Abandoned mine gravity energy storage integrated monitoring method based on digital twin model

Abstract

The application discloses a kind of abandoned mine gravity energy storage integration monitoring methods based on digital twin model, comprising: establishing the three-dimensional digital twin model of abandoned mine underground space and energy storage facility and assigning initial physical and mechanical parameters;Based on model preliminary analysis and identification seepage prone area and structure key bearing area and deploy sensor;Collect seepage field and structure response real-time monitoring data;Using monitoring data through parameter inversion algorithm dynamically updates digital twin model;Based on the updated model, fluid-structure coupling evolution simulation is carried out to predict the evolution trend of seepage field and stress field and the safety state of structure in the future period;According to the prediction result, judge whether to trigger graded early warning and execute coordinated response measures.The application dynamically assimilates digital twin and real-time monitoring data, realizes modeling-monitoring-inversion-prediction-decision-making whole-chain closed-loop management and control, can predict underground space safety risk in advance, effectively improves the safety operation level of abandoned mine gravity energy storage system.

Description

Technical Field

[0001] This invention relates to the fields of resource utilization of abandoned mines, safety monitoring of gravity energy storage systems, digital twins and multi-field coupling simulation of underground space, and specifically to an integrated monitoring method for gravity energy storage in abandoned mines based on a digital twin model. Background Technology

[0002] With the rapid development of the new energy industry, gravity energy storage technology, with its advantages of large capacity, long lifespan, and low cost, has become one of the key paths to solve the spatial and temporal mismatch between energy supply and demand. Abandoned mines, due to their naturally existing huge underground spaces, are considered ideal carriers for the transformation of gravity energy storage systems. However, the underground spaces of abandoned mines face two major core safety challenges: First, the problem of mine water leakage. Long-term seepage not only corrodes energy storage electromechanical equipment, but also significantly reduces the bearing capacity of the surrounding rock and support structure through increased pore water pressure and softening of the rock mass. Second, the problem of structural stability. The periodic storage and retrieval of mass blocks in the gravity energy storage system generates complex dynamic fatigue loads, which, combined with the aging and deterioration of the original mine support structure and water seepage erosion, can easily induce structural deformation, crack propagation, and even instability and collapse accidents.

[0003] Existing monitoring technologies have significant limitations: traditional piezometers or flowmeters can only acquire single permeability parameters and cannot quantitatively assess the degree of degradation of the surrounding rock's mechanical properties by seepage; microseismic or displacement monitoring systems only focus on the structure's own response and do not consider the dynamic changes in the structural mechanical boundary conditions caused by seepage. In recent years, digital twin technology has been initially applied in underground engineering, but existing solutions are mostly static modeling or simple visualization, lacking the ability to dynamically assimilate with real-time monitoring data and predict fluid-structure interaction evolution, and are not customized for the dynamic loads unique to gravity energy storage. Therefore, there is an urgent need to develop an integrated method based on digital twin models, integrating real-time monitoring, parameter inversion, multi-field prediction, and collaborative early warning to achieve full life-cycle safety management of underground gravity energy storage spaces in abandoned mines. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an integrated monitoring method for waterproofing and structural safety of underground spaces in abandoned mine gravity energy storage facilities based on a digital twin model. This method establishes a 1:1 digital twin model of the mine and energy storage facility, deploys seepage field and structural response sensors in key areas, utilizes real-time monitoring data to invert the physical and mechanical parameters of the surrounding rock, drives the digital twin model to perform fluid-structure interaction evolution simulation, dynamically predicts future risks, and achieves coordinated early warning across the entire chain from causes to consequences.

[0005] To achieve the above objectives, the specific technical solution of the present invention is as follows: This invention provides an integrated monitoring method for gravity energy storage in abandoned mines based on a digital twin model, comprising the following steps: S1: Establish a three-dimensional digital twin model of the underground space and energy storage facilities of the abandoned mine, and assign initial physical and mechanical parameters to the digital twin model; S2: Based on the preliminary analysis of the digital twin model, identify areas prone to seepage and critical structural load-bearing areas, deploy seepage field sensors in the areas prone to seepage, and deploy structural response sensors in the critical structural load-bearing areas; S3: Collect seepage field data through the seepage field sensor and collect structural response data through the structural response sensor to obtain real-time monitoring data; S4: Using the real-time monitoring data, the physical and mechanical parameters of the digital twin model are dynamically updated through a parameter inversion algorithm to obtain the updated digital twin model; S5: Based on the updated digital twin model, perform fluid-structure interaction evolution simulation, embed the mechanical parameter attenuation relationship caused by seepage into the fluid-structure interaction evolution simulation, predict the evolution trend of seepage field and stress field and structural safety status in future time period, and calculate the structural safety factor and comprehensive safety evaluation index to obtain the prediction results; S6: Based on the prediction results, determine whether a graded early warning is triggered and execute corresponding collaborative response measures.

[0006] In one embodiment, in step S1, the geometric dimensions of the three-dimensional digital twin model are established at a 1:1 scale according to the geometric dimensions of the actual underground space and energy storage facilities in the mine. The construction basis of the three-dimensional digital twin model includes geological survey reports, structural as-built drawings, and gravity energy storage engineering design data. The initial physical and mechanical parameters include elastic modulus, Poisson's ratio, cohesion, internal friction angle, tensile strength, permeability coefficient, porosity, and initial groundwater level.

[0007] In one embodiment, in step S2, the seepage-prone area includes at least one of the following: fault fracture zone, rock fissure development zone, boundary of the original water accumulation zone, and dynamic load disturbance sensitive area along the track; the critical load-bearing area includes at least one of the following: track support beam connection node, bottom foundation of energy storage silo, steel-concrete composite node, surrounding rock anchoring zone, and critical section on the dynamic load transmission path. The critical section refers to a representative cross-section located on the dynamic load transmission path, where stress or deformation response is concentrated and the risk of damage is high.

[0008] In one embodiment, in step S3, the seepage field sensor collects data at a cycle of 1 to 24 hours, and the seepage rate is calculated by the pressure difference between adjacent measuring points and Darcy's law; in the structural response sensor, the micro-vibration sensor continuously collects waveforms, and the displacement gauge and strain gauge collect displacement and strain data at a set cycle.

[0009] In one embodiment, in step S4, the parameter inversion algorithm employs ensemble Kalman filtering or Bayesian inversion methods to assimilate the observed data into the digital twin model. The inversion target parameters include the attenuation coefficients of the permeability coefficient field, elastic modulus field, and shear strength parameters in the seepage-affected zone; the inversion objective function is: , In the formula, The parameter vector to be inverted, Let be the prior mean vector of the parameters to be inverted. For the orthogonal operator, For observation data, , These are the observation error and the prior covariance matrix of the parameters, respectively. The forward modeling operator is used to perform fluid-structure interaction simulation to calculate the theoretical observation values ​​corresponding to the model parameters. After each inversion, the updated parameter field and its variance are obtained, and the updated parameter field is used as the input for the forward modeling simulation at the next time step.

[0010] In one embodiment, step S5 specifically includes: simulating the fluid-structure interaction evolution process over the next 7 to 30 days, and outputting the time series of pore water pressure, permeability, cumulative displacement of key points, microseismic event rate and energy release prediction during the prediction period; The mechanical parameter attenuation relationship includes the attenuation coefficient of the elastic modulus of the surrounding rock, which is defined as: , In the formula, The elastic modulus of the surrounding rock in a dry state. This is the critical pore water pressure; beyond this value, the effective stress approaches zero. The rock mass sensitivity coefficient is determined by indoor triaxial tests. Let be the elastic modulus of the surrounding rock under water-bearing conditions at time t. Let t be the real-time pore water pressure monitored at time t.

[0011] The pore water pressure is used to calculate the elastic modulus attenuation coefficient of the surrounding rock to quantify the degree of rock softening. It also serves as the input data source for the inversion model parameters and is used to determine the early warning level and response decision.

[0012] In one embodiment, in step S5, the structural safety factor is calculated using the following formula: , In the formula, The cumulative microseismic energy (J) in the 24 hours preceding time t. To allow maximum energy; The cumulative displacement of key points up to time t; To allow the maximum displacement; , As weight, and ; This is a corrosion correction factor.

[0013] In one embodiment, the corrosion correction factor The value is determined based on the mine water corrosion environment level: when the mine water corrosion environment level is assessed as Level I (slight corrosion) or Level II (weak corrosion), The criteria for judgment are that the pH value of the mine water is between 6.5 and 8.5, and the Cl... - Concentration less than 200 mg / L, SO4² - When the concentration is less than 300 mg / L and there are no significant signs of oxidation of sulfur-containing minerals in the surrounding rock; and the mine water environment is assessed as Class III (moderate corrosion) or Class IV (strong corrosion), ,in The corrosion tendency comprehensive evaluation index is determined by the weighted sum of the influence indices of each corrosion influencing factor, and ; The value range is (0, 1.0], and it should be dynamically updated over time, with an update cycle not exceeding 6 months.

[0014] In one embodiment, in step S5, the comprehensive security evaluation index Calculate using the following formula: , in, The coefficient for attenuation of the elastic modulus of the surrounding rock. For structural safety factor, , As weight, and .

[0015] In one embodiment, step S6 includes a tiered early warning system comprising a Level 1 early warning, a Level 2 early warning, and a Level 3 early warning: Level 1 Warning: Meets at least one of the following conditions: Comprehensive Safety Assessment Index Or, real-time permeation rate ; in To allow the maximum permeation rate; Level 2 Warning: Comprehensive Safety Assessment Index and or structural safety factor Or the microseismic event rate increases by more than 50% for more than one hour compared to the sliding average of the previous 24 hours; Level 3 Warning: Comprehensive Security Assessment Index and ,or ;in, This represents the accumulated microseismic energy over the preceding 24 hours up to time t. The maximum allowable cumulative microseismic energy set for the system; The cumulative displacement of the key points up to time t, To allow the maximum displacement; If the predicted indicators of the digital twin model at any time within the future warning time window meet at least one of the criteria for a Level 2 warning, the system issues a Level 2 predictive warning; if at least one of the criteria for a Level 3 warning is met, the system issues a Level 3 predictive warning; when the predicted data simultaneously meets the criteria for two or more warning levels, the system issues a warning based on the highest risk level among all the currently met warning levels.

[0016] In one embodiment, step S6 includes the following coordinated response measures: When a Level 1 warning is triggered, voice prompts and encrypted inspections will be activated. When a Level II warning is triggered, an audible and visual alarm is triggered and heavy-load operation is suspended, and grouting or crack sealing is implemented; When a Level 3 warning is triggered, an emergency shutdown, regional lockdown, and comprehensive structural reinforcement will be implemented.

[0017] In one embodiment, the digital twin model is also used for scenario simulation and decision support: simulating the effects of different grouting schemes, drainage strategies, or suspending heavy-load operation on improving the safety status, and assisting in selecting the optimal handling measures.

[0018] The core mechanism of the above-mentioned technical solution of the present invention lies in: 1. Digital twin-driven full-chain safety management: Integrating mine geological, engineering and monitoring data into a 1:1 digital twin model to achieve a closed loop of modeling-monitoring-inversion-prediction-decision, overcoming the fundamental defects of traditional methods such as separation of monitoring and evaluation and lack of predictive ability.

[0019] 2. Data assimilation and inversion: Using sparse field sensor data (seepage pressure, displacement, microseismic data), the spatial distribution of key mechanical parameters of the surrounding rock is inverted through ensemble Kalman filtering or Bayesian inversion methods, so that the digital twin model can approximate the physical entity state in real time.

[0020] 3. Fluid-Structure Coupling Evolution Prediction: The constitutive relationship of effective stress attenuation and rock softening caused by seepage is embedded in the digital twin model to dynamically simulate the evolution of seepage and stress fields in the next few days to weeks, thus upgrading the early warning from post-event alarm to pre-event prediction.

[0021] 4. Dynamic load adaptation of gravity energy storage: The digital twin model takes into account the dynamic fatigue load path generated by the periodic storage and retrieval of heavy blocks, which is significantly different from the monitoring methods of ordinary mines.

[0022] Compared with the prior art, the beneficial technical effects of this invention are reflected in: (1) Predictive maintenance capability: Through multi-field evolution simulation of digital twin model, the development trend of seepage field and stress field can be predicted 7 to 30 days in advance, and an early warning can be issued before the actual deterioration of the structure, realizing the leap from passive alarm to active prevention.

[0023] (2) Model self-evolution: The model parameters are continuously updated by using real-time monitoring data and inversion algorithm. The digital twin model always maintains the same physical state as the actual underground space, overcoming the shortcomings of traditional numerical models that are distributed offline and never updated.

[0024] (3) Decision simulation platform: Supports rapid simulation of the effects of different intervention measures (grouting, drainage, load reduction) on the safety status, providing quantitative basis for operation and maintenance decisions.

[0025] (4) High sensitivity and early warning: The piezometer can sense pore water pressure changes at the level of 0.001 MPa, and the microseismic monitoring can capture rock micro-fracture signals several days to several weeks before the structure exhibits millimeter-level macroscopic deformation. The early warning timeliness is significantly better than that of traditional methods.

[0026] (5) Strong environmental and working condition adaptability: The sensor has intrinsic explosion-proof and anti-electromagnetic interference characteristics, and can operate stably for a long time in high humidity (RH>85%) and corrosive mine environments, perfectly adapting to the monitoring needs of intermittent strong dynamic loads of gravity energy storage systems.

[0027] (6) High versatility: It does not depend on specific chemical ions and is applicable to abandoned mines with various water qualities (fresh water, acidic water, high mineralization water). There is no need to recalibrate the chemical sensor for different mines, which greatly improves the engineering scalability. Attached Figure Description

[0028] Figure 1 This is a flowchart illustrating the overall steps of the integrated monitoring method for gravity energy storage in abandoned mines based on a digital twin model, as described in this invention. Figure 2 This is a flowchart illustrating the overall architecture of the integrated monitoring method for waterproofing and structural safety of underground space in abandoned mine gravity energy storage, as described in this invention. Figure 3 Flowchart for building and dynamically updating digital twin models. Detailed Implementation

[0029] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0030] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0031] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0032] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0033] like Figure 1 As shown, this invention provides an integrated monitoring method for gravity energy storage in abandoned mines based on a digital twin model, comprising the following steps: S1: Establish a three-dimensional digital twin model of the underground space and energy storage facilities of the abandoned mine, and assign initial physical and mechanical parameters to the digital twin model; S2: Based on the preliminary analysis of the digital twin model, identify areas prone to seepage and critical load-bearing areas of the structure, deploy seepage field sensors in areas prone to seepage, and deploy structural response sensors in critical load-bearing areas of the structure. S3: Collect seepage field data through seepage field sensors and structural response data through structural response sensors to obtain real-time monitoring data; S4: Using real-time monitoring data, the physical and mechanical parameters of the digital twin model are dynamically updated through a parameter inversion algorithm to obtain the updated digital twin model; S5: Based on the updated digital twin model, perform fluid-structure interaction evolution simulation, embed the mechanical parameter attenuation relationship caused by seepage into the fluid-structure interaction evolution simulation, predict the evolution trend of seepage field and stress field and structural safety status in future time period, and calculate the structural safety factor and comprehensive safety evaluation index to obtain the prediction results. S6: Based on the prediction results, determine whether a tiered early warning is triggered and execute corresponding coordinated response measures.

[0034] This embodiment provides an integrated monitoring method for waterproofing and structural safety of underground gravity energy storage spaces in abandoned mines based on digital twin models, comprising the following parts: Part 1: Digital Twin Model Construction and Initial Parameter Assignment ① Geometric Model Construction: Based on the geological survey report of the abandoned mine (including borehole columnar section, seismic profile, rock strata interface, fault attitude, and fracture zone distribution), structural as-built drawings (tunnel cross-sectional dimensions, support type, anchor bolt / cable arrangement), and gravity energy storage engineering design data (weight block size and mass, track layout, support beam position and connection method), a three-dimensional geometric model of the mine's underground space and energy storage facilities is established. The geometric dimensions of the three-dimensional digital twin model are established at a 1:1 scale according to the actual geometric dimensions of the mine's underground space and energy storage facilities, including strata, faults, fracture zones, tunnels, support structures (anchor bolts, shotcrete, steel arches), tracks, and weight blocks.

[0035] ② Initial physical and mechanical parameter field assignment: based on rock mechanics tests (elastic modulus E, Poisson's ratio ν, cohesion c, internal friction angle φ, tensile strength). ), permeability test (permeability coefficient) k Initial properties (porosity n, unsaturated parameter van Genuchten, parameter α, m) and initial groundwater level are assigned to each rock stratum unit in the geometric model. For fault fracture zones or fractured areas with large parameter spatial variability, geostatistical methods (Kriging or sequential Gaussian simulation) are used to generate a stochastic parameter field, and the mean field is used as the input to the initial model.

[0036] ③ Boundary conditions and dynamic load definitions: Set the model's surface boundary (displacement constraint), lateral boundary (normal constraint or far-field stress applied based on measured ground stress values), and bottom boundary (fixed constraint). Based on the gravity storage operation plan, define the storage and retrieval speed, residence time, and corresponding dynamic load time history F(t) on the track support structure for the heavy block. Seepage boundary conditions include known head boundaries (groundwater level), drainage boundaries (tunnel drainage ditches), and possible impermeable boundaries.

[0037] ④ Mesh Generation and Numerical Solver Configuration: A hexahedral or tetrahedral mesh discretization model is used, with local refinement applied in the sensor deployment area, fault fracture zone, and expected damage area. A fluid-structure interaction solver (based on the finite element method or finite volume method) is configured, an appropriate constitutive model (such as the Mohr-Coulomb or Drucker-Prager model) is selected, and the initial time step and solution convergence criteria are set.

[0038] Part Two: Preliminary Preparations and Monitoring Area Delineation ① Multidimensional Information Acquisition and Risk Identification: Based on preliminary analysis using a digital twin model, high-risk areas with large seepage gradients, rapid decreases in effective stress, and concentrated displacement are identified. Combining geological data and energy storage operation parameters, the following two types of areas are precisely delineated: Areas prone to seepage include: fault fracture zones, areas with developed rock fissures, boundaries of former water accumulation areas, and areas sensitive to dynamic load disturbances along the railway line. Changes in seepage field parameters will be the primary focus of monitoring in these areas.

[0039] Critical load-bearing areas of the structure include: track support beam connection nodes, the bottom foundation of the energy storage silo, steel-concrete composite nodes, the surrounding rock anchorage zone, and key sections along the dynamic load transfer path. These critical sections refer to representative cross-sections located along the dynamic load transfer path where stress or deformation responses are concentrated and the risk of damage is high. Stress fields and structural response parameters are primarily monitored in this area.

[0040] ② Sensor selection and calibration: Waterproofing monitoring (seepage field monitoring): A fiber optic grating piezometer (range 0-2 MPa, accuracy 0.1% FS) or a time-domain reflectometry moisture sensor is used to continuously monitor pore water pressure Pw and medium moisture content θ. Optional fiber optic strain sensor cables can be buried in the borehole to invert the seepage state by sensing the matrix suction response caused by changes in surrounding rock moisture.

[0041] Structural monitoring (stress field and structural response): piezoelectric micro-vibration sensors (sampling rate ≥10 kHz, frequency response 1–1000 Hz), high-precision displacement gauges (accuracy ±0.1 mm), and strain gauges (accuracy ±1 με) were selected.

[0042] Parameter calibration: The permeability coefficient k of the monitoring area is calibrated through on-site pumping or injection tests, and the constitutive relationship between pore water pressure and seepage rate is established; the mapping relationship between microseismic energy and structural damage degree is established through artificial vibration tests; and the propagation velocity of microseismic waves in rock strata, concrete and steel components is measured on-site.

[0043] Part Three: Monitoring System Setup and Sensor Deployment ① Deployment in seepage-prone areas: Drill boreholes horizontally in seepage-prone areas (50–80 mm diameter, 3–8 m depth) to eliminate gravity interference with moisture transport monitoring. Deploy 3–5 fiber optic piezometers or moisture content sensors in each area, spaced 5–8 m apart. Seal the borehole openings with waterproof foam, and lay the sensor armored cables through protective conduits to the ground-based data acquisition center.

[0044] ② Deployment in critical load-bearing areas: Microseismic sensors are deployed using a combination of fixed-point placement at key locations and full-area coverage. The spacing between critical load-bearing nodes is 5–10 m. A triangular positioning array (≥4 sensors per layer) is deployed every 50 m along the mine rock wall to ensure three-dimensional positioning redundancy. Displacement gauges and strain gauges are installed at key sections along the track support beam connection nodes, the bottom foundation of the energy storage silo, steel-concrete composite nodes, surrounding rock anchoring zones, and dynamic load transfer paths. The sensors are tightly coupled to the measured medium via expansion bolts or drilled cement grouting.

[0045] ③ System Integration and Calibration: Construct an integrated architecture consisting of "sensor array – data acquisition terminal – analysis server – digital twin platform – early warning terminal". Before system startup, perform positioning accuracy calibration (seismic source positioning error ≤ 1 m) and zero-point calibration of the pressure sensor.

[0046] Part Four: Multi-Source Data Acquisition and Collaborative Processing ① Data Synchronization Acquisition Strategy Waterproofing data (seepage field): Pore water pressure Pw and water content θ are continuously collected by piezometer or moisture content sensor at a cycle of 1–24 h; the seepage rate is calculated by the pressure difference between adjacent measuring points and Darcy's law.

[0047] Structural data: Microseismic sensors continuously acquire waveforms for 7×24 h; displacement gauges acquire three-dimensional coordinate changes every 5 min; strain gauges acquire strain values ​​every 1 h.

[0048] ② Waterproofing data processing Step 1: Abnormal fluctuations in pore water pressure monitoring values ​​are eliminated after temperature compensation.

[0049] Step 2: Calculate the real-time permeation rate based on Darcy's Law: ,in The permeability coefficient (calibrated through field tests). The difference in pore water pressure between adjacent measuring points. This represents the distance between measuring points.

[0050] Step 3: Based on the effective stress principle, calculate the attenuation coefficient of seepage on the mechanical parameters of the surrounding rock. :

[0051] In the formula, , These are the elastic moduli of the surrounding rock in dry and water-bearing states, respectively. The critical pore water pressure (beyond which the effective stress tends to zero) is the critical pore water pressure. This is the rock mass sensitivity coefficient (which can be calibrated through indoor triaxial tests).

[0052] ③Structured data processing flow Signal denoising and feature extraction: Wavelet transform is used to remove microseismic background noise and extract feature parameters such as peak amplitude, duration, and energy.

[0053] Source location: Based on multi-channel time difference data, the Geiger iterative algorithm is used to invert the three-dimensional coordinates of the microseismic source. The multi-channel time difference data refers to the data obtained by extracting the arrival time of the first wave of each channel and calculating the arrival time difference between channels after multiple microseismic sensor channels deployed at different spatial locations receive the same microseismic event signal.

[0054] Cumulative displacement and strain calculation: Real-time calculation of cumulative displacement and strain at key sections.

[0055] Part 5: Parameter Inversion and Dynamic Update of Digital Twin Model ① Data Extraction: Extract key variables for inversion from real-time data acquired by the sensor array, including pore water pressure. (Pyrometer), Displacement (Displacement gauge), microseismic event location and energy (used to calibrate the damage zone). Further processing of the microseismic data calculates the microseismic event rate and cumulative released energy per unit time. The microseismic event rate refers to the number of microseismic events detected per unit time (e.g., 1 hour).

[0056] ② Parameter Inversion Algorithm: Using ensemble Kalman filtering (EnKF) or Bayesian inversion methods, the observed data are assimilated into a digital twin model to invert the spatial distribution of key surrounding rock parameters. The target parameters for inversion include: Permeability field , Elastic modulus field , Shear strength parameters Attenuation coefficient in the seepage-affected zone The inversion objective function is:

[0057] In the formula, The parameter vector to be inverted, Let be the prior mean vector of the parameters to be inverted. For the orthogonal operator, For observation data, , These are the observation error and the prior covariance matrix of the parameters, respectively; the theoretical observation values ​​corresponding to the model parameters are calculated by performing fluid-structure interaction simulation using the forward modeling operator.

[0058] ③ Model Update and Uncertainty Quantification: After each inversion, the updated parameter field and its variance are obtained. The updated parameter field is used as the input for the forward simulation at the next time step. The parameter variance is used to quantify the uncertainty of the prediction results and to provide a confidence interval for risk warning.

[0059] Part VI. Multi-stage Evolution Simulation and Risk Prediction ① Forward modeling prediction: Based on the inverted and updated digital twin model, simulate the next T days (recommended T=7). The fluid-structure interaction evolution process over 30 days. Outputs the predicted pore water pressure and permeability. The time series of cumulative displacement at key points, microseismic event rate, and energy release prediction are used. The pore water pressure is used to calculate the elastic modulus attenuation coefficient of the surrounding rock to quantify the degree of rock softening, and also serves as the input data source for the inversion model parameters, and is used to determine the early warning level and response decisions.

[0060] ② Dynamic surrounding rock deterioration model: This model incorporates the attenuation relationship of mechanical parameters caused by water seepage into a fluid-structure interaction simulation. The attenuation coefficient of the surrounding rock elastic modulus is defined as follows: , In the formula, The elastic modulus of the surrounding rock in a dry state. The critical pore water pressure (beyond which the effective stress tends to zero) is the critical pore water pressure. The rock mass sensitivity coefficient is related to the rock mass properties and affects the material stiffness matrix in the numerical model. It is calibrated through indoor triaxial tests.

[0061] ③ Structural safety factor prediction: Calculate the current and predicted structural safety factor by combining the cumulative energy and displacement of microseismic events. , in, The cumulative microseismic energy (J) in the 24 hours preceding time t. To allow maximum energy; The cumulative displacement of key points up to time t; To allow the maximum displacement; , As weight, and The determination should be based on the properties of the surrounding rock and test data; if specific data is unavailable, it is recommended to... , ; This is a corrosion correction factor.

[0062] The corrosion correction factor The value is determined based on the corrosion environment level of the mine water, as follows: (1) Generally, a value of 1.0 is used, which means that the mine water corrosion environment level is rated as Level I (slight corrosion) or Level II (weak corrosion). The judgment criteria are: the pH value of the mine water is between 6.5 and 8.5, and the Cl... - Concentration less than 200 mg / L, SO4² - The concentration is less than 300 mg / L, and there are no significant signs of oxidation of sulfur-containing minerals (such as pyrite) in the surrounding rock. In such environments, the deterioration effect of water quality on the anchoring structure and the long-term strength of the surrounding rock is negligible; therefore, [the following is taken as the optimal concentration]. .

[0063] (2) When the mine water environment exceeds the above range, i.e., is rated as Class III (moderate corrosion) or Class IV (strong corrosion), the following formula shall be used to classify it. Quantitative reduction:

[0064] in, The comprehensive evaluation index of corrosion tendency ( ), is determined by the following formula:

[0065] In the formula, The influence index (0-1) of the i-th corrosion influencing factor includes, but is not limited to: pH influence index, Cl... - Concentration effect index, SO4² - Concentration influence index and surrounding rock oxidation environment index; The corresponding weights were determined using the analytic hierarchy process (AHP) combined with actual mine measurement data, and each weight satisfies the following conditions: Each Influence Index Calculated based on measured values ​​using a piecewise linear interpolation function.

[0066] (3) The value range is (0, 1.0). When the corrosive environment is most severe, As the value approaches the lower limit, the structural safety factor is reduced according to the most unfavorable working condition. In practical engineering applications, if systematic water quality and surrounding rock chemical analysis data are lacking, A conservative value of 0.85 should be chosen to ensure a safety margin. Furthermore, It should be updated dynamically over time, with each update cycle not exceeding 6 months. The update should be based on the water quality test report and the sampling inspection data of anchor bolt / anchor cable corrosion during the same period.

[0067] ④ Comprehensive safety evaluation index: Construct a comprehensive safety evaluation index Integrating the deterioration of surrounding rock caused by water seepage and the structural response:

[0068] In the formula, The value is the attenuation coefficient of the elastic modulus of the surrounding rock (taken as 0–1). The structural safety factor (values ​​0–1). , As weight, and ,recommend = 0.4, =0.6. The closer to 1, the safer the overall system; the lower the value, the higher the risk.

[0069] ⑤ Predictive risk assessment: Substitute the pore water pressure, seepage rate, cumulative displacement and other predictable state quantities during the prediction period into the corresponding evaluation relationship to calculate the elastic modulus attenuation coefficient of the surrounding rock, the structural state evaluation index and the comprehensive safety evaluation index during the prediction period.

[0070] If the predicted indicators of the digital twin model at any time within the future warning time window meet at least one of the criteria for a Level II warning, the system issues a Level II predictive warning; if at least one of the criteria for a Level III warning is met, the system issues a Level III predictive warning; when the predicted data simultaneously meets the criteria for two or more warning levels (including Level I and Level II, Level II and Level III, Level I and Level III, and Level I, Level II and Level III simultaneously), the system issues a warning based on the highest risk level among all the currently met warning levels.

[0071] ⑥ Scenario simulation and decision support: The digital twin model can simulate the improvement effect of different grouting schemes, drainage strategies or suspension of heavy-load operation on the safety status, and assist operation and maintenance personnel in selecting the optimal handling measures.

[0072] Part VII. Hierarchical Collaborative Early Warning and Maintenance Optimization ① Setting the three-level early warning threshold: Level 1 Warning: Meets at least one of the following conditions: Comprehensive Safety Assessment Index Or, real-time permeation rate ; in To allow the maximum permeation rate; Level 2 Warning: Comprehensive Safety Assessment Index and or structural safety factor Or the microseismic event rate increases by more than 50% for more than one hour compared to the sliding average of the previous 24 hours; Level 3 Warning: Comprehensive Security Assessment Index and ,or ; in, This represents the accumulated microseismic energy over the preceding 24 hours up to time t. The maximum allowable cumulative microseismic energy set for the system; The cumulative displacement of the key points up to time t, To allow the maximum displacement; If the predicted indicators of the digital twin model at any time within the future warning time window meet at least one of the criteria for a Level II warning, the system issues a Level II predictive warning; if at least one of the criteria for a Level III warning is met, the system issues a Level III predictive warning; when the predicted data simultaneously meets the criteria for two or more warning levels (including Level I and Level II, Level II and Level III, Level I and Level III, and Level I, Level II and Level III simultaneously), the system issues a warning based on the highest risk level among all the currently met warning levels.

[0073] ② Collaborative response mechanism: When a Level 1 warning is triggered, voice prompts and encrypted inspections will be activated. When a Level II warning is triggered, an audible and visual alarm is triggered and heavy-load operation is suspended, and grouting or crack sealing is implemented; When a Level 3 warning is triggered, an emergency shutdown, regional lockdown, and comprehensive structural reinforcement will be implemented.

[0074] ③ Dynamic maintenance and optimization: Regularly calibrate sensors (every 3 months) and update digital twin model parameters (every 6 months or automatically triggered based on model error). Dynamically optimize sensor deployment density and early warning thresholds based on microseismic event clustering and seepage paths to achieve self-evolution of the digital twin model.

[0075] The above detailed description is a specific description of one of the feasible embodiments of the present invention. This embodiment is not intended to limit the patent scope of the present invention. All equivalent implementations or modifications that do not depart from the present invention should be included within the scope of the technical solution of the present invention.

[0076] It should be particularly noted that the various embodiments listed in this specification and accompanying drawings are intended to illustrate the technical solutions and advantages of the present invention, and not to limit the scope of protection of the present invention. Without departing from the core ideas and technical effects of the present invention, those skilled in the art can make any form of improvement, substitution, combination, or modification to the structural arrangement, process parameters, material selection, control logic, etc., of the described embodiments; any obvious changes based on the same concept should be considered equivalent solutions of the present invention and should be included within the scope of protection defined by the claims of the present invention. The actual scope of protection of the present invention is determined by the appended claims and should be correctly understood in conjunction with the specification and accompanying drawings.

Claims

1. A method for integrated monitoring of gravity energy storage in abandoned mines based on digital twin models, characterized in that, Includes the following steps: S1: Establish a three-dimensional digital twin model of the underground space and energy storage facilities of the abandoned mine, and assign initial physical and mechanical parameters to the digital twin model; S2: Based on the preliminary analysis of the digital twin model, identify areas prone to seepage and critical load-bearing areas of the structure, deploy seepage field sensors in areas prone to seepage, and deploy structural response sensors in critical load-bearing areas of the structure. S3: Collect seepage field data through seepage field sensors and structural response data through structural response sensors to obtain real-time monitoring data; S4: Using real-time monitoring data, the physical and mechanical parameters of the digital twin model are dynamically updated through a parameter inversion algorithm to obtain the updated digital twin model; S5: Based on the updated digital twin model, perform fluid-structure interaction evolution simulation, embed the mechanical parameter attenuation relationship caused by seepage into the fluid-structure interaction evolution simulation, predict the evolution trend of seepage field and stress field and structural safety status in future time period, and calculate the structural safety factor and comprehensive safety evaluation index to obtain the prediction results. S6: Based on the prediction results, determine whether a tiered early warning is triggered and execute corresponding coordinated response measures.

2. The method according to claim 1, characterized in that, In step S1, the geometric dimensions of the three-dimensional digital twin model are established at a 1:1 scale according to the actual underground space and energy storage facilities of the mine. The construction basis of the three-dimensional digital twin model includes geological survey reports, structural as-built drawings and gravity energy storage engineering design data. The initial physical and mechanical parameters include elastic modulus, Poisson's ratio, cohesion, internal friction angle, tensile strength, permeability coefficient, porosity and initial groundwater level.

3. The method according to claim 1, characterized in that, In step S2, the seepage-prone area includes at least one of the following: fault fracture zone, rock fissure development zone, boundary of original water accumulation zone, and dynamic load disturbance sensitive area along the track; the key structural bearing area includes at least one of the following: track support beam connection node, bottom foundation of energy storage silo, steel-concrete composite node, surrounding rock anchoring zone, and key section on the dynamic load transmission path.

4. The method according to claim 1, characterized in that, In step S3, the seepage field sensor collects data at a cycle of 1 to 24 hours, and the seepage rate is calculated by the pressure difference between adjacent measuring points and Darcy's law; in the structural response sensor, the micro-vibration sensor continuously collects waveforms, and the displacement gauge and strain gauge collect displacement and strain data at a set cycle.

5. The method according to claim 1, characterized in that, In step S4, the parameter inversion algorithm uses ensemble Kalman filtering or Bayesian inversion methods to assimilate the observed data into the digital twin model. The inversion target parameters include the attenuation coefficients of the permeability coefficient field, elastic modulus field, and shear strength parameters in the seepage-affected zone. The inversion objective function is: ， In the formula, The parameter vector to be inverted, Let be the prior mean vector of the parameters to be inverted. For the orthogonal operator, For observation data, , These are the observation error and the prior parameter covariance matrix, respectively; the theoretical observation values ​​corresponding to the model parameters are calculated by performing fluid-structure interaction simulation using the forward modeling operator. After each inversion, the updated parameter field and its variance are obtained, and the updated parameter field is used as the input for the forward simulation at the next time step.

6. The method according to claim 1, characterized in that, In step S5, the fluid-structure interaction evolution simulation specifically includes: simulating the fluid-structure interaction evolution process over the next 7 to 30 days, and outputting the time series of pore water pressure, permeability, cumulative displacement of key points, microseismic event rate and energy release prediction during the prediction period; The mechanical parameter attenuation relationship includes the attenuation coefficient of the elastic modulus of the surrounding rock, which is defined as: ， in, The elastic modulus of the surrounding rock in a dry state. This is the critical pore water pressure; beyond this value, the effective stress approaches zero. The rock mass sensitivity coefficient is determined by indoor triaxial tests. Let be the elastic modulus of the surrounding rock under water-bearing conditions at time t. Let t be the real-time pore water pressure monitored at time t.

7. The method according to claim 1, characterized in that, In step S5, the structural safety factor is calculated using the following formula: ， in, The cumulative microseismic energy (J) in the 24 hours preceding time t. To allow maximum energy; The cumulative displacement of key points up to time t; To allow the maximum displacement; , As weight, and ; The corrosion correction factor is determined based on the corrosion environment level of the mine water, and its value range is (0, 1.0).

8. The method according to claim 1, characterized in that, In step S5, the comprehensive safety evaluation index is calculated using the following formula: ， in, The elastic modulus attenuation coefficient of the surrounding rock. For structural safety factor, , As weight, and .

9. The method according to claim 1, characterized in that, In step S6, the tiered early warning system includes Level 1, Level 2, and Level 3 early warning systems: Level 1 Warning: Meets at least one of the following conditions: Comprehensive Safety Assessment Index ; or real-time permeation rate ; in To allow the maximum permeation rate; Level 2 Warning: Comprehensive Safety Assessment Index and or structural safety factor Or the microseismic event rate increases by more than 50% for more than one hour compared to the sliding average of the previous 24 hours; Level 3 Warning: Comprehensive Security Assessment Index and ,or ; in, This represents the accumulated microseismic energy over the preceding 24 hours up to time t. The maximum allowable cumulative microseismic energy set for the system; The cumulative displacement of the key points up to time t, To allow the maximum displacement; If the predicted indicators of the digital twin model at any time within the future warning time window meet at least one of the criteria for a Level 2 warning, the system issues a Level 2 predictive warning; if at least one of the criteria for a Level 3 warning is met, the system issues a Level 3 predictive warning; when the predicted data simultaneously meets the criteria for two or more warning levels, the system issues a warning based on the highest risk level among all the currently met warning levels.

10. The method according to claim 9, characterized in that, In step S6, the coordinated response measures include: When a Level 1 warning is triggered, voice prompts and encrypted inspections will be activated. When a Level II warning is triggered, an audible and visual alarm is triggered and heavy-load operation is suspended, and grouting or crack sealing is implemented; When a Level 3 warning is triggered, an emergency shutdown, regional lockdown, and comprehensive structural reinforcement will be implemented.

Images