Packer electric control setting optimization system based on big data analysis

Abstract

The present application provides an electric-controlled setting optimization system for a packer based on big data analysis, which relates to the technical field of packer optimization, including: a data acquisition module constructs a multi-source data warehouse; an influencing factor identification module uses data visualization to analyze the data warehouse to identify a set of key influencing factors that affect the setting effect of the packer; a feature matrix construction module extracts key features of the influencing factors and constructs a feature matrix; a model training module uses a random forest algorithm to train the feature matrix to obtain a setting effect prediction model; and a setting control module uses this model to optimize the control strategy to achieve precise setting control of the packer. This application can solve the problems in the prior art that the inability to effectively identify the factors affecting the setting, resulting in inaccurate prediction results, inflexible control strategies, and inability to adapt to changes in the environment and equipment status in a timely manner, resulting in unstable setting effects, thereby improving the success rate of setting of the packer and improving the overall efficiency and safety of oil and gas production.

Description

Technical Field

[0001] The present application relates to the technical field of packer optimization, and in particular to a packer electrically controlled setting optimization system based on big data analysis. Background Art

[0002] The packer is used to seal the annulus between the tubing and casing or isolate the target formation to ensure the normal operation and safe production of oil and gas wells. With the advancement of technology, the control method of the packer has gradually changed from mechanical to electronic control. The electronic control setting technology has poor control effect due to the interference of environmental factors, the complexity of the control algorithm and the requirement of system stability.

[0003] At present, the existing technology cannot effectively identify the factors affecting the sealing, resulting in inaccurate prediction results, inflexible control strategies, and inability to adapt to changes in the environment and equipment status in a timely manner, resulting in unstable sealing effects, which further affects the efficiency and safety of mine production. Summary of the invention

[0004] The purpose of this application is to provide an electronically controlled sealing optimization system for a packer based on big data analysis, in order to solve the problems in the prior art that the factors affecting the sealing cannot be effectively identified, resulting in inaccurate prediction results, inflexible control strategies, and inability to adapt to changes in the environment and equipment status in a timely manner, resulting in unstable sealing effects.

[0005] In view of the above problems, the present application provides a packer electric control setting optimization system based on big data analysis.

[0006] The present application provides an electric-controlled sealing optimization system for a packer based on big data analysis, wherein the electric-controlled sealing optimization system for a packer based on big data analysis includes: a data acquisition module, used to construct a data warehouse, the data warehouse includes historical sealing records from multiple data sources, downhole environmental parameters, and equipment sensor data; an influencing factor identification module, used to analyze the data distribution and data trends of the data warehouse using data visualization, and identify the set of influencing factors that affect the sealing effect of the packer; a feature matrix construction module, used to extract the key features of the set of influencing factors and construct a feature matrix; a model training module, used to use random forest to perform model training on the feature matrix to obtain a sealing effect prediction model, and the sealing effect prediction model is used to predict the sealing success rate under input control conditions; a sealing control module, used to use the sealing effect prediction model as an evaluation function to construct a search optimization space, the search optimization space includes a search center, a search direction and an optimization target, and the search optimization space is deployed in the electric control center of the packer, and iterative optimization is performed through the search optimization space to obtain a packer optimization control strategy, and the packer optimization control strategy is used to perform packer sealing control.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The data acquisition module is used to construct a data warehouse, which includes historical sealing records from multiple data sources, downhole environmental parameters, and equipment sensor data; the influencing factor identification module is used to analyze the data distribution and data trend of the data warehouse by using data visualization work, and identify the influencing factor set that affects the sealing effect of the packer; the feature matrix construction module is used to extract the key features of the influencing factor set and construct a feature matrix; the model training module is used to use random forest to train the feature matrix to obtain a sealing effect prediction model, and the sealing effect prediction model is used to predict the sealing success rate under input control conditions; the sealing control module is used to convert the The sealing effect prediction model is used as an evaluation function to construct a search optimization space, which includes a search center, a search direction and an optimization target. The search optimization space is deployed in the electrical control center of the packer, and iterative optimization is performed through the search optimization space to obtain the packer optimization control strategy. The packer optimization control strategy is used to control the sealing of the packer, which effectively solves the problems in the prior art that the factors affecting the sealing cannot be effectively identified, resulting in inaccurate prediction results, inflexible control strategies, and inability to adapt to changes in the environment and equipment status in a timely manner, resulting in unstable sealing effects. The sealing success rate of the packer is improved, and the overall efficiency and safety of oil and gas production are improved.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0011] Figure 1 This is a schematic diagram of the structure of the packer electric control setting optimization system based on big data analysis in this application;

[0012] Figure 2This is a flow chart of obtaining a data warehouse in the packer electric control setting optimization system based on big data analysis in this application.

[0013] Description of reference numerals:

[0014] Data acquisition module 11, influencing factor identification module 12, characteristic matrix construction module 13, model training module 14, and sealing control module 15. DETAILED DESCRIPTION

[0015] The present application provides an electronically controlled sealing optimization system for packers based on big data analysis, thereby solving the problems in the prior art that the factors affecting the sealing cannot be effectively identified, resulting in inaccurate prediction results, inflexible control strategies, and inability to adapt to changes in the environment and equipment status in a timely manner, resulting in unstable sealing effects. This improves the sealing success rate of the packers and enhances the overall efficiency and safety of oil and gas production.

[0016] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0017] This application provides a packer electric control setting optimization system based on big data analysis. Figure 1 The packer electric control setting optimization system based on big data analysis includes:

[0018] The data acquisition module 11 is used to construct a data warehouse, which includes historical sealing records, downhole environmental parameters, and equipment sensor data from multiple data sources.

[0019] Specifically, historical setting records include the time, location, results (success or failure), control parameters used, etc. of past setting operations of the packer. Downhole environmental parameters include temperature, pressure, fluid properties, etc. Equipment sensor data comes from sensors installed on the packer or its related equipment, including real-time measurements such as vibration, displacement, current, and voltage. Clean the collected data, remove duplicate, erroneous, or inconsistent records, and format the data to ensure that all records follow the same structure and format. Design the architecture of the data warehouse, including table structure, field definition, relational model, etc., select appropriate data warehouse technology, such as SQL database, NoSQL database, or data lake, and load the preprocessed data into the data warehouse.

[0020] The influencing factor identification module 12 is used to analyze the data distribution and data trend of the data warehouse by using data visualization, and identify the influencing factor set that affects the effect of the packer setting.

[0021] Specifically, extract historical sealing records, downhole environmental parameters, and equipment sensor data from the data warehouse, and convert the data into a format suitable for visualization, such as time series, categorical data, or aggregated data. Select appropriate visualization tools according to data types and requirements, such as Excel, Python using matplotlib, seaborn, and other libraries. Draw time series graphs to show the changing trend of sealing operations over time, make scatter plots or heat maps to reveal the relationship between different environmental parameters and sealing effects, use bar charts or pie charts to show the distribution of categorical data, such as the proportion of successful sealing, and create multidimensional data visualizations, such as parallel coordinates or radar charts, to simultaneously display multiple influencing factors. Use interactive visualization tools, such as Dashboards, to allow users to explore data dynamically. Observe trends in visualization charts, such as whether the sealing success rate increases over time, or whether environmental parameters are significantly associated with the sealing effect. Identify the key influencing factors that affect the sealing effect of the packer through correlation analysis, regression analysis, or machine learning algorithms. Use data visualization tools to analyze the data in the data warehouse, identify the key influencing factor set that affects the sealing effect of the packer, and provide strong data support for subsequent optimization work.

[0022] The feature matrix construction module 13 is used to extract key features of the impact factor set and construct a feature matrix.

[0023] Specifically, select the key feature that best represents the characteristics of each influencing factor. For example, if temperature is an influencing factor, the key feature is the average temperature or the temperature fluctuation range. Determine the type of each key feature, such as numerical type, such as temperature, pressure, categorical type, such as location, equipment model, or time series type, such as timestamp. Remove records containing missing values ​​or outliers, or perform appropriate interpolation or estimation on these values, and standardize or normalize numerical features. Encode categorical features, such as using unique hot encoding or label encoding. Use feature selection techniques, such as principal component analysis and mutual information to further reduce the number of features while retaining the most important information. Create a table or matrix in which each row represents a sample, such as a sealing operation, and each column represents a key feature. For each sample, fill in the corresponding key feature value into the corresponding position of the feature matrix.

[0024] The model training module 14 is used to use random forest to perform model training on the feature matrix to obtain a setting effect prediction model, and the setting effect prediction model is used to predict the setting success rate under input control conditions.

[0025] Specifically, extract the target variables corresponding to the feature matrix from the data warehouse, and divide the feature matrix and target variables into training and test sets. You can use the RandomForestClassifier or RandomForestRegressor class in the scikit-learn library. Configure the parameters of the random forest, such as the number of trees, the depth of the tree, the minimum number of samples required for splitting, etc., initialize the random forest model with the selected parameters, train the random forest model using the feature matrix and target variables of the training set, monitor the performance indicators during the training process, such as accuracy, recall, F1 score, etc., evaluate the trained random forest model using the feature matrix of the test set, and obtain unbiased performance indicators. If the performance of the model is not good, you can adjust the parameters and retrain the model, and use cross-validation and other techniques to verify the stability and generalization ability of the model.

[0026] The packer setting control module 15 is used to use the packer setting effect prediction model as an evaluation function to construct a search optimization space, wherein the search optimization space includes a search center, a search direction and an optimization target, and the search optimization space is deployed in the packer electrical control center. Iterative optimization is performed through the search optimization space to obtain a packer optimization control strategy, and the packer optimization control strategy is used for packer setting control.

[0027] Specifically, select an initial control strategy as the search center, which can be obtained from historical data, determine the control parameters that can be adjusted during the search process, such as setting pressure, setting time, fluid properties, etc., define a search range and direction for each control parameter, that is, the amplitude by which the parameter can be increased or decreased, and clarify the optimization goal, such as maximizing the setting success rate, minimizing the setting time, or considering multiple goals at the same time. Use the trained setting effect prediction model as an evaluation function to evaluate the setting success rate under any given control strategy. According to the optimization goal, define evaluation indicators, such as the improvement of the setting success rate, the reduction of the setting time, etc., and integrate the relevant algorithms and models of the search optimization space into the software of the electric control center of the packer. Select an optimization algorithm, such as gradient descent, genetic algorithm, or simulated annealing. The selected algorithm uses an evaluation function to evaluate each control strategy, and updates the search center and search direction based on the evaluation results. The initial search center and search direction are set. In each iteration, a new set of control strategies is generated based on the current search center and search direction. These control strategies are evaluated using the evaluation function, and the strategy with the best performance is selected as the new search center. The search direction is updated so that more favorable control strategies can be explored in the next iteration. A termination condition is set, such as reaching the maximum number of iterations, the success rate of setting reaches a predetermined threshold, or the improvement of the search direction is less than a certain threshold. After the iterative optimization process is completed, the control strategy with the best performance is extracted as the packer optimization control strategy, and the optimized control strategy is deployed to the packer electrical control center and used in the actual packer setting control. Using the setting effect prediction model as the evaluation function, iterative optimization is performed by searching the optimization space to obtain the packer's optimized control strategy, and it is applied to the actual setting control to improve the setting success rate and efficiency.

[0028] Further, if Figure 2 As shown, the data acquisition module 11 is also used for:

[0029] Collect historical sealing records, including sealing pressure, speed, and time parameters, and record fault information and repair data; collect environmental parameters and geological data at different depths of the wellbore, including temperature, pressure, formation characteristics, and wellbore stability data, to obtain the downhole environmental parameters; obtain equipment sensor data, including pressure, position, current data, equipment wear status data, and sensing time during the sealing process; perform timestamp and unit standardization conversion on the historical sealing records, downhole environmental parameters, and equipment sensor data, build storage partitions according to different data sources, and set data indexes for each partition, the partition data indexes are used to establish storage paths and search paths; build a distributed database based on the storage partitions and the data indexes of each partition to obtain the data warehouse.

[0030] Specifically, record the pressure, speed and time parameters of each sealing operation, record any fault information and its subsequent repair data. At different depths of the wellbore, collect environmental parameters such as temperature and pressure, collect geological data, covering formation characteristics and wellbore stability information, obtain pressure, position and current data in real time during the sealing process, monitor and record the wear status of the equipment and the sensing time, and uniformly format all collected data with timestamps to ensure that the units of all data are standardized and consistent. Divide storage partitions according to different sources of data, such as historical sealing records, downhole environmental parameters, and equipment sensor data. Set clear data indexes for each partition to establish data storage paths and search paths. Select a suitable distributed database system based on factors such as data volume and access speed. Create a database architecture, determine the data storage method, such as column storage or row storage, and import the preprocessed data into a distributed database. Create corresponding indexes in the database based on the set partition data indexes, and perform performance tuning on the database based on actual application scenarios, such as adjusting the cache size and optimizing query statements.

[0031] Furthermore, the impact factor identification module 12 is also used for:

[0032] By drawing a histogram of the data in the data warehouse, the influence of the data frequency distribution on the sealing effect of the packer is determined; the outliers and outliers in the data are identified through box plots, the sealing success rates of different quantiles of each feature are analyzed, and the quantile intervals that affect the sealing effect of the packer are determined; a time series graph is drawn to identify the changing trend of parameters over time, the changing relationship of the success rate in different time periods is compared, and the influence of the characteristic time series changing relationship on the sealing success rate is determined; a scatter plot is drawn to analyze the correlation between the features, and the influence of the correlation between the features on the sealing success rate is identified; based on the identified influencing information of the histogram, box plot, time series graph, and scatter plot, the data parameters whose influence on the sealing success rate exceeds the preset baseline are obtained to form the influencing factor set.

[0033] Specifically, draw a histogram of each data feature to intuitively display its frequency distribution, analyze the changes in the packer setting effect under different data frequency distributions, determine the impact of data frequency on the setting success rate, use box plots to detect outliers and outliers in the data, and analyze the impact of outliers and outliers on the packer setting effect. Calculate different quantiles of each feature, such as 25%, 50%, 75%, etc. Analyze the setting success rate under different quantiles and determine the quantile interval that has a significant impact on the setting effect. Draw a time series graph to show the trend of parameter changes over time, compare the setting success rate in different time periods, analyze the impact of time series changes on the setting effect, determine the correlation between the time series change relationship of the feature and the setting success rate, and identify the key features that have a time series impact on the setting success rate. Draw a scatter plot, analyze the correlation between the features, and identify the feature combination that has a significant impact on the setting success rate. Based on the analysis results of histograms, box plots, time series graphs, and scatter plots, comprehensively identify the data parameters that have a significant impact on the setting success rate. A preset baseline of the success rate of setting the packer is set by technical personnel in the field according to actual needs. Data parameters that have an impact on the success rate of setting the packer exceeding the preset baseline are selected to form an influencing factor set, which includes key features that have a significant impact on the setting effect of the packer, providing strong support for subsequent optimization and control strategies.

[0034] Furthermore, the setting control module 15 is also used for:

[0035] Activate downhole monitoring equipment to obtain downhole environmental monitoring data, including temperature, pressure, formation characteristics, and wellbore stability data at different depths underground; activate equipment sensors to obtain equipment monitoring data during the sealing process, including pressure, position, current data, and equipment wear status data; obtain sealing task parameters, including sealing position, sealing time, and sealing success target; obtain an initial control strategy based on the sealing position and sealing time; evaluate the initial control strategy based on the downhole environmental monitoring data and equipment monitoring data through the sealing effect prediction model to obtain an evaluation success rate; determine whether the evaluation success rate reaches the sealing success target, and if so, use the initial control strategy as the optimized control strategy for the packer.

[0036] Specifically, activate the downhole monitoring equipment to ensure the normal operation of the equipment, and obtain real-time environmental monitoring data at different depths in the well, including key information such as temperature, pressure, formation characteristics, and wellbore stability. At the same time, activate the equipment sensors to obtain various equipment monitoring data during the setting process, including pressure, position, current, and equipment wear status. Obtain the key parameters of the setting task, including the setting position, setting time, and the target of successful setting. According to the setting position and setting time, combined with historical data and experience, formulate an initial control strategy. Using the trained setting effect prediction model, the downhole environmental monitoring data, equipment monitoring data, and initial control strategy are used as inputs for model prediction. The model will output an evaluation success rate, which reflects the possibility of successful setting operation under given data and initial strategy. Compare the evaluation success rate with the target of successful setting. If the evaluation success rate reaches or exceeds the target of successful setting, it means that the initial control strategy is effective and can be used as the optimization control strategy of the packer.

[0037] Furthermore, the setting control module 15 is also used for:

[0038] When the evaluation success rate does not reach the sealing success target, the deviation is determined according to the evaluation success rate and the sealing success target; the search direction is determined according to the deviation, and the initial control strategy is used as the search center to perform control parameter mutation and add a new control strategy; the new control strategy is evaluated by an evaluation function to obtain a new evaluation success rate; the loss value of the new evaluation success rate and the original evaluation success rate is calculated with the sealing success target as the target, and the control strategy with the smallest success rate loss value is selected as the current optimal solution. At the same time, the optimization stop condition is set according to the loss value. When the condition is met, the current optimal solution is used as the optimized control strategy for the packer.

[0039] Specifically, according to the gap between the evaluation success rate and the sealing success target, the deviation is determined, which reflects the gap between the current control strategy and the expected target. According to the deviation, the search direction is determined, which is to point to the control strategy space that is more likely to improve the sealing success rate. Taking the initial control strategy as the search center, the control parameters are mutated to generate a new control strategy. The newly added control strategy is evaluated using the evaluation function to obtain the newly added evaluation success rate. Based on the sealing success target, the loss value of the newly added evaluation success rate and the original evaluation success rate is calculated, which reflects the gap between the control strategy and the target and is a key indicator for optimization. The control strategy with the smallest success rate loss value is selected as the current optimal solution, which is the control strategy closest to the target found in the current search process. The optimization stop condition is set according to the loss value, which can be that the loss value is less than a certain threshold, or that no better control strategy is found after multiple consecutive iterations. When the optimization stop condition is met, the current optimal solution is used as the optimized control strategy for the packer, which is the control strategy that is most likely to achieve sealing success after a series of optimization steps.

[0040] Furthermore, the setting control module 15 is also used for:

[0041] The control parameter sensitivity in the control strategy is obtained, wherein the control parameter sensitivity is the degree of influence of each control parameter on the success rate of sealing; a search sequence is set according to the control parameter sensitivity, wherein the search sequence is sorted from large to small sensitivity; a coarse-grained single-parameter search mutation is performed according to the search sequence to obtain a variant control strategy, and the variant control strategy is evaluated through an evaluation function, wherein the number of coarse-grained searches is not greater than 10; an evaluation success rate of the variant control strategy is associated with a variant single-parameter mapping, and the search direction is determined according to the contribution value of the evaluation success rate to the deviation.

[0042] Specifically, the change in the success rate of setting when each control parameter changes is quantified, and the sensitivity of each control parameter is determined, which reflects the degree of influence of the control parameter on the success rate of setting. According to the sensitivity of the control parameters, they are sorted from the most sensitive parameter to the least sensitive parameter. The sorted parameter list constitutes a search sequence. Following the search sequence, the control parameter with the highest sensitivity is first mutated to generate a new control strategy. The evaluation function is used to evaluate these mutated control strategies to obtain the evaluation success rate. Repeat the above steps, but the number of mutations does not exceed 10 times. Preferably, the number of mutations does not exceed 5 times. The specific number of mutations is specifically determined according to the number of control parameters and the constraint range. For example, when the number of mutations is 4, preliminary screening is performed through four mutations. Based on the evaluation of the variation adjustment of the preliminary control parameters, the deviation of the variation adjustment of the control parameters is used to determine the search direction, i.e., the contribution, to ensure the search efficiency. By establishing a mapping association between the evaluation success rate of the variation control strategy and the variation single parameter, it can be clearly seen which parameter change has a significant impact on the evaluation success rate. According to the contribution value of the evaluation success rate to the deviation, the subsequent search direction can be determined.

[0043] Furthermore, the setting control module 15 is also used for:

[0044] After completing all single parameter search mutations, perform coarse-grained parameter combination search mutation based on the search sequence; obtain the minimum combination sensitivity based on the parameter combination, and use the minimum combination sensitivity as the search step; search and mutate the parameter combination based on the search step to obtain a combination mutation strategy; evaluate the success rate of the combination mutation strategy through the evaluation function, compare the deviation contribution values ​​of the parameter combination and the mutated single parameter based on the evaluation success rate, and determine the search direction.

[0045] Specifically, based on a determined search sequence, multiple control parameters are selected for combination, and these parameter combinations are searched and mutated in a coarse-grained manner to generate new combination mutation strategies. For each parameter combination, its sensitivity is evaluated, that is, the degree of influence of parameter changes on the success rate of sealing. From all parameter combinations, find the combination with the least sensitivity and determine its minimum sensitivity. Using a determined search step, perform a more detailed search and mutation on the parameter combinations. Generate a new combination mutation strategy, and use an evaluation function to evaluate the success rate of the combination mutation strategy. Obtain the evaluation success rate of each combination mutation strategy and compare it with the previous strategy. Based on the evaluation success rate, calculate the contribution value of each parameter combination and single mutated parameter to the deviation. Compare these contribution values ​​to determine which parameter or parameter combination has the greatest potential to improve the success rate of sealing.

[0046] Furthermore, the setting control module 15 is also used for:

[0047] The search directions are arranged in positive order according to the contribution values ​​of the deviations to construct a search direction sequence; based on the search direction sequence, fine-grained control strategy search and variation are performed in order for each search direction, and the current optimal solution is updated according to the evaluation success rate obtained by the evaluation function, wherein the number of fine-grained searches is repeated until there is no current optimal solution update; when the number of fine-grained searches is reached, the search direction is switched according to the search direction sequence, and the fine-grained control strategy search and variation are repeated based on the switched search direction until all search directions in the search direction sequence are completed, and the current optimal solution is used as the optimized control strategy for the packer.

[0048] Specifically, the search directions are arranged in positive order according to the calculated deviation contribution values. First, explore the directions that contribute the least to the deviation, that is, the gap between the current strategy and the target, and then gradually turn to the directions that contribute more. Construct a search direction sequence. Based on the search direction sequence, start fine-grained control strategy search and mutation, and slightly adjust the control parameters in each search direction to generate a new control strategy. Use an evaluation function, such as a sealing effect prediction model, to evaluate these mutated control strategies and obtain the evaluation success rate. If the evaluation success rate of the new control strategy is higher than the current optimal solution, update the current optimal solution. Repeat the above steps until no better control strategy can be found in the current search direction. When the fine-grained search limit of a certain search direction is reached, switch to the next direction in the search direction sequence, and repeat the fine-grained control strategy search, mutation and evaluation process in the new search direction, and continue this process until all directions in the search direction sequence are completed. After completing the fine-grained search of all search directions, you will have the optimal control strategy currently found, that is, the current optimal solution. The optimal control strategy is used as the optimal control strategy for the packer and used for subsequent sealing operations.

[0049] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0050] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. The packer electric control setting optimization system based on big data analysis is characterized by: The packer electric control setting optimization system based on big data analysis includes: A data acquisition module is used to build a data warehouse, which includes historical sealing records, downhole environmental parameters, and equipment sensor data from multiple data sources; An influencing factor identification module, used to analyze the data distribution and data trend of the data warehouse by using data visualization, and identify the influencing factor set that affects the packer setting effect; A feature matrix construction module is used to extract key features of the impact factor set and construct a feature matrix; A model training module, used for performing model training on the feature matrix using random forest to obtain a setting effect prediction model, wherein the setting effect prediction model is used for predicting a setting success rate under input control conditions; A packer setting control module is used to use the packer setting effect prediction model as an evaluation function to construct a search optimization space, wherein the search optimization space includes a search center, a search direction and an optimization target. The search optimization space is deployed in the packer electrical control center, and iterative optimization is performed through the search optimization space to obtain a packer optimization control strategy, which is used for packer setting control.

2. The packer electric control setting optimization system based on big data analysis according to claim 1, characterized in that: The data acquisition module is also used for: Collect historical sealing records, including sealing pressure, speed, time parameters, and record fault information and repair data; Collect environmental parameters and geological data at different depths of the wellbore, including temperature, pressure, formation characteristics, and wellbore stability data, to obtain the downhole environmental parameters; Obtain equipment sensor data, including pressure, position, current data, equipment wear status data, and sensing time during the setting process; Perform timestamp and unit standardization conversion on the historical sealing records, downhole environmental parameters, and equipment sensor data, build storage partitions according to different data sources, and set data indexes for each partition, which are used to establish storage paths and search paths; A distributed database is constructed based on the storage partitions and the data indexes of each partition to obtain the data warehouse.

3. The packer electric control setting optimization system based on big data analysis according to claim 2, characterized in that: The impact factor identification module is also used for: By drawing a histogram of the data in the data warehouse, the influence of data frequency distribution on the effect of packer setting is determined; Identify outliers and outliers in the data through box plots, analyze the success rate of each feature at different quantiles, and determine the quantile interval that affects the packer setting effect; Draw a time series graph to identify the changing trend of parameters over time, compare the changing relationship of success rate in different time periods, and determine the impact of characteristic time series changing relationship on the success rate of sealing; Draw a scatter plot, analyze the correlation between the features, and identify the impact of the correlation between the features on the success rate of setting. Based on the identification influence information of the histogram, box plot, time series diagram, and scatter plot, data parameters whose influence on the sealing success rate exceeds a preset baseline are obtained to form the influence factor set.

4. The packer electric control setting optimization system based on big data analysis according to claim 1, characterized in that: The setting control module is also used for: Activate downhole monitoring equipment to obtain downhole environmental monitoring data, including temperature, pressure, formation characteristics, and wellbore stability data at different depths in the well; Activate equipment sensors to obtain equipment monitoring data during the sealing process, including pressure, position, current data, and equipment wear status data; Get the parameters of the sealing task, including the sealing position, sealing time, and sealing success target; Acquire an initial control strategy according to the sealing position and sealing time; The initial control strategy is evaluated according to the downhole environment monitoring data and the equipment monitoring data by using the setting effect prediction model to obtain an evaluation success rate; It is determined whether the evaluation success rate reaches the packer setting success target, and if so, the initial control strategy is used as the packer optimization control strategy.

5. The packer electric control setting optimization system based on big data analysis according to claim 4, characterized in that: The setting control module is also used for: When the evaluation success rate does not reach the setting success target, determining a deviation amount according to the evaluation success rate and the setting success target; Determine the search direction according to the deviation, take the initial control strategy as the search center, and perform control parameter variation to add a new control strategy; Evaluate the newly added control strategy through an evaluation function to obtain a newly added evaluation success rate; Taking the sealing success target as the goal, the loss value of the newly added evaluation success rate and the original evaluation success rate is calculated, and the control strategy with the smallest success rate loss value is selected as the current optimal solution. At the same time, the optimization stop condition is set according to the loss value. When it is met, the current optimal solution is used as the optimized control strategy of the sealing device.

6. The packer electric control setting optimization system based on big data analysis according to claim 5, characterized in that: The setting control module is also used for: Acquire the control parameter sensitivity in the control strategy, where the control parameter sensitivity is the influence degree of each control parameter on the setting success rate; Setting a search sequence according to the sensitivity of the control parameter, wherein the search sequence is sorted from the highest sensitivity to the lowest sensitivity; Perform a coarse-grained single-parameter search mutation according to the search sequence to obtain a mutation control strategy, and evaluate the mutation control strategy through an evaluation function, wherein the number of coarse-grained searches is no more than 10; An association between the evaluation success rate of the variation control strategy and the variation single parameter mapping is established, and the search direction is determined according to the contribution value of the evaluation success rate to the deviation amount.

7. The packer electric control setting optimization system based on big data analysis according to claim 6, characterized in that: The setting control module is also used for: After all single parameter search mutations are completed, a coarse-grained parameter combination search mutation is performed based on the search sequence; According to the parameter combination, a minimum combined sensitivity is obtained, and the minimum combined sensitivity is used as a search step length; Search and mutate the parameter combination based on the search step to obtain a combination mutation strategy; The success rate of the combined mutation strategy is evaluated by the evaluation function, and the deviation contribution values ​​of the parameter combination and the mutation single parameter are compared according to the evaluation success rate to determine the search direction.

8. The packer electric control setting optimization system based on big data analysis according to claim 7, characterized in that: The setting control module is also used for: Arrange the search directions in positive order according to the contribution values ​​of the deviations to construct a search direction sequence; Based on the search direction sequence, the fine-grained control strategy search variation of each search direction is performed in sequence, and the current optimal solution is updated according to the evaluation success rate obtained by the evaluation function, wherein the number of fine-grained searches is until there is no current optimal solution update; When the number of fine-grained searches is reached, the search direction is switched according to the search direction sequence, and the fine-grained control strategy search variation is repeated based on the switched search direction until all search directions in the search direction sequence are completed, and the current optimal solution is used as the packer optimization control strategy.

Images