Oil and gas well yield increase measure time point identification method, device and equipment

Abstract

The invention relates to the field of oil and gas engineering, and discloses an oil and gas well yield increase measure time point identification method, device and equipment. According to the method, multiple pieces of production parameter time sequence data in one-to-one correspondence with multiple production parameters of the target oil and gas well can be obtained, and yield increase change point detection is conducted on each piece of production parameter time sequence data so as to determine the yield increase change point parameter value in each piece of production parameter time sequence data; determining a time point corresponding to each yield increasing change point parameter value as an initial yield increasing measure time point, and determining each initial yield increasing measure time point as an effective yield increasing measure time point when determining that the interval duration between every two initial yield increasing measure time points is not greater than a preset duration threshold value; and selecting an initial yield increasing measure time point as a final yield increasing measure time point. The method can achieve the intelligent recognition of the time point of the yield increase measure of the target oil and gas well, and avoids the problems of low recognition efficiency and the like of manual recognition.

Description

Technical Field

[0001] The present invention relates to the field of petroleum and natural gas engineering, and in particular to a method, device and equipment for identifying the time point of oil and gas well production increase measures. Background Art

[0002] In the exploration, mining, production and transportation of oil and gas, monitoring and analyzing the production data of oil and gas wells is crucial to optimizing the production process of oil and gas wells, increasing output and reducing costs, such as identifying the timing of production-increasing measures.

[0003] Specifically, the relevant technology may be that technical personnel analyze the production data of the oil and gas wells to identify the timing of production increase measures for the oil and gas wells.

[0004] However, when technicians manually identify the time points for production-increasing measures, they need to analyze the data one by one. When the amount of production data is large, more human resources are consumed and the efficiency of manual identification is low. Summary of the invention

[0005] The present invention provides a method, device and equipment for identifying the time point of oil and gas well production increase measures, which are used to solve the defects of manual identification of the time of oil and gas well production increase measures in the related technology, and can realize intelligent identification of the time point of the production increase measures of the target oil and gas well, avoiding the problems of low recognition efficiency in manual identification.

[0006] In a first aspect, the present invention provides a method for identifying the time of oil and gas well production stimulation measures, comprising:

[0007] Acquire multiple production parameter time series data corresponding to multiple production parameters of the target oil and gas wells; wherein each of the production parameter time series data includes parameter values ​​of the production parameter at N time points in the target period, where N is an integer greater than 1;

[0008] Performing production increase change point detection on each of the production parameter time series data respectively to determine the production increase change point parameter value in each of the production parameter time series data, and determining the time point corresponding to each of the production increase change point parameter value as the initial production increase measure time point;

[0009] Determine that the time interval between any two of the initial production increase measure time points is no greater than a preset time threshold, then determine that each of the initial production increase measure time points is a valid production increase measure time point, and select one of the initial production increase measure time points as the final production increase measure time point; wherein the preset time threshold is greater than 0.

[0010] Optionally, the determining that the time interval between the two initial production-increasing measure time points is not greater than a preset time threshold, then determining each of the initial production-increasing measure time points as an effective production-increasing measure time point, includes:

[0011] Determine that the time interval between any two of the initial production-increasing measure time points is 0, and then determine that all of the initial production-increasing measure time points are overlapping effective production-increasing measure time points;

[0012] The selecting one of the initial production-increasing measures time points as the final production-increasing measures time point comprises:

[0013] An initial production-increasing measure time point is arbitrarily selected as a final production-increasing measure time point.

[0014] Optionally, the determining that the time interval between the two initial production-increasing measure time points is not greater than a preset time threshold, then determining each of the initial production-increasing measure time points as an effective production-increasing measure time point, includes:

[0015] Determine that the time interval between at least two of the initial production-increasing measures time points is greater than 0, and the time interval between any two of the initial production-increasing measures time points is not greater than the preset time threshold, then perform an effective change point score on each of the production-increasing change point parameter values ​​according to each of the production parameter time series data, and obtain an effective change point score value corresponding to each of the production-increasing change point parameter values;

[0016] Obtaining the scoring weight of each of the production parameters and using them as the scoring weights corresponding to the production increase change point parameter values;

[0017] Based on the scoring weight of each of the yield-increasing change point parameter values, weighted summing is performed on the effective change point scoring values ​​of each of the yield-increasing change point parameter values ​​to obtain a comprehensive score;

[0018] If it is determined that the comprehensive score is greater than the score threshold, each of the initial production-increasing measures time points is determined to be an effective production-increasing measures time point.

[0019] Optionally, performing effective change point scoring on each of the production increase change point parameter values ​​according to each of the production parameter time series data to obtain an effective change point score value corresponding to each of the production increase change point parameter values ​​includes:

[0020] For any of the production increase change point parameter values, in the corresponding production parameter time series data, determine a first peak point parameter value that is before the production increase change point parameter value and is closest to the production increase change point parameter value, and determine a second peak point parameter value that is after the production increase change point parameter value and is closest to the production increase change point parameter value;

[0021] Determine a first average value of all parameter values ​​from the first peak point parameter value to the production increase change point parameter value in the production parameter time series data; and determine a second average value of all parameter values ​​from the second peak point parameter value to the production increase change point parameter value in the production parameter time series data;

[0022] The absolute value of the difference between the first average value and the second average value is determined, and the absolute value of the difference is divided by the first average value to obtain a corresponding value, which is used as the effective change point score value corresponding to the yield increase change point parameter value.

[0023] Optionally, selecting one of the initial production-increasing measures time points as the final production-increasing measures time point includes:

[0024] According to the scoring weight of each of the yield-increasing change point parameter values, determining the yield-increasing change point parameter value with the largest scoring weight, and using it as the target yield-increasing change point parameter value;

[0025] The initial production-increasing measure time point corresponding to the target production-increasing change point parameter value is used as the final production-increasing measure time point.

[0026] Optionally, performing production increase change point detection on each of the production parameter time series data to determine the production increase change point parameter value in each of the production parameter time series data includes:

[0027] Input each of the production parameter time series data into the trained change point detection model to perform production increase change point detection to determine the production increase change point parameter value in each of the production parameter time series data;

[0028] Among them, the change point detection model is obtained by training a pre-trained model using training data and corresponding label data, the training data includes multiple sample production parameter time series data corresponding one by one to the multiple production parameters, the sample production parameter time series data includes the parameter values ​​of the production parameters at N time points in the sample period, and the label data includes the real production increase measure time points corresponding to the multiple sample production parameter time series data.

[0029] Optionally, the training process of the change point detection model includes:

[0030] Input each of the sample production parameter time series data into the pre-training model to perform production increase change point detection, so as to obtain a production increase change point parameter value group corresponding to each of the sample production parameter time series data; wherein each of the production increase change point parameter value groups includes a plurality of sample production increase change point parameter values ​​in the sample production parameter time series data;

[0031] Respectively determine a production increase measure time point group corresponding to each of the production increase change point parameter value groups, wherein the production increase measure time point group includes a time point corresponding to each sample production increase change point parameter value in the production increase change point parameter value group;

[0032] Determining the final production-increasing measure time point of all samples in all the production-increasing measure time point groups;

[0033] Among all the sample final production-increasing measure time points, determine all target production-increasing measure time points that match the actual production-increasing measure time points, and determine a first total number of the target production-increasing measure time points;

[0034] Calculating a second total number of the sample yield-increasing change point parameter values ​​in all the yield-increasing change point parameter value groups, dividing the first total number by the second total number to obtain a corresponding value as the accuracy of the pre-trained model;

[0035] With the goal of improving the accuracy of the pre-trained model, the model parameters in the pre-trained model are updated to obtain a trained model, and the trained model is continued to be trained until a model with an accuracy of a preset accuracy threshold is obtained, and is used as a trained change point detection model.

[0036] Optionally, the multiple production parameters include at least two of the monthly daily average liquid production, the monthly daily average water cut, the monthly cumulative oil production and the monthly daily average bottom hole flowing pressure.

[0037] In a second aspect, the present invention provides a device for identifying time of oil and gas well production increase measures, comprising:

[0038] An acquisition unit, used to acquire a plurality of production parameter time series data corresponding one-to-one to a plurality of production parameters of a target oil and gas well; wherein each of the production parameter time series data includes parameter values ​​of the production parameter at N time points in a target period, where N is an integer greater than 1;

[0039] A detection unit, used for performing an increase production change point detection on each of the production parameter time series data, so as to determine the increase production change point parameter value in each of the production parameter time series data;

[0040] A first determining unit, configured to determine the time point corresponding to each of the production increase change point parameter values ​​as an initial production increase measure time point;

[0041] The second determination unit is used to determine that the time interval between the two initial production increase measure time points is not greater than a preset time threshold, and then determine each of the initial production increase measure time points as an effective production increase measure time point; wherein the preset time threshold is greater than 0;

[0042] As a unit, it is used to select one of the initial production-increasing measures time points as the final production-increasing measures time point.

[0043] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method for time identification of oil and gas well production increase measures according to the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0044] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for time identification of oil and gas well production increase measures of the first aspect or any corresponding embodiment thereof.

[0045] The oil and gas well production increase measure time point identification method, device and equipment provided by the present invention can obtain multiple production parameter time series data corresponding to multiple production parameters of the target oil and gas well, respectively perform production increase change point detection on each production parameter time series data to determine the production increase change point parameter value in each production parameter time series data, and determine the time point corresponding to each production increase change point parameter value as the initial production increase measure time point, determine that the interval between the two initial production increase measure time points is not greater than the preset time threshold, then determine each initial production increase measure time point as an effective production increase measure time point, and select an initial production increase measure time point as the final production increase measure time point. The present invention can realize the intelligent identification of the production increase measure time point of the target oil and gas well, improve the identification efficiency of the production increase measure time point, and avoid the problems of high human resource consumption and low identification efficiency in manual identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 A flow chart of a method for identifying the time of oil and gas well production increase measures provided by an embodiment of the present invention;

[0048] Figure 2 A schematic diagram of a parameter value variation curve corresponding to a production parameter time series data provided by an embodiment of the present invention;

[0049] Figure 3 A schematic diagram of the structure of a device for identifying time of production enhancement measures for oil and gas wells provided by an embodiment of the present invention;

[0050] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] Combine the following Figure 1-Figure 2 The present invention describes the method for identifying the time of oil and gas well production enhancement measures.

[0053] like Figure 1 As shown, this embodiment proposes a first method for identifying the time of oil and gas well production increase measures, which may include the following steps:

[0054] S101. Acquire multiple production parameter time series data corresponding one-to-one to multiple production parameters of target oil and gas wells; wherein each production parameter time series data includes parameter values ​​of the production parameter at N time points in a target time period, where N is an integer greater than 1.

[0055] The target oil and gas well may be an oil and gas well for which the time point of production increase measures needs to be identified.

[0056] The production parameters may refer to various parameters used to describe and measure the performance and status of oil and gas wells during the oil and gas production process of the oil and gas wells.

[0057] Optionally, the above-mentioned multiple production parameters include at least two of the monthly daily average liquid production, the monthly daily average water cut, the monthly cumulative oil production and the monthly daily average bottom hole pressure. It should be noted that the monthly daily average data is selected as the production parameter because the change frequency of the monthly daily average data is lower than that of the daily data, and it has a richer data volume than the annual data, and can better reflect the impact of the production increase measures on the dynamic changes of production.

[0058] Specifically, the target period may be a period after the target oil and gas well is mined. Optionally, the target period may be a year after the target oil and gas well is mined. Optionally, the target period may be several years after the target oil and gas well is mined. Optionally, the target period may be a period from the time the target oil and gas well is mined to the present.

[0059] Specifically, the production parameter time series data may be obtained by arranging the parameter values ​​of the production parameters at N time points in the target period in chronological order.

[0060] Optionally, when the above-mentioned multiple production parameters include monthly daily average liquid production, monthly daily average water cut, monthly cumulative oil production and monthly daily average bottom hole flowing pressure, the above-mentioned N time points may include each end of the month in the target period.

[0061] S102, performing an increase-in-production change point detection on each production parameter time series data respectively to determine the increase-in-production change point parameter value in each production parameter time series data.

[0062] The production increase change point parameter value may be a parameter value occurring at the time point of the production increase measure in the production parameter time series data.

[0063] Specifically, this embodiment can perform production increase change point detection on the production parameter time series data corresponding to each production parameter, and determine the production increase change point parameter value in each production parameter time series data.

[0064] S103: Determine the time point corresponding to each production-increasing change point parameter value as the initial production-increasing measure time point.

[0065] Specifically, in this embodiment, the time point corresponding to each production-increasing change point parameter value may be determined as an initial production-increasing measure time point.

[0066] S104: Determine that the time interval between any two initial production-increasing measure time points is not greater than a preset time threshold, and then determine that each initial production-increasing measure time point is an effective production-increasing measure time point, wherein the preset time threshold is greater than 0.

[0067] Among them, the preset time threshold can be set by technical personnel according to actual conditions, for example, it can be set to three months.

[0068] Specifically, this embodiment can determine whether the time interval between the two initial production increase measure time points is greater than a preset time threshold. If it is determined that the time interval between the two initial production increase measure time points is not greater than the preset time threshold, each initial production increase measure time point can be determined as an effective production increase measure time point.

[0069] If it is determined that the time interval between at least two initial production-increasing measure time points is greater than a preset time threshold, it may be prohibited to determine each initial production-increasing measure time point as a valid production-increasing measure time point.

[0070] S105, selecting an initial production-increasing measure time point as the final production-increasing measure time point.

[0071] Specifically, in this embodiment, an initial production-increasing measure time point may be selected as the final production-increasing measure time point, that is, the production-increasing measure time point that is finally identified.

[0072] Optionally, in the second oil and gas well production stimulation measure time identification method proposed in this embodiment, step S104 may include:

[0073] It is determined that the time interval between any two initial production-increasing measure time points is 0, and then all initial production-increasing measure time points are determined to be overlapping effective production-increasing measure time points.

[0074] At this time, step S105 may include:

[0075] An initial production-increasing measure time point is arbitrarily selected as the final production-increasing measure time point.

[0076] It is understandable that if the interval between the two initial production increase measure time points is determined to be 0, it can be determined that all the initial production increase measure time points coincide with each other, and all the initial production increase measure time points are effective production increase measure time points. At this time, this embodiment can arbitrarily select an initial production increase measure time point to be determined as the final identified production increase measure time point.

[0077] The oil and gas well production increase measure time identification method proposed in this embodiment can obtain multiple production parameter time series data corresponding to multiple production parameters of the target oil and gas well, perform production increase change point detection on each production parameter time series data respectively, to determine the production increase change point parameter value in each production parameter time series data, and determine the time point corresponding to each production increase change point parameter value as the initial production increase measure time point, determine that the time interval between two initial production increase measure time points is not greater than the preset time threshold, then determine each initial production increase measure time point as an effective production increase measure time point, and select an initial production increase measure time point as the final production increase measure time point. This embodiment can realize the intelligent identification of the production increase measure time point of the target oil and gas well, improve the identification efficiency of the production increase measure time point, and avoid the problems of high human resource consumption and low identification efficiency in manual identification.

[0078] based on Figure 1 This embodiment proposes a third method for identifying the time of oil and gas well production increase measures. In this method, step S104 may include:

[0079] Determine that the time interval between at least two initial production-increasing measures is greater than 0, and the time interval between any two initial production-increasing measures is not greater than a preset time threshold, then perform an effective change point score on each production-increasing change point parameter value according to each production parameter time series data, and obtain an effective change point score value corresponding to each production-increasing change point parameter value;

[0080] Obtain the scoring weight of each production parameter and use it as the scoring weight of the corresponding production increase change point parameter value;

[0081] Based on the scoring weight of each yield-increasing change point parameter value, the effective change point scoring values ​​of each yield-increasing change point parameter value are weighted and summed to obtain a comprehensive score;

[0082] If it is determined that the comprehensive score is greater than the score threshold, each initial production-increasing measure time point is determined to be an effective production-increasing measure time point.

[0083] The scoring threshold may be set by a technician according to actual conditions and is not limited in this embodiment.

[0084] Optionally, the above-mentioned effective change point scoring is performed on each production increase change point parameter value according to each production parameter time series data to obtain the effective change point scoring value corresponding to each production increase change point parameter value, including:

[0085] For any production increase change point parameter value, in the corresponding production parameter time series data, determine the first peak point parameter value that is before the production increase change point parameter value and is closest to the production increase change point parameter value, and determine the second peak point parameter value that is after the production increase change point parameter value and is closest to the production increase change point parameter value;

[0086] Determine a first average value of all parameter values ​​between a first peak parameter value and an increase-yield change-point parameter value in the production parameter time series data; and determine a second average value of all parameter values ​​between a second peak parameter value and an increase-yield change-point parameter value in the production parameter time series data;

[0087] The absolute value of the difference between the first average value and the second average value is determined, and the absolute value of the difference is divided by the first average value to obtain a corresponding value, which is used as the effective change point score value corresponding to the yield increase change point parameter value.

[0088] At this time, step S105 may include:

[0089] According to the scoring weight of each yield-increasing change point parameter value, the yield-increasing change point parameter value with the largest scoring weight is determined and used as the target yield-increasing change point parameter value;

[0090] The initial production-increasing measure time point corresponding to the target production-increasing change point parameter value is taken as the final production-increasing measure time point.

[0091] Specifically, in this embodiment, the initial production-increasing measure time point corresponding to the production-increasing change point parameter value with the largest score weight may be determined as the final production-increasing measure time point.

[0092] The method for identifying the time of oil and gas well stimulation measures proposed in this embodiment can determine the final stimulation time point through effective change point scoring and scoring weights when the interval between at least two initial stimulation time points is greater than 0 and the interval between any two initial stimulation time points is not greater than a preset time threshold, thereby effectively identifying the final stimulation time point and enhancing the reliability of the final stimulation time point.

[0093] based on Figure 1 This embodiment proposes a third method for identifying the time of oil and gas well production increase measures. In this method, step S102 may include:

[0094] Input each production parameter time series data into the trained change point detection model to perform production increase change point detection to determine the production increase change point parameter value in each production parameter time series data;

[0095] Among them, the change point detection model is obtained by training the pre-trained model using training data and corresponding label data. The training data includes multiple sample production parameter time series data corresponding to multiple production parameters one by one. The sample production parameter time series data includes the parameter values ​​of the production parameters at N time points in the sample period. The label data includes the real production increase measure time points corresponding to the multiple sample production parameter time series data.

[0096] Optionally, the training process of the change point detection model includes:

[0097] Input each sample production parameter time series data into the pre-training model to perform production increase change point detection, so as to obtain a production increase change point parameter value group corresponding to each sample production parameter time series data; wherein each production increase change point parameter value group includes multiple sample production increase change point parameter values ​​in the sample production parameter time series data;

[0098] Determine the production increase measure time point group corresponding to each production increase change point parameter value group respectively, the production increase measure time point group includes the time point corresponding to each sample production increase change point parameter value in the production increase change point parameter value group;

[0099] Determine the final production-increasing measure time point for all samples in all production-increasing measure time point groups;

[0100] Among all the sample final production-increasing measure time points, determine all target production-increasing measure time points that match the actual production-increasing measure time points, and determine a first total number of the target production-increasing measure time points;

[0101] Calculate the second total number of sample yield-increasing change point parameter values ​​in all yield-increasing change point parameter value groups, divide the first total number by the second total number, and obtain a corresponding value as the accuracy of the pre-trained model;

[0102] With the goal of improving the accuracy of the pre-trained model, the model parameters in the pre-trained model are updated to obtain a trained model, and the trained model is continued to be trained until a model with an accuracy of a preset accuracy threshold is obtained, and is used as a trained change point detection model.

[0103] Specifically, this embodiment can connect to the oilfield database through an Extract Transform Load (ETL) tool to extract and synchronize data to obtain sample production parameter time series data.

[0104] Specifically, this embodiment can obtain the dynamic change data of production parameters from the oil field database, and then perform data cleaning on the dynamic change data, process missing values ​​and abnormal values, so as to obtain the corresponding sample production parameter time series data to ensure data quality. Among them, this embodiment can use interpolation, deletion and other methods to process missing data.

[0105] Specifically, when processing missing values, this embodiment can identify missing values ​​in the dynamically changing data of production parameters and select an appropriate method for processing. If the missing value is less than 0.5%, the missing value is directly deleted, otherwise the median is used to fill the missing value. When filling the missing value with the median, the data is first arranged in order from small to large. If the number of data n is an odd number, the median is the data in the middle position after sorting. If the number of data n is an even number, the median is the average of the two middle numbers after sorting.

[0106] Specifically, when performing outlier processing in this embodiment, the box plot method can be used to detect outliers in dynamically changing data, and appropriate processing methods can be selected, such as deleting outliers, smoothing, and replacing outliers. Among them, when using a box plot to detect outliers, the first quartile Q1 and the third quartile Q3 in the dynamically changing data can be calculated first, and the interquartile range IQR = Q3-Q1, and then the boundaries of the outliers are calculated, the lower limit = Q1-1.5×IQR, the upper limit = Q3+1.5×IQR. By comparing each data point with the upper and lower limits, outliers are identified. If the number of values ​​beyond the abnormal boundary is less than 5 / 1000, the outliers are deleted, and if the number is more than 5 / 1000, they are replaced with the median.

[0107] Specifically, when performing duplicate data processing, this embodiment can detect and delete duplicate data to ensure the uniqueness and consistency of the data.

[0108] Specifically, this embodiment can select 30 oil well samples, use a binary segmentation change point detection algorithm to identify the time of implementation of production-increasing measures in the oil well production performance curve, and combine the oil well measure historical records, use multi-parameter change point fusion and change point comprehensive scoring methods to evaluate and calibrate the change point detection model (production-increasing measures time detection model). The steps are as follows:

[0109] Data preparation: 30 oil wells were selected as samples, and sample production parameter time series data corresponding to multiple production parameters of the sample oil wells were collected, such as monthly daily average liquid production, monthly daily average water content, cumulative oil production, and monthly daily average bottom hole pressure. As shown in Table 1 below, the implementation time and type of production increase measures for each oil well in the production increase measures document were sorted out, and the implementation time of the production increase measures was counted to the month.

[0110] Table 1 Statistics of oil well production increase measures types and production increase measures time

[0111] Oil well name Types of production enhancement measures Production increase measures time PL-x1 fracture 2017.5 PL-x1 Water injection 2019.8 PL-x1 acidification 2021.4 PL-x2 fracture 2015.3 PL-x2 Sand control 2016.9 PL-x3 acidification 2020.8 ……

[0112] For each oil well, this embodiment can use the change point detection model to be trained to perform change point identification on the sample production parameter time series data of the monthly daily average liquid production, the monthly daily average water cut and the monthly daily average bottom hole pressure, and obtain the production increase change point parameter value and change point time under the three sample production parameter time series data of each oil well.

[0113] like Figure 2 The monthly average daily production of a certain oil well is shown in the dynamic change curve corresponding to the production parameter time series data from December 2009 to June 2021. The horizontal axis is the date and the vertical axis is the specific parameter value of the monthly average daily production. In this embodiment, the production parameter time series data is input into the change point detection model to be trained to perform production increase change point detection, and determine the corresponding multiple sample production increase change point parameter values, such as Figure 2 The monthly average daily liquid production corresponding to the changing points 1, 2 and 3 in the curve shown.

[0114] Afterwards, this embodiment can perform multi-parameter change point fusion, and use the change point results of multiple parameters of the same oil well to comprehensively judge the validity of the change point detection result to obtain the final change point detection result. Specifically, it includes the following steps:

[0115] Step a. Change point overlap processing. Compare whether the change points of the production dynamic curves of different parameters of the same oil well overlap in time, and the overlapping change points are directly confirmed as effective production increase measures. For example, when the change points 1 in the production dynamic parameter curves of the three production parameters overlap, they can be directly confirmed as effective production increase measures. The time point corresponding to one of the change points 1 is arbitrarily selected as the final production increase measure time point of the sample. For another example, when the change points 2 in the production dynamic parameter curves of the three production parameters overlap, they can be directly confirmed as effective production increase measures. The time point corresponding to one of the change points 1 is arbitrarily selected as the final production increase measure time point of the sample.

[0116] Step b. Change point weights: For the close change points identified under different production dynamic parameter curves of the same oil well, if the change point time window is within 2 months (i.e., the above-mentioned preset time threshold), different weights are assigned to each production parameter according to the importance of different parameters.

[0117] Step c. Calculation of parameter change point scores. On the production dynamic data of each parameter of the oil well, find the peak points of the nearest neighbor data before and after each change point. The specific method is to find the first peak point of the data before the change point time as the front peak point, and the first peak point after the change point time as the back peak point (such as Figure 2 Calculate the average value V of the parameter data points from the time of each production dynamic parameter curve change point to the peak point before and after 变点前 and V 变点后 , and calculate the score S of the parameter corresponding to the change point 月度日平均产液 , S 月度日平均含水率 and S 月度日平均井底流压 .

[0118]

[0119] n 前 and n 后 are the number of data points from the change point to the peak points before and after. 前 (i) is the score of the i-th data point between the change point and the previous peak point, V 后 (i) is the score of the i-th data point between the change point and the latter peak point.

[0120]

[0121] Step d. Comprehensive score of change points. Calculate the comprehensive score for each change point, comprehensive score = W1×S 月度日平均产液 +W2×S 月度日平均含水率 +W3×S 月度日平均井底流压 Among them, W1, W2 and W3 are the weights of each parameter, S 月度日平均产液 , S 月度日平均含水率 and S 月度日平均井底流压is the score of each parameter change point.

[0122] Step e. Threshold judgment: Set the threshold of the comprehensive score. The change point where the comprehensive score exceeds the threshold is considered to be a valid change point. The change point time of the production dynamic parameter curve with the largest parameter weight is taken as the time to implement the production increase measures, that is, the final production increase measure time point of the sample.

[0123] Afterwards, this embodiment can count the number of changes in the change point time in the oil well measure record and the number of changes identified by the change point detection algorithm, and calculate the change point detection accuracy. The change point detection accuracy can be obtained by dividing the number of changes that overlap by the number of changes identified by the change point detection model, and the number of changes that overlap is the number of changes that match the change point record document among the number of changes identified by the change point detection model.

[0124] After that, this embodiment can use the change point detection accuracy as the optimization target and adjust the change point detection model parameters to optimize the accuracy of the change point detection model. In this embodiment, 30 oil wells were tuned through the change point detection model parameters, and the final accuracy reached 94%, which is relatively high. 79 change points were identified and 74 coincident change points were identified. The corresponding tuning parameters of the model are as follows:

[0125] Initial cost function: Select the residual sum of squares as the cost function.

[0126] Cutting point selection strategy: Dynamic programming method is used to select cutting points.

[0127] Stop condition: Set the minimum sub-time period length to 10 and the minimum cost function value to 0.01.

[0128] Smoothing parameter: Set the moving average window size smoothing parameter to 5.

[0129] Parameter weights: The monthly daily average liquid production parameter weight W1 is 0.5, the monthly daily average water cut parameter weight W2 is 0.3, and the monthly daily average bottom hole pressure parameter weight W3 is 0.2.

[0130] Comprehensive score threshold: Comprehensive score threshold is 0.8.

[0131] It can be understood that in this embodiment Figure 1 In the method shown, the production parameter time series data corresponding to any production parameter is subjected to production increase change point detection. When there are multiple detected production increase change point parameter values, this embodiment can determine one or more corresponding final production increase measure time points based on the above steps a, b, c, d and e.

[0132] The method for identifying the time of oil and gas well production increase measures proposed in this embodiment can train and optimize the change point detection model to obtain a trained change point detection model, and use the trained change point detection model to perform production increase change point detection, detect the production increase change point parameter values ​​in the production parameter time series data, improve the accuracy of change point detection, and thus improve the accuracy of identifying the time point of the final production increase measure.

[0133] like Figure 3 As shown, this embodiment proposes a device for identifying the time of oil and gas well production increase measures, which may include:

[0134] The acquisition unit 301 is used to acquire a plurality of production parameter time series data corresponding to a plurality of production parameters of a target oil and gas well; wherein each production parameter time series data includes parameter values ​​of the production parameter at N time points in a target period, where N is an integer greater than 1;

[0135] A detection unit 302 is used to perform an increase production change point detection on each production parameter time series data to determine the increase production change point parameter value in each production parameter time series data;

[0136] The first determining unit 303 is used to determine the time point corresponding to each production increase change point parameter value as the initial production increase measure time point;

[0137] The second determination unit 304 is used to determine that the time interval between two initial production-increasing measure time points is not greater than a preset time threshold, and then determine each initial production-increasing measure time point as a valid production-increasing measure time point; wherein the preset time threshold is greater than 0;

[0138] As unit 305, it is used to select an initial production-increasing measure time point as a final production-increasing measure time point.

[0139] It should be noted that the processing of the acquisition unit 301, the detection unit 302, the first determination unit 303, the second determination unit 304 and the processing unit 305 and the beneficial effects thereof can be respectively referred to in Figure 1 Steps S101 to S105 in the above are not described in detail.

[0140] Optionally, the second determining unit 304 is further configured to:

[0141] Determine that the time interval between any two initial production-increasing measure time points is 0, and then determine that all initial production-increasing measure time points are overlapping effective production-increasing measure time points;

[0142] As unit 305, it is also used for:

[0143] An initial production-increasing measure time point is arbitrarily selected as the final production-increasing measure time point.

[0144] Optionally, the second determining unit 304 is further configured to:

[0145] Determine that the time interval between at least two initial production-increasing measures is greater than 0, and the time interval between any two initial production-increasing measures is not greater than a preset time threshold, then perform an effective change point score on each production-increasing change point parameter value according to each production parameter time series data, and obtain an effective change point score value corresponding to each production-increasing change point parameter value;

[0146] Obtain the scoring weight of each production parameter and use it as the scoring weight of the corresponding production increase change point parameter value;

[0147] Based on the scoring weight of each yield-increasing change point parameter value, the effective change point scoring values ​​of each yield-increasing change point parameter value are weighted and summed to obtain a comprehensive score;

[0148] If it is determined that the comprehensive score is greater than the score threshold, each initial production-increasing measure time point is determined to be an effective production-increasing measure time point.

[0149] Optionally, the second determining unit 304 is further configured to:

[0150] For any production increase change point parameter value, in the corresponding production parameter time series data, determine the first peak point parameter value that is before the production increase change point parameter value and is closest to the production increase change point parameter value, and determine the second peak point parameter value that is after the production increase change point parameter value and is closest to the production increase change point parameter value;

[0151] Determine a first average value of all parameter values ​​between a first peak parameter value and an increase-yield change-point parameter value in the production parameter time series data; and determine a second average value of all parameter values ​​between a second peak parameter value and an increase-yield change-point parameter value in the production parameter time series data;

[0152] The absolute value of the difference between the first average value and the second average value is determined, and the absolute value of the difference is divided by the first average value to obtain a corresponding value, which is used as the effective change point score value corresponding to the yield increase change point parameter value.

[0153] Optionally, as unit 305, it is also used for:

[0154] According to the scoring weight of each yield-increasing change point parameter value, the yield-increasing change point parameter value with the largest scoring weight is determined and used as the target yield-increasing change point parameter value;

[0155] The initial production-increasing measure time point corresponding to the target production-increasing change point parameter value is taken as the final production-increasing measure time point.

[0156] Optionally, the detection unit 302 is further configured to:

[0157] Input each production parameter time series data into the trained change point detection model to perform production increase change point detection to determine the production increase change point parameter value in each production parameter time series data;

[0158] Among them, the change point detection model is obtained by training the pre-trained model using training data and corresponding label data. The training data includes multiple sample production parameter time series data corresponding to multiple production parameters one by one. The sample production parameter time series data includes the parameter values ​​of the production parameters at N time points in the sample period. The label data includes the real production increase measure time points corresponding to the multiple sample production parameter time series data.

[0159] Optionally, the training process of the change point detection model is set as:

[0160] Input each sample production parameter time series data into the pre-training model to perform production increase change point detection, so as to obtain a production increase change point parameter value group corresponding to each sample production parameter time series data; wherein each production increase change point parameter value group includes multiple sample production increase change point parameter values ​​in the sample production parameter time series data;

[0161] Determine the production increase measure time point group corresponding to each production increase change point parameter value group respectively, the production increase measure time point group includes the time point corresponding to each sample production increase change point parameter value in the production increase change point parameter value group;

[0162] Determine the final production-increasing measure time point for all samples in all production-increasing measure time point groups;

[0163] Among all the sample final production-increasing measure time points, determine all target production-increasing measure time points that match the actual production-increasing measure time points, and determine a first total number of the target production-increasing measure time points;

[0164] Calculate the second total number of sample yield-increasing change point parameter values ​​in all yield-increasing change point parameter value groups, divide the first total number by the second total number, and obtain a corresponding value as the accuracy of the pre-trained model;

[0165] With the goal of improving the accuracy of the pre-trained model, the model parameters in the pre-trained model are updated to obtain a trained model, and the trained model is continued to be trained until a model with an accuracy of a preset accuracy threshold is obtained, and is used as a trained change point detection model.

[0166] The oil and gas well production increase measure time identification device proposed in this embodiment can obtain multiple production parameter time series data corresponding to multiple production parameters of the target oil and gas well, and perform production increase change point detection on each production parameter time series data to determine the production increase change point parameter value in each production parameter time series data, and determine the time point corresponding to each production increase change point parameter value as the initial production increase measure time point, and determine that the time interval between the two initial production increase measure time points is not greater than the preset time threshold, then determine each initial production increase measure time point as an effective production increase measure time point, and select an initial production increase measure time point as the final production increase measure time point. This embodiment can realize the intelligent identification of the production increase measure time point of the target oil and gas well, improve the identification efficiency of the production increase measure time point, and avoid the problems of high human resource consumption and low identification efficiency in manual identification.

[0167] The oil and gas well production enhancement measure time identification device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0168] The embodiment of the present invention also provides a computer device having the above Figure 3 The device for identifying the time of production enhancement measures for oil and gas wells is shown.

[0169] See also Figure 4 , a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication, and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 4 A processor 10 is taken as an example.

[0170] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0171] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0172] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function. The data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage devices. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0173] The memory 20 may include a volatile memory, such as a random access memory. The memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive. The memory 20 may also include a combination of the above-mentioned types of memory.

[0174] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0175] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0176] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying the time of oil and gas well production increase measures, characterized in that: include: Acquire multiple production parameter time series data corresponding to multiple production parameters of the target oil and gas wells; wherein each of the production parameter time series data includes parameter values ​​of the production parameter at N time points in the target period, where N is an integer greater than 1; Performing production increase change point detection on each of the production parameter time series data respectively to determine the production increase change point parameter value in each of the production parameter time series data, and determining the time point corresponding to each of the production increase change point parameter value as the initial production increase measure time point; Determine that the time interval between any two of the initial production increase measure time points is no greater than a preset time threshold, then determine that each of the initial production increase measure time points is a valid production increase measure time point, and select one of the initial production increase measure time points as the final production increase measure time point; wherein the preset time threshold is greater than 0.

2. The method according to claim 1, characterized in that The determining that the time interval between the two initial production-increasing measure time points is not greater than a preset time threshold, and then determining each of the initial production-increasing measure time points as an effective production-increasing measure time point, includes: Determine that the time interval between any two of the initial production-increasing measure time points is 0, and then determine that all of the initial production-increasing measure time points are overlapping effective production-increasing measure time points; The selecting one of the initial production-increasing measures time points as the final production-increasing measures time point comprises: An initial production-increasing measure time point is arbitrarily selected as a final production-increasing measure time point.

3. The method according to claim 1, characterized in that The determining that the time interval between the two initial production-increasing measure time points is not greater than a preset time threshold, and then determining each of the initial production-increasing measure time points as an effective production-increasing measure time point, includes: Determine that the time interval between at least two of the initial production-increasing measures time points is greater than 0, and the time interval between any two of the initial production-increasing measures time points is not greater than the preset time threshold, then perform an effective change point score on each of the production-increasing change point parameter values ​​according to each of the production parameter time series data, and obtain an effective change point score value corresponding to each of the production-increasing change point parameter values; Obtaining the scoring weight of each of the production parameters and using them as the scoring weights corresponding to the production increase change point parameter values; Based on the scoring weight of each of the yield-increasing change point parameter values, weighted summing is performed on the effective change point scoring values ​​of each of the yield-increasing change point parameter values ​​to obtain a comprehensive score; If it is determined that the comprehensive score is greater than the score threshold, each of the initial production-increasing measures time points is determined to be an effective production-increasing measures time point.

4. The method according to claim 3, characterized in that The step of performing effective change point scoring on each of the production increase change point parameter values ​​according to each of the production parameter time series data to obtain an effective change point scoring value corresponding to each of the production increase change point parameter values ​​includes: For any of the production increase change point parameter values, in the corresponding production parameter time series data, determine a first peak point parameter value that is before the production increase change point parameter value and is closest to the production increase change point parameter value, and determine a second peak point parameter value that is after the production increase change point parameter value and is closest to the production increase change point parameter value; Determine a first average value of all parameter values ​​from the first peak point parameter value to the production increase change point parameter value in the production parameter time series data; and determine a second average value of all parameter values ​​from the second peak point parameter value to the production increase change point parameter value in the production parameter time series data; The absolute value of the difference between the first average value and the second average value is determined, and the absolute value of the difference is divided by the first average value to obtain a corresponding value, which is used as the effective change point score value corresponding to the yield increase change point parameter value.

5. The method according to claim 3, characterized in that: The selecting one of the initial production-increasing measures time points as the final production-increasing measures time point comprises: According to the scoring weight of each of the yield-increasing change point parameter values, determining the yield-increasing change point parameter value with the largest scoring weight, and using it as the target yield-increasing change point parameter value; The initial production-increasing measure time point corresponding to the target production-increasing change point parameter value is used as the final production-increasing measure time point.

6. The method according to claim 1, characterized in that The step of performing production increase change point detection on each of the production parameter time series data to determine the production increase change point parameter value in each of the production parameter time series data comprises: Input each of the production parameter time series data into the trained change point detection model to perform production increase change point detection to determine the production increase change point parameter value in each of the production parameter time series data; Among them, the change point detection model is obtained by training a pre-trained model using training data and corresponding label data, the training data includes multiple sample production parameter time series data corresponding one by one to the multiple production parameters, the sample production parameter time series data includes the parameter values ​​of the production parameters at N time points in the sample period, and the label data includes the real production increase measure time points corresponding to the multiple sample production parameter time series data.

7. The method according to claim 6, characterized in that The training process of the change point detection model includes: Input each of the sample production parameter time series data into the pre-training model to perform production increase change point detection, so as to obtain a production increase change point parameter value group corresponding to each of the sample production parameter time series data; wherein each of the production increase change point parameter value groups includes a plurality of sample production increase change point parameter values ​​in the sample production parameter time series data; Respectively determine a production increase measure time point group corresponding to each of the production increase change point parameter value groups, wherein the production increase measure time point group includes a time point corresponding to each sample production increase change point parameter value in the production increase change point parameter value group; Determining the final production-increasing measure time point of all samples in all the production-increasing measure time point groups; Among all the sample final production-increasing measure time points, determine all target production-increasing measure time points that match the actual production-increasing measure time points, and determine a first total number of the target production-increasing measure time points; Calculating a second total number of the sample yield-increasing change point parameter values ​​in all the yield-increasing change point parameter value groups, dividing the first total number by the second total number to obtain a corresponding value as the accuracy of the pre-trained model; With the goal of improving the accuracy of the pre-trained model, the model parameters in the pre-trained model are updated to obtain a trained model, and the trained model is continued to be trained until a model with an accuracy of a preset accuracy threshold is obtained, and is used as a trained change point detection model.

8. The method according to any one of claims 1 to 7, characterized in that The multiple production parameters include at least two of the monthly daily average liquid production, the monthly daily average water cut, the monthly cumulative oil production and the monthly daily average bottom hole flowing pressure.

9. A device for identifying the time of production enhancement measures for oil and gas wells, characterized in that: include: An acquisition unit, used to acquire a plurality of production parameter time series data corresponding one-to-one to a plurality of production parameters of a target oil and gas well; wherein each of the production parameter time series data includes parameter values ​​of the production parameter at N time points in a target period, where N is an integer greater than 1; A detection unit, used for performing an increase production change point detection on each of the production parameter time series data, so as to determine the increase production change point parameter value in each of the production parameter time series data; A first determining unit, configured to determine the time point corresponding to each of the production increase change point parameter values ​​as an initial production increase measure time point; The second determination unit is used to determine that the time interval between the two initial production increase measure time points is not greater than a preset time threshold, and then determine each of the initial production increase measure time points as an effective production increase measure time point; wherein the preset time threshold is greater than 0; As a unit, it is used to select one of the initial production-increasing measures time points as the final production-increasing measures time point.

10. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for time identification of oil and gas well production increase measures according to any one of claims 1 to 8 by executing the computer instructions.

Images