Hybrid end-member inversion and temperature calibration for geothermal geochemical analysis method and system

Abstract

The present application provides a kind of mixed end element inversion and geothermal geochemical analysis method and system of temperature scale correction, method includes: according to the regional hydrogeological background determination fluid end element characteristic parameter;According to geothermal water chemical sample observation data and end element characteristic parameter, inversion each end element quantitative contribution rate and determine deep circulating hot water end element contribution rate;Using the contribution rate correction geothermal temperature scale, obtain correction result;The correction result is screened in coordination, and based on the temperature scale coordination subset obtained by screening is fused, and the reservoir temperature fusion solution is output.The present application improves the calculation precision of reservoir temperature through the depth linkage of end element inversion and temperature scale correction, adapts complex hydrogeological scene, is simple and practical to operate, and can be widely applied in geothermal resource exploration and evaluation field.

Description

Technical Field

[0001] This invention relates to the field of geothermal geochemical interpretation technology, specifically providing a geothermal geochemical analysis method and system based on hybrid endmember inversion and temperature scale correction. Background Technology

[0002] In geothermal exploration and development, reservoir temperature estimation is a crucial step in evaluating reservoir potential and development pathways. Commonly used methods such as the quartz temperature scale, Na-K temperature scale, and K-Mg temperature scale are good indicators for deep geothermal reservoirs, but systematic biases often occur in shallow areas with significant mixing.

[0003] When deep circulating hot water mixes with shallow cold water, the chemical components (such as SiO2, Cl) and isotopic composition (such as δ) of the fluid change. 18 O and δD) are superimposed proportionally. If the mixed spring water or shallow well water sample data are directly substituted into the temperature scale formula for calculation, the temperature scale input parameters will deviate significantly from the original deep circulation end-member composition, resulting in a systematic underestimation of the temperature scale calculation results and a serious underestimation of the deep geothermal reservoir potential. Different temperature scales have different sensitivities to mixing (for example, the quartz temperature scale is very sensitive to mixing, while the Na-K temperature scale is relatively insensitive), which will increase the dispersion of multi-temperature scale calculation results and make it difficult to reach a consistent judgment. When performing conceptual model discrimination of geothermal systems, mixed water samples may also be incorrectly misclassified as low-temperature systems or peripheral systems.

[0004] Currently, traditional methods for handling mixing problems mainly rely on graphical methods, such as silicon-enthalpy mixing models or chlorine-enthalpy mixing models. These graphical methods typically require manual intervention for tangent or extrapolation, resulting in low automation and a focus on estimating single components, making it difficult to coordinate multi-component data. Furthermore, silicon-enthalpy or chlorine-enthalpy methods are highly dependent on enthalpy values, boiling assumptions, or ideal mixing paths, making it difficult to reliably obtain relevant steam separation data during the exploration phase, thus limiting their applicability. Among existing tools, AquaChem focuses on hydrochemical discrimination and conventional statistical analysis, while PHREEQC and Geochemist's Workbench focus on water-rock reaction equilibrium and reaction path calculation. These tools often process mixing identification, equilibrium calculation, and temperature scale estimation in separate segments within the engineering workflow, lacking an automated linkage mechanism that directly feeds back the quantitative inversion results of mixing ratios to the multi-temperature scale calculation stages, leading to a long-standing disconnect between "mixture identification" and "temperature correction" in the calculation process.

[0005] Therefore, there is an urgent need for a method that deeply couples quantitative decomposition of hybrid endmembers with automatic temperature scale calibration. Summary of the Invention

[0006] To overcome the above-mentioned shortcomings, this invention is proposed to provide a solution or at least a partial solution to the problems of ineffective elimination of shallow mixing leading to underestimation of temperature and low accuracy and reliability in identifying geothermal systems in complex mixing zones.

[0007] In a first aspect, the present invention provides a geothermal geochemical analysis method with hybrid end-member inversion and temperature scale correction, comprising the following steps: obtaining characteristic parameters of multiple fluid end-members based on the regional hydrogeological background, wherein the multiple fluid end-members include at least deep circulating hot water end-members and shallow cold water end-members; Based on the observation data of geothermal chemical samples and the characteristic parameters of each fluid end-member, the quantitative contribution rate of each fluid end-member to the mixed sample is calculated by inversion, and the contribution rate of the deep circulation hot water end-member is determined. The contribution rate of the deep circulation hot water terminal element is used to correct at least one geothermal temperature scale to obtain the corrected temperature scale result; The calibration temperature scale results are subjected to coordination screening, and the resulting coordinated subset of temperature scales is fused to output the reservoir temperature fusion solution.

[0008] Preferably, when the regional hydrogeological background is affected by steam or seawater, the plurality of fluid end units further include a steam condensation end unit or a seawater influence end unit.

[0009] Preferably, determining the contribution rate of the deep-circulation hot water end-member includes: constructing a mass balance equation system based on multiple conserved or semi-conserved components, wherein the components include Cl, SiO2, and stable isotope δ. 18 One or more of O and δD are used to solve the problem. The solution is obtained by using a joint optimization algorithm. The dynamic weights are determined by adaptive optimization based on the observation error variance of each component, the component conservatism score, or the iterative residuals. The objective is to minimize the sum of squared residuals between the measured values ​​and the model predictions. The constraint condition is that the sum of the contribution rates of each fluid endmember is 1 and the contribution rate of each fluid endmember is between 0 and 1. The iterative residuals are adapted using the Huber robust statistical method to reduce the weights of component residuals that exceed the truncation threshold. The truncation constant is taken as 1.0 to 1.5.

[0010] The quantitative contribution rate of each fluid end-member to the mixed sample is obtained; the contribution rate of the deep circulation hot water end-member is determined from the contribution rates of each fluid end-member.

[0011] Preferably, when the contribution rate of the deep-circulation hot water end-unit is lower than a preset threshold, or the relative error of the reconstructed concentration calculated based on error propagation is higher than a preset error threshold, it is determined that the component antimixing reconstruction path has a risk of numerical instability, and an antimixing incremental compensation path is used to replace the component antimixing reconstruction path, specifically including: Establish temperature correction ΔT mix Contribution rate α of deep circulation hot water terminal elementH The functional relationship is an empirical compensation model established based on regional calibration samples. The empirical compensation model is based on the fact that the pressure reduction intensity of shallow cold water on apparent temperature varies with the ratio of cold and hot end-members (1-α). H ) / α H Monotonically increasing, applicable to α H Scenarios exceeding the preset applicable lower limit: ΔT mix = κ×(1-α H ) / α H Where, ΔT mix α is the temperature correction amount. H The contribution rate of the deep circulation hot water terminal element is given by κ, which is the regional calibration coefficient. The regional calibration coefficients are determined by least squares fitting using regional calibration samples with reference reservoir temperature, and satisfy the physical constraint that they are greater than 0 and less than the upper limit of the temperature difference between the hot and cold end members. When there are insufficient regional calibration samples, the regional calibration coefficients are given prior values ​​based on the reference range of similar geothermal systems, and the corresponding correction results are marked as low-confidence auxiliary results. The temperature scale correction amount is calculated using the aforementioned functional relationship, and then superimposed onto the original apparent temperature to obtain the corrected temperature scale result.

[0012] Preferably, for two-terminal hybrid scenarios, the reconstruction concentration is calculated using the following formula: X H * = (X obs - (1-α H ) × X C ) / α H Among them, X H *Represents the reconstitution concentration in the deep circulation hot water terminal element, X obs X represents the measured component concentration of the mixed sample. C For the corresponding component characteristic value of the shallow cold water end member, α H Contribution rate to deep circulation hot water terminal components; For scenarios with three or more endmembers, the reconstructed concentration is obtained by solving the mass conservation equations for multiple endmembers.

[0013] Preferably, the geothermal temperature scale includes multiple temperature scales to form a multi-temperature scale combination; the multi-temperature scale combination includes at least a silica-based temperature scale and a cationic temperature scale; the multi-temperature scale combination is calibrated respectively to generate a calibrated temperature scale set.

[0014] Preferably, it also includes: Calculate the statistical dispersion index of the calibration temperature scale set; Based on the statistical dispersion index and combined with the applicable conditions of the temperature scale, temperature scales that meet the consistency threshold requirements are identified, and a temperature scale coordination subset is formed. Based on the consistency weight and endmember sensitivity weight of each temperature scale in the temperature scale coordination subset, the collaborative fusion weight is calculated, and the results of each corrected temperature scale in the temperature scale coordination subset are weighted and fused using the collaborative fusion weight to calculate and output the reservoir temperature fusion solution. The consistency weight Determine using the following formula: in, For the temperature scale coordination subset, the first The corrected temperature value of each temperature scale, Med The median of the temperature scale correction results for the temperature scale harmonization subset is given. To prevent positive numbers with a denominator of zero.

[0015] Secondly, the present invention provides a geothermal geochemical analysis system with hybrid end-member inversion and temperature scale correction, comprising: an acquisition unit, used to acquire characteristic parameters of multiple fluid end-members based on the regional hydrogeological background, wherein the multiple fluid end-members include at least deep circulating hot water end-members and shallow cold water end-members; The end-member inversion unit calculates the quantitative contribution rate of each fluid end-member to the mixed sample based on the observation data of the geothermal chemical sample and the characteristic parameters of each fluid end-member, and determines the contribution rate of the deep circulation hot water end-member. The temperature scale correction unit uses the contribution rate of the deep circulation hot water terminal element to correct at least one geothermal temperature scale and obtain the corrected temperature scale result; The output unit is used to perform coordination screening on the corrected temperature scale results, and to perform fusion processing on the selected coordinated subset of temperature scales to output the reservoir temperature fusion solution.

[0016] Preferably, the temperature scale correction unit includes a component antimixing reconstruction module and an antimixing incremental compensation module, and is configured to perform path switching between the component antimixing reconstruction module and the antimixing incremental compensation module when the contribution rate of the deep circulation hot water end-member is lower than a preset threshold or the relative error of the reconstruction concentration is higher than a preset error threshold; the output unit is used to calculate the collaborative fusion weight based on the consistency weight and end-member sensitivity weight of each temperature scale in the temperature scale coordination subset, and use the collaborative fusion weight to perform weighted fusion of each corrected temperature scale result in the temperature scale coordination subset to output the reservoir temperature fusion solution.

[0017] The beneficial effects of this invention are as follows: 1. This method uses the contribution rate of the deep circulating hot water end-member obtained by end-member inversion as the core basis for temperature scale correction, realizing a closed-loop linkage of "quantitative end-member inversion - accurate temperature scale correction - temperature fusion output". It eliminates the influence of interference factors such as shallow cold water, steam, and seawater on temperature scale calculation from the source, significantly improves the accuracy and reliability of reservoir temperature calculation, and solves the problem of temperature scale distortion caused by the mixing effect in traditional methods.

[0018] 2. This method can flexibly adjust the fluid end-member type according to the regional hydrogeological background. In addition to the basic deep circulation hot water end-member and shallow cold water end-member, steam condensation end-member and seawater influence end-member can be added to adapt to complex geothermal geological environments such as steam interference and seawater intrusion. It overcomes the limitation of traditional methods that can only be applied to simple two-end-member mixed scenarios. It can be widely used in different regions and different types of geothermal exploration scenarios. Whether it is a two-end-member or a three-end-member or more mixed scenario, accurate analysis can be achieved through corresponding strategies.

[0019] 3. This method is based on Cl, SiO2, and the stable isotope δ. 18 A mass balance equation system was constructed using conserved / semi-conserved components such as O and δD. Combined with a joint optimization algorithm, dynamic weighting and constraint solving were used to effectively reduce the impact of observation errors of different components on the inversion results, ensuring the quantitative accuracy of the contribution rate of end-members in deep circulation hot water.

[0020] 4. Flexible temperature scale correction path for improved stability and applicability: Addressing the numerical divergence caused by the low contribution rate of end-members in deep-circulation hot water systems, this method employs a dual-path correction scheme: a "component antimixing reconstruction path" and an "antimixing incremental compensation path," which can be flexibly switched based on the actual contribution rate. The component antimixing reconstruction path is strictly derived from the mass conservation equation, requiring no empirical parameters, ensuring mathematical self-consistency and complete traceability of the correction results. In numerically unstable scenarios, an incremental compensation path is used, guaranteeing correction accuracy under normal conditions while avoiding distortion in extreme scenarios, further adapting to fluid mixing situations of varying complexity.

[0021] 5. Multi-temperature scale fusion optimization further enhances the reliability of reservoir temperature solutions: This method adopts a combination of multiple temperature scales, identifies the coordinated subset of temperature scales through statistical dispersion index, and calculates the collaborative fusion weight by combining consistency weight and endmember sensitivity weight, so as to realize the weighted fusion of multiple corrected temperature scales and effectively reduce the limitations and errors of a single temperature scale itself. Attached Figure Description

[0022] 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 analysis method based on hybrid endmember inversion and temperature scale correction according to an embodiment of the present invention. Detailed Implementation

[0023] 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.

[0024] like Figure 1 As shown, this invention provides a geothermal geochemical analysis method based on hybrid endmember inversion and temperature scale correction, comprising the following steps: Step S1: Obtain characteristic parameters of multiple fluid end-units based on the regional hydrogeological background. The multiple fluid end-units include at least a deep circulating hot water end-unit and a shallow cold water end-unit.

[0025] In this embodiment, the knowledge base pre-collects and categorizes geochemical characteristic parameters of various typical fluid endmembers under different geothermal geological backgrounds, including the concentrations of conserved / semi-conserved components (Cl, SiO2), cation ratios, and stable isotope δ¹⁸O. 18 Typical values, regional reference values, and standard characteristic values ​​of indicators such as O / δD can be directly selected or customized as needed. Based on the regional geothermal system's recharge-runoff-discharge conditions, reservoir depth, tectonic activity characteristics, and surface hydrological environment, at least two fluid end-members are selected from the preset geothermal fluid end-member knowledge base. Deep circulation hot water end element H: represents the original high-temperature hot fluid in the deep thermal reservoir that has not been significantly mixed and disturbed. The characteristic parameters of the hot fluid under the corresponding geothermal background and tectonic conditions are selected from the knowledge base. If there is no matching typical value, it can be configured according to the measured high-temperature hot well data in the region. Shallow cold water terminus C: Represents cold water sources such as regional atmospheric precipitation and shallow low-temperature groundwater. It directly selects shallow cold water characteristic parameters from the knowledge base for areas with the same climate and geomorphology, or configures them based on the measured values ​​of regional cold springs and shallow groundwater.

[0026] In one embodiment, when the regional hydrogeological background is affected by steam or seawater, the plurality of fluid end-units further include: Steam condensation end-unit: When there are signs of steam activity such as boiling springs, fumaroles, and steam heating zones in the region, retrieve typical characteristic parameters of steam condensate (extremely low ion concentration and isotopic steam fractionation characteristics) from the knowledge base. Seawater impact terminology: When there are signs of seawater intrusion or mixing in coastal geothermal fields, characteristic parameters of standard seawater or regional intrusive seawater are selected from the knowledge base to adapt to the seawater mixing scenario.

[0027] Of course, the types of end-units can be added or removed according to the actual mixed scenarios in the region, and the feature parameters can be optimized by combining the measured data of the region (such as shallow cold water sampling and deep hot well testing). If there are no completely matching typical values ​​in the knowledge base, they can be calibrated by reference values ​​of similar geothermal systems in the region and adjusted as needed.

[0028] Step S2: Based on the observation data of geothermal chemical samples and the characteristic parameters of each fluid end-member, calculate the quantitative contribution rate of each fluid end-member to the mixed sample, and determine the contribution rate of the deep circulation hot water end-member.

[0029] In this embodiment, the geothermal chemical sample observation data refers to the actual chemical indicators obtained from laboratory tests of the collected geothermal well / hot spring mixed water samples, including Cl. - SiO2, δ 18 Concentrations or isotope ratios of components such as O and δD. After collection, unit standardization, charge balance verification, and outlier screening are performed first.

[0030] In one embodiment, determining the contribution rate of the deep circulation hot water terminal includes: A set of mass balance equations is constructed based on multiple conserved or semi-conserved components, including Cl, SiO2, and the stable isotope δ. 18 One or more of O and δD; A joint optimization algorithm is used to solve the problem. Dynamic weights, determined adaptively based on the observation error variance of each component, component conservatism scores, or iterative residuals, are used to minimize the sum of squared residuals between measured and predicted values. A constraint is applied that the sum of the contribution rates of each fluid endmember is 1, and the contribution rate of each fluid endmember is between 0 and 1, thus obtaining the quantitative contribution rate of each fluid endmember to the mixed sample. The iterative residual adaptation employs a robust statistical method to reduce the weights of component residuals exceeding a cutoff threshold. The cutoff constant corresponding to the cutoff threshold is between 1.0 and 1.5. The contribution rate of the deep circulation hot water terminal is determined from the contribution rates of each fluid terminal.

[0031] In this embodiment, the selection of the conserved / semi-conserved components can be optimized based on the regional hydrochemical characteristics (e.g., increasing Br in coastal areas). - (Components). The suitability criterion for using SiO2 as a semi-conservative component is: when the quartz saturation index... Including cases where there is no obvious silicon precipitation, secondary dissolution, or boiling silicon loss in the solution; When the score is between 0.5 and 1.0, the conservatism score should be reduced. After reaching 0.3 to 0.5, a lower weight is assigned; The conservatism of SiO2 was deemed insufficient, and it was not included in the inversion.

[0032] Component weights Initially assigned based on the reciprocal of the variance of the observation error, and combined with a conservatism score. and iterative residual correction term Dynamically updated: in Let V be the observation error variance of the k-th component; The components are quantified into a score between 0 and 1 based on their conservatism within the target area, sensitivity to phase transition or water-rock reaction, and historical fitting residuals. Cl ranges from 0.8 to 1.0, and δ... 18 O and δD are taken as 0.7 to 1.0, and SiO2 binding saturation index and precipitation risk are taken as 0.3 to 0.8. The conservation, phase transition or water-rock reaction sensitivity, and history fitting residual of the m-th component within the target area are comprehensively quantified into a score between 0 and 1. The weighting factor is calculated based on Huber's robust statistical method. Let be the variance of the m-th component in laboratory testing or field measurement.

[0033] The update rules adopt Huber's robust truncation: in Let be the standardized residual of the k-th component in the n-th iteration, and c be a robust cutoff constant, taken as 1.345 (the classic threshold of the Huber loss function under the normal distribution assumption). When the data quality is poor, c can be reduced to 1.0 to enhance outlier suppression; when the data quality is good, it can be increased to 1.5. When the residual of a component exceeds the cutoff threshold, its weight is adaptively reduced.

[0034] Joint optimization objective function for: in, This represents the actual observed value of the k-th chemical component (or isotope ratio) in the geothermal sample to be evaluated. is the characteristic concentration or characteristic value of the i-th pure end-member fluid (the original water body before mixing) on ​​the k-th chemical component.

[0035] The constraints are and The convergence criterion is that the relative rate of change of the sum of squared residuals is less than 1%. It may reach a maximum of 200 iterations.

[0036] After the solution is completed, the contribution rate of the deep circulating hot water endmembers is taken from the contribution rates of each endmember. The number of independent constraint components used for the solution is no less than the number of endmembers minus 1; when the number of components is insufficient, endmember value ranges, isotope priors, or regularization conditions are introduced to suppress multiple solutions.

[0037] Step S3: Use the contribution rate of the deep circulation hot water terminal element to correct at least one geothermal temperature scale to obtain the corrected temperature scale result.

[0038] In one embodiment, step S3, using the contribution rate of the deep circulating hot water terminal element to correct at least one geothermal temperature scale, and obtaining the corrected temperature scale result includes: Based on the mass conservation equation, the reconstruction concentration of the chemical components involved in the temperature scale calculation in the deep circulation hot water terminal element is calculated by using the contribution rate of the deep circulation hot water terminal element. Substituting the reconstructed concentration into the corresponding empirical formula for geothermal temperature scale, the corrected temperature scale result is obtained.

[0039] When the contribution rate of the deep circulation hot water terminal element is lower than a preset threshold, or the relative error of the reconstructed concentration calculated based on error propagation is higher than a preset error threshold, and it is determined that the component antimixing reconstruction path has a risk of numerical instability, an antimixing incremental compensation path is used to replace the component antimixing reconstruction path, specifically including: Establish temperature correction ΔT mix Contribution rate α of deep circulation hot water terminal element H The functional relationship is an empirical compensation model established based on regional calibration samples. Its selection criterion is that the pressure reduction intensity of shallow cold water on apparent temperature varies with the ratio of hot and cold end-members (1-α). H ) / α H Monotonically increasing, applicable to α H Scenarios exceeding the preset applicable lower limit: ΔT mix = κ×(1-α H ) / α H Where, ΔT mix α is the temperature correction amount. H κ is the contribution rate of the deep circulation hot water end-member, and κ is the regional calibration coefficient. κ is determined by least squares fitting using regional calibration samples with reference reservoir temperature, and satisfies the physical constraint of being greater than 0 and less than the upper limit of the temperature difference between the hot and cold end-members. When there are insufficient regional calibration samples, κ is given a priori value based on the reference range of similar geothermal systems, and the corresponding correction result is marked as a low-confidence auxiliary result. The temperature scale correction amount is calculated using the aforementioned functional relationship, and then superimposed onto the original apparent temperature to obtain the corrected temperature scale result.

[0040] In this embodiment, a dual-path correction strategy is employed. The conventional path is component antimixing reconstruction: the original concentrations of temperature-scale-related components in the deep-cycle endmember are calculated using the mass conservation equation and then substituted into the empirical temperature-scale formula. For the two-endmember mixing scenario, the reconstructed concentration is: in The measured component concentrations of the mixed sample are as follows. This represents the characteristic value of the shallow cold water end-member. This formula is directly derived algebraically from the mass conservation equation, without empirical parameters. For scenarios with three or more end-members, the solution is based on a system of multi-end-member equations. Substituting the reconstructed concentration into the temperature scale formula yields the corrected temperature.

[0041] The uncertainty of the reconstructed concentration needs to be quantified. For the two-endmember scenario, the input variables are taken. Sensitivity is: The sensitivity matrix J and the input covariance matrix Calculate variance In the implementation of the project, it is assumed that the three input quantities are independent. Take the diagonal matrix; By estimating the diagonal elements of the Hessian matrix inverse of the objective function after joint optimization convergence, Bootstrap resampling can also be used when the sample size is sufficient.

[0042] Its presence in the denominator indicates that the lower the contribution rate, the more severe the error amplification effect. (Definition of relative error) ,when Higher than a preset threshold (e.g., 30%), or If the value is below a preset threshold (e.g., 0.15), the component reconstruction path is deemed unreliable, and the system switches to the incremental compensation path. The region calibration coefficient (dimension °C) represents the effective temperature reduction when hot and cold end-members are mixed in equal proportions. This model is based on a first-order approximation that the apparent temperature reduction is positively correlated with the volume ratio of hot and cold end-members under two-end-member mechanical mixing conditions. Determined through least-squares fitting of calibration wells, satisfying... Upper limit of temperature difference between hot and cold terminals; when calibration data is insufficient, take the reference value of a similar system and mark the output as low confidence. When the value falls below the applicable lower limit, compensation calculation is stopped and a risk warning is output.

[0043] The above corrections are performed on various temperature scales (including at least silica-based and cationic scales) to generate a set of corrected temperature scales.

[0044] Furthermore, the method also includes calculating the statistical dispersion index of the calibration temperature scale set.

[0045] In this embodiment, the set of calibration temperature scales Calculate the mean, standard deviation, and coefficient of variation. The endmember correction was compared with the original temperature scale set before correction to evaluate the degree of improvement in dispersion.

[0046] Furthermore, the method also includes constructing a temperature scale coordination subset.

[0047] First, a screening process is performed to determine applicability: Na-K scale correction results below 150℃ are excluded; K-Mg scale results above 300℃ are marked as abnormal; and quartz scale results in SiO2 concentration exceeding the solubility limit during reconstruction are marked as abnormal. Scales that do not meet the criteria are not considered for inclusion in the candidate list.

[0048] Temperature scales that meet the conditions are sorted by temperature value, and the temperature difference between adjacent temperature scales is calculated. Scales with a difference less than a preset threshold are grouped into the same candidate subset. If multiple candidate subsets meet the consistency requirement, the subset containing more temperature scales is selected first; if the number of scales is the same, the subset with the smallest dispersion is selected.

[0049] Step S4: Perform coordination screening on the corrected temperature scale results, and perform fusion processing on the coordinated subset of temperature scales obtained from the screening to output the reservoir temperature fusion solution.

[0050] In this embodiment, the fusion temperature for: in, For the temperature scale coordination subset, the first The correction temperature value of each temperature scale, weight , For consistency weight, This is the endmember sensitivity weight.

[0051] Consistency weight Determined by the reciprocal of the number of bits deviating from the coordinating subset: Among them, Med The median of the temperature scale calibration results for the harmonized subset of the temperature scale. To prevent positive numbers with a denominator of zero.

[0052] Endmember sensitivity weights As follows: in, For the first Correction temperature function for a temperature scale, , Take 0.1℃. For The smaller the change in temperature scale after the disturbance, The larger.

[0053] When a regional reference temperature exists, Verification of deviation from reference temperature: If If an output deviation warning is issued, it is recommended to review the end-member parameter assignments, check for any missing end-members, or initiate end-member parameter sensitivity analysis.

[0054] This method is preferably applicable to medium- and high-temperature geothermal systems dominated by conduction and convection, where fluid mixing is primarily mechanical and no significant water-rock reaction occurs. For scenarios with strong water-rock reaction, significant boiling separation, or non-conservative component behavior, its application can be extended by adjusting the semi-conservative component selection and conservation score. When applied to the extended scenarios, the method corrects the mass balance equation in the endmember inversion stage by reducing the conservation score of the affected components, adding a steam condensation endmember, or replacing it with an inert component tracer within the region. In the temperature scale correction stage, a reaction degree factor or a steam separation heat effect correction term is used to impose secondary constraints on the reconstructed temperature, thus extending the applicability of the method to non-ideal mixing geothermal systems.

[0055] The following scenarios are used as examples: 1. Two-end-member mixed scenario: A set of geothermal water samples were collected from a geothermal area in a rift basin. The following observation data were obtained from a typical representative sample: Cl obs =2400 mg / L, SiO2_obs=145 mg / L, δ 18 O obs =-7.2‰. Using these uncorrected observation data directly, the original temperature scale results are: quartz scale 136℃, Na-K scale 198℃, K-Mg scale 170℃, with a mean of 168.0℃, a standard deviation of 25.4℃, and a coefficient of variation (CV) of 15.1%. The significant dispersion between the temperature scales, especially the significantly lower value of the quartz scale, suggests the presence of shallow cold water contamination.

[0056] Retrieve configured endpoint metadata from the regional hydrogeological knowledge base: The representative value of the deep circulating hot water endpoint H is: Cl H =3200 mg / L, SiO2_H=210 mg / L, δ 18 O H =-5.8‰; The representative value of C for the shallow cold water end-member is: Cl C =180mg / L, SiO2_C=38 mg / L, δ18 O C =-10.6‰.

[0057] Based solely on the Cl conservation equation, αH = (Clobs - ClC) / (ClH - ClC) = (2400 - 180) / (3200 - 180) ≈ 0.735. α is then calculated using a single-component Cl inversion method. H The value was corrected to 0.735. The corrected mean was 204.7℃, with a CV of 4.8%. The quartz scale was 191℃, the Na-K scale was 215℃, and the K-Mg scale was 208℃.

[0058] Due to the limited reliability of single-component inversion, the system initiates multi-component joint optimization inversion, assigning Cl, SiO2, and δ to the components. 18 O with different weights (e.g., Cl is 0.47, SiO2 is 0.15, δ) 18 (O = 0.38), by jointly solving the objective function to minimize it, the optimized deep loop endmember contribution rate α is finally obtained. H ≈0.708 (take approximately 0.70 for engineering applications). Let α... H Substituting 0.70 into the incremental correction models for each temperature scale, the predicted values ​​for each scale improved after correction. The corrected quartz scale value increased by approximately 197℃ (from 136℃ before correction, an increase of 61℃). The Na-K scale value increased from 198℃ to approximately 218℃, and the K-Mg scale value increased from 170℃ to approximately 213℃. The corrected mean was 209.3℃, the standard deviation was approximately 9.0℃, and the CV decreased to 4.3%.

[0059] Based on the "cooperative weighted fusion" algorithm using consistency and sensitivity weights, the system assigns higher weights to temperature scales with strong anti-mixing capabilities and tending towards consistency after correction, ultimately calculating a cooperative fusion temperature of 211.2℃. This corrected temperature closely matches the measured temperatures of subsequent deep wells drilled in the area, providing an extremely reliable input for geothermal reservoir evaluation.

[0060] 2. Three-terminal-component mixed scenario: In a coastal geothermal area, three fluids—deep circulating hot water, shallow groundwater, and seawater intrusion—coexist simultaneously, which cannot be accurately described by a two-terminal-component model. Based on the regional geological background, the system constructs and utilizes a three-terminal-component model of "deep circulating hot water (H) - shallow cold water (C) - seawater (M)". Indicators significantly affected by seawater characteristics (such as Cl and Br concentrations and the Cl / Br ratio) and δ0 are selected. 18 Joint optimization inversion was performed using O and δD isotope data.

[0061] By solving the problem, the system obtained the quantitative contribution rates of the three endmembers in each sample. For example, the inversion result for a typical water sample is: deep circulation endmember contribution rate α H= 0.62, contribution rate α of shallow cold water end-member C = 0.28, contribution rate of seawater end-members α M = 0.10. The original apparent temperature scales were 125℃ for quartz, 185℃ for Na-K, and 158℃ for K-Mg, with a mean of 156.0℃ and a CV of 15.6%.

[0062] Using deep loop endmember ratio α H =0.62 Temperature scale anti-mixing correction was performed to isolate the combined cooling and dilution effects of cold water and seawater on the sample. The correction was then applied to various temperature scales, including quartz and Na-K. The corrected quartz temperature was raised to approximately 148℃, Na-K to approximately 201℃, and K-Mg to approximately 192℃, with the CV decreasing to 12.5%. Calculation results show that the corrected harmonic temperature T... coord,corr The temperature was approximately 183℃, about 27℃ higher than the temperature calculated directly from the original mixed water sample, and the dispersion between multiple temperature scales decreased simultaneously. This result effectively eliminated the temperature interference caused by complex mixing in the coastal area, significantly improving the stability of the reservoir temperature fusion solution and the reliability of geological interpretation in this region.

[0063] 3. Production Monitoring Scenario: In the production monitoring of a developed geothermal field, the system dynamically tracks geochemical data on a monthly scale for the same production well. The system periodically extracts the conventional ion and isotope results from each sampling analysis and runs the endmember inversion and temperature scale correction workflow in real time.

[0064] In the tracking curve over a continuous 6 months, the system detected α in the 4th month. H The (time series of the proportion of deep-circulation hot water) showed a significant downward trend (from 0.85 to 0.72), and T coord,corr The (coordinated temperature time series) showed a synchronous decline. Based on this, the system triggered a production warning, indicating that "the effect of shallow cold water intrusion is increasing, and the heat storage is at risk of cooling."

[0065] Based on this early warning, the adaptive suggestion module generated targeted remedial instructions: "Reduce extraction intensity; it is recommended to collect additional deep water samples and focus on retesting stable isotopes to recalibrate end-member parameters." This automated closed-loop monitoring not only achieves static temperature assessment but also extends to engineering applications in dynamic reservoir management, providing strong data and mechanism support for well site production control.

[0066] Example 2 This invention provides a geothermal geochemical analysis system for hybrid end-member inversion and temperature scale correction, comprising: an acquisition unit for acquiring characteristic parameters of multiple fluid end-members based on regional hydrogeological background, wherein the multiple fluid end-members include at least deep circulating hot water end-members and shallow cold water end-members; an end-member inversion unit for calculating the quantitative contribution rate of each fluid end-member to the hybrid sample based on observational data of geothermal chemical samples and the characteristic parameters of each fluid end-member, and determining the contribution rate of the deep circulating hot water end-member; a temperature scale correction unit for correcting at least one geothermal temperature scale using the contribution rate of the deep circulating hot water end-member to obtain the corrected temperature scale result; and an output unit for fusing the corrected temperature scale result to output the reservoir temperature fusion solution.

[0067] In one embodiment, the temperature scale correction unit includes a component antimixing reconstruction module and an antimixing incremental compensation module, and is configured to perform path switching between the component antimixing reconstruction module and the antimixing incremental compensation module when the contribution rate of the deep circulation hot water end-member is lower than a preset threshold or the relative error of the reconstruction concentration is higher than a preset error threshold; the output unit is used to calculate the collaborative fusion weight based on the consistency weight and end-member sensitivity weight of each temperature scale in the temperature scale coordination subset, and use the collaborative fusion weight to perform weighted fusion of each corrected temperature scale result in the temperature scale coordination subset to output the reservoir temperature fusion solution.

[0068] 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 geothermal geochemical analysis method combining hybrid endmember inversion and temperature scale correction, characterized in that, Includes the following steps: Based on the regional hydrogeological background, characteristic parameters of multiple fluid end-units are obtained, and the multiple fluid end-units include at least deep circulating hot water end-units and shallow cold water end-units. Based on the observation data of geothermal chemical samples and the characteristic parameters of each fluid end-member, the quantitative contribution rate of each fluid end-member to the mixed sample is calculated by inversion, and the contribution rate of the deep circulation hot water end-member is determined. The contribution rate of the deep-circulation hot water end-unit is used to correct at least one geothermal temperature scale to obtain the corrected temperature scale result, including: based on the mass conservation equation, using the contribution rate of the deep-circulation hot water end-unit, back-calculating the reconstructed concentration of the chemical components involved in the temperature scale calculation in the deep-circulation hot water end-unit; substituting the reconstructed concentration into the corresponding empirical formula of the geothermal temperature scale to obtain the corrected temperature scale result; When the contribution rate of the deep circulation hot water terminal element is lower than a preset threshold, or the relative error of the reconstructed concentration calculated based on error propagation is higher than a preset error threshold, it is determined that the component antimixing reconstruction path has a risk of numerical instability. An antimixing incremental compensation path is then used to replace the component antimixing reconstruction path, specifically including: establishing a temperature correction amount ΔT. mix Contribution rate α of deep circulation hot water terminal element H The functional relationship is an empirical compensation model established based on regional calibration samples. The empirical compensation model is based on the fact that the pressure reduction intensity of shallow cold water on apparent temperature varies with the ratio of cold and hot end-members (1-α). H ) / α H Monotonically increasing, applicable to α H Scenarios exceeding the preset applicable lower limit: ΔT mix = κ×(1-α H ) / α H， Where, ΔT mix α is the temperature correction amount. H The contribution rate of the deep circulation hot water end-member is κ, and the regional calibration coefficient is κ. The regional calibration coefficient is determined by least squares fitting using regional calibration samples with reference reservoir temperature, and satisfies the physical constraint of being greater than 0 and less than the upper limit of the temperature difference between the hot and cold end-members. When the regional calibration samples are insufficient, the regional calibration coefficient is given by prior values ​​based on the reference range of similar geothermal systems, and the corresponding correction results are marked as low-confidence auxiliary results. The temperature scale correction amount is calculated using the functional relationship, and the temperature scale correction amount is superimposed on the original apparent temperature to obtain the corrected temperature scale result. For binary endmember hybrid scenarios, the reconstructed concentration is calculated using the following formula: X H * = (X obs - (1-a H ) × X C ) / a H Among them, X H *Represents the reconstitution concentration in the deep circulation hot water terminal element, X obs X represents the measured component concentration of the mixed sample. C For the corresponding component characteristic value of the shallow cold water end member, α H Contribution rate to deep circulation hot water terminal components; For scenarios with three or more endmembers, the reconstructed concentration is obtained by solving the mass conservation equations for multiple endmembers. The calibration temperature scale results are subjected to coordination screening, and the resulting coordinated subset of temperature scales is fused to output the reservoir temperature fusion solution.

2. The method according to claim 1, characterized in that, When the regional hydrogeological background is affected by steam or seawater, the multiple fluid end-units also include steam condensation end-units or seawater influence end-units.

3. The method according to claim 1, characterized in that, Determining the contribution rate of the deep circulation hot water terminal includes: A set of mass balance equations is constructed based on multiple conserved or semi-conserved components, including Cl, SiO2, and the stable isotope δ. 18 One or more of O and δD; A joint optimization algorithm is used to solve the problem. Dynamic weights, determined adaptively based on the observation error variance of each component, component conservatism scores, or iterative residuals, are used to minimize the sum of squared residuals between measured and predicted values. A constraint is applied that the sum of the contribution rates of each fluid endmember is 1, and the contribution rate of each fluid endmember is between 0 and 1, thus obtaining the quantitative contribution rate of each fluid endmember to the mixed sample. The iterative residual adaptation employs a robust statistical method to reduce the weights of component residuals exceeding a cutoff threshold. The cutoff constant corresponding to the cutoff threshold is between 1.0 and 1.

5. The contribution rate of the deep circulation hot water terminal is determined from the contribution rates of each fluid terminal.

4. The method according to claim 1, characterized in that, The geothermal temperature scale includes multiple temperature scales, forming a multi-scale combination; the multi-scale combination includes at least silica-based temperature scales and cationic temperature scales; The multiple temperature scale combinations are calibrated respectively to generate a set of calibrated temperature scales.

5. The method according to claim 4, characterized in that, Also includes: Calculate the statistical dispersion index of the calibration temperature scale set; Based on the statistical dispersion index and combined with the applicable conditions of the temperature scale, temperature scales that meet the consistency threshold requirements are identified, and a temperature scale coordination subset is formed. Based on the consistency weight and endmember sensitivity weight of each temperature scale in the temperature scale coordination subset, the collaborative fusion weight is calculated, and the results of each corrected temperature scale in the temperature scale coordination subset are weighted and fused using the collaborative fusion weight to calculate and output the reservoir temperature fusion solution. The consistency weight Determine using the following formula: in, For the temperature scale coordination subset, the first The corrected temperature value of each temperature scale, Med The median of the temperature scale correction results for the temperature scale harmonization subset is given. To prevent positive numbers with a denominator of zero.

6. An analytical system for performing the geothermal geochemical analysis method of hybrid endmember inversion and temperature scale correction as described in claim 1, characterized in that, include: The acquisition unit is used to acquire characteristic parameters of multiple fluid end-units based on the regional hydrogeological background. The multiple fluid end-units include at least a deep circulating hot water end-unit and a shallow cold water end-unit. The end-member inversion unit calculates the quantitative contribution rate of each fluid end-member to the mixed sample based on the observation data of the geothermal chemical sample and the characteristic parameters of each fluid end-member, and determines the contribution rate of the deep circulation hot water end-member. The temperature scale correction unit uses the contribution rate of the deep circulation hot water terminal element to correct at least one geothermal temperature scale and obtain the corrected temperature scale result; The output unit is used to perform coordination screening on the corrected temperature scale results, and to perform fusion processing on the selected coordinated subset of temperature scales to output the reservoir temperature fusion solution.

7. The system according to claim 6, characterized in that, The temperature scale correction unit includes a component antimixing reconstruction module and an antimixing incremental compensation module, and is configured to perform path switching between the component antimixing reconstruction module and the antimixing incremental compensation module when the contribution rate of the deep circulation hot water end-member is lower than a preset threshold or the relative error of the reconstruction concentration is higher than a preset error threshold; the output unit is used to calculate the collaborative fusion weight based on the consistency weight and end-member sensitivity weight of each temperature scale in the temperature scale coordination subset, and use the collaborative fusion weight to perform weighted fusion of each corrected temperature scale result in the temperature scale coordination subset to output the reservoir temperature fusion solution.

Images