A deformation monitoring system and method for aquifer energy storage surfaces

Abstract

This invention discloses a deformation monitoring system and method for aquifer energy storage surfaces, relating to the field of deformation monitoring and control. The invention processes the current surface into zones based on a model and simultaneously collects geological environmental parameters. The constructed correlation condition set can accurately output the deformation sensitivity level of each zone, significantly improving the comprehensiveness and accuracy of risk identification. When the proportion of medium-to-high sensitivity areas exceeds a threshold, the system automatically configures maintenance frequency, inspection frequency, and sensor parameters, implements maintenance, and records the time and cost until the proportion of medium-to-high sensitivity areas drops below the threshold, effectively avoiding resource waste and ensuring timely maintenance. When the current maintenance time or cost exceeds the threshold, a micro-adjustment mechanism is triggered. This mechanism iteratively optimizes sensor parameters, maintenance frequency, and inspection frequency, and re-inputs the adjusted data into the model for evaluation. If the conditions are not met, the zones are re-zoned, significantly improving maintenance efficiency and economy.

Description

Technical Field

[0001] This invention belongs to the field of deformation monitoring and control, and specifically relates to a deformation monitoring system and method for aquifer energy storage surfaces. Background Technology

[0002] The current allocation of resources for aquifer surface deformation maintenance lacks a scientific basis, and the failure to delineate sensitive areas often leads to over-investment or under-investment of resources, resulting in low maintenance efficiency and high costs. At the same time, traditional methods are difficult to achieve a dynamic balance between maintenance time and cost, which can easily lead to resource waste or untimely maintenance, affecting the safe and stable operation of aquifer energy storage facilities. Summary of the Invention

[0003] To address the problems in related technologies, this invention proposes a deformation monitoring system and method for aquifer energy storage surfaces, thereby overcoming the aforementioned technical problems in existing related technologies. To solve the above-mentioned technical problems, this invention is achieved through the following technical solution:

[0004] This invention relates to a deformation monitoring method for aquifer energy storage surfaces, comprising the following steps: S1, setting the deformation type and monitoring indicators for aquifer energy storage surfaces, and collecting historical monitoring data based on this; S2, constructing a final aquifer surface deformation type-level mapping model based on the data collected in S1; S3, dividing the current aquifer surface into zones and collecting data, inputting the data into the mapping model in S2; simultaneously collecting geological environmental parameters and constructing a set of associated conditions based on the mapping results, and outputting deformation sensitivity level data for each zone; S4, calculating the initial proportion data based on the sensitivity level data output in S3 and setting maintenance measures based on historical experience; if the proportion... If the ratio exceeds the threshold, configure the maintenance frequency, inspection frequency, and sensor parameters, implement maintenance, and record the time and cost until the proportion of medium-to-high sensitivity areas drops below the threshold; S5, collect historical deformation maintenance data and construct the final deformation maintenance time-cost mapping model; if the current maintenance time or cost exceeds the threshold, execute S6; otherwise, maintenance is complete; S6, set the threshold for the number of micro-adjustments, iteratively adjust the sensor parameters, maintenance frequency, and inspection frequency; input the adjusted data into the mapping model in S5, if the threshold for the number of micro-adjustments is reached and the total maintenance cost or time still does not meet the conditions, re-partition until both maintenance time and cost meet the conditions.

[0005] Preferably, step S1 includes the following steps: S11, setting several types of deformation and deformation degree levels for the aquifer energy storage surface, obtaining a set of energy storage surface deformation types and a set of energy storage surface deformation degree levels; the set of energy storage surface deformation types includes subsidence, landslide, expansion, and cracks; the set of energy storage surface deformation degree levels includes slight, moderate, and severe; then setting several types of indicators for monitoring the deformation of the aquifer energy storage surface, obtaining a set of energy storage surface monitoring indicator types; the set of energy storage surface monitoring indicator types includes three-dimensional coordinate data, real-time displacement data, dynamic deformation data, and macroscopic deformation trends of each monitoring point. S12. Based on the aforementioned energy storage surface monitoring index type set, energy storage surface deformation type set, and energy storage surface deformation degree level set, collect historical data on existing aquifer energy storage, including corresponding energy storage surface monitoring index data, deformation type data, deformation degree level data, and corresponding monitoring time. This yields historical surface monitoring index datasets, historical deformation type datasets, historical deformation degree level datasets, and historical deformation monitoring time datasets. By collecting historical data, data support is provided for the subsequent construction of a mapping model for deformation type and deformation degree level. This mapping model then provides solid decision support for the safe operation of aquifer energy storage facilities.

[0006] Preferably, step S2 includes the following steps: S21, based on the historical surface monitoring index dataset, historical deformation type dataset, historical deformation degree level dataset, and historical deformation monitoring time dataset, construct a mapping model with surface monitoring index data and deformation monitoring time data as inputs and deformation type data and deformation degree level data as outputs, to obtain the final aquifer surface deformation type-level mapping model; by constructing the final aquifer surface deformation type-level mapping model, the transformation from multi-source monitoring data to deformation characteristics is realized, significantly improving the accuracy and timeliness of deformation identification, and automatically associating monitoring indicators with deformation types and levels, reducing the subjective error of manual interpretation.

[0007] Preferably, step S3 includes the following steps: S31, based on the energy storage surface monitoring index type set, the current aquifer surface to be monitored is divided into zones, and the monitoring index type data and corresponding collection time of each zone are collected in real time to obtain the current energy storage surface monitoring index dataset and the current monitoring index collection time dataset; S32, the data corresponding to each zone in the current energy storage surface monitoring index dataset and the current monitoring index collection time dataset are respectively input into the final aquifer surface deformation type-level mapping model for mapping to obtain the current surface deformation type dataset and the current surface deformation degree level dataset; S33, several types of geological structural parameter types and environmental parameter types are set to obtain the geological structural parameter type set and the environmental parameter type set; based on the geological structural parameter... The data types and environmental parameter types are collected simultaneously in S32, including geological structural parameters and environmental parameter data for each zone, to obtain the current geological structural parameter dataset and the current environmental parameter dataset. In S34, a deformation sensitivity correlation condition set is constructed based on the geological structural parameter type set, environmental parameter type set, energy storage surface deformation type set, and energy storage surface deformation degree level set. Based on the deformation sensitivity correlation condition set, the deformation sensitivity level of each zone in S32 is obtained, resulting in the current deformation sensitivity level dataset. The sensitivity level of each zone is automatically determined based on the deformation sensitivity correlation condition set, considering not only the severity of the deformation itself but also incorporating geological stability and environmental dynamics, making the risk assessment more scientific and reasonable, and effectively distinguishing the actual risk differences under different geological conditions.

[0008] Preferably, S4 includes the following steps: S41, setting a threshold for the proportion of medium-to-high sensitivity zones; calculating the proportion of high-sensitivity zones and medium-sensitivity zones in each partition of S31 according to the current deformation sensitivity level dataset, obtaining the current initial proportion of medium-to-high sensitivity zones; based on the current deformation sensitivity level dataset, pre-setting maintenance measures corresponding to each deformation sensitivity level based on historical experience, obtaining a set of deformation sensitivity maintenance measures; if the current initial proportion of medium-to-high sensitivity zones is greater than or equal to the threshold for the proportion of medium-to-high sensitivity zones, executing S42; otherwise, no maintenance is required; S42, based on the set of deformation sensitivity maintenance measures, setting the initial implementation frequency of each deformation sensitivity maintenance measure in the corresponding sensitive zone and the initial manual inspection frequency for each type of sensitive zone, obtaining the current initial maintenance implementation frequency dataset and the current initial manual inspection frequency dataset; further setting the monitoring index type data and geological structural parameters and environmental parameters of each partition of the aquifer surface to be monitored, as well as the accuracy data and deployment density data of various sensors corresponding to each partition, to obtain... The system collects and maps data on the initial deployment density and initial accuracy of current monitoring sensors. Based on the set of deformation sensitivity maintenance measures, the initial frequency of current maintenance implementation, the initial frequency of current maintenance manual inspections, the initial deployment density of current monitoring sensors, and the initial accuracy of current monitoring sensors, it deploys sensors, conducts manual inspections, and implements maintenance measures for each zone in S31. In S43, based on S42, it repeats the data collection and mapping of monitoring index types, classifies each zone into sensitivity levels based on the mapping results, and calculates the proportion of medium-to-high sensitivity areas. It records the time taken and total maintenance cost when the proportion of medium-to-high sensitivity areas on the surface of the aquifer to be monitored first changes to less than the threshold, thus obtaining the current surface deformation maintenance time and total maintenance cost. By cyclically executing the monitoring-evaluation-maintenance process and recording the time and cost data, the system can optimize maintenance strategies in real time and automatically stop unnecessary maintenance when the risk drops below the threshold, thus avoiding resource waste and accumulating quantitative evidence of maintenance efficiency.

[0009] Preferably, step S5 includes the following steps: S51, collecting data on the initial proportion of high-sensitivity areas, the frequency of maintenance measures, the frequency of manual inspections, the density of monitoring sensors, the accuracy of monitoring sensors, the maintenance time, and the total maintenance cost corresponding to several historical surface deformation maintenance operations of the aquifer surface to be monitored, to obtain a historical dataset of the initial proportion of high-sensitivity areas, a historical dataset of the frequency of maintenance measures, a historical dataset of the frequency of manual inspections, a historical dataset of the density of monitoring sensors, a historical dataset of the accuracy of monitoring sensors, a historical dataset of the maintenance time, and a historical dataset of the total maintenance cost; S52, constructing a dataset based on the historical dataset of the initial proportion of high-sensitivity areas, the historical dataset of the frequency of maintenance measures, the historical dataset of the frequency of manual inspections, the historical dataset of the density of monitoring sensors, the historical dataset of the accuracy of monitoring sensors, the historical dataset of the maintenance time, and the historical dataset of the total maintenance cost, with the input being the initial... A mapping model is used to obtain the final deformation maintenance time-cost mapping model, which includes data on the proportion of medium- and high-sensitivity areas, the frequency of maintenance measures, the frequency of manual inspections, the density of monitoring sensors, the accuracy of monitoring sensors, and outputs as maintenance time data and total maintenance cost data. S53: Preset the current surface deformation maintenance time threshold and the current surface deformation maintenance total cost threshold. If the current surface deformation maintenance time data is greater than or equal to the current surface deformation maintenance time threshold or the current surface deformation maintenance total cost data is greater than or equal to the current surface deformation maintenance total cost threshold, execute S6; otherwise, maintenance is complete. The application of this model allows maintenance strategies to be dynamically adjusted based on real-time monitoring data and historical experience. While ensuring risk control effectiveness, it reduces the total cost by optimizing sensor deployment density or adjusting inspection frequency, or shortens the maintenance time in emergency situations by increasing maintenance frequency, thereby achieving an optimal balance between maintenance efficiency, cost control, and risk prevention.

[0010] Preferably, step S6 includes the following steps: S61, setting a threshold for the number of micro-adjustments; repeatedly adjusting the current initial deployment density dataset of monitoring sensors, the current initial accuracy dataset of monitoring sensors, the current initial frequency dataset of maintenance implementation, and the current initial manual inspection frequency dataset of maintenance, obtaining the current adjusted deployment density dataset of monitoring sensors, the current adjusted accuracy dataset of monitoring sensors, the current adjusted frequency dataset of maintenance implementation, and the current adjusted manual inspection frequency dataset of maintenance after each adjustment; S62, inputting the current adjusted deployment density dataset of monitoring sensors, the current adjusted accuracy dataset of monitoring sensors, the current adjusted frequency dataset of maintenance implementation, the current adjusted manual inspection frequency dataset of maintenance, and the current initial high-sensitivity area quantity ratio data into the final deformation maintenance time-cost mapping model to obtain the current adjusted maintenance time data and the current adjusted total maintenance cost data; S63, if the number of repetitions in S62 is less than or equal to the micro-adjustment... If the number of repetitions, the current adjusted maintenance time, and the current total maintenance cost are all less than the current surface deformation maintenance time threshold, the adjustment is complete. Otherwise, return to S31, re-zone the aquifer surface to be monitored, and repeat steps S31, S32, S33, S34, S41, S42, S43, S53, S61, and S62 until the number of repetitions in S62 is less than or equal to the micro-adjustment number of repetitions threshold, the current adjusted maintenance time is less than the current total surface deformation maintenance cost threshold, and the current total maintenance cost is less than the current surface deformation maintenance time threshold. By setting the micro-adjustment number of repetitions threshold, the optimal balance between time and cost is automatically found. If the conditions are not met, a full-process restart mechanism is triggered to re-perform zoning monitoring, deformation assessment, model mapping, and strategy adjustment. This not only significantly improves maintenance efficiency and economy but also provides an intelligent, robust, and practical solution for aquifer surface deformation management.

[0011] A deformation monitoring system for aquifer energy storage surfaces includes a historical aquifer surface deformation monitoring data acquisition module, an aquifer surface deformation type-level mapping model construction module, a current aquifer surface zoning sensitivity level classification module, a current surface deformation maintenance time and cost acquisition module, a current surface deformation maintenance time and cost determination module, and a current surface deformation maintenance parameter and zoning adjustment module.

[0012] The present invention has the following beneficial effects: 1. The present invention realizes intelligent and efficient management of aquifer surface deformation through data acquisition, model construction and dynamic optimization mechanism; based on the model, the current surface is zoned and geological environmental parameters are collected simultaneously. The constructed correlation condition set can accurately output the deformation sensitivity level of each zone, which significantly improves the comprehensiveness and accuracy of risk identification; the initial proportion is calculated by sensitivity level data and maintenance measures are set in combination with historical experience. When the proportion exceeds the threshold, the maintenance frequency, inspection frequency and sensor parameters are automatically configured, maintenance is implemented and the time and cost are recorded until the proportion of medium and high sensitivity areas drops below the threshold, which effectively avoids waste of resources and ensures the timeliness of maintenance; in addition, when the current maintenance time or cost exceeds the threshold, a micro-adjustment mechanism is triggered. The sensor parameters, maintenance frequency and inspection frequency are iteratively optimized, and the adjusted data is re-input into the model for evaluation. When the conditions are not met, the zones are re-zoned, which significantly improves maintenance efficiency and economy, and provides a scientific, intelligent and sustainable solution for aquifer surface deformation management. 2. This invention utilizes a set of deformation-sensitive maintenance measures based on historical experience to design differentiated maintenance schemes for areas with high, medium, and low sensitivity levels, as well as medium-intensity measures such as local reinforcement and shallow-rooted plant planting. Simultaneously, it avoids human interference in low-sensitivity areas, achieving a precise match between resource investment and risk level. By setting parameters such as the initial maintenance implementation frequency, manual inspection frequency, sensor deployment density, and accuracy, a multi-dimensional maintenance system encompassing monitoring equipment deployment, manual inspection, and maintenance measure execution is formed, ensuring that high-risk areas receive more intensive monitoring and more frequent maintenance. 3. This invention uses a mapping model built based on historical datasets to establish a quantitative correlation between input parameters such as the initial proportion of high- and medium-sensitivity areas and the frequency of maintenance measure implementation, and output results such as maintenance time and total cost. It automatically uncovers hidden patterns between parameters, achieving accurate prediction and optimization suggestions for maintenance investment. Of course, any product implementing this invention does not necessarily need to simultaneously achieve all the advantages described above. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of the invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic flowchart of a deformation monitoring method for aquifer energy storage surface according to the present invention.

[0015] Figure 2 A schematic diagram illustrating the process of constructing an aquifer surface deformation type-level mapping model for this invention;

[0016] Figure 3 This is a schematic diagram illustrating the process of classifying the sensitivity levels of the current aquifer surface zones according to the present invention;

[0017] Figure 4 This is a schematic diagram illustrating the process of recording and obtaining the maintenance time and cost of current surface deformation according to the present invention;

[0018] Figure 5 This is a schematic diagram illustrating the process of determining the maintenance time and cost of current surface deformation according to the present invention;

[0019] Figure 6 This is a schematic diagram illustrating the process of adjusting and optimizing the current surface deformation maintenance parameters and zoning according to the present invention;

[0020] Figure 7 This is a schematic diagram of a deformation monitoring system for aquifer energy storage surface according to the present invention. Detailed Implementation

[0021] Many specific details are set forth in the following description to provide a thorough 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. Furthermore, the term "an 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 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 excludes other embodiments.

[0022] Example 1

[0023] Please see Figure 1 This embodiment is a deformation monitoring method for aquifer energy storage surface, including the following steps: S1, setting the deformation type and monitoring index of aquifer energy storage surface, and collecting historical monitoring data based on this;

[0024] Please see Figure 2S1 includes the following steps: S11, setting several deformation types and deformation degree levels for the aquifer energy storage surface to obtain a set of energy storage surface deformation types and a set of energy storage surface deformation degree levels; the set of energy storage surface deformation types includes subsidence, landslide, expansion, and cracks; the set of energy storage surface deformation degree levels includes slight, moderate, and severe; for example, the contents of the set of energy storage surface deformation types and the set of energy storage surface deformation degree levels are as follows: Energy storage surface deformation type set = {"Subsidence": "Surface subsidence", "Landslide": "Slippage of rock and soil along the sliding surface", "Expansion": "Surface uplift", "Cracks": "Cracks appear on the surface"}; Energy storage surface deformation degree level set = {"Slight": {"Threshold": 0.5, "Description": "Deformation rate less than 0.5 mm / day"}, "Moderate": {"Threshold": 1.0, "Description": "Deformation rate 0.5-1.0 mm / day"}, "Severe": {"Threshold": 2.0, "Description": "Deformation rate greater than 1.0 mm / day"}}; Several types of indicators for monitoring surface deformation in aquifer energy storage are then set to obtain a set of energy storage surface monitoring indicator types; This set includes three-dimensional coordinate data of each monitoring point (the three-dimensional spatial coordinates (X, Y, Z axis coordinates) of the monitoring point can be automatically collected using a total station and a spherical prism to calculate the horizontal and vertical displacement changes of the surface; this measurement method can: automatically acquire data periodically, reducing errors introduced by manual operation; and provide millimeter-level precision positioning information), real-time displacement data (continuous displacement changes of the surface in the horizontal and vertical directions), and dynamic deformation data (i.e., the movement trajectory and rate of the monitoring point over time, real-time displacement data, and dynamic deformation data). All of these can be continuously collected through Global Navigation Satellite System (GNSS) monitoring stations, as well as macroscopic deformation trend data (the spatiotemporal distribution characteristics of large-scale surface deformation, which can be obtained using synthetic aperture radar); S12, based on the aforementioned energy storage surface monitoring index type set, energy storage surface deformation type set, and energy storage surface deformation degree level set, collect the existing aquifer energy storage corresponding to the historical energy storage surface monitoring index data, deformation type data, deformation degree level data, and corresponding monitoring time, to obtain the historical surface monitoring index dataset, historical deformation type dataset, historical deformation degree level dataset, and historical deformation monitoring time dataset; for example, the contents of the historical surface monitoring index dataset, historical deformation type dataset, and historical deformation degree level dataset are shown in Table 1 below:

[0025] Table 1. Schematic diagram of historically collected deformation data

[0026]

[0027] By defining the deformation types (subsidence, landslide, expansion, cracking) and deformation severity levels (slight, moderate, severe) of the aquifer energy storage surface, a scientific monitoring classification system was constructed. The monitoring index type set covers three-dimensional coordinate data, real-time displacement data, dynamic deformation data, and macroscopic deformation trend data, ensuring multi-dimensional and high-precision deformation monitoring. Through time-series recording, a comprehensive correlation between deformation type, severity level, and monitoring time was achieved, forming a structured dataset. This enabled full-range monitoring from microscopic deformation to macroscopic trends, providing data support for the subsequent construction of a mapping model for deformation types and severity levels. Ultimately, this mapping model provides solid decision support for the safe operation of aquifer energy storage facilities. This effectively prevents the risk of permanent deformation caused by groundwater extraction or recharge, ensuring the stability and safety of the project; S2, based on the data collected in S1, constructs the final aquifer surface deformation type-level mapping model; S2 includes the following steps: S21, based on the historical surface monitoring index dataset, historical deformation type dataset, historical deformation degree level dataset, and historical deformation monitoring time dataset, constructs a mapping model with surface monitoring index data and deformation monitoring time data as inputs and deformation type data and deformation degree level data as outputs, to obtain the final aquifer surface deformation type-level mapping model; S21 includes the following steps: S211, constructs the initial aquifer surface deformation type - A grade mapping model is established, and a first training data ratio is set (e.g., 8:2 or 7:3, which can be adaptively adjusted according to the actual training situation). Based on the first training data ratio, the historical surface monitoring index dataset, historical deformation type dataset, historical deformation degree grade dataset, and historical deformation monitoring time dataset are divided into a first training dataset and a first test dataset. S212, A first training error threshold is set (10%~15%, which can be adaptively adjusted according to the actual training situation). The first training dataset is input into the initial aquifer surface deformation type-grade mapping model for training. During training, if the training error is less than the first training error threshold... If the training error falls below the first training error threshold, training is stopped, and a well-trained aquifer surface deformation type-level mapping model is obtained; otherwise, training continues until the training error is less than the first training error threshold; S213, a first test accuracy threshold is set (90%~95%, which can be adjusted adaptively according to the actual test situation); the first test dataset is input into the well-trained aquifer surface deformation type-level mapping model for testing; after the test is completed, the first test accuracy data is obtained; if the first test accuracy data is greater than or equal to the first test accuracy threshold, the well-trained aquifer surface deformation type-level mapping model is used as the final aquifer surface deformation type-level mapping model;Otherwise, return to S212 to continue training the trained aquifer surface deformation type-level mapping model and repeat S213 until the first test accuracy data is greater than or equal to the first test accuracy threshold.

[0028] The structure of the initial aquifer surface deformation type-level mapping model can be referenced in Table 2 below:

[0029] Table 2. Schematic diagram of the mapping model for aquifer surface deformation types and levels.

[0030]

[0031] Long Short-Term Memory (LSTM) networks, through a carefully designed gating structure, can effectively capture long-term dependencies, overcoming the gradient vanishing problem of traditional RNNs and maintaining information transmission over long time spans. Specifically designed for sequential data, they can handle variable-length sequence inputs, making them suitable for analyzing dynamic evolution patterns in aquifer energy storage surface deformation monitoring. Furthermore, they can learn temporal patterns and dynamic features in time series, possessing a natural advantage in processing continuous monitoring data. They can identify trend changes and anomalous patterns during deformation processes and support feature extraction at multiple time scales. In aquifer energy storage surface deformation monitoring, LSTM can analyze the evolution trajectory of historical monitoring data, predict future deformation trends, and identify sudden deformation events. This is achieved by integrating historical surface monitoring index datasets, historical deformation type datasets, and historical deformation degrees. By constructing a mapping model using multi-source datasets and historical deformation monitoring time datasets, a model is built that takes surface monitoring index data and deformation monitoring time data as inputs and outputs deformation type data and deformation degree level data, forming the final aquifer surface deformation type-level mapping model. This model realizes the transformation from multi-source monitoring data to deformation characteristics, significantly improving the accuracy and timeliness of deformation identification. It can automatically associate monitoring indicators with deformation types and levels, reducing the subjective errors of manual interpretation. By incorporating time series data, the model can capture the dynamic evolution of deformation, providing a scientific basis for early warning and risk prevention of aquifer energy storage facilities, and effectively supporting engineering safety decisions. In addition, this mapping model has good scalability and can adapt to different geological conditions and monitoring scenarios, laying an intelligent foundation for long-term surface deformation monitoring and disaster prevention.

[0032] S3. Divide the current aquifer surface into zones and collect data, then input the mapping model from S2 for mapping; simultaneously collect geological environmental parameters and, after combining the mapping results, construct a set of associated conditions and output the deformation sensitivity level data for each zone; please refer to [link to relevant documentation]. Figure 3S3 includes the following steps: S31. Based on the energy storage surface monitoring index type set, the current aquifer surface to be monitored is divided into zones, and the monitoring index type data and corresponding collection time of each zone are collected in real time to obtain the current energy storage surface monitoring index dataset and the current monitoring index collection time dataset; S32. The data corresponding to each zone in the current energy storage surface monitoring index dataset and the current monitoring index collection time dataset are respectively input into the final aquifer surface deformation type-level mapping model for mapping to obtain the current surface deformation type dataset and the current surface deformation degree level dataset; S33. Several types of geological structural parameters and environmental parameters are set to obtain geological structural parameters. The geological structure parameter type set includes lithology (such as loose rock layers, medium-sized rock layers, and stable bedrock); the environmental parameter type set includes rainfall, etc. Based on the geological structure parameter type set and the environmental parameter type set, geological structure parameter and environmental parameter data for each partition in S32 are collected synchronously to obtain the current geological structure parameter dataset and the current environmental parameter dataset; S34, a deformation sensitivity correlation condition set is constructed based on the geological structure parameter type set, the environmental parameter type set, the energy storage surface deformation type set, and the energy storage surface deformation degree level set; based on the deformation sensitivity correlation condition set, the deformation sensitivity level of each partition in S32 is obtained to obtain the current deformation sensitivity... A deformation sensitivity dataset is provided; for example, the contents of the deformation sensitivity correlation condition set can be referenced as follows: High sensitivity area: severe deformation + loose rock layer + high rainfall; Medium sensitivity area: medium deformation + medium rock layer stability + medium rainfall; Low sensitivity area: slight deformation + stable bedrock + low rainfall; wherein, the stability of the rock layer and the magnitude of rainfall can be adaptively determined according to the actual situation; by integrating energy storage surface monitoring indicators, geological structural parameters and environmental parameters, a deformation sensitivity assessment system is constructed; specifically, by collecting monitoring indicators in different areas and inputting them into the deformation type-level mapping model, accurate identification and quantitative classification of aquifer surface deformation are achieved, providing a basis for subsequent risk assessment. It provides basic data support; secondly, it simultaneously collects geological structural parameters (such as lithology and fault distribution) and environmental parameters (such as rainfall and groundwater level), and combines deformation type and level data to construct a multi-dimensional deformation sensitivity condition set, which significantly improves the comprehensiveness and accuracy of risk identification; thirdly, it automatically determines the sensitivity level (high, medium, and low) of each zone based on the deformation sensitivity correlation condition set, which not only considers the severity of the deformation itself, but also incorporates geological stability and environmental dynamic factors, making the risk assessment more scientific and reasonable, and effectively distinguishing the actual risk differences under different geological conditions; in addition, by adaptively determining rock layer stability and rainfall level, it avoids the misjudgment problem caused by fixed thresholds in traditional methods.Finally, the resulting deformation sensitivity level dataset can be directly used to guide engineering site selection, disaster early warning and resource allocation, providing a basis for decision-making for the safe operation and long-term management of aquifer energy storage projects, while reducing the risk of secondary disasters caused by deformation, and has significant economic and social benefits.

[0033] S4. Calculate the initial percentage data based on the sensitivity level data output in S3 and set maintenance measures according to historical experience; if the percentage exceeds the threshold, configure the maintenance frequency, inspection frequency, and sensor parameters, implement maintenance, and record the time and cost, until the percentage of medium-to-high sensitivity areas drops below the threshold; please refer to Figure 4 S4 includes the following steps: S41, setting a threshold for the proportion of medium-to-high sensitivity areas (which can be adaptively set according to actual monitoring requirements); calculating the proportion of high-sensitivity areas and medium-sensitivity areas in each partition of S31 according to the current deformation sensitivity level dataset, to obtain the current initial proportion of medium-to-high sensitivity areas; based on the current deformation sensitivity level dataset, pre-setting maintenance measures corresponding to each deformation sensitivity level based on historical experience, to obtain a set of deformation sensitivity maintenance measures; if the current initial proportion of medium-to-high sensitivity areas is greater than or equal to the threshold for the proportion of medium-to-high sensitivity areas, executing S42; otherwise, no maintenance is required.

[0034] For example, the contents of the deformation sensitivity maintenance measures set can be referred to as Table 3 below:

[0035] Table 3: Schematic Diagram of Deformation Sensitivity Maintenance Measures

[0036]

[0037] S42. Based on the set of deformation-sensitive maintenance measures, set the initial implementation frequency of each deformation-sensitive maintenance measure in the corresponding sensitive area and the initial manual inspection frequency for each type of sensitive area to obtain the current maintenance implementation initial frequency dataset and the current maintenance initial manual inspection frequency dataset; then set the monitoring index type data, geological structural parameters, and environmental parameters of each zone corresponding to the monitoring surface of the aquifer to be monitored in S31 and S33, and the accuracy data and deployment density data of various types of sensors corresponding to each zone to obtain the current monitoring sensor initial deployment density dataset and the current monitoring sensor initial accuracy dataset. Based on the deformation sensitivity maintenance measures set, the current maintenance implementation initial frequency dataset, the current maintenance initial manual inspection frequency dataset, the current monitoring sensor initial deployment density dataset, and the current monitoring sensor initial accuracy dataset, sensor deployment, manual inspection, and maintenance measures are implemented for each zone in S31; S43, based on S42, the monitoring index type data collection and mapping are repeated, and the sensitivity level of each zone is divided according to the mapping results, and the proportion of medium-high sensitivity areas is calculated. The first time the proportion of medium-high sensitivity areas on the surface of the aquifer to be monitored changes to less than that of medium-high sensitivity areas, the data is recorded. By analyzing the time consumption data corresponding to the quantity percentage threshold and the total maintenance cost data (including sensor deployment costs, human resource costs, and maintenance implementation costs), the current surface deformation maintenance time data and the current surface deformation maintenance total cost data are obtained. Through the integration of a dynamic threshold triggering mechanism and a tiered maintenance strategy, intelligent optimization of aquifer surface deformation monitoring is achieved. Specifically, firstly, by setting a quantity percentage threshold for medium- and high-sensitivity areas and calculating the initial percentage data in real time, an automatic decision-making mechanism based on risk level is constructed. When the risk in the monitored area exceeds the preset threshold, the system automatically triggers the maintenance process, significantly improving the timeliness and accuracy of risk response. Secondly, based on historical experience, a set of deformation-sensitive maintenance measures was established. Differentiated maintenance schemes, such as grouting support, drainage facility construction, and vegetation restoration, were designed for areas with different sensitivity levels (high, medium, and low), as well as medium-intensity measures such as local reinforcement and shallow-rooted plant planting. At the same time, human interference in low-sensitivity areas was avoided, achieving a precise match between resource input and risk level. Furthermore, by setting parameters such as the initial frequency of maintenance implementation, the frequency of manual inspections, and the density and accuracy of sensor deployment, a multi-dimensional maintenance system was formed, including the deployment of monitoring equipment, manual inspections, and the execution of maintenance measures, ensuring that high-risk areas receive more intensive monitoring and more frequent maintenance.Finally, by cyclically executing the monitoring-evaluation-maintenance process and recording time and cost data, the system can optimize maintenance strategies in real time and automatically stop unnecessary maintenance when the risk drops below the threshold. This not only avoids waste of resources but also accumulates quantitative evidence of maintenance efficiency, providing data support for subsequent strategy adjustments. It realizes intelligent, precise, and economical management of aquifer surface deformation, effectively improving the efficiency and sustainability of geological disaster prevention and control.

[0038] S5. Collect historical deformation maintenance data and construct the final deformation maintenance time-cost mapping model; if the current maintenance time or cost exceeds the threshold, execute S6; otherwise, maintenance is complete; please refer to [link / reference]. Figure 5S5 includes the following steps: S51, collecting data on the initial proportion of high-sensitivity areas, the frequency of maintenance measures, the frequency of manual inspections, the density of monitoring sensors, the accuracy of monitoring sensors, the maintenance time, and the total maintenance cost of the aquifer surface to be monitored during several historical surface deformation maintenance operations, to obtain a historical dataset of the initial proportion of high-sensitivity areas, a historical dataset of the frequency of maintenance measures, a historical dataset of the frequency of manual inspections, a historical dataset of the density of monitoring sensors, a historical dataset of the accuracy of monitoring sensors, a historical dataset of the maintenance time, and a historical dataset of the total maintenance cost; S52, based on the historical proportion of high-sensitivity areas... The datasets include historical maintenance implementation frequency datasets, historical manual inspection frequency datasets, historical monitoring sensor deployment density datasets, historical monitoring sensor accuracy datasets, historical maintenance time datasets, and historical total maintenance cost datasets. A mapping model is constructed with the initial proportion of high-sensitivity areas, maintenance implementation frequency data, manual inspection frequency data, monitoring sensor deployment density data, and monitoring sensor accuracy data as inputs, and maintenance time data and total maintenance cost data as outputs, to obtain the final deformation maintenance time-cost mapping model. Step S52 includes the following steps: S521, Constructing the initial deformation maintenance time-cost mapping model and setting a second training data ratio (e.g., 8:2 or 7:3). (Specific adjustments can be made adaptively based on actual training conditions); Based on the second training data ratio, the datasets are divided into the following datasets: historical initial high-sensitivity area quantity ratio, historical maintenance measure implementation frequency, historical manual inspection frequency, historical monitoring sensor deployment density, historical monitoring sensor accuracy, historical maintenance time, and historical maintenance total cost, resulting in the second training dataset and the second test dataset; S522, Set the second training error threshold (10%~15%, which can be adjusted adaptively based on actual training conditions); Input the second training dataset into the initial deformation maintenance time-cost mapping model for training; During training, if If the training error is less than the second training error threshold, training stops, and a trained deformation maintenance time-cost mapping model is obtained; otherwise, training continues until the training error is less than the second training error threshold; S523, set a second test accuracy threshold (90%~95%, which can be adjusted adaptively according to the actual test situation); input the second test dataset into the trained deformation maintenance time-cost mapping model for testing; after the test is completed, the second test accuracy data is obtained; if the second test accuracy data is greater than or equal to the second test accuracy threshold, the trained deformation maintenance time-cost mapping model is used as the final deformation maintenance time-cost mapping model;Otherwise, return to S522 to continue training the trained deformation maintenance time-cost mapping model and repeat S523 until the second test accuracy data is greater than or equal to the second test accuracy threshold.

[0039] The structure of the initial deformation maintenance time-cost mapping model can be seen in Table 4 below:

[0040] Table 4: Schematic diagram of the deformation maintenance time-cost mapping model.

[0041]

[0042] Fully connected structures are suitable for handling linear and nonlinear relationships of numerical features; the ReLU activation function is used to accelerate convergence and avoid gradient vanishing; Dropout regularization is used to enhance the model's generalization ability; the Adam optimizer is used for adaptive learning rate adjustment to improve training efficiency; MSE loss is used to ensure numerical consistency between predicted results and actual values; S53, preset the current surface deformation maintenance time threshold and the current surface deformation maintenance total cost threshold (which can be adaptively set based on actual maintenance investment); if the current surface deformation maintenance time data is greater than or equal to the current surface deformation maintenance time threshold or the current surface deformation maintenance total cost data is greater than or equal to the current surface deformation maintenance total cost threshold, execute S6; otherwise, maintenance is completed; by constructing a maintenance time-cost mapping model based on historical data, a foundation is laid for subsequent intelligent optimization and dynamic control of the aquifer surface deformation maintenance process; specifically, by preset maintenance time and total cost thresholds and monitoring current data in real time, an automatic triggering mechanism based on cost-effectiveness is established, when maintenance investment exceeds a reasonable range, The automatic initiation of historical data analysis effectively avoids resource waste and over-maintenance, significantly improving the economic efficiency of maintenance management. Secondly, by collecting key parameters such as the initial proportion of high-sensitivity areas, frequency of maintenance measures, frequency of manual inspections, and sensor deployment density and accuracy from several historical maintenance tasks, a multi-dimensional historical dataset is constructed. This provides rich real-world data support for model training, enhancing the model's generalization ability and prediction accuracy. Thirdly, the mapping model built based on the historical dataset can establish a quantitative correlation between input parameters such as the initial proportion of high-sensitivity areas and the frequency of maintenance measures and the output results of maintenance time and total cost, automatically uncovering hidden patterns between parameters and achieving accurate prediction and optimization suggestions for maintenance investment. Finally, the application of this model enables maintenance strategies to be dynamically adjusted based on real-time monitoring data and historical experience. For example, while ensuring risk control effectiveness, the total cost can be reduced by optimizing sensor deployment density or adjusting inspection frequency, or the maintenance time can be shortened by increasing maintenance frequency in emergency situations, thereby achieving an optimal balance between maintenance efficiency, cost control, and risk prevention.

[0043] S6. Set a threshold for the number of micro-adjustments, and iteratively adjust sensor parameters, maintenance frequency, and inspection frequency; input the adjusted data into the mapping model in S5. If the threshold for the number of micro-adjustments is reached but the total maintenance cost or time still does not meet the conditions, re-partition the system until both maintenance time and cost meet the conditions; please refer to [link to relevant documentation]. Figure 6 S6 includes the following steps: S61, setting a threshold for the number of micro-adjustments (which can be adaptively set according to actual conditions); repeatedly adjusting the current initial deployment density dataset of monitoring sensors, the current initial accuracy dataset of monitoring sensors, the current initial frequency dataset of maintenance implementation, and the current initial manual inspection frequency dataset of maintenance, obtaining the current adjusted deployment density dataset of monitoring sensors, the current adjusted accuracy dataset of monitoring sensors, the current adjusted frequency dataset of maintenance implementation, and the current adjusted manual inspection frequency dataset of maintenance after each adjustment; S62, inputting the current adjusted deployment density dataset of monitoring sensors, the current adjusted accuracy dataset of monitoring sensors, the current adjusted frequency dataset of maintenance implementation, the current adjusted manual inspection frequency dataset of maintenance, and the current initial high-sensitivity area quantity ratio data into the final... The deformation maintenance time-cost mapping model is used to obtain the current adjusted maintenance time data and the current adjusted total maintenance cost data. S63: If the number of repetitions in S62 is less than or equal to the micro-adjustment number threshold, the current adjusted maintenance time data is less than the current total surface deformation maintenance cost threshold, and the current adjusted total maintenance cost data is less than the current surface deformation maintenance time threshold, the adjustment is complete. Otherwise, return to S31, re-partition the current aquifer surface to be monitored, and repeat S31, S32, S33, S34, S41, S42, S43, S53, S61, and S62 until the number of repetitions in S62 is less than or equal to the micro-adjustment number threshold, the current adjusted maintenance time data is less than the current total surface deformation maintenance cost threshold, and the current adjusted total maintenance cost data is less than the current surface deformation maintenance time threshold.

[0044] For example, as follows: A surface monitoring area for aquifer energy storage is divided into several zones, each approximately 1 square kilometer; monitoring index data and collection time for each zone are collected in real time and mapped; geological structural parameters of the zone are collected simultaneously, revealing loose rock strata with a relatively dense distribution of faults; environmental parameters show high rainfall and a high groundwater level; a deformation sensitivity correlation condition set is constructed, setting a threshold of 30% for the proportion of medium-to-high sensitivity zones; the initial proportion of medium-to-high sensitivity zones in each zone is calculated, currently at 35%, exceeding the threshold; based on historical experience, maintenance measures for high-sensitivity zones are set. To stabilize the soil and rock mass using grouting technology, the grouting pressure is 0.5-0.8 MPa; intercepting ditches and blind drains are installed, with the intercepting ditches being 0.5 m deep and the blind drains 0.3 m wide; deep-rooted shrubs and trees are planted, with a shrub planting density of 3 trees / square meter and a tree planting spacing of 5 m; a geomembrane is used for covering, with a geomembrane thickness of 0.5 mm, forming a set of deformation-sensitive maintenance measures; the initial implementation frequency of grouting measures in high-sensitivity areas is set to twice a week, and the frequency of manual inspection is set to once a day; the accuracy of the sensors used to collect monitoring index data is set to ±1 mm for displacement sensors and ±0 mm for dynamic deformation measuring instruments.The displacement sensor was deployed at a density of 1 unit per 100m, and the dynamic deformation measuring instrument at a density of 1 unit per 200m. The sensors used to collect geological structural and environmental parameters had an accuracy of ±5% for the lithology detector and ±2mm for the rainfall sensor, with a deployment density of 1 unit per 500m for the lithology detector and 1 unit per 1km for the rainfall sensor. Based on these data, sensors were deployed within the designated area, and manual inspections and maintenance measures were implemented. Data collection and mapping of monitoring index types were repeated. After mapping, the deformation type remained subsidence, the deformation degree level changed to medium, and the sensitivity level changed to medium-sensitive zone. The recalculated percentage of medium-to-high sensitivity zones was 28%, which was less than the threshold for medium-to-high sensitivity zones for the first time. The maintenance time was recorded as 15 days, and the total maintenance cost was 1.2 million yuan (including 300,000 yuan for sensor deployment, 500,000 yuan for human resources, and 400,000 yuan for maintenance measures). Data on the surface of the aquifer currently under monitoring was collected historically. The percentages of initial high-sensitivity areas during each surface deformation maintenance were 38%, 42%, 35%, 40%, and 33%, respectively. The frequency of maintenance measures was 3 times per week, 2 times per week, 2 times per week, 3 times per week, and 2 times per week, respectively. The frequency of manual inspections was 2 times per day, 1 time per day, 1 time per day, 2 times per day, and 1 time per day, respectively. The density of monitoring sensors was 1 displacement sensor per 80m, 1 per 100m, and 1 per 100m, respectively. One device is placed at each of the following locations: one every 80m and one every 100m. One dynamic deformation measuring instrument is placed at each of the following locations: one every 150m, one every 200m, one every 200m, one every 150m, and one every 200m. The accuracy data for the monitoring sensors are: displacement sensor accuracy ±1.2mm, ±1mm, ±1mm, ±1.2mm, ±1mm; dynamic deformation measuring instrument accuracy ±0.6mm, ±0.5mm, ±0.5mm, ±0.6mm, ±0.The maintenance time data for 5mm deformation is 18 days, 15 days, 15 days, 17 days, and 15 days respectively, and the total maintenance cost data is 1.5 million yuan, 1.2 million yuan, 1.2 million yuan, 1.4 million yuan, and 1.2 million yuan respectively. The preset thresholds for current surface deformation maintenance time are 18 days and the current thresholds for total surface deformation maintenance cost are 1.3 million yuan. If the current surface deformation maintenance time data is 15 days (less than the current surface deformation maintenance time threshold) and the current surface deformation maintenance cost data is 1.2 million yuan (less than the current surface deformation maintenance cost threshold), maintenance is completed. If the current surface deformation maintenance time data is 20 days (more than the current surface deformation maintenance time threshold), micro-adjustment is set. The threshold for the number of adjustments is 3. Adjustments are needed. After the first adjustment, the density of displacement sensors will be adjusted to 1 per 80m, the density of dynamic deformation measuring instruments will be adjusted to 1 per 150m, the maintenance frequency will be adjusted to 3 times per week, and the frequency of manual inspections will be adjusted to 2 times per day. The adjusted data will be input into the final deformation maintenance time-cost mapping model. The current adjusted maintenance time is 18 days, the current adjusted total maintenance cost is 1.35 million yuan, and the number of repetitions is 1, which is less than the threshold for the number of micro-adjustments. However, the current adjusted total maintenance cost of 1.35 million yuan is still greater than the current threshold for the total cost of surface deformation maintenance of 1.3 million yuan. Adjustments will continue. After the second adjustment, the displacement sensor deployment density was adjusted to one per 90m, the dynamic deformation measuring instrument deployment density was adjusted to one per 180m, the maintenance frequency was adjusted to 2.5 times per week, and the manual inspection frequency was adjusted to 1.5 times per day. The adjusted maintenance time was 16 days, the total maintenance cost was 1.15 million yuan, and the number of repetitions was 2, which is less than the micro-adjustment number threshold. Furthermore, the current adjusted maintenance time of 16 days is less than the current surface deformation maintenance time threshold of 18 days, and the current adjusted total maintenance cost of 1.15 million yuan is less than the current total surface deformation maintenance cost threshold of 1.3 million yuan. The adjustment is complete.

[0045] By setting thresholds for the number of micro-adjustments and establishing an iterative optimization process for sensor deployment density, accuracy, maintenance frequency, and inspection frequency, the system can automatically find the optimal balance between time consumption and cost. After each adjustment, the updated parameters and the initial data on the proportion of high-sensitivity areas are input into the deformation maintenance time-cost mapping model to achieve real-time dynamic calibration of the maintenance strategy. When the adjusted maintenance time is lower than the cost threshold and the adjusted maintenance cost is lower than the time threshold, the optimization is automatically terminated to ensure optimal maintenance results under resource constraints. If the conditions are not met, a full-process restart mechanism is triggered to re-perform zonal monitoring, deformation assessment, model mapping, and strategy adjustment. This not only significantly improves maintenance efficiency and economy but also effectively responds to changes in geological conditions and sudden risks through adaptive adjustment capabilities, providing an intelligent, robust, and practical solution for aquifer surface deformation management.

[0046] Example 2

[0047] Please see Figure 7 This embodiment discloses a deformation monitoring system for aquifer energy storage surfaces. The system implements the methods described in the above embodiments, including a historical aquifer surface deformation monitoring data acquisition module, an aquifer surface deformation type-level mapping model construction module, a current aquifer surface zoning sensitivity level classification module, a current surface deformation maintenance time and cost acquisition module, a current surface deformation maintenance time and cost determination module, and a current surface deformation maintenance parameter and zoning adjustment module. The historical aquifer surface deformation monitoring data acquisition module sets the aquifer energy storage surface deformation type and monitoring indicators, and collects historical monitoring data based on these. The aquifer surface deformation type-level mapping model construction module constructs the final aquifer surface deformation type-level mapping model based on the data collected by the historical aquifer surface deformation monitoring data acquisition module. The current aquifer surface zoning sensitivity level classification module divides the current aquifer surface into zones and collects data, inputting this data into the mapping model in the aquifer surface deformation type-level mapping model construction module. Geological environmental parameters are simultaneously collected and associated conditions are constructed based on the mapping results. The system outputs deformation sensitivity level data for each zone. The current surface deformation maintenance time and cost acquisition module calculates the initial proportion data based on the sensitivity level data output from the current aquifer surface zone sensitivity level classification module and sets maintenance measures based on historical experience. If the proportion exceeds a threshold, the system configures maintenance frequency, inspection frequency, and sensor parameters, implements maintenance, and records the time and cost until the proportion of medium-to-high sensitivity zones drops below the threshold. The current surface deformation maintenance time and cost determination module collects historical deformation maintenance data and constructs a final deformation maintenance time-cost mapping model. If the current maintenance time or cost exceeds the threshold, the current surface deformation maintenance parameter and zone adjustment module is executed; otherwise, maintenance is complete. The current surface deformation maintenance parameter and zone adjustment module sets a threshold for the number of micro-adjustments and iteratively adjusts sensor parameters, maintenance frequency, and inspection frequency. The adjusted data is input into the mapping model in the current surface deformation maintenance time and cost determination module. If the micro-adjustment number threshold is reached and the total maintenance cost or time still does not meet the conditions, the zone is re-divided until both maintenance time and cost meet the conditions.

[0048] In the description of this specification, the references to terms such as "an embodiment," "example," and "specific example" indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described can be combined in any suitable manner in one or more embodiments or examples. It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all described as a series of actions; however, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to the present invention.

Claims

1. A method for monitoring surface deformation in aquifer energy storage, characterized in that, Includes the following steps: S1. Define the surface deformation type and monitoring indicators for aquifer energy storage, and collect historical monitoring data based on this; S2. Construct the final aquifer surface deformation type-level mapping model based on the data collected in S1; S3. Divide the current aquifer surface into zones and collect data, input the mapping model in S2 for mapping; synchronously collect geological environmental parameters and combine the mapping results to construct the associated condition set, then output the deformation sensitivity level data of each zone to obtain the current deformation sensitivity level dataset; S4. Calculate the initial percentage data based on the sensitivity level data output in S3 and set maintenance measures based on historical experience; if the percentage exceeds the threshold, configure the maintenance frequency, inspection frequency and sensor parameters, implement maintenance and record the time and cost, until the percentage of medium and high sensitivity areas drops below the threshold. Specifically, it includes: S41, setting a preset set of deformation sensitivity maintenance measures and setting a threshold for the proportion of medium and high sensitive areas; based on the current deformation sensitivity level dataset, calculating the proportion of high and medium sensitive areas in each partition to obtain the current initial proportion of medium and high sensitive areas. S5. Collect historical deformation maintenance data and construct the final deformation maintenance time-cost mapping model; if the current maintenance time or cost exceeds the threshold, execute S6; otherwise, maintenance is complete. S6. Set a threshold for the number of micro-adjustments, and iteratively adjust the sensor parameters, maintenance frequency, and inspection frequency. Input the adjusted data into the mapping model in S5. If the threshold for the number of micro-adjustments is reached and the total maintenance cost or time still does not meet the conditions, re-partition the system until both maintenance time and cost meet the conditions.

2. The deformation monitoring method for aquifer energy storage surface according to claim 1, characterized in that, S1 includes the following steps: S11. Define several types of deformation and deformation severity levels for the surface of aquifer energy storage; deformation types include subsidence, landslide, expansion, and cracks; deformation severity levels include slight, moderate, and severe; further define several types of indicators for monitoring surface deformation of aquifer energy storage, including three-dimensional coordinate data, real-time displacement data, dynamic deformation data, and macroscopic deformation trend data for each monitoring point; S12. Based on S11, collect data on the surface monitoring indicators, deformation types, deformation levels, and corresponding monitoring times of existing aquifer energy storage in history.

3. The deformation monitoring method for aquifer energy storage surface according to claim 2, characterized in that, S2 includes the following steps: S21. Based on the historical data collected in S12, construct a mapping model with surface monitoring index data and deformation monitoring time data as inputs and deformation type data and deformation degree level data as outputs, to obtain the final aquifer surface deformation type-level mapping model.

4. The deformation monitoring method for aquifer energy storage surface according to claim 3, characterized in that, S3 includes the following steps: S31. Based on the energy storage surface monitoring index type set, the current aquifer surface to be monitored is divided into zones. The monitoring index type data and the corresponding collection time of each zone are collected in real time and respectively input into the final aquifer surface deformation type-level mapping model to obtain the current surface deformation type dataset and the current surface deformation degree level dataset. S32. Synchronously collect geological structural parameters and environmental parameter data of each partition in S31 to obtain the current geological structural parameter dataset and the current environmental parameter dataset; S33. Construct a set of deformation sensitivity association conditions and obtain the deformation sensitivity level of each partition in S31 to obtain the current deformation sensitivity level dataset.

5. A method for deformation monitoring of aquifer energy storage surface according to claim 4, characterized in that, S4 further includes the following steps: If the current initial proportion of medium-to-high sensitivity areas is greater than or equal to the threshold for the proportion of medium-to-high sensitivity areas, execute S42; otherwise, no maintenance is required. S42. Set the initial implementation frequency of each deformation-sensitive maintenance measure in the corresponding sensitive area and the initial manual inspection frequency for each type of sensitive area to obtain the current maintenance implementation initial frequency dataset and the current maintenance initial manual inspection frequency dataset; then set the monitoring index type data, geological structure parameters and environmental parameters of each zone corresponding to the monitoring surface of the aquifer to be monitored in S31 and S32, and the accuracy data and deployment density data of various sensors corresponding to each zone to obtain the current monitoring sensor initial deployment density dataset and the current monitoring sensor initial accuracy dataset. Then, sensors are deployed, manual inspections are conducted, and maintenance measures are implemented in each of the S31 zones. S43. Based on S42, repeat the data collection and mapping of monitoring index types, classify the sensitivity level of each zone according to the mapping results, and calculate the proportion of medium and high sensitive areas. Record the time data and total maintenance cost data corresponding to the first time the proportion of medium and high sensitive areas on the surface of the aquifer to be monitored changes to less than the threshold of the proportion of medium and high sensitive areas. Obtain the current surface deformation maintenance time data and the current surface deformation maintenance total cost data.

6. A method for deformation monitoring of aquifer energy storage surface according to claim 5, characterized in that, S5 includes the following steps: S51. Collect data on the percentage of initial medium-to-high sensitivity zones, the frequency of maintenance measures, the frequency of manual inspections, the density of monitoring sensors, the accuracy of monitoring sensors, the maintenance time, and the total maintenance cost of the aquifer surface to be monitored during several historical surface deformation maintenance operations. S52. Based on the historical data collected in S51, construct a mapping model with the inputs being the initial proportion of high-sensitivity areas, the frequency of maintenance measures, the frequency of manual inspections, the density of monitoring sensors, and the accuracy of monitoring sensors, and the outputs being maintenance time and total maintenance cost data, to obtain the final deformation maintenance time-cost mapping model.

7. A method for monitoring deformation of aquifer energy storage surface according to claim 6, characterized in that, S5 further includes the following steps: S53. Preset the current surface deformation maintenance time threshold and the current surface deformation maintenance total cost threshold; if the current surface deformation maintenance time data is greater than or equal to the current surface deformation maintenance time threshold or the current surface deformation maintenance total cost data is greater than or equal to the current surface deformation maintenance total cost threshold, execute S6; otherwise, maintenance is completed.

8. A method for deformation monitoring of aquifer energy storage surface according to claim 7, characterized in that, S6 includes the following steps: S61. Set a threshold for the number of micro-adjustments; repeatedly adjust the current monitoring sensor initial deployment density dataset, the current monitoring sensor initial accuracy dataset, the current maintenance implementation initial frequency dataset, and the current maintenance initial manual inspection frequency dataset. After each adjustment, obtain the current monitoring sensor adjusted deployment density dataset, the current monitoring sensor adjusted accuracy dataset, the current maintenance implementation adjusted frequency dataset, and the current maintenance adjusted manual inspection frequency dataset.

9. A method for monitoring deformation of aquifer energy storage surface according to claim 8, characterized in that, S6 further includes the following steps: S62. Input the current monitoring sensor deployment density dataset after adjustment, the current monitoring sensor accuracy dataset after adjustment, the current maintenance implementation frequency dataset after adjustment, the current maintenance manual inspection frequency dataset after adjustment, and the current initial high-sensitivity area quantity ratio data into the final deformation maintenance time-cost mapping model to obtain the current adjusted maintenance time data and the current adjusted total maintenance cost data. S63. If the number of repetitions in S62 is less than or equal to the micro-adjustment number threshold, the current adjusted maintenance time is less than the current total surface deformation maintenance cost threshold, and the current adjusted total maintenance cost is less than the current surface deformation maintenance time threshold, the adjustment is complete; otherwise, return to S31, re-zone the current aquifer surface to be monitored, and repeat S31, S32, S33, S41, S42, S43, S53, S61, and S62 until the number of repetitions in S62 is less than or equal to the micro-adjustment number threshold, the current adjusted maintenance time is less than the current total surface deformation maintenance cost threshold, and the current adjusted total maintenance cost is less than the current surface deformation maintenance time threshold.

10. A system for implementing the deformation monitoring method for aquifer energy storage surfaces as described in any one of claims 1-9, characterized in that: It includes a historical aquifer surface deformation monitoring data acquisition module, an aquifer surface deformation type-level mapping model construction module, a current aquifer surface zoning sensitivity level classification module, a current surface deformation maintenance time and cost acquisition module, a current surface deformation maintenance time and cost determination module, and a current surface deformation maintenance parameter and zoning adjustment module. The historical aquifer surface deformation monitoring data acquisition module sets the aquifer energy storage surface deformation type and monitoring indicators, and collects historical monitoring data based on these. The aquifer surface deformation type-level mapping model construction module constructs the final aquifer surface deformation type-level mapping model based on the data collected by the historical aquifer surface deformation monitoring data acquisition module. The current aquifer surface zoning sensitivity level classification module divides the current aquifer surface into zones and collects data. It then inputs the data into the aquifer surface deformation type-level mapping model construction module to map the data. Simultaneously, it collects geological environmental parameters and, after combining the mapping results, constructs a set of associated conditions and outputs deformation sensitivity level data for each zone. The current surface deformation maintenance time and cost acquisition module calculates the initial proportion data based on the sensitivity level data output from the current aquifer surface zoning sensitivity level classification module and sets maintenance measures based on historical experience; if the proportion exceeds the threshold, the maintenance frequency, inspection frequency and sensor parameters are configured, maintenance is implemented and the time and cost are recorded until the proportion of medium and high sensitivity areas drops below the threshold. The current surface deformation maintenance time and cost determination module collects historical deformation maintenance data and constructs a final deformation maintenance time-cost mapping model; if the current maintenance time or cost exceeds the threshold, the current surface deformation maintenance parameter and zoning adjustment module is executed; otherwise, maintenance is completed. The current surface deformation maintenance parameters and zoning adjustment module sets a threshold for the number of micro-adjustments and iteratively adjusts sensor parameters, maintenance frequency, and inspection frequency. The adjusted data is input into the mapping model of the current surface deformation maintenance time and cost determination module. If the threshold for the number of micro-adjustments is reached and the total maintenance cost or time still does not meet the conditions, the area is re-partitioned until both the maintenance time and cost meet the conditions.

Images