Method and system for monitoring stratum settlement in multi-scale geothermal exploitation and recharge process

Abstract

According to the method and system for monitoring stratum settlement in the geothermal exploitation and recharge process under multiple scales, obtained macroscopic monitoring data, mesoscopic monitoring data and microscopic monitoring data synchronously cover six types of core parameters including displacement, temperature, strain, pressure, microseismic and chemical concentration, and complete data support is provided for subsequent multi-field coupling calculation; compared with traditional single physical quantity monitoring, coupling result deviation caused by incomplete data is greatly reduced; the macroscopic monitoring data, the mesoscopic monitoring data and the microscopic monitoring data are mapped to different grid layers of a unified grid system, so that the problem that high-resolution microscopic data and low-resolution macroscopic data are difficult to cooperate is solved, and fused data better fits the actual geological law; parameters of the heat-fluid-solid-chemical coupling model are dynamically corrected through Kalman filtering, it is ensured that the parameters of the heat-fluid-solid-chemical coupling model are adjusted in real time along with dynamic changes of the stratum, and the accuracy of settlement prediction is improved.

Description

Technical Field

[0001] This invention belongs to the technical field of settlement monitoring, and particularly relates to a method and system for monitoring ground settlement during geothermal extraction and reinjection at multiple scales. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] During highway construction, shallow geothermal energy, such as ground source heat pumps, is used for heating and cooling in construction camps and for snow melting on roads, significantly reducing carbon emissions. Simultaneously, medium- and low-temperature underground hot water resources may be discovered during tunnel and deep foundation pit construction; the temporary use of these resources can improve construction efficiency. However, the extraction and reinjection of geothermal resources can easily trigger thermal-fluid-solid coupling effects in the ground, leading to localized subsidence or uplift, threatening project safety. Therefore, timely detection of ground subsidence during geothermal extraction and reinjection is of great significance for engineering safety.

[0004] Currently, conventional geothermal extraction settlement monitoring technologies have the following main shortcomings: 1. The single monitoring scale is difficult to adapt to the needs of multiple scenarios. For example, traditional methods rely on GPS or level instruments, downhole sensors, etc. for strain monitoring, which is difficult to capture micro-strains of less than 1 mm caused by temperature changes in shallow soil. It is also easily affected by vibrations from construction machinery. When underground hot water is exposed during construction, the existing technology lacks the ability to expand the monitoring scale from shallow to deep layers. 2. Weak data fusion and modeling capabilities, separation of physical models and data, traditional numerical simulations such as TOUGH2 rely on static parameters, cannot assimilate sensor data in real time, cannot reflect temperature fluctuations caused by intermittent operation of ground source heat pumps, cannot reflect dynamic disturbances during construction, and have low utilization of multi-source heterogeneous data without establishing thermo-mechanical coupling weight association. 3. The reinjection control and settlement early warning are disconnected. Existing reinjection strategies are mostly based on empirical thresholds, such as a fixed reinjection rate of ≥80%, without dynamic adjustment based on real-time settlement data. This leads to repeated expansion and contraction of soil and rock caused by the alternation of hot and cold fluids during shallow geothermal reinjection. The sudden extraction of geothermal resources lacks a "monitoring-reinjection" linkage mechanism. After reinjection, the pore pressure imbalance is aggravated, inducing settlement rates to exceed limits, such as >10mm / month. Summary of the Invention

[0005] To overcome the shortcomings of the prior art, this invention provides a method and system for monitoring ground subsidence during geothermal extraction and reinjection at multiple scales. Based on multi-scale data, combined with physical models and Kalman filtering, it enables real-time monitoring and dynamic prediction of ground subsidence during geothermal extraction and reinjection at multiple scales.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for monitoring ground subsidence during geothermal extraction and reinjection processes at multiple scales, including: Acquire macroscopic, mesoscopic, and microscopic monitoring data for the monitoring area; wherein, the macroscopic monitoring data includes regional surface deformation data and surface temperature data; the mesoscopic monitoring data includes soil and rock strain, temperature gradient, and seepage pressure; and the microscopic monitoring data includes rock mass fracture signal data and geothermal water chemical concentration data. The acquired macroscopic, mesoscopic, and microscopic monitoring data are mapped to different grid layers of a unified grid system. Key physical quantities are extracted from different grid layers to calculate risk values ​​reflecting rock mass damage risk, quantify the contribution of thermal stress effects, and reflect the trend of settlement. The calculated risk value reflecting rock mass damage risk, the contribution of quantified thermal stress effect, and the trend value reflecting settlement trend are mapped into the thermal-fluid-solid-chemical coupling model. The parameters of the thermal-fluid-solid-chemical coupling model are corrected in real time using the Kalman filter algorithm. Settlement prediction results are obtained based on the corrected thermal-fluid-solid-chemical coupling model, and the reinjection process is regulated according to the settlement prediction results.

[0007] Secondly, the present invention provides a multi-scale geothermal extraction and reinjection process ground subsidence monitoring system, comprising: The acquisition module is configured to acquire macroscopic monitoring data, mesoscopic monitoring data, and microscopic monitoring data of the monitoring area; wherein, the macroscopic monitoring data includes regional surface deformation data and surface temperature data; the mesoscopic monitoring data includes soil and rock strain, temperature gradient, and seepage pressure; and the microscopic monitoring data includes rock mass fracture signal data and geothermal water chemical concentration data. The extraction module is configured to: map the acquired macroscopic monitoring data, mesoscopic monitoring data and microscopic monitoring data to different grid layers of a unified grid system, extract key physical quantities from different grid layers to calculate risk values ​​reflecting rock mass damage risk, quantify the contribution of thermal stress effect and trend values ​​reflecting settlement trend; The prediction module is configured to: map the calculated risk value reflecting the risk of rock mass damage, the contribution of the quantified thermal stress effect, and the trend value reflecting the settlement trend into the thermal-fluid-solid-chemical coupling model; use the Kalman filter algorithm to correct the parameters of the thermal-fluid-solid-chemical coupling model in real time; obtain the settlement prediction result based on the corrected thermal-fluid-solid-chemical coupling model; and regulate the reinjection process according to the settlement prediction result.

[0008] Thirdly, the present invention provides an electronic device including a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein the computer instructions, when executed by the processor, perform the method described in the first aspect.

[0009] Fourthly, the present invention provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in the first aspect.

[0010] Fifthly, the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0011] The above one or more technical solutions have the following beneficial effects: In this invention, the acquired macroscopic, mesoscopic, and microscopic monitoring data simultaneously cover six core parameters: displacement, temperature, strain, pressure, microseismic activity, and chemical concentration. This provides complete data support for subsequent multi-field coupling calculations, significantly reducing coupling result deviations caused by incomplete data compared to traditional single-physical-quantity monitoring. By mapping macroscopic, mesoscopic, and microscopic monitoring data to different grid layers of a unified grid system, the problem of difficulty in coordinating high-resolution microscopic data with low-resolution macroscopic data is solved, making the fused data more consistent with actual geological patterns. By dynamically correcting the parameters of the thermal-fluid-solid-chemical coupling model through Kalman filtering, the parameters of the thermal-fluid-solid-chemical coupling model are adjusted in real time with the dynamic changes of the strata, improving the accuracy of settlement prediction.

[0012] In this invention, a multi-objective optimization model is constructed with the goal of minimizing settlement risk, cold shock, and chemical scaling. The optimal reinjection parameters are obtained by solving the multi-objective optimization model with the constraints of reinjection pressure limit, temperature gradient limit, and chemical concentration safety threshold. By constructing a closed-loop process of settlement prediction and adaptive control, active risk suppression is achieved.

[0013] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0014] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0015] Figure 1 This is an overall block diagram of the multi-scale geothermal extraction and reinjection process ground subsidence monitoring method in Embodiment 1 of the present invention. Detailed Implementation

[0016] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0017] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0018] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0019] Example 1 This embodiment discloses a method for monitoring ground subsidence during geothermal extraction and reinjection processes at multiple scales, including: Acquire macroscopic, mesoscopic, and microscopic monitoring data for the monitoring area; among which, macroscopic monitoring data includes regional surface deformation data and surface temperature data; mesoscopic monitoring data includes soil and rock strain, temperature gradient, and seepage pressure; microscopic monitoring data includes rock mass fracture signal data and geothermal water chemical concentration data. The acquired macroscopic, mesoscopic, and microscopic monitoring data are mapped to different grid layers of a unified grid system. Key physical quantities are extracted from different grid layers to calculate risk values ​​reflecting rock mass damage risk, quantify the contribution of thermal stress effects, and reflect the trend of settlement. The calculated risk value reflecting the risk of rock mass damage, the contribution of the quantified thermal stress effect, and the trend value reflecting the settlement trend are mapped into the thermo-fluid-solid-chemical coupling model. The parameters of the thermo-fluid-solid-chemical coupling model are corrected in real time using the Kalman filter algorithm. The settlement prediction results are obtained based on the corrected thermo-fluid-solid-chemical coupling model. A multi-objective optimization model was constructed with the goal of minimizing settlement risk, cold shock, and chemical scaling. The multi-objective optimization model was solved with the constraints of reinjection pressure limit, temperature gradient limit, and chemical concentration safety threshold to obtain the optimal reinjection parameters. The geothermal reinjection process was dynamically adjusted based on the optimal reinjection parameters.

[0020] The macroscopic, mesoscopic, and microscopic monitoring data acquired in this embodiment simultaneously cover six core parameters: displacement, temperature, strain, pressure, microseismic activity, and chemical concentration. This provides complete data support for subsequent multi-field coupling calculations, significantly reducing coupling result deviations caused by incomplete data compared to traditional single-physical-quantity monitoring. By mapping macroscopic, mesoscopic, and microscopic monitoring data to different grid layers of a unified grid system, the problem of difficulty in coordinating high-resolution microscopic data with low-resolution macroscopic data is solved, making the fused data more consistent with actual geological patterns. The parameters of the thermal-fluid-solid-chemical coupling model are dynamically corrected using Kalman filtering, ensuring that the parameters of the thermal-fluid-solid-chemical coupling model are adjusted in real time with dynamic changes in the formation, thus improving the accuracy of settlement prediction. A multi-objective optimization model is constructed with the goals of minimizing settlement risk, cold shock, and chemical scaling. The optimal reinjection parameters are obtained by solving the multi-objective optimization model under constraints of reinjection pressure limits, temperature gradient limits, and chemical concentration safety thresholds. By constructing a closed-loop process of settlement prediction and adaptive control, proactive risk suppression is achieved.

[0021] The following is combined Figure 1 This embodiment provides a detailed description of the multi-scale geothermal extraction and reinjection process ground subsidence monitoring method proposed in this example: First, this embodiment constructs a comprehensive and efficient data acquisition hardware system. This system integrates various advanced sensors and data acquisition devices to accurately collect key data during shallow geothermal extraction and reinjection in highway construction, providing sufficient and reliable data support for subsequent data analysis and coupled calculations. Second, a highly efficient multiphysics coupling algorithm is used to perform multiphysics coupling calculations on data from different data sources and of different types. Then, information display equipment is used to monitor and predict settlement in real time, and the predicted data is the result of the multiphysics coupling calculation. The system utilizes multiphysics coupling equations and multi-scale data fusion algorithms to accurately monitor and predict settlement caused by the current geothermal extraction and reinjection process.

[0022] The data acquisition hardware system in this embodiment includes a macroscopic layer, a mesoscopic layer, and a microscopic layer. These three modules provide real-time monitoring data such as displacement field, surface temperature field, well group temperature and strain, inversion data, and monitored mineral concentration. The collected data is imported into intelligent analysis for multi-scale detection network and equation coupling, performing temperature field-seepage field-stress field-chemical field coupling to achieve real-time monitoring, prediction, and dynamic control of reinjection flow for formation subsidence.

[0023] In this embodiment, the macroscopic layer includes InSAR recharge response monitoring and thermal infrared remote sensing. Differential Interferometric Radar (DInSAR) captures regional ground deformation caused by recharge with millimeter-level sensitivity; thermal infrared remote sensing captures surface temperature data. Both datasets are imported into the final intelligent analysis to constrain the boundary conditions of the temperature field-seepage field-stress field-chemical field coupling equation, i.e., the THMC equation, thus constraining the range of key data.

[0024] Specifically, the macroscopic layer combines InSAR satellites with ground-based DAS / DTS, monitoring parameters such as surface displacement and temperature, with a spatial resolution of 5m and a temporal resolution of 7 days, and deployment method of laying optical fibers in a serpentine pattern along the boundary of the geothermal field.

[0025] In this embodiment, the mesoscopic layer includes a ground-based distributed optical fiber sensor (DTS / DAS) and a downhole array pressure sensor. DAS, or distributed acoustic sensor, captures soil and rock deformation and monitors microseismic activity, while DTS, or distributed temperature sensor, captures temperature distribution and temperature gradients, directly capturing wellbore temperature and strain. The DAS is laid at a depth of 1-2m along the surface or shallow layers in a serpentine or grid pattern, covering geothermal areas or construction areas such as along highways, forming a continuous spatial sampling network. The DTS and DAS share the same optical fiber, deployed around reinjection wells or geothermal pipelines to monitor temperature field changes in shallow soil or rock.

[0026] Specifically, the mesoscopic layer is a combination of downhole optical fiber and pressure / temperature sensors, monitoring parameters such as seepage pressure (MPa) and strain (με), with a spatial resolution of 1m and a time resolution of 1 minute, and the deployment method is a radial arrangement along the reinjection well to the production well.

[0027] (1) Deploy distributed optical fibers (DAS / DTS) along the boundary of the geothermal field or the construction area, with a burial depth of 1-2m and a spacing of ≤500m. (2) Install high-temperature resistant optical fiber (DTS) (temperature resistance ≥300℃) and array-type pressure sensor (range 0-30MPa, accuracy ±0.1MPa) in the reinjection well and production well. (3) A monitoring node is set up every 10m along the well shaft to collect temperature and strain in real time.

[0028] In this embodiment, the microlayer includes a microfluidic chip that monitors the mineral concentration C of geothermal water flow at the pore scale, corrects the reaction rate, predicts the risk of chemical fouling, and considers the impact of mineral dissolution and precipitation, such as silica, on sedimentation.

[0029] Specifically, the micro-layer consists of a micro-seismic array and a microfluidic chip, with monitoring parameters including rock fracture signal (Hz) and mineral concentration (mg / L), a spatial resolution of 0.1m and a temporal resolution of 1ms, and deployment method of embedding it into the rock mass near the well or the inner wall of the reinjection pipeline.

[0030] The macroscopic, mesoscopic, and microscopic layers employ a multi-type sensor collaborative operation approach, integrating sensors for temperature monitoring, strain mechanical property monitoring, fluid pressure change monitoring, and chemical substance concentration monitoring. The deployment is optimized based on the characteristics of geothermal extraction projects and reinjection control layouts, and a stable transmission link is constructed using wireless transmission technology.

[0031] This embodiment uses the spatiotemporal pyramid algorithm to process data at different spatiotemporal resolutions in a hierarchical manner, fusing them step by step to extract multi-scale features. The specific steps are as follows: Outlier data points are detected and removed based on statistical methods (3σ criterion) or the Isolation Forest algorithm; specifically, the data dimension normalization formula is as follows:

[0032] Where X represents the collected data, and Xmax and Xmin represent the maximum and minimum values ​​of the collected data, respectively. Abnormal data will be marked by the data processing system. Multi-source data types are shown in Table 1: Table 1:

[0033] Step 1: Data layer alignment: Divide the monitoring area into multiple grids according to spatiotemporal resolution.

[0034] (1) Spatial stratification: from coarse to fine (1km×1km→100m×100m→10m×10m). (2) Time stratification: from long to short (month → week → day → hour); Step 2: Data mapping and mapping rules: Interpolate data from different sensors onto a unified grid.

[0035] (1) InSAR data (low spatiotemporal resolution): mapped to the top coarse grid (1km, monthly level), the average value of grid points is aggregated to be the average displacement of all points in the grid, reflecting the daily regional subsidence trend.

[0036]

[0037]

[0038] in, For position The top-level mesh displacement value at that location, The data is the k-th InSAR displacement point within the grid (original resolution approximately 5m), where N is the total number of displacement points within the grid. The high-resolution InSAR point cloud is aggregated into a coarse grid, for example, 100 InSAR points within a 1km grid. The average value is taken to represent the overall displacement of the region, thus suppressing local noise.

[0039] (2) DAS / DTS data (medium to high resolution): mapped to a medium-level grid (100m, day level), bilinear interpolation, weighted average of the nearest four nodes, reflecting the thermal-mechanical coupling effect of the well group.

[0040]

[0041]

[0042] in, The strain value at the target grid point. The values ​​represent the strain values ​​of the four adjacent DAS sensor nodes. The distance from the target point to the nth node. The weights are the inverse of the distance to neighboring nodes. Discrete sensor data is interpolated into a continuous field to capture well cluster-scale thermo-mechanical coupling effects.

[0043] (3) Microseismic / microfluidic data (high resolution): mapped to the underlying fine grid (10m, minute level), reflecting the risk of local crack propagation and scaling, and reflecting the density of underlying microseismic events.

[0044]

[0045] Where h=5m is the bandwidth control smoothness. for( x, y Microseismic event density at coordinates The coordinates of the i-th microseismic event (positioning accuracy 0.1m), M is the total number of events within the time window, such as 1 hour, and h is the bandwidth parameter.

[0046] Step 3: Feature extraction and weight allocation.

[0047] (1) Feature extraction (extracting key physical quantities from each mesh layer): 1. Top layer (InSAR): Vertical displacement of the ground surface (mm), displacement rate (mm / month); the formula for vertical displacement is as shown in formula (1), and the displacement rate is a linear fit of the time series. Example: The displacement of a certain grid increases from 0 mm to 15 mm in 30 days, at a rate of 0.5 mm / day = 15 mm / month.

[0048] 2. Middle layer (DAS / DTS): Strain (με), temperature gradient (°C / m), seepage pressure (MPa); strain extraction is performed by bandpass filtering of the raw DAS signal (0.1-10Hz), reflecting the elastic deformation of the soil and rock; temperature gradient differential ∇T along the depth direction. It reflects the migration speed of cold and hot fronts; the seepage pressure is directly read from the pressure sensor.

[0049] 3. Bottom Layer (Microseismic / Chemical): Microseismic event density (times / hour), SiO2 concentration (mg / L). Microseismic event density is statistically analyzed using a time window. (The density of microseismic events reflects the rock mass fracture strength).

[0050] (2) Dynamic weight allocation: Weights are allocated based on data confidence level, logistics, and relationships.

[0051]

[0052] Multi-field coupling strength factor Determined based on physical mechanisms and historical data, D is used to quantify the interaction strength between different physical fields. i With D j This does not refer to two specific values, but essentially means the i-th type of data and the j-th type of physical field. For example, D1 might represent temperature data T, and D2 might represent strain data ϵ. This involves calculating the strength of the coupling relationship between the temperature field and the stress field. The relevant data source depends on the associated physical field data, which comes from the aforementioned feature extraction layer.

[0053] in, , All of these are empirical coefficients; in this embodiment... =0.6, =0.4; The sensor accuracy factor is determined based on the sensor accuracy. For example, the InSAR error ±3mm has a weight of 0.8, and the DTS error ±1℃ has a weight of 0.9. The multi-field coupling strength factor is determined based on the multi-field correlation of the THMC equation, such as the temperature-strain coupling strength.

[0054] Step 4: Cross-scale fusion and model input.

[0055] (1) Pyramid fusion: aggregate layer by layer from bottom to top to generate multi-scale feature maps.

[0056] 1. Bottom-level integration: Microseismic event density + SiO2 concentration → fracture propagation risk level (0-10); microseismic event density reflects the rock mass fracturing activity (events / hour·m²), while increased SiO2 concentration exacerbates fracture blockage (mg / L), both jointly increasing the risk of rock mass damage. The quantification formula is:

[0057] Where α and β are the weighting coefficients dominated by microseismic events, α=0.7, β=0; reference threshold =1000 times / hour·m², critical scaling value =200mg / L. The density of microseismic events is represented by data from a microseismic sensor array at the micro-layer, and the unit is events / (hour·m²).

[0058]

[0059] 2. Mid-layer integration: Strain + temperature gradient → thermal-mechanical coupling settlement contribution (%); In the physical correlation, strain reflects soil deformation, temperature gradient reflects the migration rate of cold and hot fronts, and the product of the two quantifies the thermal stress effect.

[0060]

[0061] K = 0.05 calibration coefficient, which is related to the soil and rock expansion coefficient. The output unit is percentage (%), which represents the contribution ratio to the total settlement. The strain + temperature gradient reflects the contribution of thermo-mechanical coupling settlement.

[0062] 3. Top-level integration: Displacement rate + seepage pressure → regional settlement trend (mm / month); displacement rate reflects the settlement development speed (mm / month), seepage pressure P affects the effective stress (MPa), and the linear combination of the two predicts the long-term trend. The trend model is as follows:

[0063] Where a and b are regression coefficients, and their specific values ​​are determined using historical data, such as a=0.8 and b=0.2; ΔP=P-P0 represents the difference between the current pressure and the initial pressure. Displacement rate + seepage pressure reflects the regional settlement trend.

[0064] (2) Output unified field: Input multi-scale features into the THMC model to predict future settlement.

[0065] The input method for multi-scale features is as follows: the bottom layer risk value (Equation (6)) is mapped to the damage factor D in the momentum equation, and D corrects the rock elastic stiffness matrix C; the middle layer contribution (Equation (7)) is mapped to the heat source term in the energy equation. Top-level trends That is, formula (8) maps to the boundary pressure P in the mass equation, constraining the extraction / recharge flow rate. .

[0066] In this embodiment, a multi-field coupled inversion model is used as the core algorithm for the data obtained above. The goal is to limit the numerical range of key detection parameters in real time through the THMC equation, namely the temperature-permeability-mass-chemical four-field coupled equation, and to predict and calculate the sedimentation amount by recovering data from the macroscopic, microscopic, and mesoscopic layers.

[0067] The THMC equation, a coupling equation for the temperature field, seepage field, stress field, and chemical field, is established and integrated with data from InSAR, distributed optical fiber (DAS / DTS), microseismic sensors, and microfluidic chips. The specific coupling relationship is as follows: 1. Heat-fluid (TH) coupling: Temperature changes (reinjection of cold fluid) lead to changes in fluid density ( Changes in viscosity (μ) and permeability affect the seepage velocity ( ); seepage carries heat and changes the temperature field (T); 2. Fluid-Solid (HM) Coupling: Changes in pore pressure (P) affect the effective stress of the rock, leading to deformation (u); deformation alters porosity (u). ) and penetration rate (K); 3. Thermo-solid (TM) coupling: Temperature changes (ΔT) cause thermal expansion / contraction of rocks, which is superimposed on mechanical strain; 4. Chemical field coupling (C): Mineral dissolution / precipitation (R(C,T)) alters porosity ( ) and permeability (K); chemical reactions release / absorb heat, affecting the temperature field (the chemical model initially only considers SiO2 scaling); To focus on the core coupling effect, the following assumptions are made: 1. Constitutive relation assumptions of materials: It is assumed that the rock mass behaves as a linear elastic material during the recharge process, and the elastic matrix C is calibrated by laboratory core testing; 2. Assumptions of homogeneity and isotropy of porous media: It is assumed that the thermal reservoir is an isotropic homogeneous porous medium, and the permeability K is dynamically corrected only with temperature and strain; 3. Fluid and thermodynamic assumptions: It is assumed that the reinjection fluid is unidirectional incompressible liquid water, and its density is only related to temperature; 4. Simplification of chemical reaction: Assume that the precipitation rate of SiO2 is slower than the percolation rate.

[0068] Multiphysics coupling, the calculation process and coupling method are as follows: The THMC coupled model consists of four physical governing equations (Temperature field, Hydraulic field, Mechanical field, and Chemical field; THMC). The THMC equations serve as physical constraints, limiting parameter ranges, such as permeability K > 10⁻¹⁶ m². The governing equations for each physical field are as follows: Conservation of mass (Hydraulic flow field):

[0069] in, Porosity; For fluid density; The seepage velocity is determined by Darcy's law and is affected by the seepage coefficient K. For extraction / reinjection flow rate, source and sink terms per unit volume.

[0070] Energy conservation (thermal field):

[0071] in Equivalent heat capacity; Equivalent thermal conductivity, affected by crack aperture; As a heat source, The specific heat capacity of a liquid at constant pressure. For fluid density, The seepage velocity is [value missing].

[0072] Conservation of momentum (Mechanical stress field):

[0073] in, The rock elastic stiffness matrix (affected by temperature T and damage D). For Biot coefficient, For strain tensor and , It is a displacement vector. The contribution of pore pressure to total stress is characterized by P, where P is the pore pressure and I is the unit tensor. Chemical coupling (considering the dissolution-precipitation kinetics of minerals such as SiO2):

[0074] Where C represents the mineral concentration. The reaction rate is temperature-dependent, and C is monitored by a microfluidic chip sensor.

[0075] In this embodiment, the parameters of the heat-fluid-solidification coupling model are corrected in real time using the Kalman filter algorithm. Settlement prediction results are then obtained based on the corrected heat-fluid-solidification coupling model. Specifically: A state vector is constructed based on permeability, damage factor, equivalent thermal conductivity and chemical reaction rate; an observation vector is constructed based on soil strain, temperature, seepage pressure and SiO2 concentration. A parameter evolution model is established based on physical mechanisms, and physical evolution calculations are performed on all parameters in the state vector to obtain the state prediction value at time k. The deviation between the predicted values ​​of the heat-fluid-solid-chemical coupling model and the measured values ​​at multiple scales is calculated, and the state vector prediction value is dynamically adjusted to obtain the optimal parameter value at the current moment. The optimal parameter values ​​obtained by Kalman filtering are physically verified, and the verified optimal parameter values ​​are substituted into the heat-fluid-solid-chemical coupling model to calculate the current settlement and the predicted future settlement.

[0076] Specifically, the THMC parameters are corrected in real time using a Kalman filter algorithm. The specific parameters to be corrected are: permeability K, which reflects the key parameter most affected by temperature changes and structure; damage factor D, which reflects the degree of microfracture in the rock mass; and thermal conductivity, which varies with fracture aperture. Chemical scaling rate - reaction rate .

[0077] The parameters of the THMC model are dynamically updated using Kalman filtering, which includes the state vector, observation vector, prediction step, and update step.

[0078] State vector definition:

[0079] Definition of observation vector:

[0080] Where T is the surface temperature of the macroscopic layer.

[0081] State prediction based on physical evolution models:

[0082] Establish a penetration rate evolution model:

[0083] Establish a penetration evolution rate model: (16) For example, .

[0084] The residuals are calibrated using pressure observations (the residual values ​​only reflect the deviation from the actual observations and are not included in dynamic updates):

[0085] Constrained permeability physical range 10 -16 m 2 <K<10 -12 m 2。

[0086] The residuals reflect the discrepancy between model predictions and actual observations. The update step solves the THMC equations:

[0087] When predicting ground subsidence, the geothermal reinjection process will be considered in real time. Based on this, an adaptive control strategy for reinjection will be further adopted. Based on the prediction results of the real-time corrected THMC model, reinjection parameters (flow rate, temperature, and chemical treatment) will be dynamically adjusted to minimize subsidence risk. The implementation method of the adaptive recharge regulation strategy is as follows: 1. Control Logic Design: The objective is to dynamically adjust reinjection parameters (flow rate, temperature, chemical treatment) based on real-time corrected THMC model predictions to minimize settlement risk. The control variable is the reinjection flow rate. Recharge temperature Anti-fouling agent injection rate

[0088] 2. Multi-objective optimization model.

[0089] Objective function:

[0090] in, For the settling rate, For reservoir temperature, The concentration of silica. For safe concentration, To mitigate the risk of settlement, For cold shock suppression, To prevent and control scaling Constraints: Reinjection pressure limit P inj ≤P max (Injection pressure less than or equal to maximum pressure), temperature gradient limit less than or equal to 5℃ / m (to suppress sudden thermal stress), chemical concentration safety threshold. ≤200mg / L.

[0091] 3. Real-time decision-making process.

[0092] Data input: Real-time monitoring data z k (Settlement value), the settlement rate predicted by the THMC coupled calculation model. Temperature field T, SiO2 concentration .

[0093] Optimization Solution: A model predictive control framework is adopted, with rolling time-domain optimization. The prediction time domain is the next 6 hours, and the interior-point method (IPOPT) or a reinforcement learning surrogate model is used as the solver. Execution strategy: Staged cooling; when blockage is detected (e.g., sudden increase in DAS strain + DTS temperature gradient exceeding limits), trigger a high-pressure pulse for cleaning (20MPa for 10 minutes); when >200mg / L, according to Inject anti-scaling agent.

[0094] In this embodiment, based on multi-source data coupling calculation, multi-source data fusion and spatiotemporal pyramid algorithm are used for further data processing. By fusion of multi-scale data in a hierarchical manner, the accuracy and real-time performance of geothermal reinjection settlement monitoring are significantly improved. The range of key parameters in the fusion process is constrained by the THMC equation to avoid the risk of overfitting driven by pure data. The output is an interpretable multi-field contribution analysis to guide the dynamic optimization of reinjection parameters.

[0095] Example 2 The purpose of this embodiment is to provide a multi-scale geothermal extraction and reinjection process ground subsidence monitoring system, including: The acquisition module is configured to acquire macroscopic monitoring data, mesoscopic monitoring data, and microscopic monitoring data of the monitoring area; wherein, the macroscopic monitoring data includes regional surface deformation data and surface temperature data; the mesoscopic monitoring data includes soil and rock strain, temperature gradient, and seepage pressure; and the microscopic monitoring data includes rock mass fracture signal data and geothermal water chemical concentration data. The extraction module is configured to: map the acquired macroscopic monitoring data, mesoscopic monitoring data and microscopic monitoring data to different grid layers of a unified grid system, extract key physical quantities from different grid layers to calculate risk values ​​reflecting rock mass damage risk, quantify the contribution of thermal stress effect and trend values ​​reflecting settlement trend; The prediction module is configured to: map the calculated risk value reflecting the risk of rock mass damage, the contribution of the quantified thermal stress effect, and the trend value reflecting the settlement trend into the thermal-fluid-solid-chemical coupling model; use the Kalman filter algorithm to correct the parameters of the thermal-fluid-solid-chemical coupling model in real time; obtain the settlement prediction result based on the corrected thermal-fluid-solid-chemical coupling model; and regulate the reinjection process according to the settlement prediction result.

[0096] In further embodiments, the following is also provided: An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When executed by the processor, the computer instructions perform the method described in Embodiment 1. For brevity, further details are omitted here.

[0097] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0098] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0099] A computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in Embodiment 1.

[0100] The method in Embodiment 1 can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not provided here.

[0101] A computer program product includes a computer program that, when executed by a processor, implements the method described in Embodiment 1.

[0102] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods described above. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.

[0103] The computer program code used to implement the methods of the present invention may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the computer or other programmable data processing device, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0104] In the context of this invention, computer program code or related data may be carried by any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0105] Those skilled in the art will recognize that the units and algorithm steps described in conjunction with the embodiments herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0106] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for monitoring ground subsidence during geothermal extraction and reinjection at multiple scales, characterized in that, include: Acquire macroscopic, mesoscopic, and microscopic monitoring data for the monitoring area; wherein, the macroscopic monitoring data includes regional surface deformation data and surface temperature data; the mesoscopic monitoring data includes soil and rock strain, temperature gradient, and seepage pressure; and the microscopic monitoring data includes rock mass fracture signal data and geothermal water chemical concentration data. The acquired macroscopic, mesoscopic, and microscopic monitoring data are mapped to different grid layers of a unified grid system. Key physical quantities are extracted from different grid layers to calculate risk values ​​reflecting rock mass damage risk, quantify the contribution of thermal stress effects, and reflect the trend of settlement. The calculated risk value reflecting rock mass damage risk, the contribution of quantified thermal stress effect, and the trend value reflecting settlement trend are mapped into the thermal-fluid-solid-chemical coupling model. The parameters of the thermal-fluid-solid-chemical coupling model are corrected in real time using the Kalman filter algorithm. Settlement prediction results are obtained based on the corrected thermal-fluid-solid-chemical coupling model, and the reinjection process is regulated according to the settlement prediction results.

2. The method for monitoring ground subsidence during geothermal extraction and reinjection at multiple scales as described in claim 1, characterized in that, The acquired macroscopic, mesoscopic, and microscopic monitoring data are mapped to different grid layers of a unified grid system. Key physical quantities are extracted from each grid layer to calculate risk values ​​reflecting rock mass damage risk, quantify the contribution of thermal stress effects, and determine trend values ​​reflecting settlement trends. Specifically: The acquired macroscopic, mesoscopic, and microscopic monitoring data are mapped to different grid layers of a unified grid system, resulting in a top-level grid layer corresponding to the macroscopic monitoring data, a middle-level grid layer corresponding to the mesoscopic monitoring data, and a bottom-level grid layer corresponding to the microscopic monitoring data. The key physical quantity extracted from the top-level grid layer is displacement rate; the key physical quantities extracted from the middle-level grid layer are rock strain, temperature gradient, and seepage pressure; and the key physical quantities extracted from the bottom-level grid layer are microseismic event density and SiO2 concentration. Trend values ​​reflecting the settlement trend are obtained based on displacement rate and seepage pressure; the contribution of thermal stress effect is quantified based on rock strain and temperature gradient; and risk values ​​reflecting the risk of rock mass damage are obtained based on microseismic event density and SiO2 concentration.

3. The method for monitoring ground subsidence during geothermal extraction and reinjection at multiple scales as described in claim 1, characterized in that, The calculated risk value reflecting the rock mass damage risk is mapped to the damage factor in the momentum equation of the thermo-fluid-solid-chemical coupling model to correct the rock elastic stiffness matrix; the calculated contribution of the quantified thermal stress effect is mapped to the heat source term in the energy equation of the thermo-fluid-solid-chemical coupling model. The calculated trend value reflecting the settlement trend is mapped to the boundary pressure in the mass equation of the thermo-fluid-solid-chemical coupling model to constrain the extraction / reinjection flow rate.

4. The method for monitoring ground subsidence during geothermal extraction and reinjection at multiple scales as described in claim 1 or 3, characterized in that, The parameters of the heat-fluid-solidification coupling model are corrected in real time using the Kalman filter algorithm. Settlement prediction results are then obtained based on the corrected heat-fluid-solidification coupling model. Specifically: A state vector is constructed based on permeability, damage factor, equivalent thermal conductivity and chemical reaction rate; an observation vector is constructed based on soil strain, temperature, seepage pressure and SiO2 concentration. A parameter evolution model is established based on physical mechanisms, and physical evolution calculations are performed on all parameters in the state vector to obtain the state prediction value at time k. The deviation between the predicted values ​​of the heat-fluid-solid-chemical coupling model and the measured values ​​at multiple scales is calculated, and the state vector prediction value is dynamically adjusted to obtain the optimal parameter value at the current moment. The optimal parameter values ​​obtained by Kalman filtering are physically verified, and the verified optimal parameter values ​​are substituted into the heat-fluid-solid-chemical coupling model to calculate the current settlement and the predicted future settlement.

5. The method for monitoring ground subsidence during geothermal extraction and reinjection at multiple scales as described in claim 1, characterized in that, The recharge process is regulated based on the settlement prediction results, specifically as follows: A multi-objective optimization model is constructed with the goal of minimizing settlement risk, cold shock, and chemical scaling. The multi-objective optimization model is solved with the constraints of reinjection pressure limit, temperature gradient limit, and chemical concentration safety threshold to obtain the optimal reinjection parameters. The geothermal reinjection process is then dynamically adjusted based on the optimal reinjection parameters.

6. The method for monitoring ground subsidence during geothermal extraction and reinjection at multiple scales as described in claim 1, characterized in that, The thermo-fluid-solid-chemical coupling model includes the mass conservation equation for the seepage field, the energy conservation equation for the temperature field, the momentum conservation equation for the stress field, and the chemical field coupling equation considering the dissolution-precipitation kinetics of SiO2 minerals.

7. A multi-scale geothermal extraction and reinjection process ground subsidence monitoring system, characterized in that, include: The acquisition module is configured to acquire macroscopic monitoring data, mesoscopic monitoring data, and microscopic monitoring data of the monitoring area; wherein, the macroscopic monitoring data includes regional surface deformation data and surface temperature data; the mesoscopic monitoring data includes soil and rock strain, temperature gradient, and seepage pressure; and the microscopic monitoring data includes rock mass fracture signal data and geothermal water chemical concentration data. The extraction module is configured to: map the acquired macroscopic monitoring data, mesoscopic monitoring data and microscopic monitoring data to different grid layers of a unified grid system, extract key physical quantities from different grid layers to calculate risk values ​​reflecting rock mass damage risk, quantify the contribution of thermal stress effect and trend values ​​reflecting settlement trend; The prediction module is configured to: map the calculated risk value reflecting the risk of rock mass damage, the contribution of the quantified thermal stress effect, and the trend value reflecting the settlement trend into the thermal-fluid-solid-chemical coupling model; use the Kalman filter algorithm to correct the parameters of the thermal-fluid-solid-chemical coupling model in real time; obtain the settlement prediction result based on the corrected thermal-fluid-solid-chemical coupling model; and regulate the reinjection process according to the settlement prediction result.

8. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, perform the method described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, implements the method described in any one of claims 1-6.

Images