A geothermal well real-time data acquisition and processing system and method

Abstract

The present invention relates to the field of geothermal well data processing technology, specifically a real-time data acquisition and processing system and method for geothermal wells. The method comprises: obtaining processing parameters, performing characteristic analysis on the processing parameters, constructing a regression model, and constructing a dynamic causal network between the geothermal well processing parameters based on the regression model; correcting the processing parameters based on the causal network and updating the corrected processing parameters; calculating input costs and energy output; obtaining importance scores for the input costs and energy output, and calculating an economic benefit index based on the importance scores; obtaining multidimensional data related to the geothermal well and calculating a safety benefit index; and optimizing the control of the geothermal well using hierarchical decision rules based on the economic benefit index and the safety benefit index. The present invention integrates dual-dimensional assessments of economic benefit and safety risk to improve the one-sidedness of geothermal well assessments; and utilizes hierarchical decision rules to achieve accurate and comprehensive decision-making and optimize decision-making capabilities.

Description

Technical Field

[0001] The present invention relates to the technical field of geothermal well data processing, and in particular to a geothermal well real-time data acquisition and processing system and method. Background Art

[0002] As the scale of geothermal resource development expands, the safe and economical and efficient operation of geothermal wells is of vital importance. During the operation of geothermal wells, parameters such as temperature, pressure, and flow are interrelated and change dynamically. Minor anomalies may cause production accidents or waste of resources. Traditional methods are difficult to fully reflect the operating status.

[0003] At the same time, with the intensified competition in the energy market, there is an urgent need to accurately evaluate the economic benefits of geothermal well mining. Therefore, real-time data collection and processing of geothermal wells is very necessary. By collecting multi-dimensional data and conducting in-depth analysis, safety risk warning and economic benefit optimization can be achieved, ensuring the sustainable development and utilization of geothermal resources and enhancing industry competitiveness. Summary of the Invention

[0004] The purpose of the present invention is to address the problems existing in the background technology and to propose a geothermal well real-time data acquisition and processing system and method.

[0005] The technical solution of the present invention is a method for real-time data collection and processing of geothermal wells, comprising the following steps:

[0006] S1. Real-time acquisition of temperature, pressure, flow, water level, pH value and conductivity to obtain real-time parameters, and pre-processing of the real-time parameters to obtain processing parameters;

[0007] S2. Analyze the characteristics of the processing parameters, preliminarily determine the range of the lag order, select the optimal lag order for the processing parameters using the Akaike Information Criterion and construct a regression model;

[0008] S3. Based on the regression model, the causal relationship between the various treatment parameters of the geothermal well is hypothesized, the causal relationship is determined, and a dynamic causal network between the geothermal well treatment parameters is constructed;

[0009] S4. Modifying the processing parameters based on the causal network and updating the modified processing parameters;

[0010] S5. Obtaining mining cost-related data, performing calculations based on the mining cost-related data to obtain input costs; obtaining energy output-related data, performing calculations based on the energy output-related data to obtain energy output;

[0011] S6. Obtaining importance scores for input costs and energy output, calculating a scoring matrix based on the importance scores, performing a consistency check on the scoring matrix, performing data processing on the scoring matrix to obtain a first processing result, and calculating an economic benefit index based on the first processing result;

[0012] S7. Acquire multidimensional data related to the geothermal well, perform pressure risk modeling, temperature risk modeling, and corrosion risk modeling, output corresponding risk values, perform data processing on the risk values ​​to obtain a second processing result, and calculate a safety benefit index based on the second processing result;

[0013] S8. Based on the economic benefit index and safety benefit index, the improved TOPSIS algorithm and hierarchical decision rules are used to optimize the control of geothermal wells.

[0014] Preferably, constructing a regression model includes constructing an unrestricted model and a restricted model.

[0015] Preferably, the method for assuming a causal relationship between various treatment parameters of a geothermal well includes:

[0016] By formula Calculate the statistic F;

[0017] Where, is the residual sum of squares of the restricted model; is the residual sum of squares of the unrestricted model; T is the number of samples; P is the optimal lag order;

[0018] Perform data analysis on the statistic F. If the statistic F is greater than , then determine the independent variable For dependent variables There is a causal effect; otherwise, determine the independent variable For dependent variables No causal influence.

[0019] Preferably, S41, constructing a 6×6 zero matrix W(0), wherein the rows and columns of the matrix correspond to the six types of processing parameters for geothermal well monitoring, and setting the diagonal elements of the matrix to 1;

[0020] Using the formula Perform weight update;

[0021] Where, represents the causal weight of processing parameter i on processing parameter j at time t; λ is the forgetting factor; It represents the normalized result of the statistic F obtained by the causal relationship test between the processing parameters i and j at the current moment.

[0022] Preferably, data analysis is performed on the causal relationship weights to obtain analysis results and to execute directed edge construction decisions based on the analysis results, including when When >0.5, a directed edge processing parameter i is established to point to processing parameter j, and the edge weight is ;when When ≤0.5, no operation is performed; the causal network is visualized based on the directed edges between the processing parameters.

[0023] Preferably, the method for performing consistency check on the scoring matrix includes calculating the maximum eigenvalue of the scoring matrix and calculating the consistency index based on the maximum eigenvalue. The consistency index CI is calculated by the formula Calculated; among them, is the maximum eigenvalue; m is the order of the scoring matrix;

[0024] According to the order of the scoring matrix, the random consistency index is obtained, the consistency ratio is calculated according to the random consistency index and the consistency index, the consistency ratio is subjected to data analysis to obtain the analysis results, and the importance score is judged according to the analysis results to determine whether it passes the consistency test.

[0025] Preferably, S81, taking the time window as a unit, taking the economic benefit index and the safety benefit index as decision indicators, constructing a decision matrix, standardizing the decision matrix to obtain a standard matrix, and determining positive and negative ideal solutions based on the standard matrix.

[0026] Preferably, S82, each row of the standard matrix corresponds to an evaluation sample of a time window, the Euclidean distance between the evaluation sample of each time window and the positive and negative ideal solutions is calculated, and the evaluation score Scroe is calculated based on the Euclidean distance between the evaluation sample and the positive and negative ideal solutions.

[0027] Preferably, the hierarchical decision rules are as follows:

[0028] Compare the evaluation score Score with the preset first safety threshold, second safety threshold, third safety threshold and fourth safety threshold respectively to obtain a comparison result;

[0029] If the evaluation score Score is not less than the first safety threshold, a first instruction is generated, including maintaining the current operation and periodically recording data;

[0030] If the evaluation score Score is less than the first safety threshold and not less than the second safety threshold, a second instruction is generated, including starting a parameter optimization algorithm;

[0031] If the evaluation score Score is less than the second safety threshold and not less than the third safety threshold, a third instruction is generated, including analyzing the cost structure and planning equipment upgrades;

[0032] If the evaluation score Score is less than the third safety threshold and not less than the fourth safety threshold, a fourth instruction is generated, including increasing the frequency of safety monitoring and deploying temporary sensors;

[0033] If the evaluation score Score is less than the fourth safety threshold, a fifth instruction is generated, including automatically triggering the shutdown process and starting the emergency plan.

[0034] The present invention also discloses a geothermal well real-time data acquisition and processing system, which applies the above-mentioned geothermal well real-time data acquisition and processing method, specifically comprising:

[0035] The first data acquisition and preprocessing module is used to acquire temperature, pressure, flow, water level, pH value and conductivity in real time to obtain real-time parameters, and preprocess the real-time parameters to obtain processing parameters;

[0036] The data analysis module is used to analyze the characteristics of the processing parameters, preliminarily determine the value range of the lag order, select the optimal lag order for the processing parameters using the Akaike Information Criterion, and build a regression model;

[0037] A causal network construction module is used to make causal relationship assumptions among various geothermal well treatment parameters based on the regression model, determine the causal relationship, and construct a dynamic causal network among geothermal well treatment parameters;

[0038] The second data acquisition and calculation module is used to acquire mining cost related data, perform calculations based on the mining cost related data to obtain input costs; acquire energy output related data, perform calculations based on the energy output related data to obtain energy output;

[0039] A parameter correction module, used to correct the processing parameters based on the causal network and update the corrected processing parameters;

[0040] a third data acquisition and calculation module, configured to obtain importance scores for input costs and energy output, calculate a scoring matrix based on the importance scores, perform a consistency check on the scoring matrix, perform data processing on the scoring matrix to obtain a first processing result, and calculate an economic benefit index based on the first processing result;

[0041] a fourth data acquisition and calculation module, configured to acquire multidimensional data related to the geothermal well, perform pressure risk modeling, temperature risk modeling, and corrosion risk modeling, output corresponding risk values, perform data processing on the risk values ​​to obtain a second processing result, and calculate a safety benefit index based on the second processing result;

[0042] The decision-making control module is used to optimize the control of geothermal wells based on the economic benefit index and the safety benefit index using the improved TOPSIS algorithm and hierarchical decision rules.

[0043] Compared with the prior art, the above technical solution of the present invention has the following beneficial technical effects:

[0044] The present invention uses time-varying weights to characterize the dynamic changes in parameter causal relationships, thereby solving the problem of lagging parameter correlation analysis. It constructs a multidimensional evaluation index system that integrates the dual-dimensional evaluation of economic benefits and safety risks to improve the one-sidedness of single-index evaluation of geothermal wells. It utilizes an improved TOPSIS algorithm and hierarchical decision-making rules to achieve accurate and comprehensive decision-making and optimize decision-making capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a method block diagram of embodiment 1 proposed by the present invention. DETAILED DESCRIPTION

[0046] Example 1, as Figure 1 As shown, the present invention proposes a method for real-time data collection and processing of geothermal wells, comprising the following steps:

[0047] S1. Real-time acquisition of temperature, pressure, flow, water level, pH value, and conductivity to obtain real-time parameters, and pre-processing of the real-time parameters to obtain processed parameters. It should be noted that six different types of sensors can be vertically deployed at the wellhead and five key depths downhole to obtain a three-dimensional monitoring network, respectively acquiring six types of sensor data.

[0048] S11, preprocessing includes randomly selecting a time series point of the parameter time series as a feature, and randomly selecting a split value between the maximum and minimum values ​​of the feature; for example, for temperature data within a certain period of time, randomly selecting a temperature value at a time point as a division basis, and dividing the data set into two parts greater than the value and less than the value;

[0049] S12, recursively divide the data and generate nodes until each node contains only one sample or reaches the maximum depth of the tree, and construct an isolated tree; each tree is constructed independently, and a total of 100 isolated trees are generated; it should be noted that the above operation of randomly selecting features and split values ​​is repeated for the two parts of the divided data, and recursive division is performed; this process is to construct a binary tree until each node contains only one sample or reaches the pre-set maximum depth of the tree, which is set as , N is the number of samples; each tree is constructed independently, and a total of 100 isolated trees are generated;

[0050] S13. Calculate the path length h(x) of sample x in the isolation tree: the number of path edges from the root node to the leaf node containing x. It should be noted that this refers to the number of path edges from the root node to the leaf node containing the sample. The path length reflects the location characteristics of the sample in the data distribution. Generally, in areas with dense data distribution, the path length of the sample is short; while in areas with sparse data distribution, the path length of the sample is long.

[0051] S14. Calculate the average path length based on the path lengths of the samples. The formula for calculating the average path length is c(n) = 2H(n-1) - (2(n-1) / n), where H is the harmonic number and n is the number of samples. It should be noted that the average path length is used to normalize the path lengths of the samples to facilitate comparison in datasets of different sizes.

[0052] S15. Calculate the anomaly score of sample x. The anomaly score is calculated by the formula It is calculated that s(x, n)∈[0,1], when s>0.7, it is determined to be an outlier, and 0.7 is the score threshold optimized based on historical data;

[0053] S16. Use the deep adjacent point interpolation algorithm to repair the data of the outliers. The repair formula is: Where, is the repaired sample data; and are normal data at time t-1 and time t+1 respectively; for example, in the temperature data at a certain depth, if the data at time t is determined to be an abnormal value, and the data at time t-1 and time t+1 are normal, the average value of the data at these two adjacent times is used to replace the abnormal value;

[0054] S2. Analyze the characteristics of the processing parameters to preliminarily determine the range of lag orders. Use the Akaike Information Criterion to select the optimal lag order for the processing parameters and construct a regression model. For example, pressure changes may take some time to affect flow, so it is necessary to consider lags within a certain time range.

[0055] The lag order can range from 1 to 5. The AIC values ​​for each processing parameter are calculated for lag orders of 1, 2, 3, 4, and 5. During the calculation process, the corresponding regression models are constructed, including unrestricted models and restricted models. The criterion value is determined based on statistics such as the residual sum of squares of the model.

[0056] Constructing a regression model includes constructing an unrestricted model and a restricted model;

[0057] For example, for processing parameters and , when the lag order is 1, construct and Unrestricted model with lagged terms and only For the restricted model of the lag term, the AIC value at this time is calculated, and then the lag order is changed in turn, and the above modeling and calculation process is repeated. The AIC values ​​under different lag orders are compared, and the lag order that minimizes the AIC value is selected as the optimal lag order p. For example, if the AIC value reaches the minimum when the lag order is 3, then in the subsequent regression modeling for the causal test of this pair of processing parameters, the model with a lag order of 3 is used, that is, the influence of the processing parameter values ​​of the previous 3 time steps on the processing parameter value at the current moment is considered;

[0058] The expression of the unrestricted model is as follows:

[0059] ;

[0060] Where, represents the measured value of the dependent variable treatment parameter j at time t; Dependent variable The sum of its own lag terms; p is the optimal lag order; k is the lag time step; is the autoregressive coefficient, reflecting the dependent variable The degree and direction of the impact of the past k time steps on the current moment value; Indicates independent variables The sum of the lagged terms of ; is the cross-regression coefficient, which reflects the effect of the independent variable on the dependent variable in the past k time steps. The intensity and nature of the impact of the current moment value; represents the unrestricted random error term; it should be noted that the independent variable It is possible that the dependent variable Another processing parameter that produces a causal effect;

[0061] Autoregressive coefficient and the cross-regression coefficient The following exemplary supplements are made:

[0062] like >0, indicating that the dependent variable The value of the previous moment has a positive effect on the current moment; if <0, indicating a negative inhibitory effect;

[0063] like >0, indicating that the independent variable The larger the value two time steps ago, the greater the dependent variable at the current moment. The value of will also tend to be larger; if <0, then the dependent variable The value of will also tend to be smaller;

[0064] The restricted model is expressed as follows:

[0065] Where, represents the restricted random error term;

[0066] S3. Based on the regression model, we hypothesize the causal relationship between the various treatment parameters of geothermal wells, determine the causal relationship, and construct a dynamic causal network between the geothermal well treatment parameters. This provides a basis for accurately characterizing the interaction mechanism between the treatment parameters, and further serves the analysis of the geothermal well operation status and the assessment of economic benefits and safety risks:

[0067] Methods for making causal hypothesis and determining causal relationship include:

[0068] By formula Calculate the statistic F;

[0069] Where, is the residual sum of squares of the restricted model; is the residual sum of squares of the unrestricted model; T is the number of samples; P is the optimal lag order;

[0070] Perform data analysis on the statistic F. If the statistic F is greater than , then determine the independent variable For dependent variables There is a causal effect; otherwise, determine the independent variable For dependent variables No causal influence;

[0071] S31. Construct a 6×6 zero matrix W(0), where the rows and columns of the matrix correspond to the six types of processing parameters for geothermal well monitoring, and set the diagonal elements of the matrix to 1. It should be noted that this means that each processing parameter has a complete causal relationship with itself, that is, the current processing parameter value is 100% affected by its own value at the previous moment. This is set based on the basic characteristics of the processing parameter time series data.

[0072] S32, using formula Perform weight update;

[0073] Where, represents the causal weight of processing parameter i on processing parameter j at time t; λ is the forgetting factor; represents the normalized result of the statistic F obtained by the causal relationship test between the processing parameters i and j at the current moment. It should be noted that the forgetting factor can be set to 0.98 to reflect the dominant role of recent data in the weight calculation, so that the model pays more attention to the causal relationship between the processing parameters at the current moment, while giving a certain attenuation to historical data. The value range of is [0,1], which is used to quantify the causal influence of processing parameter i on processing parameter j at the current moment;

[0074] S33, perform data analysis on the causal relationship weights, obtain analysis results and execute directed edge construction decisions based on the analysis results, including when When >0.5, a directed edge processing parameter i is established to point to processing parameter j, and the edge weight is ;when When ≤0.5, no operation is performed; the causal network is visualized based on the directed edges between the processing parameters;

[0075] S4. Modify the processing parameters based on the causal network and update the modified processing parameters to optimize the accuracy of the benefit evaluation; for example, the mining flow Make corrections to get the corrected flow , and update the electricity cost based on the modified flow, including determining the relevant mining flow according to the causal network Directed edges, when the mining flow in the directed edges When the influential processing parameters change, the corresponding edge weight is marked as the flow-affecting edge weight, and the maximum flow-affecting edge weight is obtained and marked as , mining traffic through the following formula Make corrections:

[0076] ; Where d is the flow correction coefficient, which is determined by fitting historical data;

[0077] S5. Obtaining mining cost-related data, performing calculations based on the mining cost-related data to obtain input costs; obtaining energy output-related data, performing calculations based on the energy output-related data to obtain energy output;

[0078] Mining cost-related data include fluid density, mining flow, pump head, pump efficiency, motor efficiency, basic maintenance costs, sensor repair costs, sensor repair times, recharge flow, and water resource fees:

[0079] S51. Calculation based on mining cost related data, including the formula Calculate the electricity cost ;

[0080] Where, is the fluid density, in kg / m³; H is the pump head, in meters; g is the acceleration due to gravity; is the pump efficiency; is the motor efficiency; is the electricity price, in yuan / kWh;

[0081] S52, through the formula Calculate maintenance costs ;

[0082] Where, As basic maintenance fee; is the maintenance cost of the jth type of sensor; is the number of maintenance times of the jth type sensor;

[0083] S53, through the formula Calculate the cost of water resources Where, is the recharge flow; Water resource fee;

[0084] S54. Electricity cost , maintenance costs and water resource costs Perform sum calculation to obtain input cost;

[0085] S55, energy output E is calculated by the formula Calculated;

[0086] Where c is the specific heat capacity of the fluid; is the temperature difference between the wellhead temperature and the underground constant temperature layer; is the energy conversion efficiency; it should be noted that the temperature of the underground isothermal layer is the average underground temperature;

[0087] S6. Obtain importance scores for input costs and energy output, calculate a scoring matrix based on the importance scores, perform a consistency check on the scoring matrix, perform data processing on the scoring matrix to obtain a first processing result, and calculate an economic benefit index based on the first processing result. It should be noted that the importance scores are obtained based on a 1-9 scale, with increasing numbers representing increasing importance. The scoring matrix can be obtained by having experts in different industries perform importance scoring and then performing mean calculation.

[0088] For example, the scoring matrix A is as follows:

[0089] ; represents the importance score of cost i relative to cost j, For the same cost and equal importance, the score is 1; It means cost 1 is more important than cost 2, and the score is 3; It means that cost 1 is more important than cost 3, and the score is 5. According to the reciprocal rule, , thereby determining the scoring matrix A;

[0090] S61. The method for performing consistency test on the scoring matrix includes calculating the maximum eigenvalue of the scoring matrix and calculating the consistency index based on the maximum eigenvalue. The consistency index CI is calculated by the formula Calculated; among them, is the maximum eigenvalue; m is the order of the scoring matrix;

[0091] It should be noted that the maximum eigenvalue of the rating matrix can be calculated by first normalizing each column of the rating matrix to obtain a normalized matrix, then summing the normalized matrix row by row to obtain a row vector, and finally normalizing the row vector to obtain an approximate value of the weight vector and calculate the maximum eigenvalue. ;

[0092] S62. Obtain a random consistency index based on the order of the scoring matrix, calculate a consistency ratio based on the random consistency index and the consistency index, perform data analysis on the consistency ratio to obtain an analysis result, and judge the importance score based on the analysis result to determine whether the consistency test passes;

[0093] The method for data analysis of consistency ratio includes: calculating the consistency ratio by the formula CR = RI / CI. If CR < 0.1, it is considered that the scoring matrix A has satisfactory consistency, indicating that the importance score is reasonably obtained. If CR ≥ 0.1, it is necessary to re-obtain the importance score and adjust the scoring matrix A until it passes the consistency test.

[0094] S63. Processing the scoring matrix to obtain a processing result. The method for calculating the economic benefit index based on the processing result includes:

[0095] Normalize the eigenvector of the rating matrix to obtain a list of benefit weights , which is taken as the first processing result;

[0096] Mark the benefit weight elements in the benefit weight list as , h is the benefit weight element number, h is a positive integer; the economic benefit index is calculated based on the equity weight list , economic benefit index Calculated by the following formula:

[0097] ;

[0098] is the economic benefit index; is the current cost, is the historical optimal value; it should be noted that the economic benefit index It can comprehensively reflect the economic performance of current geothermal well mining. The larger the value, the better the economic benefit.

[0099] S7. Acquire multidimensional data related to the geothermal well, perform pressure risk modeling, temperature risk modeling, and corrosion risk modeling, output corresponding risk values, perform data processing on the risk values ​​to obtain a second processing result, and calculate a safety benefit index based on the second processing result;

[0100] S71. The expression of the pressure risk model is as follows:

[0101] ;

[0102] Where, is the pressure risk value; P is the real-time pressure; The median value of normal pressure range; is the maximum design pressure; is the pressure change rate; It is the security response time window;

[0103] The expression of the temperature risk model is as follows:

[0104] ;

[0105] Where, is the temperature risk value; T is the real-time temperature; The median of the normal temperature range; Design maximum temperature; is the duration coefficient; it should be noted that when the overheating time is greater than 10 minutes, the duration coefficient is 1.5; when the overheating time is not more than 10 minutes, the duration coefficient is 1;

[0106] The corrosion risk model is expressed as follows:

[0107] ;

[0108] Where, is the corrosion risk value; age is the well age, in years; E is the real-time conductivity; is the normal value of conductivity; H is the real-time pH value; is the neutral pH value; α1, α2 and α3 are weight coefficients, which are fitted by historical corrosion data;

[0109] Stress Risk Value , Temperature Risk Value and corrosion risk value Combining as a second processing result;

[0110] S72. The method for calculating the safety benefit index based on the second processing result includes: objectively assigning weights to the risk values ​​using the entropy weight method, and calculating a safety weight list based on the degree of variation of each risk value. ;

[0111] The risk values ​​are marked as , mark the security weight element in the security weight list as , z is the security weight element number, z is a positive integer; the security benefit index is calculated based on the security weight list , safety benefit index Calculated by the following formula:

[0112] It should be noted that the safety benefit index ,The higher the value, the better the security;

[0113] S8. Based on the economic benefit index and safety benefit index, the improved TOPSIS algorithm and hierarchical decision rules are used to optimize the control of geothermal wells;

[0114] Methods for decision-making and control modeling based on economic benefit index and safety benefit index include:

[0115] S81. Using the time window as a unit, the economic benefit index and the safety benefit index are used as decision indicators to construct a decision matrix. The decision matrix is ​​standardized to obtain a standard matrix. The positive and negative ideal solutions are determined based on the standard matrix. It should be noted that the positive ideal solution is a vector composed of the maximum values ​​of each indicator, and the negative ideal solution is a vector composed of the minimum values ​​of each indicator.

[0116] For example, the decision matrix is ​​as follows: ;

[0117] Obtain the maximum economic benefit index and the maximum safety benefit index in the decision matrix and mark them as and , and Construct a positive ideal solution; obtain the minimum value of the economic benefit index and the minimum value of the safety benefit index in the decision matrix, and mark them as and , and Constitute a negative ideal solution;

[0118] S82. Each row of the standard matrix corresponds to an evaluation sample of a time window. The Euclidean distance between the evaluation sample and the positive and negative ideal solutions of each time window is calculated. The evaluation score Scroe is calculated based on the Euclidean distance between the evaluation sample and the positive and negative ideal solutions.

[0119] The Euclidean distance between the evaluation result and the positive ideal solution and the negative ideal solution is calculated using the following formulas:

[0120] ;

[0121] ;

[0122] By formula Calculate the evaluation score Score of the evaluation sample;

[0123] The hierarchical decision rules are as follows:

[0124] Compare the evaluation score Score with the preset first safety threshold, second safety threshold, third safety threshold and fourth safety threshold respectively to obtain a comparison result;

[0125] If the evaluation score Score is not less than the first safety threshold, a first instruction is generated, including maintaining the current operation and periodically recording data;

[0126] If the evaluation score Score is less than the first safety threshold and not less than the second safety threshold, a second instruction is generated, including starting a parameter optimization algorithm;

[0127] If the evaluation score Score is less than the second safety threshold and not less than the third safety threshold, a third instruction is generated, including analyzing the cost structure and planning equipment upgrades;

[0128] If the evaluation score Score is less than the third safety threshold and not less than the fourth safety threshold, a fourth instruction is generated, including increasing the frequency of safety monitoring and deploying temporary sensors;

[0129] If the evaluation score Score is less than the fourth safety threshold, a fifth instruction is generated, including automatically triggering the shutdown process and starting the emergency plan.

[0130] In a second embodiment, a geothermal well real-time data acquisition and processing system proposed by the present invention is applied to a geothermal well real-time data acquisition and processing method proposed in the first embodiment, and specifically includes:

[0131] The first data acquisition and preprocessing module is used to acquire temperature, pressure, flow, water level, pH value and conductivity in real time to obtain real-time parameters, and preprocess the real-time parameters to obtain processing parameters;

[0132] The data analysis module is used to analyze the characteristics of the processing parameters, preliminarily determine the value range of the lag order, select the optimal lag order for the processing parameters using the Akaike Information Criterion, and build a regression model;

[0133] A causal network construction module is used to make causal relationship assumptions among various geothermal well treatment parameters based on the regression model, determine the causal relationship, and construct a dynamic causal network among geothermal well treatment parameters;

[0134] A parameter correction module, used to correct the processing parameters based on the causal network and update the corrected processing parameters;

[0135] The second data acquisition and calculation module is used to acquire mining cost related data, perform calculations based on the mining cost related data to obtain input costs; acquire energy output related data, perform calculations based on the energy output related data to obtain energy output;

[0136] a third data acquisition and calculation module, configured to obtain importance scores for input costs and energy output, calculate a scoring matrix based on the importance scores, perform a consistency check on the scoring matrix, perform data processing on the scoring matrix to obtain a first processing result, and calculate an economic benefit index based on the first processing result;

[0137] a fourth data acquisition and calculation module, configured to acquire multidimensional data related to the geothermal well, perform pressure risk modeling, temperature risk modeling, and corrosion risk modeling, output corresponding risk values, perform data processing on the risk values ​​to obtain a second processing result, and calculate a safety benefit index based on the second processing result;

[0138] The decision-making control module is used to optimize the control of geothermal wells based on the economic benefit index and the safety benefit index using the improved TOPSIS algorithm and hierarchical decision rules.

[0139] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A method for real-time data collection and processing of geothermal wells, characterized in that: The following steps are involved: S1. Real-time acquisition of temperature, pressure, flow, water level, pH value and conductivity to obtain real-time parameters, and pre-processing of the real-time parameters to obtain processing parameters; S2. Analyze the characteristics of the processing parameters, preliminarily determine the range of the lag order, select the optimal lag order for the processing parameters using the Akaike Information Criterion and construct a regression model; S3. Based on the regression model, the causal relationship between the various treatment parameters of the geothermal well is hypothesized, the causal relationship is determined, and a dynamic causal network between the geothermal well treatment parameters is constructed; S4. Modifying the processing parameters based on the causal network and updating the modified processing parameters; S5. Obtaining mining cost-related data, performing calculations based on the mining cost-related data to obtain input costs; obtaining energy output-related data, performing calculations based on the energy output-related data to obtain energy output; S6. Obtaining importance scores for input costs and energy output, calculating a scoring matrix based on the importance scores, performing a consistency check on the scoring matrix, performing data processing on the scoring matrix to obtain a first processing result, and calculating an economic benefit index based on the first processing result; S7. Acquire multidimensional data related to the geothermal well, perform pressure risk modeling, temperature risk modeling, and corrosion risk modeling, output corresponding risk values, perform data processing on the risk values ​​to obtain a second processing result, and calculate a safety benefit index based on the second processing result; S71. The expression of the pressure risk model is as follows: ； Where, is the pressure risk value; P is the real-time pressure; The median value of normal pressure range; is the maximum design pressure; is the pressure change rate; It is the security response time window; The expression of the temperature risk model is as follows: ； Where, is the temperature risk value; T is the real-time temperature; The median of the normal temperature range; Design maximum temperature; is the duration coefficient; The corrosion risk model is expressed as follows: ； Where, is the corrosion risk value; age is the well age, in years; E is the real-time conductivity; is the normal value of conductivity; H is the real-time pH value; is the neutral pH value; α1, α2 and α3 are weight coefficients, which are fitted by historical corrosion data; Stress Risk Value , Temperature Risk Value and corrosion risk value Combining as a second processing result; S72. The method for calculating the safety benefit index based on the second processing result includes: objectively assigning weights to the risk values ​​using the entropy weight method, and calculating a safety weight list based on the degree of variation of each risk value. ; The risk values ​​are marked as , mark the security weight element in the security weight list as , z is the security weight element number, z is a positive integer; the security benefit index is calculated based on the security weight list , safety benefit index Calculated by the following formula: ； S8. Based on the economic benefit index and safety benefit index, the improved TOPSIS algorithm and hierarchical decision rules are used to optimize the control of geothermal wells.

2. A geothermal well real-time data acquisition and processing method according to claim 1, characterized in that: Building a regression model includes building an unrestricted model and a restricted model.

3. A geothermal well real-time data acquisition and processing method according to claim 2, characterized in that: Methods for making causal assumptions about the various treatment parameters of geothermal wells include: By formula Calculate the statistic F; Where, is the residual sum of squares of the restricted model; is the residual sum of squares of the unrestricted model; T is the number of samples; p is the optimal lag order; Perform data analysis on the statistic F. If the statistic F is greater than , then determine the independent variable For dependent variables There is a causal effect; otherwise, determine the independent variable For dependent variables No causal influence.

4. A geothermal well real-time data acquisition and processing method according to claim 3, characterized in that: S41, constructing a 6×6 zero matrix W(0), wherein the rows and columns of the matrix correspond to the six types of processing parameters for geothermal well monitoring, and setting the diagonal elements of the matrix to 1; Using the formula Perform weight update; Where, represents the causal weight of processing parameter i on processing parameter j at time t; λ is the forgetting factor; It represents the normalized result of the statistic F obtained by the causal relationship test between the processing parameters i and j at the current moment.

5. A geothermal well real-time data acquisition and processing method according to claim 4, characterized in that: Perform data analysis on causal relationship weights, obtain analysis results and make directed edge construction decisions based on the analysis results, including when When >0.5, a directed edge processing parameter i is established to point to processing parameter j, and the edge weight is ;when When ≤0.5, no operation is performed; the causal network is visualized based on the directed edges between the processing parameters.

6. A geothermal well real-time data acquisition and processing method according to claim 5, characterized in that: The method for consistency testing of the scoring matrix includes calculating the maximum eigenvalue of the scoring matrix and calculating the consistency index based on the maximum eigenvalue. The consistency index CI is calculated by the formula Calculated; among them, is the maximum eigenvalue; m is the order of the scoring matrix; According to the order of the scoring matrix, the random consistency index is obtained, the consistency ratio is calculated according to the random consistency index and the consistency index, the consistency ratio is subjected to data analysis to obtain the analysis results, and the importance score is judged according to the analysis results to determine whether it passes the consistency test.

7. A geothermal well real-time data acquisition and processing method according to claim 6, characterized in that: S81. Taking the time window as the unit, the economic benefit index and the safety benefit index are used as decision-making indicators to construct a decision matrix, standardize the decision matrix to obtain a standard matrix, and determine the positive and negative ideal solutions based on the standard matrix.

8. A geothermal well real-time data acquisition and processing method according to claim 7, characterized in that: S82. Each row of the standard matrix corresponds to an evaluation sample of a time window. The Euclidean distance between the evaluation sample and the positive and negative ideal solutions of each time window is calculated. The evaluation score Scroe is calculated based on the Euclidean distance between the evaluation sample and the positive and negative ideal solutions.

9. A geothermal well real-time data acquisition and processing method according to claim 8, characterized in that: The hierarchical decision rules are as follows: Compare the evaluation score Score with the preset first safety threshold, second safety threshold, third safety threshold and fourth safety threshold respectively to obtain a comparison result; If the evaluation score Score is not less than the first safety threshold, a first instruction is generated, including maintaining the current operation and periodically recording data; If the evaluation score Score is less than the first safety threshold and not less than the second safety threshold, a second instruction is generated, including starting a parameter optimization algorithm; If the evaluation score Score is less than the second safety threshold and not less than the third safety threshold, a third instruction is generated, including analyzing the cost structure and planning equipment upgrades; If the evaluation score Score is less than the third safety threshold and not less than the fourth safety threshold, a fourth instruction is generated, including increasing the frequency of safety monitoring and deploying temporary sensors; If the evaluation score Score is less than the fourth safety threshold, a fifth instruction is generated, including automatically triggering the shutdown process and starting the emergency plan.

10. A geothermal well real-time data acquisition and processing system, applied to a geothermal well real-time data acquisition and processing method according to any one of claims 1 to 9, characterized in that: Specifically include: The first data acquisition and preprocessing module is used to acquire temperature, pressure, flow, water level, pH value and conductivity in real time to obtain real-time parameters, and preprocess the real-time parameters to obtain processing parameters; The data analysis module is used to analyze the characteristics of the processing parameters, preliminarily determine the value range of the lag order, select the optimal lag order for the processing parameters using the Akaike Information Criterion, and build a regression model; The causal network construction module makes causal relationship assumptions among various geothermal well treatment parameters based on the regression model, determines the causal relationship, and constructs a dynamic causal network among geothermal well treatment parameters; A parameter correction module, used to correct the processing parameters based on the causal network and update the corrected processing parameters; The second data acquisition and calculation module is used to acquire mining cost related data, perform calculations based on the mining cost related data to obtain input costs; acquire energy output related data, perform calculations based on the energy output related data to obtain energy output; a third data acquisition and calculation module, configured to obtain importance scores for input costs and energy output, calculate a scoring matrix based on the importance scores, perform a consistency check on the scoring matrix, perform data processing on the scoring matrix to obtain a first processing result, and calculate an economic benefit index based on the first processing result; a fourth data acquisition and calculation module, configured to acquire multidimensional data related to the geothermal well, perform pressure risk modeling, temperature risk modeling, and corrosion risk modeling, output corresponding risk values, perform data processing on the risk values ​​to obtain a second processing result, and calculate a safety benefit index based on the second processing result; The decision-making control module is used to optimize the control of geothermal wells based on the economic benefit index and the safety benefit index using the improved TOPSIS algorithm and hierarchical decision rules.

Images