Shale gas production data analysis method and system
By constructing a network structure model and using Gibbs sampling technology, the problem of insufficient real-time feedback in shale gas production was solved, enabling dynamic optimization of the production process and risk identification, thereby improving the efficiency and safety of shale gas production.
Patent Information
- Application Number
- CN202410983343.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing data analysis technologies lack real-time feedback and dynamic optimization capabilities in shale gas production, making it difficult to respond quickly to unexpected situations during the production process. This results in a failure to maximize production efficiency and resource utilization, as well as insufficient risk assessment and management, which increases costs and safety risks.
By collecting shale gas production data, a network structure model is constructed, probability assessment and parameter optimization are performed, and Gibbs sampling is used to adjust the probability values between nodes to build an optimized network model. This model can predict changes in production efficiency in real time, identify key risk points, and optimize production strategies.
It enables dynamic optimization and risk identification of the shale gas production process, allowing for timely adjustments to production strategies, improving extraction efficiency and safety, reducing costs, and enhancing resource utilization and economic returns.
Smart Images

Figure CN118917687A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data analysis, and in particular to a shale gas production data analysis method and system. Background Art
[0002] The field of data analysis technology focuses on collecting, processing, analyzing and interpreting batch data obtained from the oil and gas production process to optimize mining strategies, improve energy output efficiency and ensure operational safety. With the development of information technology, data analysis technology has been able to predict production trends, assess potential risks and improve resource management through advanced algorithms and machine learning methods, enabling the oil and gas industry to manage its resources more effectively, reduce costs and minimize environmental impact.
[0003] Among them, the shale gas production data analysis method is used to analyze and optimize the data in shale gas production, including data collection, processing, analysis and visualization, with the aim of improving the efficiency and safety of shale gas extraction. Through in-depth analysis of production data, key parameters and potential problems in the production process can be identified, operational strategies can be adjusted, and production processes can be optimized. The main purpose is to help engineers and decision makers make more accurate operational decisions based on data-driven insights, improve resource utilization, reduce environmental impact, and increase economic returns.
[0004] The main shortcomings of existing data analysis technologies in actual operations include limited real-time feedback and dynamic optimization capabilities for production data, reliance on preset data processing modes, and lack of rapid response capabilities to real-time changes. For example, in oil and gas production, traditional data analysis methods make it difficult to adjust mining strategies in real time, and the response to emergencies in production is not fast enough, resulting in the failure to maximize production efficiency and resource utilization, which is particularly prominent in the highly variable shale gas production process. Existing technologies are relatively static in risk assessment and management, and are not flexible enough to adjust to rapidly changing geological and market conditions. The shortcomings lead to waste of resources, increased production costs, and even more serious environmental and safety risks due to delayed response to safety issues. Summary of the invention
[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a shale gas production data analysis method and system.
[0006] In order to achieve the above object, the present invention adopts the following technical scheme, a shale gas production data analysis method, comprising the following steps:
[0007] S1: Collect shale gas production data, identify geological parameters of rock hardness, porosity and permeability, record shale gas production parameters such as output, pressure and temperature, and obtain an initialization data set;
[0008] S2: Based on the initialization data set, the relationship between geological parameters and production parameters is analyzed, nodes and edges are constructed, nodes represent differentiated parameters, edges are set according to the influence between parameters, and a network structure model is constructed;
[0009] S3: Probabilistically evaluate the nodes in the network structure model, analyze the conditional probabilities for the connections between the differentiated nodes, construct a complete conditional probability table, analyze the geological parameters and production parameters in the shale gas production data, and obtain a conditional probability network;
[0010] S4: According to the conditional probability network, parameter optimization is performed, key data of nodes are selected for Gibbs sampling, probability values between nodes are adjusted, and network model parameters are updated to obtain an optimized network model;
[0011] S5: using the optimization network model, inputting real-time shale gas production data, predicting changes in shale gas production efficiency, analyzing and evaluating the impact of production parameter changes on production efficiency, and obtaining efficiency change prediction results;
[0012] S6: Use the efficiency change prediction results for real-time production decision-making, identify key risk points in shale gas production, optimize shale gas production efficiency and safety, and obtain a production adjustment plan.
[0013] As a further solution of the present invention, the initialization data set includes the hardness, porosity, permeability of the rock and the production, pressure and temperature of shale gas, the network structure model includes nodes of differentiated parameters and edges set based on the influence relationship between parameters, the conditional probability network includes a conditional probability table between geological parameters and production parameters and the probability correlation between parameters, the optimization network model includes the probability values between nodes adjusted by Gibbs sampling and the updated network parameters, the efficiency change prediction results include the production, pressure and temperature change predictions of shale gas production data, and the production adjustment plan includes optimization measures for production efficiency and safety and identification and response strategies for key risk points.
[0014] As a further solution of the present invention, the steps of collecting shale gas production data, identifying geological parameters of rock hardness, porosity and permeability, and recording shale gas production parameters such as output, pressure and temperature, and obtaining an initialization data set are specifically as follows:
[0015] S101: Collect shale gas production data, monitor rock hardness, porosity and permeability in real time, collect and record relevant data, and classify the data to generate geological parameter data sets;
[0016] S102: Based on the geological parameter data set, the production and pressure of shale gas are monitored, the wellhead temperature is recorded, the data is synchronized, timestamps are aligned and formats are standardized, and a production parameter data set is generated;
[0017] S103: Based on the production parameter data set and in combination with real-time geological parameters, the data is sorted, data deviations are corrected, data formats are unified, and data quality assessment is performed to generate an initialization data set.
[0018] As a further solution of the present invention, based on the initialization data set, the relationship between geological parameters and production parameters is analyzed, nodes and edges are constructed, nodes represent differentiated parameters, edges are set according to the influence between parameters, and the steps of constructing the network structure model are specifically as follows:
[0019] S201: using the initialization data set, extracting geological parameters and production parameters, removing outliers and noise through data processing, integrating similar data items, optimizing the efficiency of data analysis, and generating a data analysis basic set;
[0020] S202: Based on the data analysis basic set, identifying the connection and dependency between geological parameters and production parameters, recording the characteristics of the relationship, including the strength of association and the direction of action, and generating a parameter relationship map;
[0021] S203: Using the parameter relationship mapping, construct nodes representing differentiated parameters and edges showing the influence between parameters, assign weights to each edge, identify the intensity of the effects between differentiated parameters, and generate a network structure model.
[0022] As a further solution of the present invention, the nodes in the network structure model are probability evaluated, conditional probabilities are analyzed for connections between differentiated nodes, a complete conditional probability table is constructed, geological parameters and production parameters in shale gas production data are analyzed, and the steps of obtaining a conditional probability network are specifically as follows:
[0023] S301: Based on the network structure model, probability estimation is performed on each node, and the frequency of the node under differentiated production conditions is calculated by analyzing shale gas production data to generate a node probability distribution table;
[0024] S302: Calculate the conditional probability between nodes using the node probability distribution table, evaluate the impact of the node state change by monitoring and recording the conditional response of the node state change, and generate a conditional probability table between nodes;
[0025] S303: constructing a network including nodes and edges through the inter-node conditional probability table, identifying and recording the probability relationship between the differentiated production parameters and the geological parameters, and generating a conditional probability network.
[0026] As a further solution of the present invention, according to the conditional probability network, parameter optimization is performed, key data of nodes are selected for Gibbs sampling, probability values between nodes are adjusted, and network model parameters are updated to obtain the optimized network model in the following steps:
[0027] S401: Select influential nodes from the conditional probability network, identify key geological and production parameters through the nodes, extract target data sets, and generate key node data sets;
[0028] S402: using the key node data set, adopting the Gibbs sampling method, by randomly replacing the node state and calculating the conditional probability of the new state, cyclically iteratively matching the stable probability distribution, and generating an updated node probability distribution;
[0029] S403: According to the updated node probability distribution, the weights of the edges and the probability values of the nodes in the network model are adjusted, the matching of the parameters of the entire network model with the real-time shale gas production data is checked, and an optimized network model is generated.
[0030] As a further solution of the present invention, the formula of the Gibbs sampling method is as follows:
[0031]
[0032] Among them, P(X i ) is node X i The probability distribution value of X i For nodes that update their status in real time, X -i To divide X i Node states other than x and x -i X i and X -i The state of the control unit is shown in Figure 2, T is the temperature parameter, α is the adjustment coefficient, and β is the bias term.
[0033] As a further solution of the present invention, the optimization network model is used to input real-time shale gas production data, predict changes in shale gas production efficiency, analyze and evaluate the impact of production parameter changes on production efficiency, and obtain the efficiency change prediction results in the following steps:
[0034] S501: inputting real-time shale gas production data into the optimization network model, the data including rock hardness, porosity, production, pressure and temperature, analyzing the data associated with shale gas production, and generating a model input data set;
[0035] S502: simulating the model input data set in the optimization network model, calculating the prediction results under the differentiated production parameter changes, including the expected output and production efficiency of shale gas, and generating efficiency change analysis results;
[0036] S503: Analyze the efficiency change analysis results, evaluate the impact of differentiated production parameters on shale gas production efficiency, predict the production cycle and volatility of shale gas, and generate efficiency change prediction results.
[0037] As a further solution of the present invention, the efficiency change prediction result is used for real-time production decision-making, key risk points in shale gas production are identified, and the production efficiency and safety of shale gas are optimized. The steps of obtaining a production adjustment plan are specifically as follows:
[0038] S601: review multiple data in the efficiency change prediction results, identify key risk parameters associated with shale gas production efficiency and safety, including abnormal pressure and temperature readings, and record key data points to generate a key risk point data set;
[0039] S602: extract risk points from the key risk point data set, perform impact analysis on the risk points, identify the impact of the risk points on shale gas production, record the resulting equipment failures and production delays, and generate risk identification results;
[0040] S603: Based on the risk identification results, formulate shale gas production adjustment measures, including adjusting production equipment settings and working parameters, verifying the implementation of production adjustment measures in shale gas production, optimizing overall production efficiency and safety, and generating a production adjustment plan.
[0041] A shale gas production data analysis system, the shale gas production data analysis system is used to execute the above-mentioned shale gas production data analysis method, the system comprises:
[0042] The data collection module records rock hardness, porosity, permeability, shale gas production, pressure and temperature production parameters to generate an initialization data set;
[0043] The model building module identifies the difference between geological parameters and production parameters based on the initialization data set, sets connecting edges for each pair of parameter relationships, builds a network structure of nodes and edges, and obtains a network structure model;
[0044] The probability evaluation module analyzes the nodes in the network through the network structure model, sets conditional probabilities for the connections between the nodes, constructs a conditional probability table, analyzes the correlation between geological parameters and production parameters, and obtains a conditional probability network;
[0045] The network adjustment module adopts the conditional probability network to extract key data of nodes, calculate the probability values between nodes, and adjust the parameters of the network structure to obtain an optimized network model;
[0046] The efficiency prediction module uses the optimization network model to input real-time shale gas production data, predict changes in production efficiency, analyze the impact of production parameter changes on efficiency, optimize shale gas production efficiency and safety, and obtain a production adjustment plan.
[0047] Compared with the prior art, the advantages and positive effects of the present invention are:
[0048] In the present invention, by deeply analyzing shale gas production data, geological parameters and production parameters are obtained, a network structure model is constructed, probabilistic evaluation of nodes is performed, and network model parameters are optimized to achieve dynamic prediction of production efficiency and risk identification, and real-time optimization and adjustment of key parameters in the shale gas production process can be performed, so that potential risk points and efficiency bottlenecks can be identified in the early stage of production, allowing operators to adjust production strategies in a timely manner, optimize resource management, reduce costs, and improve the overall mining efficiency and safety of shale gas. By constructing and maintaining a network model based on conditional probability, a refined prediction and control method is provided, and model parameters can be adjusted according to real-time data, so that the decision-making process is more accurate and the response speed is faster. The optimization strategy based on real-time data feedback effectively improves resource utilization and economic returns, while reducing environmental impact. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] Figure 1 It is a schematic diagram of the workflow of the present invention;
[0050] Figure 2 This is a detailed flow chart of S1 of the present invention;
[0051] Figure 3 This is a detailed flow chart of S2 of the present invention;
[0052] Figure 4 This is a detailed flow chart of S3 of the present invention;
[0053] Figure 5 This is a detailed flow chart of S4 of the present invention;
[0054] Figure 6 This is a detailed flow chart of S5 of the present invention;
[0055] Figure 7 This is a detailed flow chart of S6 of the present invention;
[0056] Figure 8 It is a system flow chart of the present invention. DETAILED DESCRIPTION
[0057] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0058] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.
[0059] See also Figure 1 The present invention provides a technical solution, a shale gas production data analysis method, comprising the following steps:
[0060] S1: Collect shale gas production data, identify geological parameters such as rock hardness, porosity and permeability, record shale gas production parameters such as output, pressure and temperature, organize the data, correct data deviations, and obtain an initialization data set;
[0061] S2: Based on the initialization data set, analyze the relationship between geological parameters and production parameters, construct nodes and edges, nodes represent differentiated parameters, set edges through the influence between parameters, record the characteristics of the relationship, and construct a network structure model;
[0062] S3: Probabilistically evaluate the nodes in the network structure model, analyze the conditional probabilities for the connections between differentiated nodes, construct a complete conditional probability table, analyze the geological parameters and production parameters in the shale gas production data, and obtain a conditional probability network;
[0063] S4: According to the conditional probability network, perform parameter optimization, select key data of nodes for Gibbs sampling, adjust the probability values between nodes, and update the network model parameters to obtain the optimized network model;
[0064] S5: Using the optimization network model, inputting real-time shale gas production data, calculating through the model, predicting the change of shale gas production efficiency, analyzing and evaluating the impact of production parameter changes on production efficiency, and obtaining the efficiency change prediction results;
[0065] S6: Use the efficiency change prediction results for real-time production decision-making, identify key risk points in shale gas production, analyze the impact of risk points on shale gas production, optimize shale gas production efficiency and safety, and obtain production adjustment plans.
[0066] The initialization data set includes the hardness, porosity, permeability of rocks and the production, pressure and temperature of shale gas. The network structure model includes nodes of differentiated parameters and edges set based on the influence relationship between parameters. The conditional probability network includes the conditional probability table between geological parameters and production parameters and the probabilistic correlation between parameters. The optimized network model includes the probability values between nodes adjusted by Gibbs sampling and the updated network parameters. The efficiency change prediction results include the prediction of production, pressure and temperature changes of shale gas production data. The production adjustment plan includes optimization measures for production efficiency and safety and identification and response strategies for key risk points.
[0067] See also Figure 2 , collect shale gas production data, identify geological parameters of rock hardness, porosity and permeability, record shale gas production parameters such as production volume, pressure and temperature, and obtain the initialization data set in the following steps:
[0068] S101: Collect shale gas production data, monitor rock hardness, porosity and permeability in real time, collect and record related data, and classify the data to generate the geological parameter data set. The execution process is as follows;
[0069] The S101 substep collects shale gas production data, including the hardness, porosity and permeability of the rock, by deploying sensors in key rock layers, monitoring key parameters in real time, and collecting production data associated with them. Data is essential for understanding the behavior of shale layers and optimizing the gas extraction process. The collected data will be systematically classified and organized into geological parameter data sets to provide a basis for subsequent analysis. Comprehensive data collection is to ensure that there is enough information to evaluate the production efficiency and feasibility of shale gas wells, and to generate geological parameter data sets using the formula:
[0070]
[0071] Among them, G param Represents the comprehensive index of geological parameter dataset, H i , P i and K i are the rock hardness, porosity and permeability of the ith data point, respectively, and n is the total number of data points.
[0072] S102: Based on the geological parameter data set, by monitoring the production and pressure of shale gas and recording the wellhead temperature, the data is synchronized, timestamps are aligned and the format is standardized, and the execution process of generating the production parameter data set is as follows;
[0073] Based on the geological parameter data set, sub-step S102 further monitors the production and wellhead pressure of shale gas, and records the wellhead temperature. Production parameters are key indicators for evaluating the efficiency of shale gas extraction. The accuracy and availability of data can be ensured by aligning the timestamps and standardizing the format after data synchronization. The standardized data set provides real-time and accurate decision support for gas field management, helps optimize production operations and prevent potential production problems, and generates a production parameter data set using the formula:
[0074]
[0075] Among them, P sync represents the synchronization index of the production parameter dataset, Q j , T j and ψ j represent the production, temperature and pressure of the jth data point respectively, and m is the total number of data points.
[0076] S103: According to the production parameter data set, combined with the real-time geological parameters, the data is sorted, the data deviation is corrected, the data format is unified and the data quality is evaluated. The execution process of generating the initialization data set is as follows;
[0077] Substep S103 organizes and corrects the data according to the production parameter data set, including correcting data deviations, unifying data formats, and conducting data quality assessments. The comprehensive data processing process ensures that all data meet the quality standards for analysis and reporting. It is used for further analysis and modeling to predict shale gas production and adjustments to development strategies, providing a scientific basis for decision-making for the management team, and generating an initialization data set using the formula:
[0078]
[0079] Among them, D init represents the consistency index of the initialization data set, is the original value of the lth data point, is the corrected value, and k is the total number of data points.
[0080] See also Figure 3 ,Based on the initialization data set, the relationship between geological parameters and production parameters is analyzed, nodes and edges are constructed, nodes represent differentiated parameters, and edges are set by the influence between parameters.,The specific steps of constructing the network structure model are:
[0081] S201: Using the initialization data set, extracting geological parameters and production parameters, removing outliers and noise through data processing, integrating similar data items, optimizing the efficiency of data analysis, and generating the data analysis basic set. The execution process is as follows;
[0082] Substep S201 uses the initialization data set to extract key geological parameters and production parameters, and removes outliers and noise in the data through advanced data processing technology to ensure the accuracy of data analysis. Classify and integrate the data, and group similar data items together to optimize the efficiency of subsequent data analysis. Not only does it clean the data, but it also improves the speed and quality of data processing, and generates a basic set of data analysis, using the formula:
[0083]
[0084] Among them, D base represents the basic set of data analysis, P i is the value of the ith data point, μ P is the mean of the data set, σ P is the standard deviation and n is the total number of data points.
[0085] S202: Based on the data analysis basic set, identify the connection and dependency between geological parameters and production parameters, record the characteristics of the relationship, including the strength of association and the direction of action, and generate the execution process of parameter relationship mapping as follows;
[0086] Substep S202 identifies the connections and dependencies between geological parameters and production parameters based on the data analysis foundation set, and identifies and records the strength and direction of association between parameters through data mining techniques such as correlation analysis and factor analysis. This is essential for understanding how different geological and production factors affect each other, helping to develop more effective resource extraction strategies and generating parameter relationship mappings using the formula:
[0087]
[0088] Among them, R map represents the parameter relationship mapping, G j and Q j represent geological parameters and production parameters respectively, Cov is the covariance, σ is the standard deviation of the parameters, and m is the total number of parameter pairs.
[0089] S203: Using parameter relationship mapping, construct nodes representing differentiated parameters and edges showing the influence between parameters, assign weights to each edge, identify the strength of the effects between differentiated parameters, and generate the network structure model. The execution process is as follows;
[0090] Substep S203 uses parameter relationship mapping to construct nodes representing differentiated parameters and edges showing the influence between parameters, and assigns weights to each edge to quantify the intensity of the effects between different parameters. The network structure model allows visualization and quantification of the interactions between differentiated parameters, providing decision makers with intuitive data interpretation and support for future strategy formulation, and generates a network structure model using the formula:
[0091]
[0092] Among them, N model represents the network structure model, w k is the weight of edge k, R k1, is the correlation coefficient connecting nodes k1 and k2, and o is the total number of edges.
[0093] See also Figure 4 , perform probability evaluation on nodes in the network structure model, analyze conditional probabilities for connections between differentiated nodes, construct a complete conditional probability table, analyze geological parameters and production parameters in shale gas production data, and obtain the conditional probability network in the following steps:
[0094] S301: Based on the network structure model, the probability of each node is estimated, and the frequency of the node under differentiated production conditions is calculated by analyzing the shale gas production data, and the execution process of generating the node probability distribution table is as follows;
[0095] Sub-step S301 estimates the probability of each node based on the network structure model and analyzes the shale gas production data, especially considering the frequency of node performance under differentiated production conditions. Through statistical methods such as frequency analysis and probability density function estimation, the probability distribution of each node is accurately calculated and recorded. It provides basic data for further conditional probability analysis, can reveal the behavior patterns of various geological parameters under different production conditions, and generate a node probability distribution table using the formula:
[0096]
[0097] Among them, P node represents the probability distribution of nodes, is an indicator function, which takes the value 1 when node i is in state x and 0 otherwise, and N is the total number of observations.
[0098] S302: Calculate the conditional probability between nodes using the node probability distribution table, evaluate the impact of the node state change by monitoring and recording the conditional response of the node state change, and generate the execution flow of the conditional probability table between nodes as follows;
[0099] Substep S302 uses the node probability distribution table to calculate the conditional probability between nodes, which is performed by monitoring and recording the conditional response of node state changes. By analyzing the conditional response data, the impact of node state changes can be evaluated and how different production parameters interact with each other can be understood. Revealing how the state of one node affects the state of another node under certain conditions is crucial for predicting system behavior and formulating intervention strategies. The conditional probability table between nodes is generated using the formula:
[0100]
[0101] Among them, P(A|B) is the conditional probability of node A's state given node B's state, P(A∩B) is the joint probability that two nodes are in a specific state at the same time, and P(B) is the marginal probability of node B's state.
[0102] S303: constructing a network including nodes and edges through the conditional probability table between nodes, identifying and recording the probability relationship between the differentiated production parameters and the geological parameters, and generating the execution flow of the conditional probability network is as follows;
[0103] In substep S303, a conditional probability network including nodes and edges is constructed through the conditional probability table between nodes. The network model not only shows the direct relationship between nodes, but also records the probability relationship between differentiated production parameters and geological parameters in detail. It can reveal the potential dependencies and impact paths in the shale gas production system, which is of great significance for optimizing production strategies and preventing production risks. The conditional probability network is generated using the formula:
[0104]
[0105] Among them, C net represents the conditional probability network, w k is the weight of edge k, P(A k |B k ) is the conditional probability between nodes, and m is the total number of edges.
[0106] See also Figure 5 , according to the conditional probability network, parameter optimization is performed, key data of nodes are selected for Gibbs sampling, probability values between nodes are adjusted, and network model parameters are updated. The specific steps to obtain the optimized network model are as follows:
[0107] S401: Select influential nodes from the conditional probability network, identify key geological and production parameters through the nodes, extract the target data set, and generate the key node data set. The execution process is as follows;
[0108] Substep S401 selects influential nodes from the conditional probability network, and the nodes represent key geological and production parameters. By identifying the nodes, we can have a deeper understanding of the factors in the production process that have a decisive impact on the production and safety of shale gas, which can be used for further statistical analysis and model optimization to ensure that the monitoring and control measures of the production process are targeted at the most important variables, and generate key node data sets using the formula:
[0109]
[0110] Among them, K set represents the key node dataset, P iis the probability value of node i, θ is the selected influence threshold, and N is the total number of nodes in the network.
[0111] S402: Using the key node data set, adopting the Gibbs sampling method, by randomly replacing the node state and calculating the conditional probability of the new state, cyclically iteratively matching the stable probability distribution, the execution process of generating an updated node probability distribution is as follows;
[0112] Substep S402 uses the key node data set and the Gibbs sampling method to update the probability distribution. Gibbs sampling is a sampling technique based on the Markov chain Monte Carlo method. It achieves a stable probability distribution by randomly replacing the node state and calculating the conditional probability of the new state. The model is allowed to estimate the conditional distribution of each parameter under the joint distribution of all relevant variables, iteratively update and optimize the probability estimate of the network model, and generate an updated node probability distribution.
[0113] The formula for the Gibbs sampling method is as follows:
[0114]
[0115] Among them, P(X i ) is node X i The probability distribution value of X i For nodes that update their status in real time, X -i To divide X i Node states other than x and x -i X i and X -i The state of the control unit is shown in Figure 2, T is the temperature parameter, α is the adjustment coefficient, and β is the bias term.
[0116] The execution flow is as follows:
[0117] Calculate the weighted joint probability in the numerator, weighting the original joint probability P(X i =x,X -i =x -i ) and multiply it by T to adjust the influence, plus the bias β to avoid the case where the probability is zero. The denominator calculates the sum of the weighted joint probabilities under all states. Similarly, α and T are applied, and β is added. The weighted numerator is divided by the denominator to obtain the adjusted conditional probability. The values of the parameters α and β are determined through an optimization process, such as using cross-validation or historical data fitting to adjust the coefficients to ensure optimal sampling efficiency and accuracy.
[0118] S403: According to the updated node probability distribution, the weights of the edges and the probability values of the nodes in the network model are adjusted, and the matching of the parameters of the entire network model with the real-time shale gas production data is checked. The execution process of generating the optimized network model is as follows;
[0119] Substep S403 adjusts the edge weights and node probability values in the network model according to the updated node probability distribution, and ensures that the parameters of the network model more accurately reflect the real-time shale gas production data through adjustment. This helps to improve the accuracy of prediction, enable the network model to reflect changes in production conditions in real time, and provide more accurate support for operational decisions, and generate an optimized network model using the formula:
[0120]
[0121] Among them, W ij is the edge weight between nodes i and j, α is the normalization coefficient, P x i ,x j ) is the joint probability of nodes i and j, P(x i ) and P x j ) are their respective marginal probabilities.
[0122] See also Figure 6 , using the optimization network model, inputting real-time shale gas production data, predicting the changes in shale gas production efficiency, analyzing and evaluating the impact of production parameter changes on production efficiency, and obtaining the efficiency change prediction results in the following steps:
[0123] S501: Input the real-time shale gas production data into the optimization network model, the data includes the hardness, porosity, production, pressure and temperature of the rock, analyze the data related to shale gas production, and generate the execution process of the model input data set as follows;
[0124] The S501 sub-step inputs real-time shale gas production data into the optimization network model. The data includes key geological and production parameters, such as rock hardness, porosity, production, pressure and temperature. Data is the basis for analyzing shale gas production efficiency and formulating strategies. Input data into the model, analyze the relationship between data through the network model, ensure that all relevant data are comprehensively considered, and reflect the interaction and influence in simulation and prediction, and generate the model input data set. The formula used is:
[0125]
[0126] Among them, M input represents the model input data set, f(x i ) is a data processing function that processes the i-th input parameter x i , n is the total number of input parameters.
[0127] S502: The model input data set is simulated in the optimization network model to calculate the prediction results under the differential production parameter changes, including the expected production and production efficiency of shale gas, and the execution process of generating the efficiency change analysis results is as follows;
[0128] Substep S502 simulates the model input data set in the optimized network model to calculate the prediction results under different production parameter changes, including the expected production and production efficiency of shale gas. The simulation process uses the statistical and machine learning algorithms of the network model to predict future production and efficiency changes, helping decision makers understand how different parameter changes affect production efficiency, so as to formulate more effective production strategies and generate efficiency change analysis results. The formula used is:
[0129]
[0130] Among them, E predict represents the predicted efficiency change, β is the model coefficient, P x j |x -j ) is the conditional probability of parameter j given the parameter state, and m is the total number of parameters.
[0131] S503: Analyze the efficiency change analysis results, evaluate the impact of differentiated production parameters on shale gas production efficiency, predict the production cycle and volatility of shale gas, and generate the efficiency change prediction results. The execution process is as follows;
[0132] Sub-step S503 analyzes the efficiency change analysis results, evaluates how different production parameters affect the production efficiency of shale gas, predicts the production cycle and volatility of shale gas, and uses statistical analysis methods such as regression analysis and time series analysis to predict future production trends and fluctuations, providing a scientific basis for adjusting production plans and responding to market changes, and generating efficiency change prediction results. The formula used is:
[0133] P output =γ·∫P efficiency (t)dt
[0134] Among them, P output represents the predicted output of production efficiency, γ is the adjustment factor, P efficiency (t) is the efficiency probability function at time t.
[0135] See also Figure 7 , the efficiency change prediction results are used for real-time production decision-making, key risk points in shale gas production are identified, and the production efficiency and safety of shale gas are optimized. The specific steps for obtaining the production adjustment plan are as follows:
[0136] S601: Review multiple data in the efficiency change prediction results, identify key risk parameters associated with shale gas production efficiency and safety, including abnormal pressure and temperature readings, and record key data points. The execution process of generating a key risk point data set is as follows;
[0137] Substep S601 reviews multiple data in the efficiency change prediction results, especially focusing on key risk parameters that are closely related to shale gas production efficiency and safety. Advanced data analysis techniques, such as anomaly detection algorithms, identify key risk points such as abnormal pressure and temperature readings. The identification of risk points is critical because it indicates potential safety hazards or production efficiency issues. Record key data points and summarize them into a key risk point data set using the formula:
[0138]
[0139] Among them, R key represents the key risk point dataset, r i is the risk index value of the i-th data point, τ is the set risk threshold, I is an indicator function, when r i It takes the value of 1 when it exceeds τ, otherwise it is 0, and n is the total number of monitoring points.
[0140] S602: extract risk points from the key risk point data set, perform impact analysis on the risk points, identify the impact of the risk points on shale gas production, record the resulting equipment failures and production delays, and generate the risk identification results. The execution process is as follows;
[0141] Substep S602 extracts risk points from the key risk point data set and conducts in-depth impact analysis on risk points. By analyzing how risk points affect shale gas production, potential factors leading to equipment failures, production delays, and safety accidents can be identified. This helps develop specific prevention and mitigation strategies to reduce risks and improve production efficiency, and generates risk identification results using the formula:
[0142]
[0143] Among them, F risk represents the risk identification result, C j is the impact cost of the j-th risk point, C total is the total impact cost, m is the total number of risk points, I impact It is an indicator of the severity of the impact of the risk point.
[0144] S603: Based on the risk identification results, formulate shale gas production adjustment measures, including adjusting production equipment settings and working parameters, checking the implementation of production adjustment measures in shale gas production, optimizing overall production efficiency and safety, and generating the execution process of the production adjustment plan as follows;
[0145] Sub-step S603 formulates adjustment measures for shale gas production based on the risk identification results, including adjusting the settings and operating parameters of production equipment to adapt to the identified risks. Through adjustments, the overall production efficiency and safety can be optimized to ensure the stability and continuity of the production process. The implementation of production adjustment measures will be recorded in detail, and the performance in actual production will be checked to verify the effectiveness and adaptability, and generate a production adjustment plan using the formula:
[0146]
[0147] Among them, S adjust represents the effect of the production adjustment plan, Δx k is the parameter x k The adjustment amount, X k is the parameter x k is the original value of , α is the adjustment efficiency coefficient, and p is the total number of adjustment parameters.
[0148] See also Figure 8 , a shale gas production data analysis system, the shale gas production data analysis system is used to execute the above shale gas production data analysis method, the system includes:
[0149] The data collection module records rock hardness, porosity, permeability, shale gas production, pressure and temperature production parameters to generate an initialization data set;
[0150] The model building module identifies the differences between geological parameters and production parameters based on the initialization data set, sets connecting edges for each pair of parameter relationships, builds a network structure of nodes and edges, and obtains a network structure model;
[0151] The probability assessment module analyzes the nodes in the network through the network structure model, sets conditional probabilities for the connections between nodes, constructs a conditional probability table, analyzes the correlation between geological parameters and production parameters, and obtains a conditional probability network;
[0152] The network adjustment module uses a conditional probability network to extract key data of nodes, calculate the probability values between nodes, and adjust the parameters of the network structure to obtain an optimized network model;
[0153] The efficiency prediction module uses an optimized network model to input real-time shale gas production data, predict changes in production efficiency, analyze the impact of changes in production parameters on efficiency, optimize shale gas production efficiency and safety, and obtain production adjustment plans.
[0154] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.
Claims
1. A shale gas production data analysis method, characterized in that: The following steps are involved: Collect shale gas production data, identify geological parameters of rock hardness, porosity and permeability, record shale gas production parameters such as output, pressure and temperature, and obtain an initial data set; Based on the initialization data set, the relationship between geological parameters and production parameters is analyzed, nodes and edges are constructed, nodes represent differentiated parameters, edges are set according to the influence between parameters, and a network structure model is constructed; Probabilistically evaluating the nodes in the network structure model, analyzing the conditional probabilities for the connections between the differentiated nodes, constructing a complete conditional probability table, analyzing the geological parameters and production parameters in the shale gas production data, and obtaining a conditional probability network; According to the conditional probability network, parameter optimization is performed, key data of nodes are selected for Gibbs sampling, probability values between nodes are adjusted, and network model parameters are updated to obtain an optimized network model; The optimization network model is used to input real-time shale gas production data, predict changes in shale gas production efficiency, analyze and evaluate the impact of production parameter changes on production efficiency, and obtain efficiency change prediction results; The efficiency change prediction results are used for real-time production decision-making, key risk points in shale gas production are identified, shale gas production efficiency and safety are optimized, and production adjustment plans are obtained.
2. The shale gas production data analysis method according to claim 1, characterized in that: The initialization data set includes the hardness, porosity, permeability of the rock and the production, pressure and temperature of shale gas. The network structure model includes nodes of differentiated parameters and edges set based on the influence relationship between parameters. The conditional probability network includes the conditional probability table between geological parameters and production parameters and the probability correlation between parameters. The optimized network model includes the probability values between nodes adjusted by Gibbs sampling and the updated network parameters. The efficiency change prediction results include the prediction of production, pressure and temperature changes of shale gas production data. The production adjustment plan includes optimization measures for production efficiency and safety and identification and response strategies for key risk points.
3. The shale gas production data analysis method according to claim 1, characterized in that: The steps to collect shale gas production data, identify geological parameters such as rock hardness, porosity and permeability, and record shale gas production parameters such as production volume, pressure and temperature are as follows: Collect shale gas production data, monitor rock hardness, porosity and permeability in real time, collect and record relevant data, and classify the data to generate geological parameter data sets; Based on the geological parameter data set, the production and pressure of shale gas are monitored, the wellhead temperature is recorded, the data is synchronized, timestamps are aligned and formats are standardized, and a production parameter data set is generated; According to the production parameter data set, combined with real-time geological parameters, the data is sorted, data deviations are corrected, the data format is unified, and data quality assessment is performed to generate an initialization data set.
4. The shale gas production data analysis method according to claim 1, characterized in that: Based on the initialization data set, the relationship between geological parameters and production parameters is analyzed, nodes and edges are constructed, nodes represent differentiated parameters, edges are set according to the influence between parameters, and the steps of constructing the network structure model are as follows: Using the initialization data set, extracting geological parameters and production parameters, removing outliers and noise through data processing, integrating similar data items, optimizing the efficiency of data analysis, and generating a basic set for data analysis; Based on the data analysis basis set, identifying the connection and dependency between geological parameters and production parameters, recording the characteristics of the relationship, including the strength of association and the direction of action, and generating a parameter relationship map; By using the parameter relationship mapping, nodes representing differentiated parameters and edges showing the influence between parameters are constructed, a weight is assigned to each edge, the intensity of the effect between differentiated parameters is identified, and a network structure model is generated.
5. The shale gas production data analysis method according to claim 1, characterized in that: The steps of probabilistically evaluating the nodes in the network structure model, analyzing the conditional probabilities for the connections between the differentiated nodes, constructing a complete conditional probability table, analyzing the geological parameters and production parameters in the shale gas production data, and obtaining the conditional probability network are as follows: Based on the network structure model, probability estimation is performed on each node, and the frequency of nodes under differentiated production conditions is calculated by analyzing shale gas production data to generate a node probability distribution table; Using the node probability distribution table, calculate the conditional probability between nodes, evaluate the impact of node state changes by monitoring and recording the conditional response of node state changes, and generate a conditional probability table between nodes; Through the inter-node conditional probability table, a network including nodes and edges is constructed, the probability relationship between the differentiated production parameters and the geological parameters is identified and recorded, and a conditional probability network is generated.
6. The shale gas production data analysis method according to claim 1, characterized in that: According to the conditional probability network, parameter optimization is performed, key data of nodes are selected for Gibbs sampling, probability values between nodes are adjusted, and network model parameters are updated to obtain the optimized network model in the following steps: Select influential nodes from the conditional probability network, identify key geological and production parameters through the nodes, extract target data sets, and generate key node data sets; Using the key node data set and adopting the Gibbs sampling method, by randomly replacing the node state and calculating the conditional probability of the new state, cyclically iteratively matching the stable probability distribution, an updated node probability distribution is generated; According to the updated node probability distribution, the weights of the edges and the probability values of the nodes in the network model are adjusted, the matching of the parameters of the entire network model with the real-time shale gas production data is checked, and an optimized network model is generated.
7. The shale gas production data analysis method according to claim 6, characterized in that: The formula for the Gibbs sampling method is as follows: Among them, P(X i ) is node X i The probability distribution value of X i For nodes that update their status in real time, X -i To divide X i Node states other than x and x -i X i and X -i The state of the control unit is shown in Figure 2, T is the temperature parameter, α is the adjustment coefficient, and β is the bias term.
8. The shale gas production data analysis method according to claim 1, characterized in that: The steps of adopting the optimization network model, inputting real-time shale gas production data, predicting changes in shale gas production efficiency, analyzing and evaluating the impact of production parameter changes on production efficiency, and obtaining efficiency change prediction results are as follows: Inputting real-time shale gas production data into the optimization network model, the data including rock hardness, porosity, production, pressure and temperature, analyzing the data associated with shale gas production, and generating a model input data set; The model input data set is simulated in the optimization network model to calculate the prediction results under the differential production parameter changes, including the expected production and production efficiency of shale gas, and generate efficiency change analysis results; Analyze the efficiency change analysis results, evaluate the impact of differentiated production parameters on shale gas production efficiency, predict the production cycle and volatility of shale gas, and generate efficiency change prediction results.
9. The shale gas production data analysis method according to claim 1, characterized in that: The efficiency change prediction results are used for real-time production decision-making, key risk points in shale gas production are identified, and the production efficiency and safety of shale gas are optimized. The specific steps for obtaining a production adjustment plan are as follows: Reviewing multiple data in the efficiency change prediction results, identifying key risk parameters associated with shale gas production efficiency and safety, including abnormal pressure and temperature readings, and recording key data points to generate a key risk point data set; Extract risk points from the key risk point data set, perform impact analysis on the risk points, identify the impact of the risk points on shale gas production, record the resulting equipment failures and production delays, and generate risk identification results; Based on the risk identification results, formulate shale gas production adjustment measures, including adjusting production equipment settings and operating parameters, verify the implementation of production adjustment measures in shale gas production, optimize overall production efficiency and safety, and generate a production adjustment plan.
10. A shale gas production data analysis system, characterized in that: According to the shale gas production data analysis method according to any one of claims 1 to 9, the system comprises: The data collection module records rock hardness, porosity, permeability, shale gas production, pressure and temperature production parameters to generate an initialization data set; The model building module identifies the difference between geological parameters and production parameters based on the initialization data set, sets connecting edges for each pair of parameter relationships, builds a network structure of nodes and edges, and obtains a network structure model; The probability evaluation module analyzes the nodes in the network through the network structure model, sets conditional probabilities for the connections between the nodes, constructs a conditional probability table, analyzes the correlation between geological parameters and production parameters, and obtains a conditional probability network; The network adjustment module adopts the conditional probability network to extract key data of nodes, calculate the probability values between nodes, and adjust the parameters of the network structure to obtain an optimized network model; The efficiency prediction module uses the optimization network model to input real-time shale gas production data, predict changes in production efficiency, analyze the impact of changes in production parameters on efficiency, optimize shale gas production efficiency and safety, and obtain a production adjustment plan.
Citation Information
Cited By
Detection method and system for pore structure of shale gas reservoir
CN120105965A
