A comprehensive geothermal geochemical discrimination method, system, and medium

Abstract

This invention provides a comprehensive geothermal geochemical discrimination method, system, and medium. The method includes: obtaining basic prior scores for conceptual models of each candidate geothermal system; for each candidate model, calculating its mechanism matching score with actual samples from at least two different physicochemical mechanism dimensions, and aggregating the matching scores of each mechanism dimension into a mechanism consistency factor; using the mechanism consistency factor to perform soft correction on the basic prior scores, and adjusting the retention degree of the basic prior scores through a preset fidelity coefficient to obtain the final score of each candidate model; re-ranking each candidate model according to the final score, and outputting the ranking results and the corresponding geothermal system discrimination conclusion. This invention, through multi-mechanism coupling, soft correction control, and quantitative scoring design, improves the scientificity, accuracy, and practicality of geothermal system discrimination, reduces the subjectivity of discrimination, adapts to various geothermal system types, and can provide precise technical support for geothermal resource exploration and development.

Description

Technical Field

[0001] This invention relates to the field of geothermal geochemical comprehensive interpretation technology, specifically providing a geothermal geochemical comprehensive discrimination method, system, and medium. Background Technology

[0002] Current geothermal geochemical interpretation processes typically rely on temperature scale calculations and empirical graphical interpretations. Traditional graphical interpretations, such as Piper diagrams and Giggenbach Na-K-Mg triangulation diagrams, while visually reflecting water type classifications or local equilibrium states, struggle to generate comprehensive regional and systematic conclusions. In recent years, some cross-map weighting or machine learning methods have been introduced into geothermal data analysis, improving automation to some extent. However, most remain purely data-driven or rely on simple empirical rule matching, lacking effective constraints on the actual deep-seated thermo-water-rock reaction mechanisms.

[0003] During the ascent of geothermal fluids, the dramatic changes in temperature, pressure, and mixing environment often result in seemingly contradictory geochemical characteristics in the same water sample. For example: (1) Cation geothermal scales (such as the Na-K scale) give a strong signal of the presence of medium- to high-temperature geothermal reservoirs at depth, but the same sample falls completely into the "immature water" region on the Na-K-Mg triangulation, indicating that the fluid has not yet reached equilibrium with the surrounding rocks; (2) Stable isotope δ 18 O and δD showed significant positive oxygen shift, strongly suggesting that the fluid underwent deep circulation and long-term water-rock isotope exchange. However, the saturation index (SI) of key high-temperature minerals (such as quartz and feldspar) calculated by software such as PHREEQC was far below zero, which does not support the conclusion of high-temperature equilibrium. (3) In the prospective area with excellent geological structure and lithological conditions, the surface spring water sample was mixed with a large amount of shallow cold water, resulting in a serious underestimation of the traditional weighted scoring result based on ion concentration, thus misjudging the potential area as a "no prospective area".

[0004] Furthermore, while general multi-attribute decision-making methods (MADM), such as TOPSIS, VIKOR, and weighted summation (WSM), are widely used in the field of multi-index comprehensive ranking, these general methods treat all evaluation indicators as homogeneous "attribute values" and mechanically weight them. They cannot distinguish between "appearance data indicators" (such as ion concentration) affected by shallow mixing and "underlying mechanism constraints" (such as mineral saturation index combinations and isotope fractionation-temperature coupling) reflecting deep equilibrium. Moreover, these methods cannot achieve hierarchical integration and dynamic control between prior geological knowledge and measured mechanism matching information, and they are also difficult to incorporate into the physicochemical mechanism verification process unique to geothermal systems. Therefore, they are unable to cope with the multiple solutions caused by fluid mixing and multi-stage evolution. Summary of the Invention

[0005] In order to overcome the above-mentioned defects, the present invention is proposed to provide a solution or at least a partial solution to the above-mentioned problems.

[0006] In a first aspect, the present invention provides a comprehensive geothermal geochemical discrimination method, comprising the following steps: obtaining basic prior scores for each candidate geothermal system conceptual model; for each candidate model, calculating the mechanism matching score between the candidate model and the actual sample from at least two different physicochemical mechanism dimensions, and aggregating the matching scores of each mechanism dimension into a mechanism consistency factor; using the mechanism consistency factor to perform soft correction on the basic prior scores to obtain the final score of each candidate model, wherein the soft correction refers to the final score being a weighted combination of the retained portion of the basic prior scores and the portion of the basic prior scores adjusted by the mechanism consistency factor, and the direct retention ratio of the basic prior scores is controlled by a preset fidelity coefficient; re-ranking each candidate model according to the final score, and outputting the ranking result and the corresponding geothermal system discrimination conclusion.

[0007] Preferably, obtaining the basic prior scores for each candidate geothermal system conceptual model includes: acquiring hydrochemical data and tectonic lithology data of the geothermal fluid to be measured, and preprocessing them; performing graphical interpretation and temperature scale calculation on the preprocessed data to obtain graphical criteria and temperature scale values; matching the graphical criteria and temperature scale values ​​with the expected characteristics of each candidate model, and assigning each candidate model a corresponding basic prior score according to the degree of matching.

[0008] Preferably, the mechanism consistency factor M m The calculation formula is: M m =Σ k (λ k *q k *g k,m *B k ) / Σ k (λ k *q k *B k ) Where k represents different physicochemical mechanism dimensions, and λ k For the prior weights of the mechanism term, q k For the mass coefficient, g k,m B represents the matching score of candidate model m on k. k This is an indicator of availability.

[0009] Preferably, calculating the mechanism matching score of the candidate model on a certain mechanism dimension includes: Based on the geothermal system type of the candidate model, determine the expected features of the model in the mechanism dimension; compare the actual mechanism calculation values ​​with the expected features; and assign a corresponding matching score based on the degree of agreement between the comparison and the expected features.

[0010] Preferably, the physicochemical mechanism dimension k includes: mineral saturation index combination constraints, cation exchange constraints, isotope fractionation-geothermal consistency constraints, and tectonic / lithological a priori constraints.

[0011] Preferably, the mineral saturation index combination constraint matches the saturation index combination pattern of one or more minerals among quartz, calcite, and anhydrite with the target range of the candidate model; the cation exchange constraint compares the changing directions of CAI-I, CAI-II, and related ion ratios with the expected exchange process of the candidate model; the isotope fractionation-geothermal consistency constraint uses δ 18 The O drift amplitude, δD characteristics, and the coordination temperature range of the candidate model are coupled for analysis; the structural / lithological prior constraints are achieved by matching the candidate model with existing structural and lithological background data, expert annotations of structural conditions, or default structural adaptation values, wherein the structural and lithological background data is regional geological survey information independent of the basic prior score.

[0012] Preferably, the soft correction is calculated in the following form: R m =S m *(η+(1-η)*M m ) Among them, R m For the final score, S m Based on the prior score, η is the fidelity coefficient of the basic prior score, 0 < η < 1, M m This is the mechanism consistency factor.

[0013] Preferably, the discrimination conclusion includes at least a candidate model ranking list, the best matching geothermal system type, the reservoir temperature estimation range, the mechanism consistency level, and supplementary test suggestions; wherein, the best matching geothermal system type and the reservoir temperature estimation range are determined based on the candidate model with the highest final score, the mechanism consistency level is determined based on the value of the mechanism consistency factor, and when the matching score of a certain mechanism dimension is lower than a set threshold, supplementary test suggestions for that mechanism dimension are generated.

[0014] Secondly, the present invention provides a geothermal geochemical integrated discrimination system, comprising: an acquisition unit for acquiring basic prior scores of conceptual models of each candidate geothermal system; a mechanism consistency unit for calculating, for each candidate model, a mechanism matching score between the candidate model and the actual sample from at least two different physicochemical mechanism dimensions, and aggregating the matching scores of each mechanism dimension into a mechanism consistency factor; a correction unit for performing soft correction on the basic prior scores using the mechanism consistency factor to obtain a final score for each candidate model, wherein the soft correction is that the final score is a weighted combination of the retained portion of the basic prior score and the portion of the basic prior score adjusted by the mechanism consistency factor, and the direct retention ratio of the basic prior score is controlled by a preset fidelity coefficient; a sorting unit for re-sorting each candidate model according to the final score; and an output unit for outputting the sorting results and the corresponding geothermal system discrimination conclusion.

[0015] Thirdly, the present invention provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; and when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the geothermal geochemical integrated discrimination method.

[0016] The beneficial effects of this invention are as follows: 1. This invention achieves multi-dimensional cross-validation by calculating the mechanism matching scores between candidate models and actual samples from at least two different physicochemical mechanism dimensions and aggregating them into a mechanism consistency factor. This method covers key evolutionary processes of geothermal systems and is adaptable to geothermal systems of different origins and types (such as magmatic, fracture, and sedimentary geothermal systems). It eliminates the need to adjust the methodological framework for specific geothermal system types, effectively avoiding the limitations of single-dimensional discrimination and significantly improving the method's versatility and applicability.

[0017] 2. This invention introduces a mass coefficient q into the calculation of the mechanism consistency factor. k Availability Indicator B k When data in a certain dimension is missing or the data quality is inconsistent, it can automatically adjust the effective participation level of each dimension (when B...). k When the value is 0, this dimension does not participate in aggregation. It also assigns lower weights to low-quality data, which solves the rigid limitation in existing technologies that require all data to be complete before operation, and ensures reliable discrimination under the condition of incomplete data.

[0018] 3. This invention uses a mechanism consistency factor to softly correct the basic prior score and regulates the retention of prior information in the final score through a preset basic prior score fidelity coefficient. This mechanism ensures that the final score retains the reference value of geological priors while incorporating mechanistic matching information from actual samples. It solves the technical problem of the disconnect between prior information and mechanism analysis in existing technologies, achieving a homogeneous fusion of geochemical evidence and structural priors, and significantly improving the flexibility and adaptability of the discrimination method.

[0019] 4. This invention clarifies the consistency factor of the mechanism and the quantitative calculation formula of the final score, and standardizes the process of obtaining the basic prior score (including data preprocessing, graphic interpretation, temperature scale calculation, matching and scoring, etc.). This makes the entire discrimination process quantifiable, traceable and verifiable, effectively reducing the influence of human subjective factors, ensuring that different users can obtain consistent discrimination results using this method, and significantly improving the repeatability and application value of the method.

[0020] 5. In complex scenarios with severe mixing of shallow cold water, traditional weighted scoring based on ion concentration is easily suppressed by the dilution effect, leading to high-potential geothermal areas being misclassified as "foregroundless areas." This invention, through multi-mechanism cross-validation and soft correction mechanisms, can identify the true characteristics of masked deep geothermal systems, ensuring that the classification and ranking are not affected by systematic shifts caused by shallow disturbances, thus providing a more reliable model foundation for subsequent reservoir temperature estimation and geothermal potential assessment. Attached Figure Description

[0021] The disclosure of this invention will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. Furthermore, similar numbers in the drawings are used to denote similar components, wherein: Figure 1 This is a schematic flowchart of a geothermal geochemical integrated discrimination method according to an embodiment of the present invention. Detailed Implementation

[0022] Some embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0023] like Figure 1 As shown, this invention provides a comprehensive geothermal geochemical discrimination method, comprising the following steps: Step S1: Obtain the basic prior scores of the conceptual models of each candidate geothermal system.

[0024] In this embodiment, the candidate geothermal system conceptual model refers to a variety of possible geothermal system theoretical models (such as different reservoir modes and circulation path models, such as hydrothermal convection type, conduction type, magma heat source type, and fracture-controlled heat type) constructed based on the preliminary understanding of regional geothermal geological background, structural characteristics, reservoir burial depth, and fluid source.

[0025] The basic prior score is an initial quantitative score for the rationality of each candidate model based on prior geological information, and the value range is [0,1].

[0026] Furthermore, the number of candidate models, the scoring rules for the basic prior score, and the score range can all be flexibly set according to the degree of exploration and the richness of geological data in the study area. The higher the degree of exploration and the more sufficient the prior information, the more refined the discrimination of the basic prior score can be set. Those skilled in the art can optimize it as needed.

[0027] In one embodiment, step S1, obtaining the basic prior scores of each candidate geothermal system conceptual model, includes: Step S11: Obtain hydrochemical data and tectonic lithology data of the geothermal fluid to be tested, and perform preprocessing.

[0028] In this embodiment, the hydrochemical data of the geothermal fluid to be tested refers to the geochemical indicators such as the main ion concentration, trace element content, and stable isotope ratio obtained by collecting geothermal fluid samples from geothermal wells, hot springs, hot springs, etc., and testing them. These are the core original data reflecting the source, evolution, and mixing characteristics of geothermal fluids. Structural and lithological data: refers to geological exploration data such as geological structure maps, fault distribution, stratigraphic lithology, reservoir lithology, and caprock conditions of the study area; Specifically, the raw hydrochemical data and tectonic lithology data are standardized, including removing abnormal test values, supplementing missing data, unifying data units, calibrating coordinates, and vectorizing lithological and tectonic information to eliminate analytical biases caused by data errors and inconsistent formats. Step S12: Perform graphical interpretation and temperature scale calculation on the preprocessed data to obtain graphical criteria and temperature scale values.

[0029] In this embodiment, graphical interpretation refers to the use of classic discriminant diagrams in the field of geothermal geochemistry (such as Piper's triangular diagram, Na-K-Mg triangular diagram, Cl-SO4-HCO3 triangular diagram, δ...). 18 (O-δD isotope diagrams, etc.) are used to perform mapping analysis on the pre-processed hydrochemical data, and the geothermal fluid type, mixing degree, water-rock reaction degree and other characteristics are determined by the location of the plot points. Temperature scale calculation: refers to the calculation of hydrochemical data using classic geothermal temperature scale formulas such as silica temperature scale and cation temperature scale (Na-K, Na-K-Ca, etc.) to obtain temperature scale values ​​that reflect the temperature level of geothermal system reservoirs; Graphical criteria: refers to qualitative and semi-quantitative judgment conclusions obtained through graphical interpretation, such as the fluid being classified as a deep circulating hot water or shallow cold water mixture, the water-rock reaction being sufficient / weak, and belonging to a convective / conductive geothermal system, etc. Temperature scale value: refers to the temperature measurement result of the geothermal reservoir obtained by temperature scale calculation, which reflects the temperature field characteristics of the geothermal system. The temperature scale value range of different types of geothermal systems has significant differences.

[0030] Step S13: Match the graphical criteria and temperature scale values ​​with the expected features of each candidate model, and assign each candidate model a corresponding basic prior score according to the degree of matching.

[0031] In this embodiment, the expected characteristics of the candidate model refer to the geochemical and temperature field characteristics that each candidate geothermal system conceptual model should theoretically possess, including the graphical landing point range, temperature scale value range, fluid mixing characteristics, water-rock reaction degree and other preset theoretical characteristics, which are the benchmark for matching and scoring. Matching degree: refers to the degree of agreement between the measured graphical criteria, temperature scale values ​​and the expected features of the candidate model. For example, it can be divided into levels such as perfect match, good match, average match and no match. The higher the matching degree, the stronger the prior rationality of the candidate model.

[0032] Alternatively, in an alternative implementation, when sufficient hydrochemical data is lacking in the early stages of exploration, prior scores can be directly assigned based on expert geological knowledge, or a combination of statistical classification scores, expert prior scores, and graphical temperature scale scores can be used. Of course, the basic prior scores for obtaining conceptual models of each candidate geothermal system are not limited to the above two methods; the specific scoring acquisition path can be flexibly selected based on data availability.

[0033] Step S2: For each candidate model, calculate the mechanism matching score between the candidate model and the actual sample from at least two different physicochemical mechanism dimensions, and aggregate the matching scores of each mechanism dimension into a mechanism consistency factor.

[0034] In this embodiment, different physicochemical mechanism dimensions refer to quantitative evaluation dimensions that focus on characterizing different operating mechanisms of the geothermal system. Each dimension constrains the candidate model from different angles such as thermodynamic balance, kinetic exchange, isotope tracing, and geological background. Although some dimensions may share underlying physical factors such as temperature, the evaluation objects, input parameters, and judgment logic of each dimension are distinguished from each other. Mechanism matching score: refers to the quantitative score of the degree of agreement between the theoretical prediction value of the candidate model and the measured value of the geothermal chemical sample under a single mechanism dimension. The higher the score, the more closely the model matches the actual geothermal fluid characteristics in this mechanism dimension. Mechanism consistency factor: This refers to the comprehensive mechanism matching quantitative index obtained by aggregating the matching scores of multiple different mechanism dimensions. It is a core indicator that reflects the overall conformity of the candidate model to the geothermal physicochemical evolution law. The higher the value, the stronger the mechanism consistency between the model and the measured data.

[0035] Furthermore, the selection of mechanism dimensions can be flexibly adjusted according to the type of geothermal system and the type of measured geochemical indicators in the study area, and the scoring aggregation method can be weighted according to the importance of each mechanism dimension.

[0036] In this embodiment, the mechanism consistency factor is aggregated using a weighted arithmetic mean. This aggregation method has the advantages of linear superposition and strong interpretability: the contribution of each dimension to the final factor can be directly traced through the weight coefficients, making it easier for geothermal technicians to understand and verify the discriminative contribution of each dimension. Meanwhile, in cases where some dimensions are missing (B... k When =0), simply remove the corresponding term; no additional adjustments to the formula structure are required.

[0037] Furthermore, the prior weight λ of the mechanism term k The determination of weights can be achieved using one or more of the following methods: (1) Analytic Hierarchy Process (AHP), where geothermal experts compare the relative importance of each mechanism dimension pairwise to construct a judgment matrix, and calculate the weights of each dimension after a consistency test; (2) Entropy weighting method, where the weight of each dimension is assigned based on the information entropy of each dimension in historical case data, with a lower information entropy (i.e., a higher degree of differentiation between different candidate models); and (3) Expert experience weighting, where technical personnel with rich experience in geothermal geochemistry directly assign weights. The weighting method can be flexibly selected based on the actual conditions of the study area and the existing data.

[0038] In one embodiment, the mechanism consistency factor M m The calculation formula is: M m =Σ k (λ k *q k *g k,m *B k ) / Σ k (λ k *q k *B k ) Where k represents different physicochemical mechanism dimensions, and λ k For the prior weights of the mechanism term, q k For the mass coefficient, gk,m B represents the matching score of candidate model m on k. k This is an indicator of availability.

[0039] It should be noted that although the aforementioned physicochemical mechanism dimensions focus on different geochemical processes, under specific geothermal system conditions, some dimensions may exhibit a certain degree of correlation (collinearity). When the number of mechanism dimensions is large and the correlation coefficient between dimensions exceeds a preset threshold (e.g., 0.7), principal component analysis (PCA) or similar dimensionality reduction methods can be selectively used to extract orthogonal factors before weighted aggregation to reduce the impact of collinearity between dimensions on the mechanism consistency factor. In the preferred four-dimensional embodiment of this invention, each dimension corresponds to four different levels of geochemical processes: mineral balance, ion exchange, isotope fractionation, and tectonic setting. The correlation between dimensions is usually weak, and weighted arithmetic averages can be directly used for aggregation.

[0040] In this embodiment, the mass coefficient q k The reliability of the input data for the kth mechanism dimension is reflected by methods including, but not limited to: (1) the ion charge balance error analyzed based on this dimension, where the smaller the error, the higher the q. k The closer to 1; (2) Based on repeated sampling, the smaller the standard deviation, the better q k The higher; (3) Standardized evaluation based on sampling conditions, such as whether sampling was performed directly at the wellhead and whether steam loss was avoided. When all dimensions of data pass the quality inspection, q k We approximate the value to 1.

[0041] In one embodiment, the physicochemical mechanism dimension k includes: mineral saturation index combination constraints, which are matched based on the saturation index combination patterns of key minerals; cation exchange constraints, which are matched based on cation exchange indices and the direction of ion ratio changes; isotope fractionation-geothermal consistency constraints, which are matched based on the coupling relationship between stable isotope drift characteristics and coordinated temperature ranges; and tectonic / lithological a priori constraints, which are matched based on regional tectonic and lithological data.

[0042] In this embodiment, the mineral saturation index (SI) characterizes the degree to which a certain mineral in the geothermal fluid reaches a dissolution-precipitation equilibrium. SI > 0 indicates supersaturation (easy to precipitate), SI ≈ 0 indicates equilibrium, and SI < 0 indicates unsaturation (can continue to dissolve). The combination of mineral saturation indices of different types of geothermal systems has typical characteristics. One or more compositional combinations of quartz, calcite, and anhydrite, which are the most indicative components in the geothermal system, are selected. Different candidate geothermal models (such as convection type, conduction type, magma heat source type, etc.) correspond to fixed saturation index target ranges.

[0043] The mineral saturation index combination constraint g SI,mThe combination of mineral saturation indices obtained from actual measurements is compared with the target intervals preset by each candidate model. A high score is given if the index falls completely within the target interval, a medium score is given if it partially matches, and a low score is given if it deviates from the target interval. Finally, the mechanism matching score for this dimension is obtained.

[0044] Cation exchange index: Geothermal geochemical indices CAI-I and CAI-II (cation exchange index) are used to quantify the Na+ exchange between geothermal fluids and surrounding rocks. + -Ca 2+ -Mg 2+ The intensity and process of plasma exchange; the direction of ion ratio change: such as the evolution trend of characteristic ion ratios like Na / K, Ca / Mg, Cl / HCO3, etc., and the candidate models corresponding to different geothermal circulation depths and migration paths, whose ion ratio change directions have fixed expectations.

[0045] The cation exchange constraint g CAI,m By comparing the changing directions of CAI-I, CAI-II and related ion ratios with the expected exchange process of the candidate model, a high score is obtained if the evolution direction is consistent and the index values ​​match, and a low score is obtained if they deviate. Finally, the mechanism matching score for this dimension is obtained.

[0046] Stable isotope drift characteristics: refers to the measured δ¹⁸O values ​​of geothermal fluids. 18 O drift amplitude and δD isotopic composition characteristics show significant differences in isotopic drift patterns among different geothermal systems (e.g., shallow mixing, deep circulation, and steam heating). Coordination temperature ranges: each candidate model corresponds to a theoretically reasonable temperature scale range. There is a strong coupling relationship between isotopic fractionation and temperature; the higher the temperature, the greater the δD drift. 18 O-shift is usually more pronounced.

[0047] The isotope fractionation-geothermal uniformity constraint g ISO,m : The measured δ 18 The O drift amplitude, δD features, and the coordinated temperature range of the candidate model are coupled and matched. A high score is obtained if the isotopic features and temperature range are consistent, and a low score is obtained if they are contradictory or mismatched. Finally, the mechanism matching score of this dimension is obtained.

[0048] The structural / lithological prior constraints g STR,m The geological settings of candidate models are compared with existing geological prior information. A high score is given for a complete match in geological structure, a medium score for a partial match, and a low score for a conflict with tectonic lithology conditions. The final mechanism matching score for this dimension is obtained. The tectonic lithology background information is regional geological survey information independent of the basic prior score.

[0049] The coordinated temperature range is the theoretical reservoir temperature corresponding to each candidate model, determined by cross-validation of the indicated temperatures of multiple geothermal temperature scales.

[0050] Step S3: Softly correct the basic prior score using the mechanism consistency factor to obtain the final score of each candidate model. The soft correction is that the final score is a weighted combination of the retained part of the basic prior score and the part of the basic prior score adjusted by the mechanism consistency factor, and the direct retention ratio of the basic prior score is controlled by a preset fidelity coefficient.

[0051] In this embodiment, the fidelity coefficient refers to a preset adjustment coefficient used to regulate the proportion of basic prior scores retained in the final score. The larger the coefficient, the higher the contribution of the basic prior scores to the final score and the stronger the retention. The smaller the coefficient, the more the final score focuses on the measured information of the mechanism consistency factor. Furthermore, the value of the fidelity coefficient and the specific calculation formula for soft correction can be flexibly set according to the exploration stage of the study area: when there is less data in the early stage of exploration, the fidelity coefficient should be larger (e.g., η=0.5-0.7) to focus on prior knowledge; when there is sufficient data in the later stage of exploration, the fidelity coefficient should be smaller (e.g., η=0.2-0.4) to focus on the matching degree of measured mechanism.

[0052] In one embodiment, the soft correction employs a method of coupling the basic prior score with a mechanistic factor to obtain the final score, where the final score R... m The calculation formula is: R m =S m *(η+(1-η)*M m ) Among them, R m For the final score, S m Based on the prior score, η is the fidelity coefficient of the basic prior score, 0 < η < 1, M m This is the mechanism consistency factor. The formula can be expanded to R0. m =η*S m +(1-η)*S m *M m , where η*S m The directly retained portion of the basic prior score, (1-η)*S m *M m The final score is the portion of the basic prior score adjusted by the mechanistic consistency factor. A larger η indicates a higher proportion of the basic prior score is directly retained, while a smaller η indicates a stronger moderating effect of the mechanistic consistency factor on the final score. When M... m When the value is in the interval [0,1], the final score R m The range of values ​​for is [η*S] m ,S m That is, the retention ratio of the basic prior score is not less than η, ensuring that the prior information is not completely covered.

[0053] Step S4: Re-rank each candidate model according to the final score, and output the ranking results and the corresponding geothermal system discrimination conclusion.

[0054] In one embodiment, the discrimination conclusion includes at least a candidate model ranking list, the best matching geothermal system type, the reservoir temperature estimation range, the mechanism consistency level, and supplementary test recommendations. The optimal matching geothermal system type and reservoir temperature estimation range are determined based on the candidate model with the highest final score. The mechanism consistency level is determined based on the value of the mechanism consistency factor. When the matching score of a certain mechanism dimension is lower than a set threshold, supplementary test suggestions for that mechanism dimension are generated.

[0055] In this embodiment, the best matching geothermal system type is the theoretical type that best matches the actual geothermal system in the study area, determined based on the candidate model with the highest final score.

[0056] The reservoir temperature estimation range is a reasonable estimate of the actual temperature of the geothermal reservoir, given by combining the calculation results of the geothermal temperature scale corresponding to the best matching model with multi-scale fusion and correction processing. For example: Range = Fusion Solution ± Reasonable Error (Sources of error: inherent error of empirical formulas of temperature scales, sample testing error, and minor deviations in the correction and fusion process); Mechanism consistency rating is a graded evaluation of the degree of physicochemical mechanism matching between candidate models (usually the top-ranked key models) based on the magnitude of the mechanism consistency factor. For example: High consistency: Mechanism consistency factor ≥ 0.8 (the model almost perfectly matches the experimental data); Medium-high consistency: 0.6 ≤ Mechanism consistency factor < 0.8 (the model matches the experimental data well); General consistency: 0.4 ≤ Mechanism consistency factor < 0.6 (the model basically matches the experimental data, but there are some deviations); Low consistency: Mechanism consistency factor < 0.4 (the model deviates significantly from the experimental data, and the reliability is low).

[0057] In this embodiment, the threshold can be determined in one or more of the following ways: (1) taking the lower quartile Q1 of the matching score of each candidate model in this dimension as the threshold; (2) empirically setting it to 0.5, that is, when the matching score is lower than 0.5, it is considered that the data support for this dimension is insufficient; (3) taking the quantile that significantly deviates from the normal distribution as the threshold according to the statistical distribution characteristics of the historical data of the study area. Those skilled in the art can flexibly select according to actual needs.

[0058] Supplementary testing recommendations: Low score in mineral saturation index dimension: It is recommended to collect additional deep geothermal reservoir core samples for detailed mineralogical identification (such as XRD analysis) to accurately calculate the saturation index.

[0059] Low score in cation exchange dimension: It is recommended to increase sampling density at different depths and different fracture zone locations to test CAI-I, CAI-II and characteristic ion ratios to clarify the ion exchange evolution path.

[0060] Low isotope fractionation dimension: It is recommended to collect more geothermal fluid samples covering different regions and depths, and remeasure the delta content. 18 O and δD isotopes were used to verify the coupling relationship between isotope drift characteristics and temperature.

[0061] Low score in tectonic lithology dimension: It is recommended to carry out high-precision geophysical exploration (such as seismic exploration) to redefine the regional tectonic and lithological distribution and verify the geological settings of the model.

[0062] Example 1 Taking the geological conditions of a deep, concealed geothermal area in a fault basin as an example, the project team collected 28 surface hot spring and shallow well water samples and completed conventional ion, SiO2, trace element, and δ-ion analysis. 18 A full range of geochemical tests, including O and δD, are performed. The system automatically calculates parameters such as temperature scale and mineral saturation index (SI).

[0063] Based solely on the seven-dimensional basic score of traditional water type and temperature scale, the scores of the top two candidate conceptual models output by the system are extremely close. This score difference makes it very easy for the ranking to be reversed due to interference from a single abnormal sample, resulting in obvious multiple solutions, as shown in Table 1.

[0064] Table 1

[0065] The system automatically triggers in-depth analysis by introducing mechanistic terms. For each model, its corresponding expected mechanistic features are extracted and compared with the calculation results of actual samples. The matching scores of the four mechanistic dimensions are shown in Table 2.

[0066] Table 2

[0067] In this case, the test data for all 28 water samples are complete, and the data for all four mechanistic dimensions are available; therefore, the availability indicator B is valid. k All values ​​are set to 1; the quality of the test data in each dimension meets the standards after preprocessing and verification, and the quality coefficient q is [value missing]. k All are approximately taken as 1.

[0068] The system is based on the preset prior weights of the mechanism items (λ). SI =0.35, λ CAI =0.20, λ ISO =0.25, λ STR =0.20), under the condition of approximately equal weighting after quality coefficient conversion, the mechanism consistency factor is calculated as follows: MA =(0.35×0.84+0.20×0.85+0.25×0.81+0.20×0.68)≈0.80; M B =(0.35×0.58+0.20×0.62+0.25×0.55+0.20×0.68)≈0.60.

[0069] Set the basic prior score fidelity coefficient η=0.4, and recalculate the final score using R=S×[η+(1-η)×M: R A ==0.52×(0.4+0.6×0.80)≈0.458; R B ==0.49×(0.4+0.6×0.60)≈0.372.

[0070] After introducing mechanistic constraints, model A has a high degree of matching across various mechanistic dimensions (M). A The final score was reduced by a small factor (scaling factor 0.88), while Model B had a lower mechanistic matching degree (M ≈ 0.80). B The final score reduction was even greater (scaling factor was 0.76), with the difference between candidate models A and B increasing from a slight 0.03 to 0.086, effectively improving the discriminative power and robustness of the identification.

[0071] According to the mechanistic consistency level classification standard, Model A's M A A value of ≈0.80 falls under the "high consistency" level (mechanistic consistency factor ≥0.8), and the M value of model B is... B A value of ≈0.60 falls within the "medium-high consistency" level (0.6 ≤ mechanism consistency factor < 0.8). Based on this, the system outputs the following conclusion: the best-matched geothermal system type is the deep-circulation fracture-controlled heat type, with a reservoir temperature estimation range of 180-220 degrees Celsius. For candidate model A, three methods—Na-K scale, quartz scale, and Na-K-Ca scale—were selected. Under the assumption of no shallow cold water mixing influence, the indicated temperature was calculated for each method. The intersection range of the three scale results [180℃, 220℃] was taken as the coordination temperature range for this model.

[0072] Example 2 Taking the analysis of a water sample extracted from an exploration well in a high-temperature geothermal prospect as an example: the water sample analysis showed a severe internal divergence in the temperature scale calculation results: the quartz temperature scale reached as high as 245 degrees Celsius, and the Na-K temperature scale also reached 232 degrees Celsius, but the K-Mg temperature scale was abnormally low, only 128 degrees Celsius. The temperature scale clusters were extremely dispersed, as shown in Table 3. When performing basic graphical analysis, because the Na-K temperature scale indicates high temperatures, the basic prior score (S... mThe water sample shows a slight tendency toward a "typical volcanic high-temperature geothermal system." However, the water sample falls entirely into the "immature water" region in the lower right corner of the Na-K-Mg triangle diagram, indicating that the water-rock reaction is extremely incomplete. This strongly contradicts the prediction of a high temperature exceeding 200 degrees Celsius.

[0073] Table 3

[0074] After detecting a data conflict, the system initiates mechanistic calculations to compare the two candidate models, as shown in Table 4. In-depth calculations of the mechanistic component show that for the "typical volcanic high-temperature system" (high-temperature model): although the saturation indices of some high-temperature minerals match at this temperature (g... SI High score, g SI ≈0.72), but the cation exchange process based on CAI-I and CAI-II does not conform to the characteristics of long-term high-temperature curing (g CAI Extremely low, g CAI ≈0.30), and δ 18 The O isotope drift is far from reaching the theoretical value for water-rock equilibrium at this temperature (g). ISO Partial match only, g ISO ≈0.45); For "intermediate temperature system superimposed with shallow, intensely cold water mixing" (intermediate temperature mixing model): the mineral SI assemblage imbalance mode of the sample (g CAI ≈0.78), incomplete cation exchange process (g CAI ≈0.78), slight isotopic drift (g ISO ≈0.70) and the specific fault zone water-conducting tectonic setting (g STR The value ≈0.75 may indicate that the deep subsurface was originally at a mesophilic temperature (e.g., around 150 degrees Celsius), but during its ascent, it encountered a large amount of magnesium-rich cold water, causing the K-Mg equilibrium to be rapidly disrupted while the Na-K equilibrium was frozen due to the slow reaction. The M value in the mesophilic mixing model... m Significantly higher than the high temperature model.

[0075] Table 4

[0076] Let the basic prior score S 高温 =0.55, S 中温 =0.50, fidelity coefficient η=0.4, substitute into the final scoring formula: R 高温 =0.55×(0.4+0.6×0.52)≈0.392; R 中温 =0.50×(0.4+0.6×0.71)≈0.413.

[0077] The high-temperature model with the leading basic prior score (S) 高温 =0.55>S 中温 =0.50), which was surpassed by the intermediate-temperature hybrid model after mechanistic constraint correction (R 中温 =0.413>R 高温 =0.392).

[0078] According to the classification of mechanism consistency levels, high-temperature model M 高温 ≈0.52 falls under the "general consistency" category, for the medium-temperature mixed model M. 中温 A score of ≈0.71 falls under the "medium-high consistency" level. Based on this, the system concludes that the best match is a medium-temperature system superimposed with the influence of shallow cold water mixing, with a reservoir temperature estimation range of 140-160 degrees Celsius. The previous model, based on temperature scales above 200 degrees Celsius, suffered from a Na-K equilibrium lag due to shallow mixing. Simultaneously, due to the high-temperature model's score in the cation exchange dimension (g... CAI If the value is ≈0.30, which is below the set threshold, the system will automatically generate supplementary test suggestions: It is recommended to supplement the N2-He-Ar inert gas diagram to track deep volatiles, and it is recommended to conduct sealed sampling and retesting of deeper layers.

[0079] Example 3 This invention provides a comprehensive geothermal geochemical discrimination system, comprising an acquisition unit, a mechanism consistency unit, a correction unit, a sorting unit, and an output unit. Taking the rift basin case in Embodiment 1 as an example, the system's operational data flow is as follows: The acquisition unit receives hydrochemical and tectonic lithological data from 28 water samples imported by the user. After preprocessing, it performs graphical interpretation and temperature scale calculation, matching the graphical criteria and temperature scale values ​​with the expected characteristics of each candidate model, and outputs the basic prior score (e.g., S) for each candidate model. A =0.52, S B =0.49), and the scoring result is passed to the mechanism consistency unit.

[0080] The mechanism consistency unit receives preprocessed hydrochemical data and calculates the mechanism matching score of each candidate model across four dimensions: mineral saturation index combination constraint, cation exchange constraint, isotope fractionation-geothermal consistency constraint, and tectonic / lithological prior constraint. Combining these with pre-defined mechanism prior weights, quality coefficients, and usability indicators, the unit aggregates the results according to the mechanism consistency factor formula to obtain the M-value of each candidate model. m Value (e.g., M) A ≈0.80, M B (≈0.60), and then pass the result to the correction unit.

[0081] Correction unit receives basic prior score S m Mechanism consistency factor M mAccording to the preset fidelity coefficient η=0.4, the final score calculation formula R is executed. m =S m ×(η+(1-η)×M m ), to obtain the final score of each candidate model (e.g., R). A ≈0.458, R B ≈0.372), and then pass the result to the sorting unit.

[0082] The sorting unit ranks the candidate models from highest to lowest according to the final score, determines the best matching model as model A (deep cycle fracture heat control type), and passes the sorting result to the output unit.

[0083] The output unit generates a complete discrimination conclusion, including: a candidate model ranking list, the best matching geothermal system type (deep circulation fracture-controlled heat type), the reservoir temperature estimation range (180-220 degrees Celsius), the mechanism consistency level (model A is highly consistent), and supplementary test suggestions.

[0084] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the original technical features, and the technical solutions resulting from these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A comprehensive geothermal geochemical discrimination method, characterized in that, Includes the following steps: Obtain the basic prior scores of the conceptual models of each candidate geothermal system; For each candidate model, the mechanism matching score between the candidate model and the actual sample is calculated from at least two different physicochemical mechanism dimensions, and the matching scores of each mechanism dimension are aggregated into a mechanism consistency factor. The mechanistic consistency factor is used to perform soft correction on the basic prior score to obtain the final score of each candidate model. The soft correction is that the final score is composed of a weighted combination of the retained part of the basic prior score and the part of the basic prior score adjusted by the mechanistic consistency factor, and the direct retention ratio of the basic prior score is controlled by a preset fidelity coefficient. Based on the final score, the candidate models are re-ranked, and the ranking results and corresponding geothermal system discrimination conclusions are output.

2. The method according to claim 1, characterized in that, The basic prior scores for obtaining the conceptual models of each candidate geothermal system include: Acquire hydrochemical and tectonic-lithological data of the geothermal fluid to be tested, and perform preprocessing. The preprocessed data is subjected to graphical interpretation and temperature scale calculation to obtain graphical criteria and temperature scale values. The graphical criteria and temperature scale values ​​are matched with the expected features of each candidate model, and each candidate model is assigned a corresponding basic prior score based on the degree of matching.

3. The method according to claim 1, characterized in that, Mechanism consistency factor M m The calculation formula is: M m =S k (l k *q k *g k,m *B k ) / S k (l k *q k *B k ) Where k represents different physicochemical mechanism dimensions, and λ k For the prior weights of the mechanism term, q k For the mass coefficient, g k,m B represents the matching score of candidate model m on k. k This is an indicator of availability.

4. The method according to claim 1, characterized in that, Calculate the mechanism matching score of the candidate model on a certain mechanism dimension, including: Based on the geothermal system type of the candidate model, determine the expected characteristics of the model in this mechanism dimension; The actual calculated value of the mechanism is compared with the expected feature; A matching score is assigned based on the degree of similarity in the comparison.

5. The method according to claim 4, characterized in that, The physicochemical mechanism dimension k includes: mineral saturation index combination constraints, cation exchange constraints, isotope fractionation-geothermal consistency constraints, and tectonic / lithological a priori constraints.

6. The method according to claim 5, characterized in that, The mineral saturation index combination constraint matches the saturation index combination pattern of one or more minerals, such as quartz, calcite, and anhydrite, with the target interval of the candidate model. The cation exchange constraint compares the direction of change of CAI-I, CAI-II and related ion ratios with the expected exchange process of the candidate model. The isotope fractionation-geothermal consistency constraint is achieved by using δ 18 The O drift amplitude, δD characteristics, and the coordination temperature range of the candidate model were coupled for analysis. The structural / lithological prior constraints are achieved by matching candidate models with existing structural and lithological background data, expert annotations of structural conditions, or default structural adaptation values. The structural and lithological background data is regional geological survey information independent of the basic prior score.

7. The method according to claim 1, characterized in that, The soft correction is calculated in the following form: R m =S m *(n+(1-n)*M m ) Among them, R m For the final score, S m Based on the prior score, η is the fidelity coefficient of the basic prior score, 0 < η < 1, M m This is the mechanism consistency factor.

8. The method according to claim 1, characterized in that, The judgment conclusions include at least a candidate model ranking list, the best matching geothermal system type, the reservoir temperature estimation range, the mechanism consistency level, and supplementary test recommendations. The optimal matching geothermal system type and reservoir temperature estimation range are determined based on the candidate model with the highest final score. The mechanism consistency level is determined based on the value of the mechanism consistency factor. When the matching score of a certain mechanism dimension is lower than a set threshold, supplementary test suggestions for that mechanism dimension are generated.

9. A comprehensive geothermal geochemical discrimination system, characterized in that, include: The acquisition unit is used to acquire the basic prior scores of the conceptual models of each candidate geothermal system. The mechanism consistency unit is used to calculate the mechanism matching score between the candidate model and the actual sample from at least two different physicochemical mechanism dimensions for each candidate model, and to aggregate the matching scores of each mechanism dimension into a mechanism consistency factor. The correction unit is used to perform soft correction on the basic prior score using the mechanism consistency factor to obtain the final score of each candidate model. The soft correction is that the final score is composed of a weighted combination of the retained part of the basic prior score and the part of the basic prior score adjusted by the mechanism consistency factor, and the direct retention ratio of the basic prior score is controlled by a preset fidelity coefficient. A sorting unit is used to re-sort each candidate model based on the final score; The output unit is used to output the sorting results and the corresponding geothermal system identification conclusions.

10. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method as described in any one of claims 1-8.

Images