 well group deployment parameter statistics:

[0128]

[0129] Example 2

[0130] This embodiment provides an injection-production parameter optimization system based on connectivity analysis, including:

[0131] The data acquisition module is used to obtain the water well stratified water injection data and oil well stratified production data of the oil field to be tested.

[0132] The connectivity analysis result determination module is used to process the water well stratified water injection data and oil well stratified production data of the oil field to be tested and input them into a pre-established water injection volume-oil production volume relationship model to obtain the connectivity analysis result of the oil field to be tested.

[0133] The injection and production parameter optimization scheme determination module is used to construct the relationship between the liquid production, water content and water injection volume of the oil field to be tested based on the water well stratified water injection data, oil well stratified production data and connectivity analysis results of the oil field to be tested, and obtain the injection and production parameter optimization scheme of the oil field to be tested.

[0134] The system provided in this embodiment is used to execute the above-mentioned method embodiments. Please refer to the above-mentioned embodiments for specific processes and detailed contents, which will not be repeated here.

[0135] Example 3

[0136] This embodiment provides a processing device corresponding to the injection-production parameter optimization method based on connectivity analysis provided in this embodiment 1. The processing device can be applicable to a client processing device, such as a mobile phone, a laptop computer, a tablet computer, a desktop computer, etc., to execute the method of embodiment 1.

[0137] The processing device includes a processor, a memory, a communication interface and a bus, and the processor, the memory and the communication interface are connected through the bus to complete mutual communication. The memory stores a computer program that can be run on the processing device, and when the processing device runs the computer program, the injection and production parameter optimization method based on connectivity analysis provided in this embodiment 1 is executed.

[0138] In some implementations, the memory may be a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage.

[0139] In some other implementations, the processor may be a central processing unit (CPU), a digital signal processor (DSP), or other general-purpose processors of various types, which are not limited herein.

[0140] In addition, the logic instructions in the above-mentioned memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on such an understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0141] Those skilled in the art will understand that the structure of the above-mentioned computing device is only a partial structure related to the scheme of the present application, and does not constitute a limitation on the computing device to which the scheme of the present application is applied. The specific computing device may include more or fewer components, or combine certain components, or have a different arrangement of components.

[0142] Example 4

[0143] This embodiment provides a computer program product corresponding to the method for optimizing injection and production parameters based on connectivity analysis provided in this embodiment 1. The computer program product may include a computer-readable storage medium on which computer-readable program instructions for executing the method for optimizing injection and production parameters based on connectivity analysis described in this embodiment 1 are loaded.

[0144] Computer readable storage media can be tangible devices that hold and store instructions used by instruction execution devices. Computer readable storage media can be, for example, but not limited to, electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any combination thereof.

[0145] The above embodiment provides a computer-readable storage medium, whose implementation principle and technical effect are similar to those of the above method embodiment, and will not be repeated here.

[0146] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0147] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0148] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0149] The above embodiments are only used to illustrate the present invention, wherein the structure, connection mode and manufacturing process of each component may be changed. Any equivalent transformations and improvements based on the technical solution of the present invention should not be excluded from the protection scope of the present invention.

Claims

1. A method for optimizing injection and production parameters based on connectivity analysis. It is characterized in that include: Obtain the water well stratified water injection data and oil well stratified production data of the oil field to be tested; After data processing, the water well stratified water injection data and oil well stratified production data of the oil field to be tested are input into a pre-established water injection volume-oil production volume relationship model to obtain the connectivity analysis results of the oil field to be tested; According to the water well stratified injection data, oil well stratified production data and connectivity analysis results of the oil field to be tested, the relationship between the liquid production, water content and water injection volume of the oil field to be tested is constructed, and the injection and production parameter optimization scheme of the oil field to be tested is obtained; The construction process of the water injection volume-oil production volume relationship model includes: Conduct static injection-production connectivity analysis based on application block input data; Based on the results of static injection-production connectivity analysis, a regression algorithm was used to establish a water injection volume-oil production volume relationship model; The static injection-production connectivity analysis based on the application block input data includes: Based on the input data of the application block, with the oil well as the center and the well spacing as the constraint, the water injection wells that have a static corresponding relationship with the central oil well within the well spacing constraint range are searched, and a static injection-production corresponding relationship statistics table is obtained; According to the static injection-production correspondence statistics table, determine the water injection wells and their corresponding layers that have a static correspondence with the central oil well; The static injection-production correspondence statistical table was corrected to obtain the static injection-production connectivity analysis results.

2. A method for optimizing injection and production parameters based on connectivity analysis as claimed in claim 1, It is characterized in that The above-mentioned method is based on the static injection-production connectivity analysis results and uses a regression algorithm to establish a water injection volume-oil production volume relationship model, including: Process the water injection data and oil well production data in the static injection-production connectivity analysis results; Select the regression algorithm to be used, set the hyperparameters of the regression algorithm, and establish a water injection volume-oil production relationship model; The water injection data and oil production data after data processing are divided into training set, validation set and test set, and the established water injection volume-oil production volume relationship model is trained, verified and tested to obtain a trained water injection volume-oil production volume relationship model.

3. A method for optimizing injection and production parameters based on connectivity analysis as claimed in claim 2, It is characterized in that The data processing of the water injection volume data and the oil production well production data in the static injection-production connectivity analysis results includes: Select the water injection volume data and oil well production data during the period of stable injection-production relationship; Eliminate production outliers in water injection data and oil well production data; Remove water well outliers from water injection data and oil well production data; Perform N-day average processing on the water injection data and the oil well production in the oil well production data; Extract the oil well operation measures data from the water injection volume data and the oil well production data, and remove the production data for N days after the measures are removed.

4. A method for optimizing injection and production parameters based on connectivity analysis as claimed in claim 2, It is characterized in that The selected regression algorithm and the hyperparameters of the regression algorithm are set to establish a water injection volume-oil production volume relationship model, including: Select the regression algorithm to be used; Set hyperparameters of the regression algorithm; The global sensitivity analysis method is used to perform parameter sensitivity analysis on the hyperparameters of the regression algorithm and obtain the dynamic connectivity coefficient; The selected regression algorithm is used to establish a water injection volume-oil production volume relationship model based on the dynamic connectivity coefficient, wherein the input of the water injection volume-oil production volume relationship model is the water injection volume injection data and the oil well production data, and the output is the connectivity analysis result.

5. The method for optimizing injection and production parameters based on connectivity analysis according to claim 1, It is characterized in that The method constructs the relationship between the production of liquid, water content and water injection volume of the oil field to be tested based on the water well stratified water injection data, oil well stratified production data and connectivity analysis results of the oil field to be tested, and obtains the injection and production parameter optimization scheme of the oil field to be tested, including: Using a regression algorithm, based on the obtained connectivity analysis results of the oil field to be tested, and according to the water well stratified water injection data and oil well stratified production data of the oil field to be tested, the relationship between the liquid production, water content and water injection volume of the oil field to be tested is constructed; The Bayesian optimization algorithm is used to optimize the injection and production parameters according to the relationship between the liquid production, water content and injection volume of the oil field to be tested, and the injection and production parameter optimization scheme of the oil field to be tested is obtained.

6. An injection and production parameter optimization system based on connectivity analysis, It is characterized in that include: A data acquisition module, used to acquire the water well stratified water injection data and oil well stratified production data of the oil field to be tested; The connectivity analysis result determination module is used to process the water well stratified water injection data and oil well stratified production data of the oil field to be tested and input them into a pre-established water injection volume-oil production volume relationship model to obtain the connectivity analysis result of the oil field to be tested; The injection and production parameter optimization scheme determination module is used to construct the relationship between the liquid production, water content and water injection volume of the oil field to be tested based on the water well stratified water injection data, oil well stratified production data and connectivity analysis results of the oil field to be tested, and obtain the injection and production parameter optimization scheme of the oil field to be tested; The construction process of the water injection volume-oil production volume relationship model includes: Conduct static injection-production connectivity analysis based on application block input data; Based on the results of static injection-production connectivity analysis, a regression algorithm was used to establish a water injection volume-oil production volume relationship model; The static injection-production connectivity analysis based on the application block input data includes: Based on the input data of the application block, with the oil well as the center and the well spacing as the constraint, the water injection wells that have a static corresponding relationship with the central oil well within the well spacing constraint range are searched, and a static injection-production corresponding relationship statistics table is obtained; According to the static injection-production correspondence statistics table, determine the water injection wells and their corresponding layers that have a static correspondence with the central oil well; The static injection-production correspondence statistical table was corrected to obtain the static injection-production connectivity analysis results.

7. A processing device, It is characterized in that It includes computer program instructions, wherein when the computer program instructions are executed by a processing device, they are used to implement the steps corresponding to the method for optimizing injection and production parameters based on connectivity analysis described in any one of claims 1 to 5.

8. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores computer program instructions, wherein the computer program instructions, when executed by a processor, are used to implement the steps corresponding to the method for optimizing injection and production parameters based on connectivity analysis according to any one of claims 1 to 5.

Images