A system for predicting the timing of polymer injection in an oil reservoir with special geological conditions

Abstract

The application relates to the technical field of oil layer polymer injection stopping time prediction, in particular to a special geological condition oil layer polymer injection stopping time prediction system. The system comprises an industrial data acquisition module, which is used for acquiring industrial bottom-hole flowing pressure data and industrial water cut data of produced liquid of an injection well and its adjacent production well in real time; an industrial data processing module, which is used for constructing displacement failure degree of the injection well, and representing the significance of polymer displacement mechanism failure of the injection well in the acquisition cycle; and a polymer injection stopping time prediction module, which is used for judging whether the current acquisition cycle has reached the polymer injection stopping time in the polymer oil displacement process by using the displacement failure degree predicted in the next acquisition cycle and the industrial water cut data of all adjacent production wells of the injection well. The application aims to fully consider the actual oil displacement state change of polymer oil displacement to predict the time, so that the polymer injection stopping time of the oil layer under special geological conditions can be more reasonably optimized.

Description

Technical Field

[0001] This application relates to the field of oil reservoir polymerization timing prediction technology, specifically to a system for predicting the timing of oil reservoir polymerization under special geological conditions. Background Technology

[0002] The Class I oil-bearing reservoirs in the Xingnan Development Area of ​​Daqing Oilfield, my country, have been the first targets for tertiary oil recovery, and mature extraction technologies have been developed and industrialized. However, the tertiary oil recovery potential and supporting adjustment technologies for Class III oil-bearing reservoirs are still incomplete. In particular, there is no relevant method for determining the timing of polymer flooding shutdown during the later stages of water cut recovery, and there is a lack of scientific and systematic basis for judging the timing of polymer flooding shutdown during Class III oil-bearing reservoir polymer flooding. Therefore, it is necessary to study reasonable and effective timing of polymer flooding shutdown during Class III oil-bearing reservoir polymer flooding and to conduct prediction of the timing of polymer flooding shutdown under special geological conditions in order to improve the polymer flooding effect and thus ensure the efficient development of the oilfield.

[0003] However, in polymer flooding, prematurely stopping polymer injection prevents the polymer from fully exerting its oil displacement effect, reducing oil recovery; while stopping too late may result in polymer waste and increased extraction costs. Current technologies often determine the timing of polymer injection stoppage based on real-time monitored industrial data parameters, such as when the water cut reaches a certain value or the injected polymer volume reaches a certain scale, adopting a "one-size-fits-all" approach to stop polymer injection across the entire reservoir. However, this method does not fully consider the actual changes in the oil displacement state during polymer flooding. Relying solely on real-time monitored industrial data parameters to determine the timing of polymer injection stoppage easily leads to problems with stopping too early or too late, failing to meet the demands of low-cost, high-efficiency extraction. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide a system for predicting the timing of polymer injection shutdown in oil reservoirs under special geological conditions. The specific technical solution adopted is as follows:

[0005] This application proposes a system for predicting the timing of polymer injection shutdown in oil reservoirs under special geological conditions, the system comprising:

[0006] The industrial data acquisition module is used to collect industrial bottom-hole flowing pressure data and industrial water content data of produced fluid from the injection well and its adjacent production well in real time.

[0007] The industrial data processing module is used to analyze the synchronicity of the upward trend of industrial water content data in produced fluid from different adjacent wells, as well as the similarity between the rising order when the rise occurs, and to construct a superimposed upward synchronicity to characterize the synchronicity of the sharp upward fluctuations in industrial water content data.

[0008] The rate of change of abrupt change points and the first-order difference of the abrupt change point sequence in the industrial injection-production pressure difference data between the injection well and its adjacent production well are identified to calculate the high-frequency anomalies in injection and production, which are used to characterize the instability of oil displacement efficiency.

[0009] The displacement failure degree of the injection well is obtained by positively correlating the superimposed rising synchronicity with the high-frequency anomalies of injection and production, which is used to characterize the significance of the failure of the polymer displacement mechanism in the injection well during the production cycle.

[0010] The polymer flooding timing prediction module is used to predict the displacement failure degree of the next acquisition cycle. It uses the displacement failure degree predicted for the next acquisition cycle and the industrial water cut data of all adjacent production wells of the injection well to set two conditions to determine whether the current acquisition cycle has reached the timing for stopping polymer flooding in the polymer flooding process.

[0011] Preferably, the superimposed ascending synchronicity is determined by positively correlating the synchronicity of the ascending trend with the similarity between the ascending position order when the ascending trend occurs.

[0012] Preferably, the upward trend is determined by the slope of the fitted data of industrial water cut in the produced fluid from different adjacent production wells.

[0013] Preferably, the synchronicity is determined by the inverse correlation fusion of the fitted slope values ​​and standard deviations calculated from all different neighboring production wells.

[0014] Preferably, the similarity between the rising sequences when an increase occurs is determined by the average similarity between the sequences of rising sequences when an increase occurs in the industrial water cut data of the produced fluid from all different neighboring wells.

[0015] Preferably, the high-frequency anomaly of injection and production is positively correlated with the rate of change of abrupt change points in the industrial injection-production pressure difference data between the injection well and its adjacent production well, and negatively correlated with the first-order difference of the abrupt change point position in the industrial injection-production pressure difference data between the injection well and its adjacent production well.

[0016] Preferably, the industrial injection-production pressure difference data is the difference between the industrial bottom-hole flowing pressure data of the injection well and the industrial bottom-hole flowing pressure data of its adjacent production well.

[0017] Preferably, the displacement failure degree is further determined by the product of the superimposed rising synchronization degree and the injection-production high-frequency anomaly.

[0018] Preferably, the method for predicting the displacement failure degree of the next acquisition cycle is as follows: the displacement failure degrees of the injection wells in the current acquisition cycle and all previous historical acquisition cycles are sorted in chronological order and recorded as the displacement failure sequence of the current acquisition cycle, so as to predict the displacement failure degree of the next acquisition cycle.

[0019] Preferably, the two conditions are set as follows:

[0020] Condition 1: The industrial water cut data of all adjacent production wells of the injection well in the current acquisition cycle are greater than the preset threshold;

[0021] Condition 2: The displacement failure rate predicted in the next acquisition cycle is higher than the upper limit of the 3 sigma range of the current acquisition cycle;

[0022] The 3sigma range of the current acquisition period is determined by using the displacement failure sequence of the current acquisition period and the displacement failure degree predicted for the next acquisition period as inputs to the 3sigma anomaly detection algorithm.

[0023] This application has at least the following beneficial effects:

[0024] (1) This application analyzes the synchronicity of the dramatic rise in industrial water cut data from both the overall and local perspectives, thereby accurately measuring the synchronicity of the superimposed rise in industrial water cut data in adjacent production wells. This more clearly reflects the oil displacement effect in the polymer flooding process, helps to more accurately predict the timing of stopping polymer injection, avoids the waste of polymer in the polymer flooding process, and improves the high utilization rate of polymer flooding.

[0025] (2) This application analyzes the abnormal high-frequency changes in the industrial injection-production pressure difference data between the injection well and its neighboring production well, thereby accurately measuring the high-frequency anomalies in the injection and production between the injection well and its neighboring production well. Combined with the superimposed rise synchronization between neighboring production wells, it accurately measures the characteristic magnitude of the failure of the polymer displacement mechanism during polymer flooding, thereby making the phenomenon of failure of the polymer displacement mechanism during polymer flooding clearer and facilitating a more accurate determination of the timing for stopping polymer injection.

[0026] (3) This application identifies the abnormal situation generated by the polymer displacement mechanism in the next collection cycle by taking into full account the high and low levels of industrial water content data and the actual oil displacement state of polymer flooding. This allows for a more accurate determination of the timing of stopping polymer injection during the polymer flooding process, thereby solving the problem of premature or late stopping of polymer injection in the prior art and achieving a more reasonable optimization of the timing of stopping polymer injection in oil layers under special geological conditions. Attached Figure Description

[0027] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a flowchart of a system for predicting the timing of polymer injection cessation in oil reservoirs under special geological conditions, provided as an embodiment of this application. Detailed Implementation

[0029] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a special geological condition oil reservoir polymerization timing prediction system proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0030] 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 pertains.

[0031] The following description, in conjunction with the accompanying drawings, details the specific scheme of the oil reservoir polymerization timing prediction system under special geological conditions provided in this application.

[0032] Please see Figure 1 The diagram illustrates a flowchart of a system for predicting the timing of polymer injection shutdown in oil reservoirs under special geological conditions, according to an embodiment of this application. The system includes:

[0033] The industrial data acquisition module is used to collect in real time the industrial bottom-hole flowing pressure data of the injection well and the industrial water content data of the produced fluid from the injection well and its adjacent production well.

[0034] To avoid the problem of stopping polymer injection too early or too late, it is necessary to fully consider the actual changes in the oil displacement state of polymer flooding, predict the timing of stopping polymer injection in oil reservoirs under special geological conditions, and thus more rationally optimize the timing of stopping polymer injection in oil reservoirs under special geological conditions.

[0035] Therefore, during polymer flooding in oil reservoirs (i.e., Class III oil reservoirs) under special geological conditions, industrial data parameters are collected in real time using IoT sensor technology. Specifically, the K nearest production wells to each injection well in terms of Euclidean distance are designated as the K nearest neighboring production wells for each injection well; in this embodiment, K is 3. The industrial water cut data of the produced fluid in each neighboring production well of each injection well is collected in real time using a capacitive water cut meter, and the industrial bottomhole flowing pressure data of each injection well and each neighboring production well is collected in real time using a downhole pressure gauge. These industrial data parameters include industrial water cut data and industrial bottomhole flowing pressure data. In this embodiment, the data collection interval is 1 minute, and the collection period is 24 hours per day.

[0036] Furthermore, all industrial data parameters within each acquisition cycle are arranged in chronological order, and linear interpolation is used to time-align different industrial data parameters during polymer flooding, resulting in the industrial bottom-hole flowing pressure sequence for each injection well, the industrial water cut sequence for each adjacent production well of each injection well, and the industrial bottom-hole flowing pressure sequence for each acquisition cycle. The use of linear interpolation for time alignment is a well-known technique, and the specific process will not be elaborated further.

[0037] Industrial data processing module: Processes industrial data parameters during polymer flooding and extracts features of polymer displacement mechanism failure during polymer flooding using feature engineering.

[0038] Generally, during polymer flooding, the more stable the industrial water cut data of the produced fluid in the production well at low levels, and the more stable the industrial injection-production pressure difference data between the injection well and its adjacent production well at high levels, the better the effect and efficiency of polymer flooding, and the more conducive it is to ensuring the efficient development of the oilfield.

[0039] However, if the industrial water cut data of the produced fluid in the adjacent wells shows a sharp increase and fluctuation, and the synchronicity of the sharp increase and fluctuation of industrial water cut data in different adjacent wells is higher, it is more likely to affect the effect of polymer flooding, thus causing waste of polymer during the polymer flooding process. The occurrence of this phenomenon indicates that the closer the collection cycle is to the time of stopping polymer injection during the polymer flooding process, the higher the utilization rate of polymer flooding can be guaranteed.

[0040] Therefore, for each injection well and K neighboring production wells in each acquisition cycle, this application analyzes the synchronicity of the upward trend of industrial water cut data in the produced fluid of different neighboring production wells, as well as the similarity between the upward order when the rise occurs, and constructs a superimposed upward synchronicity to characterize the synchronicity of the drastic upward fluctuation of industrial water cut data.

[0041] In one implementation, the superimposed synchronicity is determined by positively correlating the synchronicity of the upward trend with the similarity between the upward sequence at the time of the upward trend. The upward trend is determined by the fitted slope of the industrial water cut data in the produced fluid from different neighboring wells, and the synchronicity is determined by inversely correlating the mean and standard deviation of the fitted slopes calculated from all different neighboring wells. The similarity between the upward sequences at the time of the upward trend is determined by the average similarity between the sequences comprising the upward sequences when the industrial water cut data in the produced fluid from all different neighboring wells shows an upward trend.

[0042] In other embodiments, the upward trend can also be determined by statistics of industrial water cut data in produced fluid from different neighboring wells, and the synchronicity can also be determined solely by the inverse proportional mapping result of the standard deviation of the fitted slope calculated from all different neighboring wells.

[0043] It is understood that data fusion can be divided into positive correlation fusion and negative correlation fusion. Positive correlation fusion uses methods such as addition and multiplication of data, while negative correlation fusion uses methods such as subtraction and division of data. The specific positive correlation fusion and negative correlation fusion methods are determined by the implementer based on the actual situation, and this application does not impose any special restrictions. Optionally, the inverse proportional mapping can be implemented through negative linear mapping, negative exponential mapping, or by setting adjustment parameters, etc., which are not limited or elaborated here.

[0044] Specifically, in this embodiment, the industrial water cut sequence of the kth neighboring well is used as the input to the linear least squares method. The fitting slope of the industrial water cut data in the kth neighboring well is obtained through the linear least squares method. The larger the fitting slope, the more significant the drastic fluctuation in the industrial water cut data of the neighboring wells. In other embodiments, a moving linear regression method can also be used to determine the water cut, which can be set by the implementer according to the actual situation.

[0045] Furthermore, the mean and standard deviation of the fitted slopes of the industrial water cut data from K adjacent wells are calculated. The mean of the fitted slopes is used as the numerator, and the standard deviation and error parameter are used as the denominator. The error parameter is used to avoid the denominator being 0. It takes a value within a small data range (0.001, 0.01), and its impact on the calculation results is small and negligible. In this embodiment, the value is 0.005. The ratio of the numerator to the denominator is used as the synchronicity of the upward trend among the K adjacent wells. The synchronicity of the upward trend reflects the synchronicity of the overall industrial water cut data in the K adjacent wells showing a sharp upward fluctuation. The larger the synchronicity is greater than 0, the higher the synchronicity of the overall industrial water cut data in the K adjacent wells showing a sharp upward fluctuation. Conversely, the smaller the synchronicity is less than 0, the higher the synchronicity of the overall industrial water cut data in the K adjacent wells showing a sharp downward fluctuation. In addition, when the synchronicity is equal to 0, that is, the mean of the fitting slope is 0, it indicates that the synchronicity of the industrial water cut data in the K neighboring wells is poor, which shows drastic changes and fluctuations.

[0046] Simultaneously, the first-order difference sequence of the industrial water cut sequence of the k-th neighboring well is calculated. The sequence composed of the positions of all positive numbers within the first-order difference sequence is denoted as the rising position sequence of the k-th neighboring well. Each element in the rising position sequence represents a local element in the industrial water cut sequence where the industrial water cut data is experiencing a sharp upward fluctuation. Then, the mean of the Jaccard similarity among the rising position sequences of the K neighboring wells is calculated, denoted as the similarity between the rising positions when an upward fluctuation occurs among the K neighboring wells. The similarity between the rising positions when an upward fluctuation occurs reflects the synchronicity of the sharp upward fluctuations in the local industrial water cut data among the K neighboring wells. The greater the similarity, the higher the synchronicity of the sharp upward fluctuations in the local industrial water cut data among the K neighboring wells. In other embodiments, the absolute mean of the Pearson correlation coefficient between the rising position sequences can also be used to determine the similarity between the rising positions when an upward fluctuation occurs among the K neighboring wells. The specific method can be set by the implementer according to the actual situation. Jaccard similarity and Pearson correlation coefficient are well-known techniques and will not be elaborated further.

[0047] Specifically, in this embodiment, the superimposed rise synchronization degree among K adjacent production wells is calculated. : ; In the formula, for Normalization function, It is the synchronicity of the upward trend among K neighboring producing wells. Let K be the similarity between the ascending order of K neighboring producing wells when an ascending pattern occurs. The normalization function normalizes the overall rise synchronicity, providing accuracy in measuring the superimposed rise synchronicity among K neighboring production wells.

[0048] It should be noted that the synchronicity of the upward trend reflects the synchronicity of the sharp upward fluctuations in the overall industrial water cut data among the K neighboring production wells. The similarity between the rising positions reflects the synchronicity of the sharp upward fluctuations in the industrial water cut data locally among the K neighboring production wells. The two reflect the synchronicity of the sharp upward fluctuations in water cut from the overall and local perspectives, respectively. In order to improve the accuracy of subsequent prediction of the timing of polymer injection cessation, the formula considers the superposition between the overall and local perspectives, which more fully reflects the synchronicity of the sharp upward fluctuations in industrial water cut data.

[0049] The superimposed synchronicity reflects the synchronicity of the superimposed upward fluctuations in the industrial water cut data of K adjacent production wells. The greater the superimposed synchronicity, the higher the synchronicity of the superimposed upward fluctuations in the industrial water cut data of K adjacent production wells. At this time, it is more likely to affect the effect of polymer flooding. Moreover, the occurrence of this phenomenon indicates that it is closer to the time of stopping polymer injection in the polymer flooding process, which helps to avoid the waste of polymer in the polymer flooding process and improve the high utilization rate of polymer flooding.

[0050] Generally, the industrial injection-production pressure difference data between the injection well and its adjacent production well can reflect the efficiency of polymer flooding. The more significant the abnormal high-frequency changes in the industrial injection-production pressure difference data between the injection well and its adjacent production well, the more unstable the efficiency of polymer flooding is within that sampling period. This indicates a greater possibility of polymer displacement mechanism failure. Therefore, in order to reduce polymer waste during polymer flooding, this should be done as close as possible to the time when polymer injection is stopped.

[0051] Therefore, this application identifies the rate of change of abrupt change points and the first-order difference of the abrupt change point sequence in the industrial injection-production pressure difference data between the injection well and its adjacent production well, and calculates the high-frequency anomalies of injection and production to characterize the instability of oil displacement efficiency.

[0052] In one embodiment, the high-frequency anomaly in the injection-production process is positively correlated with the rate of change of abrupt change points in the industrial injection-production pressure differential data between the injection well and its adjacent production well, and negatively correlated with the first-order difference of the abrupt change point order in the industrial injection-production pressure differential data between the injection well and its adjacent production well. The industrial injection-production pressure differential data is the difference between the industrial bottom-hole flowing pressure data of the injection well and the industrial bottom-hole flowing pressure data of its adjacent production well.

[0053] It is understandable that a positive correlation means that the dependent variable increases as the independent variable increases, and decreases as the independent variable decreases; a negative correlation means that the dependent variable decreases as the independent variable increases, and increases as the independent variable decreases. This is determined by the actual application, and this application does not impose any special restrictions.

[0054] Specifically, this embodiment calculates the industrial injection-production pressure difference sequence between each injection well and its kth neighboring production well. The industrial injection-production pressure difference sequence is the difference sequence between the industrial bottom-hole flowing pressure sequence of the injection well and the industrial bottom-hole flowing pressure sequence of its neighboring production well. Then, the industrial injection-production pressure difference sequence between each injection well and its kth neighboring production well is used as the input to the Bernaola-Galvan segmentation algorithm. The significance level in the segmentation algorithm is set to 0.05 by default, and the preset minimum segment length is set to 10. The Bernaola-Galvan segmentation algorithm obtains all abrupt change points and their order in the industrial injection-production pressure difference sequence between each injection well and its kth neighboring production well, reflecting the abnormal abrupt changes in the industrial injection-production pressure difference data between the injection well and its neighboring production well. The Bernaola-Galvan segmentation algorithm is a known technology, and the specific process will not be described in detail.

[0055] Furthermore, the sum of the absolute values ​​of the rate of change of all abrupt changes in the industrial injection-production pressure differential sequence between each injection well and its kth neighboring production well is calculated. The sum of the absolute values ​​of the rate of change of all abrupt changes can reflect the degree of abnormal cumulative abrupt changes in the industrial injection-production pressure differential data. At the same time, the first-order difference mean of the position of all abrupt changes in the industrial injection-production pressure differential sequence is calculated. The first-order difference mean of the position of all abrupt changes can reflect the time interval of abnormal abrupt changes in the industrial injection-production pressure differential data. If the degree of abnormal cumulative abrupt changes in the industrial injection-production pressure differential data is greater and the time interval of abnormal abrupt changes in the industrial injection-production pressure differential data is smaller, it indicates that the abnormal high-frequency changes in the industrial injection-production pressure differential data are more significant, and the efficiency of polymer flooding is more unstable within this sampling period.

[0056] Furthermore, the ratio of the mean change rate of all mutation points to the mean of the first difference of the order of all mutation points is denoted as the first ratio between each injection well and its kth neighboring production well. The mean of the first ratio between each injection well and its K neighboring production wells is denoted as the injection-production high-frequency anomaly between each injection well and its K neighboring production wells. The injection-production high-frequency anomaly reflects the abnormal high-frequency changes in the industrial injection-production pressure difference data between the injection well and its K neighboring production wells. The greater the injection-production high-frequency anomaly, the more unstable the efficiency of polymer flooding is within the sampling period, which represents a greater possibility of polymer displacement mechanism failure. Therefore, in order to reduce polymer waste during polymer flooding, this should be closer to the time to stop polymer injection during the polymer flooding process.

[0057] Furthermore, since the higher the synchronicity of the superimposed upward fluctuations in the industrial water cut data of the K adjacent production wells, and the more significant the abnormal high-frequency changes in the industrial injection-production pressure difference data between the injection well and its K adjacent production wells, it can comprehensively indicate that the effect and efficiency of polymer flooding are worse, and there is a greater possibility of the failure of the polymer displacement mechanism. Therefore, the closer this collection period is to the timing of stopping polymer injection during the polymer flooding process.

[0058] Therefore, based on the above analysis, this application will positively correlate the superimposed rising synchronicity with the high-frequency anomaly of injection and production to obtain the displacement failure degree of the injection well, which is used to characterize the significance of the failure of the polymer displacement mechanism in the injection well during the production cycle.

[0059] In this embodiment, the displacement failure degree of each injection well in the t-th acquisition cycle is calculated. : ; In the formula, Let K be the superimposed rise synchronization degree between the K neighboring production wells of each injection well in the t-th acquisition cycle. Let be the high-frequency injection-production anomalies between each injection well and its K neighboring production wells during the t-th acquisition cycle.

[0060] It should be noted that the polymer flooding characteristic value is measured by the product of two dimensionless parameters. The greater the superimposed synchronicity, the worse the effect of polymer flooding; while the greater the injection-production high-frequency anomaly, the worse the efficiency of polymer flooding. The failure of the polymer displacement mechanism will simultaneously produce poor oil displacement effect and oil displacement efficiency. Therefore, the two are closely related to the failure of the uniform polymer displacement mechanism. Thus, the superimposed synchronicity and injection-production high-frequency anomaly are used to measure the degree of displacement failure.

[0061] Displacement failure degree reflects the magnitude of the failure of the polymer displacement mechanism during polymer flooding. The greater the displacement failure degree, the more significant the failure of the polymer displacement mechanism is within the collection period. In order to avoid the waste of polymer during polymer flooding, the collection period is more likely to reach the time of stopping polymer injection during polymer flooding.

[0062] The polymer flooding timing prediction module is used to predict the displacement failure degree of the next acquisition cycle. It uses the displacement failure degree predicted for the next acquisition cycle and the industrial water cut data of all adjacent production wells of the injection well to set two conditions to determine whether the current acquisition cycle has reached the timing for stopping polymer flooding in the polymer flooding process.

[0063] In order to more accurately predict the timing of polymer injection cessation in oil reservoirs under special geological conditions, this application uses the displacement failure rate of injection wells in the current collection cycle and all previous historical collection cycles to predict the displacement failure rate of the next collection cycle during polymer flooding.

[0064] Specifically, in this embodiment, the displacement failure degree of the injection wells in the current acquisition cycle and all previous historical acquisition cycles is sorted in chronological order and recorded as the displacement failure sequence of the current acquisition cycle. The displacement failure sequence of the current acquisition cycle is used as the input of the ARIMA autoregressive moving average model. In the model, the autoregressive order p is 2, the difference order d is 2, and the moving average order q is 1. The displacement failure sequence is predicted by the ARIMA autoregressive moving average model to obtain the displacement failure degree predicted for the next acquisition cycle. The ARIMA autoregressive moving average model is a well-known technology, and the specific process will not be described in detail.

[0065] Furthermore, to avoid the problem of stopping polymer injection too early or too late, this application fully considers the actual oil displacement state of polymer flooding and sets two conditions based on the displacement failure degree predicted in the next acquisition cycle and the industrial water cut data of all adjacent production wells of the injection well, in order to determine whether the current acquisition cycle has reached the time to stop polymer injection in the polymer flooding process.

[0066] The two conditions for satisfaction are set as follows:

[0067] Condition 1: The industrial water cut data of all adjacent production wells of the injection well in the current data collection period are greater than 92%;

[0068] Condition 2: The displacement failure rate predicted in the next acquisition cycle is higher than the upper limit of the 3sigma range of the current acquisition cycle.

[0069] If the current acquisition cycle meets both of the above conditions, it is determined that the current acquisition cycle has reached the time to stop polymer injection during the polymer flooding process. At this time, polymer flooding is stopped at the last moment of the current acquisition cycle to avoid the problem of stopping polymer injection too late.

[0070] Otherwise, it is determined that the current collection cycle has not reached the timing for stopping polymer injection during the polymer flooding process, thus avoiding the problem of stopping polymer injection too early.

[0071] Specifically, in this embodiment, the displacement failure sequence of the current acquisition cycle and the displacement failure degree predicted for the next acquisition cycle are used as inputs to the 3sigma anomaly detection algorithm. The 3sigma range of the current acquisition cycle is obtained through the 3sigma anomaly detection algorithm. If the displacement failure degree predicted for the next acquisition cycle is higher than the upper limit of the 3sigma range of the current acquisition cycle, it indicates that the polymer displacement mechanism in the next acquisition cycle has an abnormal condition. The 3sigma anomaly detection algorithm is a well-known technology, and the specific process will not be described in detail.

[0072] Simultaneously, by real-time monitoring of the industrial water cut data of K adjacent production wells of the injection well, if the industrial water cut data of all K adjacent production wells of the injection well in the current acquisition cycle are greater than a preset threshold, it indicates that the polymer flooding effect in the current acquisition cycle is poor, and polymer injection should be considered to be stopped in the current acquisition cycle. Since the water cut of Class C wells when polymer injection is stopped is mostly concentrated in the range of 92.5%-93.5%, this application sets the value range of the preset threshold to 92.5%-93.5%, and the value in this embodiment is 93%.

[0073] Thus, the invention of a system for predicting the timing of polymer injection cessation in oil reservoirs under special geological conditions was completed.

[0074] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0075] It should be noted that, unless otherwise specified and limited, terms such as “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a circuit structure, article, or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such article or device. Without further limitations, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the article or device that includes said element. Furthermore, the term “and / or” as used herein includes any and all combinations of one or more of the associated listed items.

[0076] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not invented in this application.

[0077] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A system for predicting the timing of polymer injection shutdown in three types of oil reservoirs, characterized in that, The system includes: The industrial data acquisition module is used to collect industrial bottom-hole flowing pressure data and industrial water content data of produced fluid from the injection well and its adjacent production well in real time. The industrial data processing module is used to analyze the synchronicity of the upward trend of industrial water content data in produced fluid from different adjacent wells, as well as the similarity between the order of the rise when the rise occurs, and to construct a superimposed synchronicity of the rise, which is used to characterize the synchronicity of the dramatic upward fluctuations in industrial water content data. The rate of change of abrupt change points and the first-order difference of the abrupt change point sequence in the industrial injection-production pressure difference data between the injection well and its adjacent production well are identified to calculate the high-frequency anomalies in injection and production, which are used to characterize the instability of oil displacement efficiency. The displacement failure degree of the injection well is obtained by positively correlating the superimposed rising synchronicity with the high-frequency anomalies of injection and production, which is used to characterize the significance of the failure of the polymer displacement mechanism in the injection well during the production cycle. The polymer flooding timing prediction module is used to predict the displacement failure degree of the next acquisition cycle; and uses the displacement failure degree predicted for the next acquisition cycle and the industrial water cut data of all adjacent production wells of the injection well to set two conditions to determine whether the current acquisition cycle has reached the timing for stopping polymer flooding in the process. The superimposed synchronicity is determined by positively correlating the synchronicity of the upward trend with the similarity between the upward sequence when the upward trend occurs; the upward trend is determined by the fitting slope of the industrial water cut data in the produced fluid of different neighboring wells; the synchronicity is determined by negatively correlating the values ​​and standard deviations of the fitting slopes calculated from all different neighboring wells; and the similarity between the upward sequences when the upward trend occurs is determined by the average similarity between the sequences composed of the upward sequences when the industrial water cut data in the produced fluid of all different neighboring wells increases. The high-frequency anomalies in injection and production are positively correlated with the rate of change of abrupt change points in the industrial injection-production pressure difference data between the injection well and its adjacent production well, and negatively correlated with the first-order difference of the abrupt change point position in the industrial injection-production pressure difference data between the injection well and its adjacent production well.

2. The three-type reservoir polymer injection timing prediction system as described in claim 1, characterized in that, The industrial injection-production pressure difference data is the difference between the industrial bottom hole flowing pressure data of the injection well and the industrial bottom hole flowing pressure data of its adjacent production well.

3. The three-type oil reservoir polymer injection timing prediction system as described in claim 1, characterized in that, The displacement failure degree is further determined by the product of the superimposed rising synchronicity and the injection-production high-frequency anomaly.

4. The three-type oil reservoir polymer injection timing prediction system as described in claim 1, characterized in that, The method for predicting the displacement failure rate of the next acquisition cycle is as follows: the displacement failure rates of the injection wells in the current acquisition cycle and all previous historical acquisition cycles are sorted in chronological order and recorded as the displacement failure sequence of the current acquisition cycle, so as to predict the displacement failure rate of the next acquisition cycle.

5. The three-type oil reservoir polymer injection timing prediction system as described in claim 4, characterized in that, The two conditions for satisfaction are set as follows: Condition 1: The industrial water cut data of all adjacent production wells of the injection well in the current acquisition cycle are greater than the preset threshold; Condition 2: The displacement failure rate predicted in the next acquisition cycle is higher than the upper limit of the 3 sigma range of the current acquisition cycle; The 3sigma range of the current acquisition period is determined by using the displacement failure sequence of the current acquisition period and the displacement failure degree predicted for the next acquisition period as inputs to the 3sigma anomaly detection algorithm.

Images