Karst geothermal development evaluation method and system based on 3E- super efficiency DEA model

Abstract

The application discloses a karst geothermal development evaluation method and system based on a 3E- super efficiency DEA model, and relates to the technical field of geothermal resource development and utilization evaluation. The method is realized through the following steps: 1) constructing an “energy-economy-environment” (3E) three-dimensional evaluation index system, collecting enterprise operation, geological exploration and environmental monitoring data; 2) adopting Z-score standardization to process the data and constructing an index matrix; 3) introducing a karst characteristic correction coefficient, triggering regional adaptation according to reservoir permeability and ecological sensitivity, and performing multiplicative correction on key index columns according to a preset coefficient; 4) establishing a super efficiency DEA model containing slack variables and residual variables, obtaining efficiency values through Matlab solving, further distinguishing and sorting effective development units, and outputting input redundancy and output deficiency to identify bottleneck directions; and 5) combining heat reservoir temperature and permeability to identify difference driving factors, and outputting popularization priority, development improvement scheme and policy countermeasures.

Description

Technical Field

[0001] This invention relates to the field of geothermal resource development and utilization evaluation technology, and more specifically, to a method and system for evaluating karst geothermal development based on the 3E-ultra-efficiency DEA model. Background Technology

[0002] Geothermal resources, as a clean and renewable energy source, are of great significance for optimizing the energy structure and protecting the ecological environment. Yibin City is located in a typical karst landform area and possesses abundant deep and shallow karst geothermal resources. However, the development and utilization of geothermal resources in this region faces the following core issues:

[0003] (1) Existing geothermal development evaluation methods are not targeted enough: Traditional evaluation methods focus on a single economic indicator (such as return on investment) and do not take into account key dimensions such as energy utilization efficiency and environmental impact. They cannot fully reflect the adaptability of karst geothermal resources with their characteristics of "high development difficulty and high ecological sensitivity".

[0004] (2) Limited accuracy of evaluation model: Conventional DEA (data envelopment analysis) model has the defect that "effective cells cannot be further sorted", making it difficult to accurately distinguish the efficiency differences of different geothermal development methods, resulting in a lack of scientific basis for the selection of development methods;

[0005] (3) Lack of regional adaptability: The existing evaluation methods do not take into account the geological characteristics of karst geothermal resources (such as reservoir permeability and thermal reservoir temperature distribution), nor do they design a differentiated evaluation system for the differences in development technologies between deep and shallow geothermal resources. This results in the evaluation results being out of touch with actual development needs and cannot provide effective guidance for the optimization of the regional geothermal industry.

[0006] Therefore, there is an urgent need for a development efficiency evaluation method that takes into account the three dimensions of "energy-economy-environment", is adapted to the characteristics of karst geothermal energy, and has high-precision sorting function. This method would address the shortcomings of existing technologies, such as poor targeting, low accuracy, and insufficient adaptability, and provide technical support for the scientific development and industrial optimization of geothermal resources in typical karst areas such as Yibin City. Summary of the Invention

[0007] To overcome the aforementioned deficiencies in existing technologies, this invention provides a karst geothermal development evaluation method based on a 3E-ultra-efficient DEA model. This method constructs a three-dimensional "energy-economy-environment" index and standardizes it using Z-scores. It introduces karst characteristic correction coefficients to multiplicatively adjust the weights of key indicators for permeability and ecological sensitivity. An ultra-efficient DEA model with relaxation / residual variables is used to solve for θ and rank the results, outputting differentiated optimization strategies to address the problems mentioned in the background.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A method for evaluating karst geothermal development based on the 3E-ultra-efficiency DEA model includes the following steps:

[0010] Step 1: Construct a 3E evaluation index system for the development of karst geothermal resources. The 3E evaluation index system includes energy dimension indicators, economic dimension indicators, and environmental dimension indicators. The raw data of each indicator are processed using the Z-score standardization method to obtain a standardized index matrix.

[0011] Step 2: Based on the standardized index matrix, construct the 3E-super-efficiency DEA evaluation model, using constraint indicators as the input set and contribution indicators as the output set. The model achieves the ranking of effective decision-making units by introducing slack variables and residual variables, and introduces the karst geothermal characteristic correction coefficient to adapt the key indicators to the region.

[0012] Step 3: For karst geothermal resources in the target area, select typical deep and shallow geothermal development methods as decision-making units, collect and verify the original data of the 3E indicators of each decision-making unit, and after completing the missing data completion and abnormal data verification, form the standardized indicator matrix according to the rules of Step 1.

[0013] Step 4: Substitute the standardized index matrix into the 3E-super-efficiency DEA evaluation model constructed in Step 2, and obtain the efficiency value of each decision unit by solving it through Matlab. Based on the efficiency value, complete the ranking and carry out difference analysis. Combine reservoir permeability and thermal reservoir temperature to identify the driving factors of efficiency differences.

[0014] Step 5: Based on the current status of the geothermal industry in the target area and the efficiency calculation results, output differentiated development optimization strategies. These strategies include at least three categories: large-scale promotion, technological improvement, and policy support.

[0015] As a further aspect of the present invention, the energy dimension indicators in step one include geothermal resource utilization rate, unit energy consumption output efficiency, and sustainable utilization coefficient of thermal storage resources; the economic dimension indicators include unit investment return rate, development and operation cost payback period, and unit production capacity economic benefits; the environmental dimension indicators include carbon dioxide emission reduction, water resource consumption intensity, and geological environmental impact level, wherein the geological environmental impact level is quantified as level 1 to level 5.

[0016] As a further aspect of the present invention, the energy dimension indicators and economic dimension indicators use a statistical period of 12 months as the data caliber, and the environmental dimension indicators use the monitoring and accounting data of the competent authorities as the caliber, and complete the unit conversion and direction consistency verification before entering the standardization process.

[0017] As a further aspect of the present invention, the karst geothermal characteristic correction coefficient in step two is generated based on the reservoir permeability and ecological sensitivity of the target area, which can increase the weight of the geothermal resource utilization rate index in the model and increase the weight of the geological environment impact level index in the model.

[0018] As a further aspect of the present invention, the weighted regional adaptation in step two is implemented in the form of "indicator column regional adaptation coefficients": the regional adaptation coefficients of the column corresponding to the geothermal resource utilization rate index are adjusted from 0.30 to 0.40, and the regional adaptation coefficients of the column corresponding to the geological environment impact level index are adjusted from 0.25 to 0.35. The adjustments are then applied to the corresponding index columns in the standardized index matrix in a multiplicative correction manner. Except for the key columns, the remaining index columns remain unchanged, thereby achieving regional adaptation oriented towards karst characteristics without changing the DEA model solution structure.

[0019] As a further aspect of the present invention, the data collection in step three includes obtaining reservoir permeability and thermal reservoir temperature data through on-site surveys, obtaining investment, operating costs and production capacity data through enterprise surveys, and obtaining environmental impact monitoring data from the competent authorities.

[0020] As a further aspect of the present invention, the data verification in step three includes unifying the time caliber, unifying the unit caliber, completing missing data, and reviewing abnormal data. The missing data completion adopts a completion rule that combines the median of samples from similar development methods with the scale coefficient.

[0021] As a further aspect of the present invention, in step four, when the efficiency value satisfies θ not less than 1, the corresponding decision unit is determined to be an effective decision unit; when the efficiency value satisfies θ less than 1, the corresponding decision unit is determined to be an invalid decision unit; and directions for improvement are given based on slack variables and residual variables.

[0022] As a further aspect of the present invention, the large-scale promotion in step five includes prioritizing the promotion of shallow geothermal heat pump heating and cooling for hot spring tourism and aquaculture, and improving the supporting heating pipeline network; technological improvement includes implementing reservoir fracturing and permeability enhancement for deep geothermal energy to improve energy utilization; and policy support includes providing technical subsidies for deep geothermal development to reduce the improvement costs for enterprises.

[0023] As a further aspect of the present invention, the difference analysis in step four includes grouping and comparing deep development methods and shallow development methods, and using reservoir permeability and thermal reservoir temperature as driving factors to explain the efficiency differences.

[0024] Furthermore, the present invention also provides a karst geothermal development evaluation system based on the 3E-ultra-efficiency DEA model. This system can be implemented by a general-purpose computing device by executing an evaluation program. The evaluation program performs standardized processing, karst characteristic regional adaptation correction, and model solving on the collected enterprise operation, geological exploration, and environmental monitoring data according to a fixed computing link, thereby outputting a ranking of development methods, bottleneck indicators, and countermeasure suggestions that can be used for engineering implementation.

[0025] A karst geothermal development evaluation system based on the 3E-super-efficiency DEA model is characterized by comprising: a data acquisition interface module for acquiring raw data of energy, economic, and environmental indicators for karst geothermal development in a target area; a preprocessing module for standardizing the time caliber, converting units, and verifying directional consistency of the raw data, and processing the raw data of each indicator using the Z-score standardization method to obtain a standardized indicator matrix; a karst adaptation module for generating correction coefficients for karst geothermal characteristics and performing multiplicative corrections on key indicator columns based on reservoir permeability and ecological sensitivity to achieve regional adaptation; a model construction and solution module for constructing a 3E-super-efficiency DEA evaluation model containing slack variables and residual variables based on the standardized indicator matrix after regional adaptation, and obtaining the efficiency values ​​of each decision-making unit through Matlab solution to achieve ranking of effective decision-making units; and an output module for performing difference analysis based on the efficiency values, identifying the driving factors of efficiency differences in conjunction with reservoir temperature and permeability, and outputting promotion, improvement, and policy countermeasures.

[0026] The technical effects and advantages of the karst geothermal development evaluation method based on the 3E-ultra-efficiency DEA model of this invention are as follows:

[0027] This invention constructs a three-dimensional indicator system of "energy-economy-environment" and couples it with the super-efficiency DEA ranking, so that the evaluation can be expanded from a single economic return to a comparison of energy efficiency and ecological constraints. It can accurately characterize the comprehensive performance of development methods in the highly sensitive background of karst areas, avoid resource waste and environmental risks caused by unilateral pursuit of benefits, and is applicable to deep and shallow layer comparison and screening.

[0028] This invention addresses the challenges of large fluctuations in geological data and nonlinear coupling between environmental indicators and productivity benefits in karst areas. It introduces a "karst characteristic correction coefficient" into the 3E-ultra-efficient DEA model, using reservoir permeability and ecological sensitivity as triggering conditions to perform traceable regional adaptation corrections on key indicator columns. The reservoir temperature is used to identify driving factors for subsequent difference analysis but is not used as a triggering condition for the correction coefficient. This ensures the comparability and interpretability of evaluation results across different karst regions and allows for direct identification of bottlenecks such as "increasing permeability, reducing energy consumption, and mitigating disturbance."

[0029] This invention adapts and optimizes the construction and solution chain of the ultra-efficient DEA model for karst geothermal evaluation scenarios. Addressing the instability of the feasible region caused by large fluctuations in geological data and the nonlinear correlation between environmental indicators and production efficiency in karst areas, it first performs directional consistency labeling and positive value processing on constraint indicators such as payback period, water consumption intensity, and impact level, ensuring that the input / output matrix satisfies the non-negativity and unidirectional constraints of linear programming. While maintaining the DEA input-output structure unchanged, a karst characteristic correction coefficient is introduced to multiplicatively correct the key indicator columns. The correction coefficient, triggering conditions, and matrix summaries before and after correction are written into the correction log, ensuring "same scope, recalculateable" feasible solutions for different karst zones from a constraint caliber perspective. Furthermore, at the objective optimization level, a solution strategy of "ultra-efficiency differentiation + relaxation improvement output" is adopted: the ultra-efficiency mechanism performs comparable ranking of effective units, and relaxation / residual variables are introduced as constraint input redundancy and output insufficiency to provide executable improvement directions. By combining Matlab solutions with the identification of differential driving factors, the final output includes strategies for large-scale promotion, technological improvement, and policy support, forming a closed loop from evaluation to implementation. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the 3E evaluation index system and constraint / contribution attribute annotation of the present invention;

[0031] Figure 2 This is a schematic diagram illustrating the data preprocessing and solution sequence of the 3E-super-efficiency DEA model in this invention.

[0032] Figure 3 This is a schematic diagram illustrating the generation and feasibility verification of countermeasures based on efficiency ranking and bottleneck identification in this invention.

[0033] Figure 4 This is a schematic diagram of the module structure of the evaluation system of the present invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Example 1

[0036] This embodiment takes the development of karst geothermal resources in Yibin City as an example, conducts a comprehensive efficiency evaluation of typical deep and shallow geothermal development methods, and outputs differentiated optimization strategies for engineering implementation. The evaluation process uses a three-dimensional indicator system of energy, economy, and environment to construct input data. Z-score standardization is used to form a standardized indicator matrix. Based on the standardized indicator matrix, a 3E-super-efficiency DEA evaluation model is constructed. The super-efficiency DEA mechanism is used to further differentiate and rank effective decision-making units. Relaxed variables and residual variables are used to identify input redundancy and output insufficiency to complete bottleneck identification. A karst geothermal characteristic correction coefficient is introduced to regionally adapt the weights of key indicators. Finally, the efficiency value θ is solved using Matlab to complete the ranking and difference analysis. Combined with reservoir permeability and reservoir temperature, the driving factors of efficiency differences are identified, and three types of countermeasures are output: large-scale promotion, technological improvement, and policy support.

[0037] The evaluation process in this embodiment consists of a series of steps: data acquisition and verification, missing data completion, Z-score standardization to form a standardized index matrix, positive value processing of input columns, multiplicative correction of key index columns, construction of the 3E-super-efficiency DEA model, solving for the output efficiency value θ and relaxation / residual information in Matlab, analysis of differences between deep and shallow groupings and identification of driving factors, and strategy generation and feasibility verification. The output of the previous step serves as the input of the next step, forming a recalcible closed-loop computational chain.

[0038] The evaluation objects are divided into decision-making units according to their development methods. The statistical boundaries of each decision-making unit are fixed as the same region, the same statistical period, and the same measurement caliber. The statistical period is uniformly 12 consecutive months. Energy and economic data are summarized using a 12-month caliber, while environmental data adopts the monitoring and accounting caliber of the competent authority and is mapped to the same statistical period. In this example, four typical methods are selected as decision-making units in the demonstration area, covering the differences between shallow and deep geothermal layers and ensuring data availability: shallow geothermal heat pump heating and cooling, medium-deep geothermal well group heating, hot spring cultural tourism integrated utilization, and geothermal agricultural and aquaculture heating.

[0039] Simultaneously, regional geological constraint variables are collected for two purposes: firstly, as input for the correction coefficient of karst geothermal characteristics; and secondly, for identifying driving factors in difference analysis. Geological constraint variables include at least reservoir permeability and reservoir temperature, supplemented by formation thickness for engineering interpretation and group determination. Permeability uses representative values ​​from well test interpretation or regional hydrogeological results, in mD; reservoir temperature uses representative values ​​from well logging, temperature recovery tests, or long-term operational monitoring, in degrees Celsius; formation thickness uses the effective thickness related to the reservoir from geological exploration results, in meters. In the demonstration area, deep reservoir temperatures are not less than 90℃, and shallow reservoir temperatures are in the range of 25–60℃; deep permeability is in the range of 0.05–0.10 mD, and shallow permeability is in the range of 0.10–0.30 mD.

[0040] The evaluation index system consists of energy, economic, and environmental dimensions. The energy dimension includes geothermal resource utilization rate, unit energy consumption output efficiency, and sustainable utilization coefficient of geothermal storage resources. The economic dimension includes unit investment return rate, development and operation cost payback period, and unit production capacity economic benefits. The environmental dimension includes carbon dioxide emission reduction, water resource consumption intensity, and geological environmental impact level, which is quantified from level 1 to level 5, with higher values ​​indicating stronger impact and stricter constraints. Figure 1 As shown, the 3E evaluation indicator system is grouped into three parts: energy, economy, and environment, with each indicator's attribute category clearly marked. This facilitates the subsequent mapping of constraint-type indicators to the input set and contribution-type indicators to the output set. In the economic dimension, raw data related to revenue are collected using an annualized net revenue caliber. Annualized net revenue consists of main business revenue and subsidy revenue, minus costs for operation and maintenance, energy purchases, necessary environmental governance, and compliance monitoring. Subsidy revenue is only included when the corresponding subsidy standard trigger conditions are met; if the trigger conditions are not met or verification materials are incomplete, subsidy revenue is set to 0, avoiding the inclusion of unattainable policy benefits in the revenue caliber. For situations where multiple subsidies may coexist within the same decision-making unit, values ​​are determined according to the superposition rules or upper limit rules stipulated by the competent authority, and the adopted clauses, applicable years, and upper limit constraints are marked in the archived records to ensure that the revenue caliber is recalculated and not artificially inflated.

[0041] Data collection is conducted from three sources: field surveys to obtain reservoir permeability, geothermal temperature, and formation thickness; enterprise surveys to obtain investment amounts, operating costs, production capacity parameters, heat and electricity consumption measurements; and environmental impact monitoring and accounting data from relevant government departments. Simultaneously, geothermal development subsidy standards from policy data are collected (e.g., standard texts and applicable conditions for subsidies based on heat supply, installed capacity, or project type). Subsidy standards are not included in the model as one of the nine indicators of the 3E framework, but are used for verifying economic benefits, explaining policy factors in difference-driven analysis, and determining the applicable conditions for policy support measures to avoid mismatches between measures and policy conditions.

[0042] The subsidy standard data collection adopts a parameterized archiving method, recording at least the subsidy type, measurement scope, subsidy unit price or quota, triggering conditions, verification materials, upper limit rules, and applicable year. Subsidy types include subsidies measured by effective heating and cooling capacity, subsidies measured by installed capacity, and quota subsidies based on project type. The measurement scope is uniformly a 12-month scope consistent with the statistical period of this embodiment, and consistent with the measurement scope of effective heating and cooling energy. Triggering conditions at least include project category, heating capacity or installed capacity threshold, monitoring compliance requirements, and additional restrictions for ecologically sensitive areas. Verification materials at least include clauses from the competent authority's documents, project filing or acceptance data, measurement ledgers, and monitoring compliance certificates. Each decision-making unit marks its subsidy availability determination result in the archiving table and records the judgment basis clauses and corresponding material numbers, making the subsidy parameters verifiable and traceable.

[0043] Before standardization, unit conversion and directional consistency verification were completed. Unit conversion requirements: energy was standardized to MWh or equivalent units and maintained consistency across the entire sample; revenue was standardized to the same currency unit and expressed on an annualized basis; water consumption was standardized to m³ / MWh; emission reduction was standardized to t / year; and impact levels were integers from 1 to 5. Directional consistency verification adopted fixed rules: payback period, water consumption intensity, and impact level were labeled as constraint indicators; utilization rate, energy efficiency, sustainability, revenue, unit capacity benefit, and emission reduction were labeled as contribution indicators. The labeling results were then kept consistent with subsequent input-output mappings to avoid directional confusion leading to solution bias.

[0044] Missing data completion employs a rule combining the median of samples from similar development methods with a scale factor. The scale factor is preferentially based on the ratio of designed heating and cooling capacity (MW). Only when design capacity is missing or inconsistent in scope is the effective heating and cooling energy (MWh) of the statistical period used as a substitute scale benchmark, with the reason and value for the substitute noted in the archived records. After completion, a consistency check of orders of magnitude is performed; if it fails, the data is returned for review. Abnormal data review is conducted based on original metering records, monitoring report numbers, and financial vouchers to ensure the traceability of input data.

[0045] The original data of the 3E indicators for the four development methods in the demonstration area of ​​this embodiment are shown in Table 1. The statistical period for all of them is 12 months.

[0046] Table 1. Original data of the 3E indicators for the four development methods.

[0047] index Shallow ground source heat pump heating and cooling Heating from medium-deep geothermal well groups Comprehensive utilization of hot spring culture and tourism Geothermal agriculture and aquaculture heating Geothermal resource utilization rate E1 (%) 72 68 75 70 Energy efficiency per unit of energy consumption, E2 (kWh / kWh) 4.60 5.20 6.00 5.50 Sustainable utilization coefficient E3 0.90 0.82 0.86 0.88 Unit investment return rate C1 (%) 11.50 9.20 13.80 12.20 Payback period C2 (years) 7 10 6 5 Economic benefit per unit of production capacity C3 (ten thousand yuan / MW) 320 410 560 380 <![CDATA[CO2 emission reduction V1 (t / year)]]> 52000 65000 48000 36000 Water consumption intensity V2 (m³ / MWh) 0.03 0.10 0.08 0.06 Geological environment impact level V3 (1-5) 2 3 4 2

[0048] To support the generation of correction coefficients and the identification of driving factors, representative permeability, reservoir temperature, effective layer thickness, and ecological sensitivity are classified and grouped as shown in Table 2.

[0049] Table 2. Collection of Geological Constraint Variables

[0050] Development Method Representative penetration (mD) Representative thermal storage temperature (°C) Effective layer thickness (m) Ecological sensitivity assessment Shallow ground source heat pump heating and cooling 0.20 40 45 middle Heating from medium-deep geothermal well groups 0.08 95 62 middle Comprehensive utilization of hot spring culture and tourism 0.12 80 55 high Geothermal agriculture and aquaculture heating 0.18 55 48 middle

[0051] The determination of ecological sensitivity is based on data such as ecological red lines, water source protection areas, and geological environmental risk zoning disclosed by the competent authorities. It is classified into high, medium, and low levels, and the determination of triggering coefficients and the applicability of countermeasures are checked.

[0052] To reflect the explanatory role of subsidy standards in the differences in economic performance, the subsidy standards are archived in a parameterized manner and their applicability is marked for four types of decision-making units, as shown in Table 3. Table 3 is not included in the model as the 3E nine indicators, but is used for income caliber verification, explanation of policy driving factors, and verification of policy feasibility.

[0053] Table 3. Subsidy Standard Parameter Archiving and Applicability Labeling

[0054] project Shallow ground source heat pump heating and cooling Heating from medium-deep geothermal well groups Comprehensive utilization of hot spring culture and tourism Geothermal agriculture and aquaculture heating Subsidy type By heat supply / By installed capacity By project type / By heating supply Subsidies by type or specific item By heat supply / By installed capacity Measuring caliber MWh / MW MWh / project Project or supervisor's perspective MWh / MW Triggering Condition Summary Heating supply or installed capacity threshold, monitoring compliance Project category, acceptance criteria, and monitoring compliance. Ecologically sensitive additional restrictions and compliance requirements Heating threshold, complete metering, and compliant monitoring Availability determination Available / Unavailable Available / Unavailable Available / Unavailable Available / Unavailable Summary of Limit / Stacking Rules Cap clause or superimposed clause Cap clause Cap clause Cap clause or superimposed clause Key points for verifying materials Document terms, measurement ledgers, monitoring certificates Filing / acceptance, measurement ledger, monitoring certificate Approval documents and monitoring compliance materials Document terms, measurement ledgers, monitoring certificates

[0055] The original data for the nine indicators in Table 1 are standardized using Z-scores for each indicator column. The standardization formula is as follows:

[0056]

[0057] in, These are the standardized indicator values; For the first The decision-making unit in the first... The original values ​​for each indicator; For the first The average of the indicators; For the first The standard deviation of each indicator. After standardization, a standardized indicator matrix is ​​obtained.

[0058] To avoid negative or zero values ​​after standardization, which would lead to constraints... "Numerically unstable, this embodiment performs uniform positive value processing on all constrained input candidate columns: for each input index column, calculate the minimum value (min) of its standardized values. When min is not greater than 0, add the same translation constant K to all standardized values ​​of the column to make the minimum value of the column 1; when min is greater than 0, do not translate. The translation constant K is archived and saved with the evaluation results for recalculation. This processing does not change the relative differences within the column, but only changes the numerical benchmark to meet the solution stability requirements. For example..." Figure 2 As shown, standardization, positiveizing of input columns, multiplicative correction of key indicator columns, and model solving are performed in a fixed order to ensure consistency in the computational chain.

[0059] This embodiment constructs a 3E-super-efficiency DEA evaluation model based on a standardized index matrix. The objective function formula is as follows:

[0060]

[0061] In the formula, This is the efficiency value; Input indicator weights; For the first The first decision-making unit One input indicator value; To output the indicator weights; For the first The first decision-making unit Each output indicator value; Enter the number of indicators; The number of output indicators is defined. The input and output sets are fixedly mapped as follows: the input set includes payback period C2, water consumption intensity V2, and impact level V3; the output set includes utilization rate E1, energy efficiency E2, sustainability E3, return on investment C1, unit production capacity benefit C3, and emission reduction V1. Among these, a larger value for impact level V3 indicates a stronger environmental constraint. Using it as an input set can make the model tend to pursue outputs while meeting the lower environmental cost, thus reflecting the inhibitory effect of constraint indicators.

[0062] The constraints are expressed by the following formula:

[0063]

[0064] When solving the 3E-super-efficient DEA evaluation model, slack variables and residual variables are introduced to address input-output redundancy and output insufficiency. The super-efficient DEA mechanism is used to further distinguish and rank effective decision-making units, so that effective units can also obtain comparable ranking results.

[0065] To reflect regional constraints such as "low permeability in karst areas leading to increased reinjection resistance, higher maintenance costs, and longer payback periods" and "more rigid control of disturbance risks in ecologically sensitive areas," this invention sets a correction coefficient for karst geothermal characteristics and adopts a determination method of "threshold-level triggering—preference intensity quantification—multiplicative effect—log archiving recalculation."

[0066] (1) Triggering variables and classification criteria: The reservoir permeability P (mD) and ecological sensitivity S (high / medium / low) of each decision unit are read. Among them, P comes from the representative value of well test interpretation or regional hydrogeological results; S is determined and classified based on the data of ecological red lines, water source protection areas, geological environmental risk zoning, etc. of the competent authorities.

[0067] (2) Triggering rules: When P < 0.10mD, it is judged as low permeability and triggers "resource utilization enhancement"; when S is high, it is judged as high sensitivity and triggers "enhanced environmental constraints"; both can be triggered simultaneously. The reason for setting the 0.10mD threshold is that: under karst low permeability conditions, the limited reinjection channel is more likely to lead to suppressed recoverable flow, increased reinjection resistance and increased maintenance frequency of scaling and blockage, thus making the improvement of utilization rate a key bottleneck; under ecologically high sensitivity conditions, the disturbance risk has a stronger constraint on the feasibility of the project, and it is necessary to increase the relative importance of environmental constraint indicators to avoid the evaluation bias of "optimal economy / energy efficiency but unacceptable environment".

[0068] (3) Quantification of preference intensity and generation of coefficients: The baseline coefficients for columns E1 and V3 in the baseline region adaptation coefficient table are preset to 0.30 and 0.25, respectively; when the low-permeability level is triggered, the region adaptation coefficient of column E1 is adjusted to 0.40; when the high-sensitivity level is triggered, the region adaptation coefficient of column V3 is adjusted to 0.35. The basis for this adjustment is that the utilization rate has a stronger explanatory power for efficiency differences in the low-permeability case, and the influence level has a stronger constraint on feasibility in the high-sensitivity case, so the preference intensity is increased respectively.

[0069] (4) Multiplicative Correction (Correction Coefficient Definition): Convert the above preference intensity into multiplicative correction coefficients, defining kE1 = 0.40 / 0.30 (kE1 = 1 if not triggered), kV3 = 0.35 / 0.25 (kV3 = 1 if not triggered). Apply column multiplicative correction to the standardized index matrix Z to obtain Z': Z'(:,E1) = Z(:,E1) × kE1, Z'(:,V3) = Z(:,V3) × kV3, with the remaining columns remaining unchanged.

[0070] (5) Traceable archiving: The source of P and S values, hierarchical results, triggering results, kE1 and kV3 values, and the modified matrix summary are written into the "adaptation correction log" so that any third party can recalculate the same result under the same input and rules.

[0071] The standardized index data, after being positive-valued in the input columns and multiplicative corrected in the key columns, are substituted into the model. Matlab is used to solve for the efficiency value θ and relaxation / residual output of each decision unit. The Matlab workflow is fixed as follows: reading the standardized matrix and translation constant record, verifying the positive values ​​of the input columns, performing multiplicative corrections on the E1 and V3 columns, constructing the solution input structure, calling the linear programming solution module to output θ and relaxation / residual, forming the sorting table and bottleneck table, and performing back-substitution verification on randomly sampled decision units to ensure that the solution output is consistent with the input data.

[0072] The efficiency judgment rule is fixed as follows: θ not less than 1 is considered valid, and θ less than 1 is considered invalid, and the sorting is completed accordingly. The output of the demonstration area solution is as follows: geothermal agricultural and aquaculture heat θ=1.180, shallow ground source heat pump heating and cooling θ=1.120, hot spring cultural tourism comprehensive utilization θ=1.050, and medium-deep geothermal well group heating θ=0.930.

[0073] To transform the relaxed / residual outputs into actionable improvement directions, a fixed mapping rule based on the index columns is adopted: a positive redundancy in the input column indicates that the constraint-type index can be compressed, and a positive insufficiency in the output column indicates that the contribution-type index can be improved. The specific generation rules are as follows: When the payback period redundancy is positive, priority is given to shortening the payback period by reducing operation and maintenance costs and maintenance frequency, and adjusting net cash flow in conjunction with subsidy standards; when the water consumption intensity redundancy is positive, priority is given to reducing new water intake and increasing the reinjection ratio or adopting closed heat exchange to reduce water consumption; when the impact level redundancy is positive, priority is given to increasing monitoring frequency and optimizing reinjection water quality and operating conditions, and setting threshold linkage to limit extraction to reduce risk level; when the utilization rate is insufficient, priority is given to using fracturing and permeability enhancement and reinjection channel optimization for deep layers, and to using well site layout and load matching optimization for shallow layers; when the energy efficiency is insufficient, priority is given to reducing external power consumption and optimizing scheduling; when the sustainability is insufficient, priority is given to improving the balance between extraction and injection and controlling temperature drop and production capacity decay; when the revenue and unit production capacity efficiency are insufficient, priority is given to improving load stability and operating hours and improving pipeline network support; when the emission reduction is insufficient, priority is given to increasing the substitution of fossil energy for heating and optimizing the power consumption structure.

[0074] After obtaining the efficiency ranking and bottleneck table, a difference analysis was conducted, comparing deep and shallow development methods. Permeability and reservoir temperature were used as driving factors to explain efficiency differences, while formation thickness was used as a supplementary variable to explain differences under the same temperature and permeability conditions. Permeability was used to explain recoverable flow rate and reinjection resistance, thus affecting utilization rate and payback period; reservoir temperature was used to explain heating capacity per unit flow rate, thus affecting unit production capacity and energy efficiency; formation thickness was used to explain the impact of differences in usable reservoir capacity on long-term sustainability. In the demonstration area, medium-deep well groups have high heating temperatures but low permeability, leading to increased reinjection resistance and maintenance costs, resulting in a longer payback period and limited utilization rate improvement, making θ less than 1; shallow heating and cooling systems have high permeability, are stable, and easily achieve economies of scale, making θ greater than 1; hot spring tourism and cultural development have strong economic output potential but are classified as highly ecologically sensitive, and increased environmental constraints suppress θ, making it effective but not ranked highly; agricultural and aquaculture use has stable heat loads, tiered utilization, and controllable water consumption and impact levels, resulting in higher overall efficiency.

[0075] The difference-driven analysis also introduces policy-driven factors to explain the differences in economic performance. The policy-driven factors consist of two parts: subsidy availability and subsidy constraint strength. Subsidy availability is determined based on triggering conditions and verification materials. Subsidy constraint strength is graded according to the upper limit rules, applicable year and additional restrictions in ecologically sensitive areas, and is explained in correspondence with differences in economic indicators such as unit investment return rate and payback period. This is used to identify differences in economic performance caused by differences in subsidy application when geological conditions are similar, so that the difference explanation chain covers both geological and policy factors.

[0076] Based on efficiency ranking, bottleneck mapping, and explanation of driving factors, differentiated optimization strategies are output, covering three categories: large-scale promotion, technological improvement, and policy support. Each strategy undergoes a feasibility check, with the check criteria including at least the applicable conditions of subsidy standards, the accessibility of pipeline infrastructure, and ecological sensitivity constraints. For example... Figure 3 As shown, the strategy generation uses efficiency value θ ranking and bottleneck indicators as inputs, and combines permeability, temperature, thickness, and ecological sensitivity interpretations to form three types of strategy output paths: effective and controllable constraints output large-scale promotion plans; ineffective plans or those where bottlenecks are concentrated in utilization rate, payback period, and impact level output technology improvement plans; and plans with high improvement costs and significant policy influence output policy support plans tied to subsidy standards and performance indicators, with restrictions on implementation or alternative solutions. Strategy feasibility verification uses a checklist-based approach: for large-scale promotion strategies, it verifies whether the accessibility of pipeline infrastructure, metering scope, and monitoring compliance conditions are met; for technology improvement strategies, it verifies whether additional restrictions in ecologically sensitive areas, impact level control targets, and monitoring and early warning configurations are met; for policy support strategies, it verifies whether subsidy triggering conditions, verification material completeness, applicable year, and upper limit rules are met. For strategies that pass verification, the subsidy type, metering scope, and verification material checklist are specified; for strategies that fail verification, subsidies are not used as an assumption for improved returns, and technology improvement or alternative paths are prioritized to avoid strategies being deemed unreasonable due to policy unavailability.

[0077] Large-scale promotion will prioritize shallow-layer heating and cooling, hot spring tourism, and agricultural and aquaculture heating, while simultaneously improving heating pipe networks and heat exchange stations to increase access load and operating hours, and establishing a standardized metering and maintenance system to reduce energy consumption fluctuations and maintenance costs. Technological improvements will address the low permeability constraints of heating in medium-deep well groups by employing reservoir fracturing to enhance permeability and improve resource utilization, along with optimized reinjection processes and wellbore anti-scaling maintenance to reduce reinjection resistance and blockage risks. Simultaneously, monitoring, early warning, and threshold-based production restrictions will mitigate the impact level. Policy support will provide technical subsidies for situations requiring high investment in deep-layer transformation to reduce enterprise improvement costs. Subsidies will be specifically targeted at fracturing and permeability enhancement, reinjection system upgrades, and monitoring and early warning system construction, and will be linked to performance indicators such as improved utilization, shortened payback period, impact level control, and monitoring compliance. Subsidy standards will follow the guidelines of relevant government documents, with document numbers and clause summaries retained. If a certain method's θ is significantly less than 1 and the improvement cost is too high, while simultaneously exhibiting high ecological sensitivity or difficulty in reducing the impact level, the countermeasures will output restrictions on advancement or alternative paths to avoid uncontrollable risks under high constraints.

[0078] Example 2

[0079] In one optional implementation, this embodiment provides a specific implementation of a karst geothermal development evaluation system based on the 3E-ultra-efficiency DEA model. This evaluation system can be deployed on a server, industrial control computer, or personal computer, and can run as a standalone software program or be embedded as an evaluation function module of a geothermal development management platform. Figure 4 As shown, the system of the present invention is implemented in a modular manner, including at least a data acquisition interface module, a preprocessing module, a karst adaptation module, a model construction and solution module, and an output module. The modules are transferred to each other through data tables, matrices and log files, forming a closed-loop computing link of "data acquisition - verification and normalization - standardization and adaptation - model solution - output landing", which facilitates archiving, recalculation and auditing.

[0080] The data acquisition interface module obtains raw data on energy, economic, and environmental indicators for karst geothermal development in the target area, and records the data source, time frame, and data collection rules. Data sources may include: geological exploration results databases (including well test interpretation, logging temperature, formation thickness, etc.), enterprise operating ledgers (including investment amount, operation and maintenance costs, heating and cooling capacity, electricity consumption, production capacity, etc.), and environmental monitoring platforms of competent authorities or third parties (including water consumption, emission reduction calculations, geological environmental impact levels, etc.). To ensure the comparability of the evaluation objects, this embodiment divides the evaluation objects into decision units (DMUs) according to their development methods, such as shallow ground source heat pump heating and cooling, medium-deep geothermal well group heating, hot spring tourism integrated utilization, and geothermal agricultural and aquaculture heating. Each decision unit corresponds to one or more data records, and the record fields must include at least: the decision unit identifier (DMU). IDThe data collection includes the development method category, the start and end dates of the statistical period, the data source identifier, the original values ​​of the nine 3E indicators, and karst constraint variables (representative permeability, reservoir temperature, ecological sensitivity classification, etc.). The original values ​​of the nine 3E indicators can be as described in the claims: Energy dimension indicators E1 geothermal resource utilization rate, E2 unit energy consumption output efficiency, E3 sustainable utilization coefficient of reservoir resources; Economic dimension indicators C1 unit investment return rate, C2 development and operation cost payback period, C3 unit production capacity economic benefit; Environmental dimension indicators V1 carbon dioxide emission reduction, V2 water resource consumption intensity, V3 geological environmental impact level (quantified into levels 1 to 5). After importing or collecting the data, the data acquisition interface module generates a "raw data archive table" and a "data retrieval description record" for subsequent traceability.

[0081] The preprocessing module performs time standardization, unit conversion, direction consistency verification, missing data completion, and anomaly verification on the raw data, and generates a standardized pre-dataset that can be used for model computation.

[0082] 1) Unified Time Scope: Energy and economic indicators adopt a unified statistical period of 12 consecutive months, while environmental indicators adopt the monitoring and accounting standards of the competent authorities and are mapped to the same statistical period. When cross-period data exists, it is converted according to daily / weekly averages or according to the proportion of heat supply, and the conversion method and parameters are recorded in the log.

[0083] 2) Unit conversion: Energy data are unified to MWh or equivalent units and kept consistent across the entire sample; monetary data are unified to the same currency unit and expressed on an annualized basis; water consumption is unified to m³ / MWh; emission reduction is unified to t / year; and the impact level is an integer from 1 to 5.

[0084] 3) Directional Consistency Verification: The preprocessing module labels payback period C2, water consumption intensity V2, and impact level V3 as constraint indicators, and utilization rate E1, energy efficiency E2, sustainability E3, return on investment C1, economic benefit per unit capacity C3, and emission reduction V1 as contribution indicators. Based on this, a mapping table between the input and output sets of the subsequent DEA model is formed. This mapping table is archived along with the data to ensure consistency in the calculation process.

[0085] 4) Missing Indicators: When a decision-making unit has missing indicators, they are filled in according to a fixed rule combining the median of similar development methods with the scale coefficient. The scale coefficient is preferentially taken as the ratio of designed heating and cooling capacity (MW). When the designed capacity is missing, the effective heating and cooling energy (MWh) of the statistical period can be used as a substitute scale benchmark. After filling, an order-of-magnitude consistency check is performed (e.g., comparison with the range of similar samples). If it fails, it is marked as needing review.

[0086] 5) Anomaly Review: When an indicator value is found to deviate significantly from similar samples (e.g., exceeding the mean ± 3 times the standard deviation or exceeding the physical / economic reasonable range), the anomaly review process is triggered, requiring the tracing back of the original measurement ledger, monitoring report number or financial voucher. Only after the review is passed can subsequent calculations proceed; the review process is recorded in the "Anomaly Review Log".

[0087] After standardizing and verifying the data, the preprocessing module performs Z-score standardization on the raw data of each indicator, generating a standardized indicator matrix Z. During standardization, the mean and standard deviation are calculated for each indicator column, and a standardized value is calculated for each decision unit. To avoid negative or zero values ​​after standardization affecting the numerical stability of the linear programming solution, this embodiment performs uniform positiveization processing on the constrained input candidate columns: for each input indicator column, the minimum standardized value (min) of that column is calculated. When min is not greater than 0, all standardized values ​​in that column are added to the same translation constant K to make the minimum value of that column 1; when min is greater than 0, no translation is performed. The translation constant K and the corresponding column index are written to the "positiveization processing record table" for recalculation. This processing does not change the relative differences within the column, but only changes the numerical benchmark to meet the stability requirements of the solution.

[0088] The karst adaptation module generates correction coefficients for karst geothermal characteristics and performs multiplicative corrections on key indicators based on reservoir permeability and ecological sensitivity to achieve regional adaptation. The karst adaptation module reads the karst constraint variables for each decision unit, including representative permeability, reservoir temperature, and ecological sensitivity classification. The correction coefficient generation adopts a tiered triggering rule: reservoir permeability is assessed with a threshold of 0.10 mD; values ​​less than 0.10 mD are considered low and trigger resource utilization enhancement; high ecological sensitivity triggers environmental constraint enhancement; both can be triggered simultaneously. Weighted regional adaptation is achieved using "indicator column regional adaptation coefficients," and is applied multiplicatively to the corresponding indicator columns in the standardized indicator matrix: the regional adaptation coefficient for the column corresponding to the geothermal resource utilization rate E1 indicator is adjusted from 0.30 to 0.40, the regional adaptation coefficient for the column corresponding to the geological environment impact level V3 indicator is adjusted from 0.25 to 0.35, and the remaining indicator columns remain unchanged. The calculation method for multiplicative correction is to uniformly multiply column E1 by the ratio of 0.40 to 0.30, and column V3 by the ratio of 0.35 to 0.25. The karst adaptation module writes the trigger threshold, grading results and final multiplicative coefficients into the "adaptation correction log" to ensure that the same data input can produce consistent adaptation output. The threshold grading basis, preference intensity quantification method, multiplicative correction method and log archiving recalculation rules for the karst geothermal characteristic correction coefficient are consistent with the determination methods described in (1) to (5) of Example 1.

[0089] The model building and solution module constructs a 3E-super-efficiency DEA evaluation model containing slack variables and residual variables based on the standardized index matrix after regional adaptation, and obtains the efficiency value θ of each decision unit through Matlab to achieve the ranking of effective decision units.

[0090] 1) Input / output set mapping: The model building and solution module maps the input and output sets according to the direction consistency verification results. The payback period C2, water consumption intensity V2, and impact level V3 are used as the input set, and the utilization rate E1, energy efficiency E2, sustainability E3, return on investment C1, economic benefit per unit capacity C3, and emission reduction V1 are used as the output set.

[0091] 2) Super-efficient DEA and relaxation / residual: Relaxation variables and residual variables are introduced into the DEA solution to characterize input redundancy and output insufficiency. A super-efficient mechanism is used to further distinguish and sort effective decision units, thereby overcoming the defect that effective units in traditional DEA cannot be further sorted.

[0092] 3) Fixed Solution Process: In this embodiment, the Matlab solution process is fixed as follows: read the standardized matrix and positive translation constant records, verify that the input column positiveization has been completed, perform multiplicative correction of key columns generated by the karst adaptation module, construct the linear programming solution input structure, call the linear programming solution module to output the efficiency value θ and relaxation / residual, form the sorting table and bottleneck table, and perform back-substitution verification on the random sampling decision units to verify that the solution output is consistent with the input data. Key parameters and solution logs in this solution process are archived and saved for recalculation and traceability.

[0093] 4) Efficiency judgment and ranking: The efficiency judgment rule is fixed as follows: if θ is not less than 1, it is judged as a valid decision unit, and if θ is less than 1, it is judged as an invalid decision unit; all decision units are ranked from largest to smallest according to θ, and the super-efficient ranking results are retained within the valid units.

[0094] The output module performs a difference analysis based on the efficiency value θ and the relaxation / residual results, identifies the driving factors of efficiency differences by combining thermal storage temperature and permeability, and outputs promotion, improvement and policy countermeasures.

[0095] 1) Output Structured Results: The output module generates at least three types of structured files: an efficiency ranking table, a bottleneck indicator table, and a countermeasures list. The efficiency ranking table includes DMU... ID The table includes: θ, effective / ineffectiveness determination, and grouping (deep / shallow / comprehensive utilization, etc.); the bottleneck indicator table includes column indexes for input redundancy and output insufficiency, corresponding indicator names, and suggested improvement directions; the countermeasure list includes countermeasure categories (large-scale promotion / technological improvement / policy support), applicable decision-making units, triggering reasons (bottleneck indicators and driving factors), implementation points, and feasibility verification conclusions.

[0096] 2) Bottleneck to Countermeasure Mapping Rules: The output module uses fixed mapping rules to transform relaxed / residual results into actionable improvement directions: a positive redundancy in the input column indicates that the constraint-type indicator can be compressed, while a negative redundancy in the output column indicates that the contribution-type indicator can be improved. For example, when the payback period redundancy is positive, it is recommended to shorten the payback period by reducing operation and maintenance costs and maintenance frequency, and improving the cash flow structure; when the water consumption intensity redundancy is positive, it is recommended to increase the reinjection ratio or adopt closed heat exchange to reduce new water intake; when the impact level redundancy is positive, it is recommended to increase the monitoring frequency and optimize reinjection water quality and operating conditions, and set threshold linkage to limit extraction to reduce the risk level; when the utilization rate is negative, for deep-layer methods, it is recommended to use fracturing and permeability enhancement and reinjection channel optimization, and for shallow-layer methods, it is recommended to use well site layout and load matching optimization. The output module fixes this mapping rule as a configuration file and archives it together with the specific output results.

[0097] 3) Difference Analysis and Driver Identification: The output module groups and compares deep and shallow development methods, using permeability and reservoir temperature as driving factors to explain efficiency differences. Permeability is used to explain recoverable flow rate and reinjection resistance, thus affecting utilization rate and payback period; reservoir temperature is used to explain heating capacity per unit flow rate, thus affecting economic benefits and energy efficiency per unit capacity; if necessary, formation thickness can be introduced as a supplementary variable to explain long-term sustainability differences. The results of driver identification are written into the "Difference Explanation Report" to support the rationality of countermeasures.

[0098] 4) Countermeasure Output and Feasibility Verification: The output module outputs three types of countermeasures based on efficiency ranking, bottleneck mapping, and driving factor interpretation, and performs feasibility verification on each type of countermeasure. Countermeasures for large-scale promotion focus on verifying the accessibility of pipeline infrastructure, consistency of metering standards, and monitoring compliance conditions; countermeasures for technological improvement focus on verifying additional restrictions in ecologically sensitive areas, impact level control targets, and monitoring and early warning configurations; countermeasures for policy support focus on verifying the triggering requirements of subsidies or policy conditions, the completeness of verification materials, and upper limit rules. Countermeasures that pass verification are marked "Feasible" in the countermeasure list, along with a list of prerequisites and required materials; countermeasures that fail verification are marked "Requires Supplementary Conditions / Postponement," and alternative paths are provided to avoid outputting unfeasible solutions.

[0099] To verify the system's feasibility, this embodiment's evaluation system can select the demonstration area and decision-making unit from Embodiment 1 as input, run according to the above module process, and output efficiency ranking, bottleneck indicators, and a countermeasure list. After each run, the system packages and archives the original data archive table, preprocessing log, positiveized shift constant record, adaptation correction log, Matlab solution log, efficiency ranking table, bottleneck indicator table, and countermeasure list. This allows any third party to obtain consistent results by recalculating under the same rules after obtaining the archive package, thereby supporting the traceability of evaluation conclusions and the auditability of engineering implementation decisions.

[0100] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0101] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for evaluating karst geothermal development based on the 3E-ultra-efficiency DEA model, characterized in that, Includes the following steps: Step 1: Construct a 3E evaluation index system for the development of karst geothermal resources. The 3E evaluation index system includes energy dimension indicators, economic dimension indicators, and environmental dimension indicators. The raw data of each indicator are processed using the Z-score standardization method to obtain a standardized index matrix. Step 2: Based on the standardized index matrix, construct the 3E-super-efficiency DEA evaluation model, using constraint indicators as the input set and contribution indicators as the output set. The model achieves the ranking of effective decision-making units by introducing slack variables and residual variables, and introduces the karst geothermal characteristic correction coefficient to adapt the key indicators to the region. Step 3: For karst geothermal resources in the target area, select typical deep and shallow geothermal development methods as decision-making units, collect and verify the original data of the 3E indicators of each decision-making unit, and after completing the missing data completion and abnormal data verification, form the standardized indicator matrix according to the rules of Step 1. Step 4: Substitute the standardized index matrix into the 3E-super-efficiency DEA evaluation model constructed in Step 2, and obtain the efficiency value of each decision unit by solving it through Matlab. Based on the efficiency value, complete the ranking and carry out difference analysis. Combine reservoir permeability and thermal reservoir temperature to identify the driving factors of efficiency differences. Step 5: Based on the current status of the geothermal industry in the target area and the efficiency calculation results, output differentiated development optimization strategies. These strategies include at least three categories: large-scale promotion, technological improvement, and policy support.

2. The method for evaluating karst geothermal development based on the 3E-ultra-efficiency DEA model according to claim 1, characterized in that, The energy dimension indicators mentioned in Step 1 include geothermal resource utilization rate, unit energy consumption output efficiency, and sustainable utilization coefficient of thermal storage resources; the economic dimension indicators include unit investment return rate, development and operation cost payback period, and unit production capacity economic benefits; the environmental dimension indicators include carbon dioxide emission reduction, water resource consumption intensity, and geological environmental impact level, wherein the geological environmental impact level is quantified as level 1 to level 5.

3. The method for evaluating karst geothermal development based on the 3E-ultra-efficiency DEA model according to claim 2, characterized in that, The energy and economic indicators use a 12-month statistical period as the data caliber, while the environmental indicators use monitoring and accounting data from the competent authorities as the caliber. Unit conversion and directional consistency verification are completed before entering the standardization process.

4. The method for evaluating karst geothermal development based on the 3E-ultra-efficiency DEA model according to claim 1, characterized in that, The karst geothermal characteristic correction coefficient in step two is generated based on the reservoir permeability and ecological sensitivity of the target area. It can increase the weight of the geothermal resource utilization rate index in the model and increase the weight of the geological environment impact level index in the model.

5. The method for evaluating karst geothermal development based on the 3E-ultra-efficiency DEA model according to claim 1, characterized in that, In step two, the weighted regional adaptation is achieved in the form of "indicator column regional adaptation coefficients": the regional adaptation coefficients of the column corresponding to the geothermal resource utilization rate index are adjusted from 0.30 to 0.40, and the regional adaptation coefficients of the column corresponding to the geological environment impact level index are adjusted from 0.25 to 0.

35. The adjustments are then applied to the corresponding index columns in the standardized index matrix in a multiplicative correction manner. Except for the key columns, the other index columns remain unchanged, thereby achieving regional adaptation oriented towards karst characteristics without changing the DEA model solution structure.

6. The method for evaluating karst geothermal development based on the 3E-ultra-efficiency DEA model according to claim 1, characterized in that, The data collection in step three includes obtaining reservoir permeability and thermal reservoir temperature data through on-site surveys, investment, operating costs and production capacity data through enterprise surveys, and environmental impact monitoring data from the competent authorities.

7. The method for evaluating karst geothermal development based on the 3E-ultra-efficiency DEA model according to claim 1, characterized in that, The data verification in step three includes unifying the time caliber, unifying the unit caliber, completing missing data, and reviewing abnormal data. Among them, the missing data completion adopts a completion rule that combines the median of samples from similar development methods with the scale coefficient.

8. The method for evaluating karst geothermal development based on the 3E-ultra-efficiency DEA model according to claim 1, characterized in that, In step four, when the efficiency value satisfies θ not less than 1, the corresponding decision unit is determined to be an effective decision unit; when the efficiency value satisfies θ less than 1, the corresponding decision unit is determined to be an invalid decision unit, and directions for improvement are given based on slack variables and residual variables.

9. The method for evaluating karst geothermal development based on the 3E-ultra-efficiency DEA model according to claim 1, characterized in that, The large-scale promotion in step five includes prioritizing the promotion of shallow ground source heat pump heating and cooling, as well as the use of heat in hot spring tourism and aquaculture, and improving the supporting heating pipeline network. Technological improvements include reservoir fracturing and permeability enhancement for deep geothermal energy to improve energy efficiency; policy support includes providing technology subsidies for deep geothermal development to reduce the improvement costs for enterprises.

10. A karst geothermal development evaluation system based on the 3E-ultra-efficiency DEA model, characterized in that, include: The data acquisition interface module obtains raw data on energy, economic, and environmental indicators for karst geothermal development in the target area. The preprocessing module unifies the time caliber, converts units, and verifies directional consistency of the raw data. It also processes the raw data of each indicator using the Z-score standardization method to obtain a standardized indicator matrix. The karst adaptation module generates correction coefficients for karst geothermal characteristics and performs multiplicative corrections on key indicator columns based on reservoir permeability and ecological sensitivity to achieve regional adaptation. The model construction and solution module constructs a 3E-super-efficiency DEA evaluation model containing slack variables and residual variables based on the standardized indicator matrix after regional adaptation. It then solves the model using Matlab to obtain the efficiency values ​​of each decision-making unit, thus ranking effective decision-making units. The output module performs difference analysis based on the efficiency values, identifies the driving factors of efficiency differences by combining reservoir temperature and permeability, and outputs promotion, improvement, and policy countermeasures.

Images