Real-time monitoring and evaluation system for leakage risk of coal mine underground reservoir

Abstract

The application discloses a real-time monitoring and evaluation system for leakage risk of underground reservoirs of coal mines, and relates to the technical field of mine safety, which uses reservoir water level monitoring information to meet the preset water level disturbance condition as the trigger basis, limits the unified response time window through the water level disturbance event result, so that the temperature response information, the sound wave response information and the osmotic pressure response information are derived from the response stage after the same water level disturbance; through the multi-source anomaly consistency discrimination result, at least two types of monitoring responses are jointly directed to the same presumed leakage process in time and space, reducing the misjudgment caused by single temperature drift, construction sound wave interference or osmotic pressure instantaneous fluctuation; through mapping the multi-source anomaly consistency discrimination result to the candidate leakage unit and forming the presumed leakage channel inversion result, the leakage risk evaluation result is generated in combination with the hydrogeological model correction result, which provides a basis for inspection and investigation, grouting plugging and water level regulation.

Description

Technical Field

[0001] This invention relates to the field of mine safety technology, specifically to a real-time monitoring and assessment system for leakage risks in underground water reservoirs in coal mines. Background Technology

[0002] In the process of coal mining shifting from simple drainage management to the utilization of mine water resources, the construction of water storage units using underground spaces such as goaf areas, coal pillar dams, roof fissure zones, and floor aquitards has become an important form of coal mine underground reservoir construction. However, during long-term water storage, replenishment, drainage, and water level fluctuations, coal mine underground reservoirs are continuously affected by the development of surrounding rock fissures, seepage in coal pillar dams, changes in the integrity of floor aquitard boundaries, and the connectivity of external aquifers. Therefore, real-time monitoring and assessment systems for leakage risks in coal mine underground reservoirs belong to the technical field of combining coal mine water hazard prevention and control, goaf water storage safety monitoring, and groundwater resource utilization assurance.

[0003] In existing coal mine underground water reservoir operation monitoring, the presence of anomalies in the water storage area is usually determined by water level monitoring, seepage pressure monitoring, temperature monitoring, or manual inspection. However, these methods mostly rely on single data thresholds or changes in local measuring points as alarm criteria. It is difficult to conduct a unified analysis of the response relationship between hydraulic disturbances caused by changes in water level and temperature, acoustic, and seepage pressure information. It is also difficult to determine whether different anomalies originate from the same leakage process. As a result, early hidden leaks are easily masked by changes in ambient temperature, pump station start-up and shutdown, roadway disturbances, or local pressure fluctuations. Therefore, a real-time monitoring and evaluation method is needed that can use changes in water level in the reservoir as a trigger condition and perform spatiotemporal consistency judgment of multi-source information within the response time window. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a real-time monitoring and assessment system for leakage risks in underground water reservoirs of coal mines, solving the problems mentioned in the background art.

[0005] This invention is achieved through the following technical solution: a real-time monitoring and assessment system for leakage risk of underground water reservoirs in coal mines, including a water level triggering module, a response evidence collection module, a collaborative discrimination module, a channel inversion module, and a feedback assessment module;

[0006] The water level triggering module is used to collect water level monitoring information in underground water reservoirs in coal mines, and generate water level disturbance event results as the triggering basis for leakage identification based on the water level monitoring information and preset water level disturbance conditions.

[0007] The response evidence collection module is used to construct a unified response time window based on the results of water level disturbance events, and extract temperature response information, acoustic response information, and seepage pressure response information from temperature monitoring information, acoustic wave monitoring information, and seepage pressure monitoring information, respectively, within the unified response time window.

[0008] The collaborative discrimination module is used to perform spatiotemporal consistency discrimination on temperature response information, acoustic response information and seepage pressure response information based on a unified response time window, and generate multi-source anomaly consistency discrimination results to indicate whether temperature response information, acoustic response information and seepage pressure response information jointly point to the same presumed leakage process.

[0009] The channel inversion module is used to divide the coal pillar dam body, roof fracture zone, floor water-proof boundary and external boundary into candidate seepage units based on the hydrogeological basic model, and to map the multi-source anomaly consistency discrimination results to the candidate seepage units, generating inversion results of the inferred seepage channel that represent the candidate seepage unit where the inferred seepage channel is located, the development direction of the inferred seepage channel, the number of anomaly response types and the number of channel units;

[0010] The feedback assessment module is used to correct the hydrogeological basic model based on the inversion results of the inferred leakage channels, generate the corrected hydrogeological model results, and feed the corrected hydrogeological model results back to the channel inversion module as the model basis for subsequent channel inversion. At the same time, it generates leakage risk assessment results based on the corrected hydrogeological model results and the inversion results of the inferred leakage channels.

[0011] Preferably, the water level triggering module includes a water level acquisition unit, a water level change acquisition unit, and a disturbance event generation unit;

[0012] The water level acquisition unit acquires the current water level value and the reference water level value in the reservoir from the reservoir water level monitoring information according to a preset acquisition interval;

[0013] The benchmark water level in the reservoir is the average value of the reservoir water level monitoring information within a preset water level benchmark time period prior to the current water level value.

[0014] The water level change acquisition unit obtains the reservoir water level change amplitude based on the absolute value of the difference between the current reservoir water level and the benchmark reservoir water level.

[0015] The disturbance event generation unit compares the magnitude of the water level change in the reservoir with the preset water level disturbance threshold in the preset water level disturbance conditions;

[0016] When the magnitude of the water level change in the reservoir reaches the preset water level disturbance threshold, the disturbance event generation unit determines the collection time when the preset water level disturbance threshold is first reached as the water level disturbance start time, and generates the water level disturbance event result based on the water level disturbance start time and the magnitude of the water level change in the reservoir.

[0017] Preferably, the response evidence collection module includes a time window determination unit, a response information extraction unit, and a response set generation unit;

[0018] The time window determination unit uses the start time of the water level disturbance as the starting point of the unified response time window.

[0019] The time window determination unit takes the time after the preset response duration following the start time of the water level disturbance as the end point of the unified response time window.

[0020] The time window determination unit determines the unified response time window based on the start and end points of the unified response time window.

[0021] The response information extraction unit extracts temperature response information, acoustic response information, and osmotic pressure response information from temperature monitoring information, acoustic wave monitoring information, and osmotic pressure monitoring information respectively within a unified response time window;

[0022] The response set generation unit consists of a disturbance-related response set composed of temperature response information, acoustic response information, and osmotic pressure response information.

[0023] Preferably, the channel inversion module includes a spatial modeling unit, a channel positioning unit, and a channel result generation unit;

[0024] The spatial modeling unit obtains the spatial range of the water storage area, the spatial range of the coal pillar dam, the spatial range of the roof fracture zone, the spatial range of the floor water-proof boundary, and the spatial range of the outward boundary from the hydrogeological basic model.

[0025] The spatial modeling unit divides candidate leakage units according to the spatial range of the coal pillar dam body, the spatial range of the roof fissure zone, the spatial range of the floor water-proof boundary, the spatial range of the outward boundary, and the preset spatial division scale.

[0026] The spatial modeling unit determines the location of each candidate seepage unit based on the basic hydrogeological model.

[0027] The spatial modeling unit defines the relationship between two candidate leakage units as having a shared boundary or the shortest distance between two candidate leakage units not exceeding a preset spatial division scale as the relationship between adjacent candidate leakage units.

[0028] The spatial modeling unit determines the relationship between two candidate seepage units that have adjacent candidate seepage unit relationships as candidate seepage unit connectivity relationships if there is a water-conducting fracture connectivity record and no water-impeding boundary isolation record in the hydrogeological basic model.

[0029] The candidate leakage unit locations and the subsequently acquired abnormal spatial locations use the same spatial coordinate reference.

[0030] Preferably, the collaborative discrimination module includes an abnormal response generation unit, an abnormal result selection unit, and a consistency result generation unit;

[0031] The abnormal response generation unit obtains the temperature residual value, acoustic energy value, and osmotic pressure change value based on the disturbance-related response set.

[0032] The temperature baseline information is the average value of temperature monitoring information at the same monitoring location within a preset baseline time period;

[0033] The seepage pressure baseline information is the average value of seepage pressure monitoring information at the same monitoring location within a preset baseline time period;

[0034] The preset benchmark time period is located before the start of the unified response time window;

[0035] The temperature residual value is the absolute value of the difference between the temperature monitoring information and the temperature reference information at the same monitoring location within a unified response time window.

[0036] The acoustic energy value is the energy value of acoustic monitoring information at the same monitoring location within a preset acoustic frequency band within a unified response time window.

[0037] The seepage pressure change value is the absolute value of the difference between the seepage pressure monitoring information and the seepage pressure reference information at the same monitoring location within a unified response time window;

[0038] The abnormal response generation unit compares the temperature residual value, the acoustic energy value, and the osmotic pressure change value with the preset temperature residual threshold, the preset acoustic energy threshold, and the preset osmotic pressure change threshold, respectively.

[0039] The abnormal response generation unit generates a temperature abnormal response result when the temperature residual value reaches a preset temperature residual threshold, generates an acoustic abnormal response result when the acoustic energy value reaches a preset acoustic energy threshold, and generates an osmotic pressure abnormal response result when the osmotic pressure change value reaches a preset osmotic pressure change threshold.

[0040] The results of the temperature anomaly response, acoustic anomaly response, and osmotic pressure anomaly response all include the time of anomaly occurrence and the spatial location of the anomaly.

[0041] The time of an anomaly is the moment when the temperature residual value, sound wave energy value, or osmotic pressure change value first reaches the corresponding threshold during data collection.

[0042] The abnormal spatial location is the location of the monitoring point where the temperature residual value, sound wave energy value, or osmotic pressure change value first reaches the corresponding threshold under the same spatial coordinate reference.

[0043] Preferably, the abnormal result selection unit selects at least two types of abnormal response results from the temperature abnormal response results, the sound wave abnormal response results, and the seepage pressure abnormal response results;

[0044] The consistency result generation unit determines whether the occurrence times of the anomalies in at least two types of abnormal response results are both within a unified response time window.

[0045] The consistency result generation unit determines, based on the abnormal spatial location and the candidate leakage unit location, whether the abnormal spatial locations of at least two types of abnormal response results are mapped to the same candidate leakage unit, or whether they are mapped to two candidate leakage units that have an adjacent candidate leakage unit relationship.

[0046] When the occurrence times of at least two types of abnormal response results are all within a unified response time window, and the spatial locations of the abnormalities of at least two types of abnormal response results are mapped to the same candidate leakage unit or to two candidate leakage units that have an adjacent candidate leakage unit relationship, the consistency result generation unit generates a multi-source abnormality consistency discrimination result.

[0047] The multi-source anomaly consistency discrimination result includes at least two types of anomaly response results, consistency candidate leakage units corresponding to the at least two types of anomaly response results, and the number of consistency candidate leakage units.

[0048] Preferably, the channel positioning unit determines the set of presumed leakage initiation units and presumed leakage extension units based on the consistency candidate leakage units, the number of consistency candidate leakage units, and the connectivity of candidate leakage units in the multi-source anomaly consistency discrimination results;

[0049] The channel positioning unit determines the shortest distance from the location of each consistent candidate leakage unit to the spatial range of the water storage area as the water storage area distance.

[0050] The channel positioning unit determines the shortest distance from the candidate leakage unit position of each consistent candidate leakage unit to the outer boundary space range as the outer boundary distance;

[0051] When the number of consistent candidate leakage units is one, the channel positioning unit determines the consistent candidate leakage unit as the presumed leakage initiation unit and determines the presumed leakage extension unit set as an empty set.

[0052] When there are at least two consistent candidate leakage units, the channel positioning unit determines the consistent candidate leakage unit with the smallest distance from the water storage area as the presumed leakage initiation unit, and determines the consistent candidate leakage units other than the presumed leakage initiation unit that have a candidate leakage unit connection relationship with the presumed leakage initiation unit as the presumed leakage extension unit set.

[0053] The channel positioning unit generates the inversion result of the estimated leakage channel based on the estimated leakage initiation unit, the estimated leakage extension unit set, and the connectivity of the candidate leakage units.

[0054] Preferably, the channel result generation unit combines the estimated leakage initiation unit and the estimated leakage extension unit to form the candidate leakage unit where the estimated leakage channel is located;

[0055] When the set of estimated leakage extension units is empty, the channel result generation unit determines the direction from the position of the candidate leakage unit of the estimated leakage initiation unit to the position in the outer boundary space range that is closest to the position of the candidate leakage unit of the estimated leakage initiation unit as the estimated leakage channel development direction.

[0056] When the set of estimated leakage extension units is not empty, the channel result generation unit determines the estimated leakage extension unit with the smallest outward boundary distance in the set of estimated leakage extension units as the direction termination unit, and determines the direction from the estimated leakage start unit to the direction termination unit as the estimated leakage channel development direction.

[0057] The channel result generation unit determines the number of abnormal response types based on the number of at least two types of abnormal response results involved in generating the multi-source abnormality consistency discrimination result.

[0058] The channel result generation unit determines the number of channel units based on the number of candidate leakage units where the estimated leakage channel is located.

[0059] The inversion results of the estimated leakage channels include the candidate leakage unit where the estimated leakage channel is located, the development direction of the estimated leakage channel, the number of abnormal response types, and the number of channel units.

[0060] Preferably, the feedback evaluation module includes a model feedback unit, a risk level determination unit, and an evaluation result generation unit;

[0061] The model feedback unit updates the connectivity state of the candidate seepage unit corresponding to the candidate seepage unit where the estimated seepage channel is located in the hydrogeological basic model to the estimated conduction state based on the inversion results of the estimated seepage channel.

[0062] The presumed conduction state is used to indicate that the corresponding candidate leakage unit has been included in the presumed leakage channel inversion result;

[0063] The model feedback unit writes the inferred development direction of the seepage channel and the connectivity of the candidate seepage units in the candidate seepage unit where the inferred seepage channel is located into the basic hydrogeological model.

[0064] The model feedback unit generates hydrogeological model correction results based on the updated candidate seepage unit connectivity status, the presumed seepage channel development direction after writing, and the candidate seepage unit connectivity relationship after writing.

[0065] The model feedback unit uses the correction results of the hydrogeological model as the model basis for the channel inversion module to generate the inversion results of the inferred leakage channel inversion next time.

[0066] Preferably, the risk level determination unit determines the leakage risk level based on the hydrogeological model correction results, the candidate leakage unit where the leakage channel is located, the estimated development direction of the leakage channel, the number of abnormal response types, and the number of channel units;

[0067] When there are two abnormal response types and one channel unit, the risk level determination unit will determine the leakage risk level as the leakage warning level.

[0068] When there are two abnormal response types and at least two channel units, the risk level determination unit will determine the leakage risk level as a continuous leakage level.

[0069] When there are three abnormal response types and one channel unit, the risk level determination unit will determine the leakage risk level as a continuous leakage level.

[0070] When there are three abnormal response types and at least two channel units, and the leakage channel is presumed to be developing towards the outward boundary space, the risk level determination unit determines the leakage risk level as the outward risk level.

[0071] The assessment result generation unit generates leakage risk assessment results based on the leakage risk level, the candidate leakage unit where the estimated leakage channel is located, and the estimated development direction of the leakage channel.

[0072] The leakage risk assessment results include leakage risk level, estimated leakage location, and estimated leakage path development direction. The estimated leakage location is determined by the candidate leakage unit where the estimated leakage path is located.

[0073] This invention provides a real-time monitoring and assessment system for leakage risk in underground water reservoirs of coal mines, which has the following beneficial effects:

[0074] (1) The water level monitoring information in the reservoir meets the preset water level disturbance conditions as the trigger basis, and the unified response time window is limited by the water level disturbance event results, so that the temperature response information, sound wave response information and seepage pressure response information all come from the response stage after the same water level disturbance; the consistency judgment results of multi-source anomalies constrain at least two types of monitoring responses to point to the same presumed leakage process in time and space, reducing misjudgments caused by single temperature drift, local construction sound wave interference or instantaneous fluctuation of seepage pressure, and converting dispersed monitoring anomalies into presumed leakage channel inversion results that can be inspected, verified and dealt with.

[0075] (2) Convert the temperature response information, sound wave response information and osmotic pressure response information into temperature residual value, sound wave energy value and osmotic pressure change value respectively, and combine the temperature reference information, osmotic pressure reference information and preset sound wave frequency band to eliminate equipment drift, slow environmental changes and mechanical noise interference; constrain the abnormal results by the time of abnormal occurrence and the abnormal spatial location, and map at least two types of abnormal response results to the same candidate leakage unit or adjacent candidate leakage units to form the inversion result of the inferred leakage channel with spatial starting point, extension range and investigation direction.

[0076] (3) The inversion results of the presumed seepage channels are written into the basic hydrogeological model, and the connectivity status of the corresponding candidate seepage units is updated to the presumed conduction status, so that the subsequent channel inversion calls the corrected spatial connectivity record; the seepage risk level is determined by the number of abnormal response types and the number of channel units, and the seepage risk level, presumed seepage location and presumed seepage channel development direction are output from the seepage risk assessment results, so as to provide a positioning basis for inspection and investigation, grouting and sealing and water level control. Attached Figure Description

[0077] Figure 1 This is a schematic diagram of the real-time monitoring and assessment system for leakage risk of underground water reservoirs in coal mines according to the present invention.

[0078] Figure 2 A schematic diagram of the process flow for a real-time leakage risk monitoring and assessment system;

[0079] Figure 3 This is a schematic diagram of channel inversion and feedback closed loop. Detailed Implementation

[0080] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings in the embodiments of the present invention.

[0081] Example 1

[0082] This embodiment is applied to the long-term water storage operation scenario of underground reservoirs in coal mine goaf areas. In this scenario, the underground reservoir uses the goaf area as the water storage space, and the coal pillar dam, roof fissure zone, floor waterproof boundary, and outward boundary are used as the relevant boundaries for water storage safety. The underground reservoir generates water level changes during water replenishment, drainage, pump station start-up and shutdown, or natural water inflow changes. Water level changes may cause seepage responses at the coal pillar dam, roof fissure zone, or floor waterproof boundary. Therefore, water level changes are selected as the triggering basis for leakage identification. In the response stage after water level changes, temperature monitoring information, acoustic monitoring information, and seepage pressure monitoring information are jointly analyzed to generate the inversion results of the inferred leakage channel and the corresponding leakage risk level in real time.

[0083] This invention provides a real-time monitoring and assessment system for leakage risk in underground water reservoirs of coal mines. Please refer to [link / reference]. Figure 1 It includes a water level triggering module, a response evidence collection module, a collaborative discrimination module, a channel inversion module, and a feedback evaluation module;

[0084] The water level triggering module is used to collect water level monitoring information in underground water reservoirs in coal mines, and generate water level disturbance event results as the triggering basis for leakage identification based on the water level monitoring information and preset water level disturbance conditions.

[0085] The response evidence collection module is used to construct a unified response time window based on the results of water level disturbance events, and extract temperature response information, acoustic response information, and seepage pressure response information from temperature monitoring information, acoustic wave monitoring information, and seepage pressure monitoring information, respectively, within the unified response time window.

[0086] The collaborative discrimination module is used to perform spatiotemporal consistency discrimination on temperature response information, acoustic response information and seepage pressure response information based on a unified response time window, and generate multi-source anomaly consistency discrimination results to indicate whether temperature response information, acoustic response information and seepage pressure response information jointly point to the same presumed leakage process.

[0087] The channel inversion module is used to divide the coal pillar dam body, roof fracture zone, floor water-proof boundary and external boundary into candidate seepage units based on the hydrogeological basic model, and to map the multi-source anomaly consistency discrimination results to the candidate seepage units, generating inversion results of the inferred seepage channel that represent the candidate seepage unit where the inferred seepage channel is located, the development direction of the inferred seepage channel, the number of anomaly response types and the number of channel units;

[0088] The feedback assessment module is used to correct the hydrogeological basic model based on the inversion results of the inferred leakage channels, generate the corrected hydrogeological model results, and feed the corrected hydrogeological model results back to the channel inversion module as the model basis for subsequent channel inversion. At the same time, it generates leakage risk assessment results based on the corrected hydrogeological model results and the inversion results of the inferred leakage channels.

[0089] In this embodiment, by using reservoir water level monitoring information that meets preset water level disturbance conditions as the trigger basis for leakage identification, it is possible to distinguish water level disturbances caused by water replenishment, drainage, pump station start-up and shutdown, or natural water inflow changes from ordinary monitoring background fluctuations. By limiting a unified response time window through the results of water level disturbance events, it is possible to ensure that temperature response information, acoustic response information, and seepage pressure response information all originate from the response stage after the same water level disturbance. By constraining at least two types of monitoring responses to point to the same presumed leakage process in both time and space through the consistency judgment results of multi-source anomalies, it is possible to reduce the impact of single temperature drift, local construction acoustic interference, or other factors. Misjudgments caused by instantaneous fluctuations in pressure sensors can be mitigated. By mapping the consistency discrimination results of multi-source anomalies to candidate leakage units and forming inversion results of presumed leakage channels, the originally scattered monitoring anomalies can be transformed into presumed leakage locations and presumed leakage channel development directions that can be inspected, verified, and handled on-site. The results of hydrogeological model correction can be used to participate in subsequent channel inversion and generate leakage risk assessment results. This can dynamically update the leakage risk level during the continuous water storage operation of underground water reservoirs in coal mines, providing a clear basis for the inspection, grouting, and water level control of coal pillar dams, roof fissure zones, floor water-proof boundaries, or external boundaries.

[0090] Example 2

[0091] In the long-term water storage operation scenario of the underground reservoir in the coal mine goaf area set in Example 1, the underground reservoir in the coal mine generates changes in water level monitoring information during water replenishment, drainage, pump station start-up and shutdown, and natural water inflow. The changes in the water level monitoring information may appear before the temperature monitoring information, acoustic monitoring information, and seepage pressure monitoring information. Therefore, this example uses the water level disturbance event result generated by the water level triggering module as the starting point for subsequent data evidence collection, and uses the response evidence collection module, collaborative discrimination module, and channel inversion module to locate the various monitoring responses after the water level change to the specific spatial boundary of the underground reservoir in the coal mine.

[0092] In this embodiment, the water level monitoring information in the reservoir is collected by water level sensors deployed in the underground water storage area of ​​the coal mine; the temperature monitoring information is collected by distributed temperature monitoring optical cables deployed near the coal pillar dam, the roof fissure zone, the floor water-proof boundary, and the outward boundary; the acoustic monitoring information is collected by distributed acoustic monitoring optical cables deployed near the aforementioned boundaries; and the seepage pressure monitoring information is collected by seepage pressure sensors deployed near the coal pillar dam, the floor water-proof boundary, or the outward boundary.

[0093] Please see Figure 2 Specifically: the water level triggering module includes a water level acquisition unit, a water level change acquisition unit, and a disturbance event generation unit;

[0094] In this embodiment, the water level acquisition unit, the water level change acquisition unit, and the disturbance event generation unit sequentially perform data reading, water level change calculation, and trigger judgment, so that subsequent processing is only initiated after the water level in the underground water reservoir of the coal mine changes in a manner that meets the preset water level disturbance conditions.

[0095] The water level acquisition unit acquires the current water level value and the reference water level value in the reservoir from the reservoir water level monitoring information according to a preset acquisition interval;

[0096] Specifically, the preset sampling interval is set according to the stable sampling period of the water level sensor, the on-site data transmission period, and the shortest control period for pump station start-up and shutdown. The preset sampling interval is less than the shortest control period for pump station start-up and shutdown, so that the current water level value in the reservoir can be obtained during a single water replenishment, drainage, or pump station start-up and shutdown process.

[0097] The benchmark water level in the reservoir is the average value of the reservoir water level monitoring information within a preset water level benchmark time period prior to the current water level value.

[0098] It should be noted that the preset water level reference time period is selected before the time corresponding to the current water level value in the reservoir, and the preset water level reference time period does not include the start time of water replenishment, the start time of drainage, the start time of pumping station, and the stop time of pumping station, so that the water level value in the reference reservoir represents the stable water level before the water level disturbance.

[0099] The water level change acquisition unit obtains the reservoir water level change amplitude based on the absolute value of the difference between the current reservoir water level and the benchmark reservoir water level.

[0100] During data processing, the water level change acquisition unit subtracts the benchmark water level from the current water level in the reservoir and takes the absolute value to obtain the water level change amplitude in the reservoir. This allows the water level change amplitude to simultaneously reflect the disturbances caused by both rising and falling water levels.

[0101] The disturbance event generation unit compares the magnitude of the water level change in the reservoir with the preset water level disturbance threshold in the preset water level disturbance conditions;

[0102] Specifically, the preset water level disturbance threshold is set based on the upper limit of water level measurement error and the upper limit of natural water level fluctuation. The upper limit of water level measurement error is determined by the calibration record of the water level sensor, and the upper limit of natural water level fluctuation is determined by the difference between the maximum and minimum values ​​of the water level monitoring information in the reservoir under the conditions of no water replenishment, no drainage, and no pump station start-stop. The preset water level disturbance threshold is not less than the sum of the upper limit of water level measurement error and the upper limit of natural water level fluctuation.

[0103] When the magnitude of the water level change in the reservoir reaches the preset water level disturbance threshold, the disturbance event generation unit determines the collection time when the preset water level disturbance threshold is first reached as the water level disturbance start time, and generates the water level disturbance event result based on the water level disturbance start time and the magnitude of the water level change in the reservoir.

[0104] In this embodiment, the result of the water level disturbance event includes at least the start time of the water level disturbance and the amplitude of the water level change in the reservoir. The start time of the water level disturbance is used to determine the subsequent unified response time window, and the amplitude of the water level change in the reservoir is used to represent the disturbance intensity corresponding to this water level change.

[0105] The response evidence collection module includes a time window determination unit, a response information extraction unit, and a response set generation unit;

[0106] In this embodiment, the response evidence module is used to extract temperature monitoring information, acoustic monitoring information and seepage pressure monitoring information within the same time range after the water level disturbance, so as to avoid including background fluctuations before the water level disturbance in subsequent judgment.

[0107] The time window determination unit uses the start time of the water level disturbance as the starting point of the unified response time window.

[0108] Specifically, the starting point of the unified response time window is directly adopted as the starting time of the water level disturbance, so that the subsequently extracted temperature response information, acoustic response information and seepage pressure response information are all established with the results of the same water level disturbance event in time.

[0109] The time window determination unit takes the time after the preset response duration following the start time of the water level disturbance as the end point of the unified response time window.

[0110] It should be noted that the preset response duration is set based on the response lag time during historical water replenishment, historical drainage, or trial operation. The response lag time is the time difference between the start of the water level disturbance and the first abnormal response of the temperature monitoring information, acoustic monitoring information, or seepage pressure monitoring information. The preset response duration is not less than the sum of the maximum response lag time in the historical record and the preset acquisition interval.

[0111] The time window determination unit determines the unified response time window based on the start and end points of the unified response time window.

[0112] During data processing, the unified response time window is a continuous time interval from the start to the end of the unified response time window. Temperature monitoring information, acoustic monitoring information, and seepage pressure monitoring information collected outside the unified response time window are not included in the disturbance-related response set.

[0113] The response information extraction unit extracts temperature response information, acoustic response information, and osmotic pressure response information from temperature monitoring information, acoustic wave monitoring information, and osmotic pressure monitoring information respectively within a unified response time window;

[0114] Specifically, the temperature response information includes the acquisition time, monitoring location, and temperature value corresponding to each temperature monitoring location within the unified response time window; the acoustic response information includes the acquisition time, monitoring location, and acoustic value corresponding to each acoustic monitoring location within the unified response time window; and the osmotic pressure response information includes the acquisition time, monitoring location, and osmotic pressure value corresponding to each osmotic pressure monitoring location within the unified response time window.

[0115] The response set generation unit consists of a disturbance-related response set composed of temperature response information, acoustic response information, and osmotic pressure response information;

[0116] In this embodiment, the disturbance-related response set retains the acquisition time and monitoring location of the temperature response information, acoustic response information, and osmotic pressure response information, so that subsequent temporal consistency can be judged based on the acquisition time and spatial consistency can be judged based on the monitoring location.

[0117] The channel inversion module includes a spatial modeling unit, a channel positioning unit, and a channel result generation unit;

[0118] In this embodiment, the channel inversion module is used to locate the subsequent multi-source anomaly consistency discrimination results to the hydrogeological space of the coal mine underground water reservoir, the spatial modeling unit is used to establish candidate leakage units, the channel positioning unit is used to determine the candidate leakage unit where the inferred leakage channel is located, and the channel result generation unit is used to generate the inversion results of the inferred leakage channel.

[0119] The spatial modeling unit obtains the spatial range of the water storage area, the spatial range of the coal pillar dam, the spatial range of the roof fracture zone, the spatial range of the floor water-proof boundary, and the spatial range of the outward boundary from the hydrogeological basic model.

[0120] Specifically, the basic hydrogeological model is established based on data on the goaf area, mine excavation engineering diagrams, coal pillar dam layout data, roof fracture zone development data, floor aquitard data, and external aquifer boundary data. The spatial range of the water storage area, the spatial range of the coal pillar dam, the spatial range of the roof fracture zone, the spatial range of the floor aquitard boundary, and the spatial range of the external boundary are all converted into a unified three-dimensional spatial coordinate system for the coal mine.

[0121] It should be noted that, in this embodiment, the hydrogeological basic model is established based on the construction data of the coal mine underground reservoir and the field survey data. Specifically, it includes the coordinates of the goaf boundary, the coordinates of the coal pillar dam layout, the development range of the roof fracture zone, the location of the floor water-proof boundary, and the location of the outward boundary. During the model establishment process, the goaf outline, the location of the coal pillar dam, and the spatial location of the roadway in the mining engineering drawings are converted to a unified spatial coordinate benchmark, and the range of the roof fracture zone, the range of the floor water-proof boundary, and the range of the outward boundary obtained from the geological survey are superimposed into the same spatial model. When there are deviations in the locations of different data sources, the field measurement coordinates and the verified engineering boundary locations are used as calibration basis to adjust the spatial boundaries within the deviation range, so that the spatial range of the water storage area, the spatial range of the coal pillar dam, the spatial range of the roof fracture zone, the spatial range of the floor water-proof boundary, and the spatial range of the outward boundary maintain a corresponding relationship.

[0122] Furthermore, the candidate leakage units in the hydrogeological basic model are not generated by calculation based on continuous parameters such as permeability coefficient and porosity distribution, but are discretely divided according to the determined spatial boundaries and preset spatial division scales. The preset spatial division scales are determined based on the spacing between monitoring points, the size of the engineering boundary, and the scope of on-site inspection, so that each candidate leakage unit corresponds to an actual locatable underground space area.

[0123] The spatial modeling unit divides candidate leakage units according to the spatial range of the coal pillar dam body, the spatial range of the roof fissure zone, the spatial range of the floor water-proof boundary, the spatial range of the outward boundary, and the preset spatial division scale.

[0124] It should be noted that the preset spatial division scale is set based on the upper limit of the monitoring and positioning error and the length of the on-site treatment section. The upper limit of the monitoring and positioning error is determined by the maximum value among the distributed optical fiber monitoring and positioning error, the seepage pressure monitoring point measurement error, and the mine measurement coordinate error. The length of the on-site treatment section is determined by the smallest construction section that can be inspected, sealed, grouted, or dewatered on-site. The preset spatial division scale is not less than the upper limit of the monitoring and positioning error and not greater than the length of the on-site treatment section.

[0125] The spatial modeling unit determines the location of each candidate seepage unit based on the basic hydrogeological model.

[0126] Specifically, the location of a candidate leakage unit includes the spatial boundary range and the spatial center position of the candidate leakage unit in the unified three-dimensional spatial coordinate system of the coal mine. The spatial boundary range is used to determine whether subsequent abnormal spatial locations fall into the candidate leakage unit, and the spatial center position is used to calculate the spatial distance between the candidate leakage unit and the spatial range of the water storage area or the spatial range of the outer boundary.

[0127] The spatial modeling unit defines the relationship between two candidate leakage units as having a shared boundary or the shortest distance between two candidate leakage units not exceeding a preset spatial division scale as the relationship between adjacent candidate leakage units.

[0128] During data processing, the shortest distance between two candidate leakage units is the shortest distance between the spatial boundary ranges of the two candidate leakage units. When the two candidate leakage units share a common boundary, the shortest distance between the two candidate leakage units is zero.

[0129] The spatial modeling unit determines the relationship between two candidate seepage units that have adjacent candidate seepage unit relationships as candidate seepage unit connectivity relationships if there is a water-conducting fracture connectivity record and no water-impeding boundary isolation record in the hydrogeological basic model.

[0130] Specifically, the water-conducting fracture connectivity record includes mining-induced fractures, roof water-conducting fracture zones, fault water-conducting zones, or water-conducting channels exposed by boreholes recorded in the basic hydrogeological model. The water-blocking boundary isolation record includes complete water-blocking layers, sealing coal pillars, grouting reinforced boundaries, or seepage prevention structures recorded in the basic hydrogeological model. Only two candidate seepage units that simultaneously satisfy the condition of having a water-conducting fracture connectivity record and not having a water-blocking boundary isolation record are determined to have a candidate seepage unit connectivity relationship.

[0131] The candidate leakage unit locations and the subsequently acquired abnormal spatial locations use the same spatial coordinate reference;

[0132] In this embodiment, by limiting the water level value in the reference reservoir to the average value of the water level monitoring information within a preset water level reference time period, and by using a preset water level disturbance threshold to exclude water level sensor measurement errors and natural fluctuations under conditions of no water replenishment, no drainage, and no pump station start-up and shutdown, the results of water level disturbance events can correspond to actual operational disturbances rather than single-point water level jumps. By determining a unified response time window based on the start time of the water level disturbance, the temperature response information, acoustic response information, and seepage pressure response information can be ensured to originate only from the effective response interval after the same water replenishment, drainage, or pump station start-up and shutdown. By defining the spatial range of the water storage area, the spatial range of the coal pillar dam, and the roof crack in the hydrogeological basic model, the results can be optimized. The spatial range of the gap zone, the spatial range of the bottom plate water-proof boundary, and the spatial range of the outward boundary are uniformly converted to the same spatial coordinate reference, and candidate leakage units are divided according to the preset spatial division scale. This can place the distributed optical fiber monitoring location and seepage pressure monitoring point in the spatial section that can be inspected, sealed, and grouted. By distinguishing the relationship between adjacent candidate leakage units and the connection relationship between candidate leakage units, it can avoid judging two candidate leakage units as water-conducting connected simply because they are spatially close. For example, when there is a complete water-proof layer or grouting reinforcement boundary near the coal pillar dam, the incorrect connection judgment can still be blocked by the water-proof boundary isolation record, thereby improving the spatial reliability of subsequent channel positioning and risk assessment.

[0133] Example 3

[0134] Based on the long-term water storage operation scenario of underground reservoir in coal mine goaf set in Example 1 and the unified response time window and candidate leakage unit established in Example 2, this example performs anomaly extraction, spatiotemporal consistency discrimination and leakage channel inversion on temperature response information, sound wave response information and seepage pressure response information in disturbance-related response set, so that multiple monitoring responses after water level disturbance can be converted into inversion results of inferred leakage channels that can be used for risk assessment.

[0135] Please see Figure 3 Specifically: the collaborative discrimination module includes an abnormal response generation unit, an abnormal result selection unit, and a consistency result generation unit;

[0136] In this embodiment, the abnormal response generation unit is used to convert the continuous monitoring data in the disturbance-related response set into temperature abnormal response results, acoustic abnormal response results and seepage pressure abnormal response results. The abnormal result selection unit is used to select the abnormal type participating in the consistency judgment from the generated abnormal response results. The consistency result generation unit is used to determine whether the selected abnormal type belongs to the same presumed leakage process after the same water level disturbance.

[0137] The abnormal response generation unit obtains the temperature residual value, acoustic energy value, and osmotic pressure change value based on the disturbance-related response set.

[0138] During the data processing, the abnormal response generation unit reads the temperature response information, acoustic response information and seepage pressure response information from the disturbance-related response set according to the monitoring location, and calculates the monitoring values ​​of the same monitoring location within the unified response time window time by time, so as to obtain the temperature residual value, acoustic energy value and seepage pressure change value corresponding to each acquisition time.

[0139] The temperature baseline information is the average value of temperature monitoring information at the same monitoring location within a preset baseline time period;

[0140] It should be noted that the preset reference time period is selected before the start of the unified response time window and does not include the start time of water replenishment, the start time of drainage, the start time of pumping station, and the stop time of pumping station. The temperature reference information is obtained by averaging multiple temperature monitoring information at the same monitoring location within the preset reference time period.

[0141] The seepage pressure baseline information is the average value of seepage pressure monitoring information at the same monitoring location within a preset baseline time period;

[0142] It should be noted that the seepage pressure benchmark information is obtained by averaging multiple seepage pressure monitoring data at the same monitoring location within a preset benchmark time period, so that the seepage pressure benchmark information represents the stable seepage pressure level at that monitoring location before water level disturbance.

[0143] The preset benchmark time period is located before the start of the unified response time window;

[0144] In this embodiment, the length of the preset reference time period is set according to the preset acquisition interval and the stability of the monitoring data. The preset reference time period should include monitoring data from multiple consecutive acquisition times, and the preset reference time period does not overlap with the starting point of the unified response time window.

[0145] The temperature residual value is the absolute value of the difference between the temperature monitoring information and the temperature reference information at the same monitoring location within a unified response time window.

[0146] During the data processing, the anomaly response generation unit subtracts the temperature reference information from each temperature monitoring information at the same monitoring location within the unified response time window, and takes the absolute value of the difference to obtain the temperature residual value at the corresponding acquisition time.

[0147] The acoustic energy value is the energy value of acoustic monitoring information at the same monitoring location within a preset acoustic frequency band within a unified response time window.

[0148] Specifically, the preset acoustic frequency band is determined based on background acoustic data under conditions of no water level disturbance, or based on water flow disturbance records formed after water replenishment, drainage, and pump station start-up and shutdown during the trial operation phase. When there are no historical water flow disturbance records, acoustic data corresponding to water flow disturbance is collected through short-term water replenishment or drainage tests on site. During data processing, spectral analysis is performed on the background acoustic data and the acoustic data corresponding to water flow disturbance, and the frequency range where the acoustic energy corresponding to water flow disturbance is higher than the background acoustic data and does not belong to the equipment operating frequency is determined as the preset acoustic frequency band. The main frequency of pump station mechanical vibration and the main frequency of ventilation equipment are determined by the peak value of the acoustic spectrum during stable operation of the pump station or ventilation equipment and are excluded from the preset acoustic frequency band. The acoustic sampling frequency is not less than twice the upper limit frequency of the preset acoustic frequency band, and the acoustic energy value is obtained by summing the squares or averaging the squares of the acoustic amplitudes within the preset acoustic frequency band according to the preset energy calculation window within the unified response time window.

[0149] The seepage pressure change value is the absolute value of the difference between the seepage pressure monitoring information and the seepage pressure reference information at the same monitoring location within a unified response time window;

[0150] During the data processing, the abnormal response generation unit subtracts the seepage pressure baseline information from each seepage pressure monitoring information at the same monitoring location within the unified response time window, and takes the absolute value of the difference to obtain the seepage pressure change value at the corresponding acquisition time.

[0151] The abnormal response generation unit compares the temperature residual value, the acoustic energy value, and the osmotic pressure change value with the preset temperature residual threshold, the preset acoustic energy threshold, and the preset osmotic pressure change threshold, respectively.

[0152] Specifically, the preset temperature residual threshold is set based on the upper limit of temperature measurement error and the upper limit of temperature stability fluctuation under the condition of no water level disturbance; the preset sound wave energy threshold is set based on the upper limit of sound wave background energy in the preset sound wave frequency band under the condition of no water level disturbance; and the preset seepage pressure change threshold is set based on the upper limit of seepage pressure measurement error and the upper limit of seepage pressure stability fluctuation under the condition of no water level disturbance.

[0153] The abnormal response generation unit generates a temperature abnormal response result when the temperature residual value reaches a preset temperature residual threshold, generates an acoustic abnormal response result when the acoustic energy value reaches a preset acoustic energy threshold, and generates an osmotic pressure abnormal response result when the osmotic pressure change value reaches a preset osmotic pressure change threshold.

[0154] In this embodiment, when multiple acquisition times at the same monitoring location reach the corresponding threshold, the abnormal response generation unit takes the acquisition time at which the corresponding threshold is first reached as the time when the abnormality occurs, and takes the position of the monitoring location under the same spatial coordinate reference as the abnormal spatial position.

[0155] The results of the temperature anomaly response, acoustic anomaly response, and osmotic pressure anomaly response all include the time of anomaly occurrence and the spatial location of the anomaly.

[0156] Specifically, the temperature anomaly response results can also record the corresponding temperature residual value, the acoustic anomaly response results can also record the corresponding acoustic energy value, and the osmotic pressure anomaly response results can also record the corresponding osmotic pressure change value, so as to verify the source of the anomaly later.

[0157] The time of an anomaly is the moment when the temperature residual value, sound wave energy value, or osmotic pressure change value first reaches the corresponding threshold during data collection.

[0158] During data processing, if the same monitoring location does not produce a temperature residual value, acoustic energy value, or osmotic pressure change value that reaches the corresponding threshold within a unified response time window, then the monitoring location will not generate the corresponding temperature anomaly response result, acoustic anomaly response result, or osmotic pressure anomaly response result.

[0159] The abnormal spatial location is the location of the monitoring point where the temperature residual value, sound wave energy value, or osmotic pressure change value first reaches the corresponding threshold under the same spatial coordinate reference.

[0160] It should be noted that the monitoring points corresponding to the distributed optical fiber are converted to the same spatial coordinate reference through the optical cable laying path calibration table, and the monitoring points corresponding to the pressure sensor are converted to the same spatial coordinate reference through the installation measurement coordinates.

[0161] The abnormal result selection unit selects at least two types of abnormal response results from the temperature abnormal response results, the sound wave abnormal response results, and the seepage pressure abnormal response results;

[0162] In this embodiment, at least two types of abnormal response results refer to any two or three of the following: abnormal temperature response results, abnormal sound wave response results, and abnormal seepage pressure response results. If only one type of abnormal response result exists, the abnormal result selection unit will not use that abnormal response result alone to generate a multi-source abnormality consistency discrimination result.

[0163] The consistency result generation unit determines whether the occurrence times of the anomalies in at least two types of abnormal response results are both within a unified response time window.

[0164] During data processing, the consistency result generation unit reads the time of occurrence of the anomaly in each type of abnormal response result and compares the time of occurrence of the anomaly with the start and end points of the unified response time window. When the time of occurrence of the anomaly is not earlier than the start and not later than the end of the unified response time window, it is determined that the time of occurrence of the anomaly is within the unified response time window.

[0165] The consistency result generation unit determines, based on the abnormal spatial location and the candidate leakage unit location, whether the abnormal spatial locations of at least two types of abnormal response results are mapped to the same candidate leakage unit, or whether they are mapped to two candidate leakage units that have an adjacent candidate leakage unit relationship.

[0166] Specifically, when an abnormal spatial location falls within the spatial boundary range of a candidate leakage unit location, the consistency result generation unit maps the abnormal spatial location to the corresponding candidate leakage unit. When the abnormal spatial location does not fall within the spatial boundary range of any candidate leakage unit location, the consistency result generation unit calculates the shortest distance from the abnormal spatial location to the spatial boundary range of each candidate leakage unit location. If the shortest distance is not greater than the upper limit of the monitoring and positioning error, the candidate leakage unit corresponding to the shortest distance is used as the mapping result.

[0167] When the occurrence times of at least two types of abnormal response results are all within a unified response time window, and the spatial locations of the abnormalities of at least two types of abnormal response results are mapped to the same candidate leakage unit or to two candidate leakage units that have an adjacent candidate leakage unit relationship, the consistency result generation unit generates a multi-source abnormality consistency discrimination result.

[0168] In this embodiment, the multi-source anomaly consistency discrimination result is used to exclude occasional fluctuations of a single monitoring type. Only at least two types of abnormal response results that are in the same unified response time window in time and fall into the same candidate leakage unit or adjacent candidate leakage units in space are used as evidence of the same presumed leakage process.

[0169] The multi-source anomaly consistency discrimination result includes at least two types of anomaly response results, consistency candidate leakage units corresponding to the at least two types of anomaly response results, and the number of consistency candidate leakage units.

[0170] Specifically, a consistent candidate leakage unit is a set of candidate leakage units obtained by mapping at least two types of abnormal response results after deduplication, and the number of consistent candidate leakage units is the number of candidate leakage units in this set.

[0171] The channel positioning unit determines the set of presumed leakage initiation units and presumed leakage extension units based on the consistency candidate leakage units, the number of consistency candidate leakage units, and the connectivity of candidate leakage units in the multi-source anomaly consistency discrimination results.

[0172] In this embodiment, the channel positioning unit preferentially uses consistent candidate leakage units close to the water storage area as the presumed leakage initiation units, and uses consistent candidate leakage units that have a candidate leakage unit connection relationship with the presumed leakage initiation units as the presumed leakage extension unit set.

[0173] The channel positioning unit determines the shortest distance from the location of each consistent candidate leakage unit to the spatial range of the water storage area as the water storage area distance.

[0174] Specifically, the distance between the water storage area and the water storage area is the shortest distance between the spatial boundary of the consistent candidate leakage unit and the spatial boundary of the water storage area. When the two overlap or are in contact, the distance between the water storage area is zero.

[0175] The channel positioning unit determines the shortest distance from the candidate leakage unit position of each consistent candidate leakage unit to the outer boundary space range as the outer boundary distance;

[0176] Specifically, the external boundary distance is the shortest distance between the spatial boundary range of the consistent candidate leakage unit and the spatial boundary range of the external boundary. When the two overlap or are in contact, the external boundary distance is zero.

[0177] When the number of consistent candidate leakage units is one, the channel positioning unit determines the consistent candidate leakage unit as the presumed leakage initiation unit and determines the presumed leakage extension unit set as an empty set.

[0178] It should be noted that when there is only one candidate leakage unit, the system can determine the location of the presumed leakage initiation unit, but has not yet obtained the adjacent extension location from the multi-source anomaly consistency discrimination result. Therefore, the set of presumed leakage extension units is determined to be an empty set.

[0179] When there are at least two consistent candidate leakage units, the channel positioning unit determines the consistent candidate leakage unit with the smallest distance from the water storage area as the presumed leakage initiation unit, and determines the consistent candidate leakage units other than the presumed leakage initiation unit that have a candidate leakage unit connection relationship with the presumed leakage initiation unit as the presumed leakage extension unit set.

[0180] During data processing, if there are multiple consistent candidate leakage units with the same distance between water storage areas, the channel positioning unit will preferentially select the consistent candidate leakage unit with the smaller distance to the outer boundary as the estimated leakage initiation unit.

[0181] The channel positioning unit generates the inversion result of the estimated leakage channel based on the estimated leakage initiation unit, the estimated leakage extension unit set, and the connection relationship of the candidate leakage units;

[0182] In this embodiment, the inversion results of the estimated leakage channel are further organized by the channel result generation unit into the candidate leakage unit where the estimated leakage channel is located, the development direction of the estimated leakage channel, the number of abnormal response types, and the number of channel units;

[0183] The channel result generation unit combines the estimated leakage initiation unit and the estimated leakage extension unit to form the candidate leakage unit where the estimated leakage channel is located.

[0184] Specifically, the candidate leakage unit where the estimated leakage channel is located is the result of merging the estimated leakage initiation unit and the estimated leakage extension unit set, and duplicate candidate leakage units are removed.

[0185] When the set of estimated leakage extension units is empty, the channel result generation unit determines the direction from the position of the candidate leakage unit of the estimated leakage initiation unit to the position in the outer boundary space range that is closest to the position of the candidate leakage unit of the estimated leakage initiation unit as the estimated leakage channel development direction.

[0186] Specifically, when the set of presumed leakage extension units is empty, the development direction of the presumed leakage channel indicates the direction of investigation from the located presumed leakage initiation unit toward the nearest external boundary, which is used to provide spatial guidance for subsequent inspections and risk assessments.

[0187] When the set of estimated leakage extension units is not empty, the channel result generation unit determines the estimated leakage extension unit with the smallest outward boundary distance in the set of estimated leakage extension units as the direction termination unit, and determines the direction from the estimated leakage start unit to the direction termination unit as the estimated leakage channel development direction.

[0188] During data processing, the direction from the estimated leakage initiation unit to the direction termination unit is determined by the spatial center position from the estimated leakage initiation unit to the spatial center position of the direction termination unit.

[0189] The channel result generation unit determines the number of abnormal response types based on the number of at least two types of abnormal response results involved in generating the multi-source abnormality consistency discrimination result.

[0190] Specifically, the number of abnormal response types is the number of abnormal response result types that participate in generating the multi-source anomaly consistency discrimination result. Temperature abnormal response result, acoustic abnormal response result, and osmotic pressure abnormal response result each correspond to one abnormal response result type.

[0191] The channel result generation unit determines the number of channel units based on the number of candidate leakage units where the estimated leakage channel is located.

[0192] Specifically, the number of channel units is the number of candidate leakage units in the candidate leakage units where the presumed leakage channel is located after deduplication;

[0193] The inversion results of the estimated leakage channels include the candidate leakage unit where the estimated leakage channel is located, the development direction of the estimated leakage channel, the number of abnormal response types, and the number of channel units;

[0194] In this embodiment, by converting the temperature response information, acoustic response information and osmotic pressure response information in the disturbance-related response set into temperature residual value, acoustic energy value and osmotic pressure change value respectively, and using the temperature reference information and osmotic pressure reference information within a preset reference time period as a stable reference before water level disturbance, it is possible to distinguish the real response after water level disturbance from equipment drift, slow change of ambient temperature or short-term fluctuation of osmotic pressure.

[0195] It should be noted that, in this embodiment, the inversion result of the estimated leakage channel is not obtained by solving the seepage parameters of underground fissures, but is determined based on the spatial location corresponding to the monitored anomaly and the spatial connectivity in the basic hydrogeological model. In the specific process, the abnormal spatial locations in the temperature anomaly response results, sound wave anomaly response results and seepage pressure anomaly response results are converted to a unified spatial coordinate reference, and the candidate leakage unit to which the abnormal spatial location belongs is determined respectively.

[0196] When the abnormal spatial locations corresponding to at least two types of abnormal response results are located in the same candidate leakage unit, the candidate leakage unit is determined as a consistent candidate leakage unit participating in channel positioning; when the abnormal spatial locations corresponding to at least two types of abnormal response results are located in different candidate leakage units with a connection relationship, multiple candidate leakage units are jointly determined as a consistent candidate leakage unit set participating in channel positioning; subsequently, the set of presumed leakage initiation units and presumed leakage extension units are determined according to the connection direction between candidate leakage units, the distance of the water storage area, and the distance of the external boundary.

[0197] Therefore, the channel inversion process in this embodiment determines the candidate spatial paths that anomaly responses may take based on established spatial relationships and multi-source monitoring evidence. It does not require solving for the number of underground fissures, fissure geometry, or permeability parameters, nor does it address the issue of multiple seepage parameter combinations corresponding to the same monitoring response leading to indeterminate results. By presetting the acoustic frequency band to avoid the dominant frequencies of pump station mechanical vibration and ventilation equipment, the interference of mechanical noise on the acoustic anomaly response results in the coal mine underground reservoir operation site can be reduced. By constraining the temperature anomaly response results, acoustic anomaly response results, and seepage pressure anomaly response results together with the anomaly occurrence time and anomaly spatial location, it is possible to avoid generating multi-source anomaly consistency judgment results simply because a single monitoring location instantaneously exceeds the limit. By mapping at least two types of anomaly response results to the same candidate seepage unit or two candidate seepage units with adjacent candidate seepage unit relationships, it is possible to detect temperature changes, acoustic enhancement, and seepage pressure changes near the coal pillar dam. When at least two types of responses are detected, they are grouped together as evidence of the same presumed leakage process. By determining the presumed leakage initiation unit, the set of presumed leakage extension units, and the development direction of the presumed leakage channel through the distance of the water storage area, the distance of the outer boundary, the number of consistent candidate leakage units, and the connectivity of the candidate leakage units, the results of the multi-source anomaly consistency discrimination can be further converted into the inversion results of the presumed leakage channel with spatial starting point, extension range, and investigation direction. By retaining the number of monitoring types and channel coverage involved in the discrimination through the number of anomaly response types and the number of channel units, a clear counting basis can be provided for the subsequent determination of leakage risk level.

[0198] Example 4

[0199] Based on the long-term water storage operation scenario of the underground reservoir in the coal mine goaf set in Example 1, the hydrogeological basic model established in Example 2, and the inversion results of the inferred leakage channels generated in Example 3, this example writes the inversion results of the inferred leakage channels into the hydrogeological basic model through the feedback evaluation module, and determines the leakage risk level according to the number of abnormal response types and the number of channel units, so that the next channel inversion can be based on the updated hydrogeological model correction results;

[0200] Please see Figure 3Specifically: the feedback evaluation module includes a model feedback unit, a risk level determination unit, and an evaluation result generation unit;

[0201] In this embodiment, the model feedback unit is used to update the state and connectivity of candidate seepage units in the hydrogeological basic model, the risk level determination unit is used to determine the seepage risk level based on the number of abnormal response types and the number of channel units in the inversion results of the estimated seepage channel, and the assessment result generation unit is used to output the seepage risk assessment result including the seepage risk level, the estimated seepage location, and the estimated seepage channel development direction.

[0202] The model feedback unit updates the connectivity state of the candidate seepage unit corresponding to the candidate seepage unit where the estimated seepage channel is located in the hydrogeological basic model to the estimated conduction state based on the inversion results of the estimated seepage channel.

[0203] Specifically, a candidate seepage unit status table is pre-established in the hydrogeological basic model. The candidate seepage unit status table records candidate seepage units and sets the candidate seepage unit connectivity status for each candidate seepage unit. When the model feedback unit reads the candidate seepage unit where the estimated seepage channel is located, it updates the candidate seepage unit connectivity status in the corresponding record from the original status to the estimated conduction status.

[0204] The presumed conduction state is used to indicate that the corresponding candidate leakage unit has been included in the presumed leakage channel inversion result;

[0205] It should be noted that the presumed continuity state does not mean that the corresponding candidate leakage unit has undergone deterministic connection, but rather that the corresponding candidate leakage unit has simultaneously met the multi-source anomaly consistency discrimination result and the channel inversion condition, and serves as the model record state for subsequent channel inversion and leakage risk level determination.

[0206] The model feedback unit writes the inferred development direction of the seepage channel and the connectivity of the candidate seepage units in the candidate seepage unit where the inferred seepage channel is located into the basic hydrogeological model.

[0207] During data processing, the model feedback unit takes the candidate seepage unit where the inferred seepage channel is located as the node and the existing candidate seepage unit connection relationship between the candidate seepage units where the inferred seepage channel is located as the edge, and writes the node, edge and the inferred seepage channel development direction into the channel record table in the hydrogeological basic model.

[0208] The model feedback unit generates hydrogeological model correction results based on the updated candidate seepage unit connectivity status, the presumed seepage channel development direction after writing, and the candidate seepage unit connectivity relationship after writing.

[0209] Specifically, the hydrogeological model correction results include an updated candidate seepage unit status table, an updated channel record table, and the corresponding generation time. The updated candidate seepage unit status table is used to indicate which candidate seepage units are marked as presumed to be in a conducting state, and the updated channel record table is used to indicate the connection order between the candidate seepage units where the presumed seepage channels are located and the development direction of the presumed seepage channels.

[0210] The model feedback unit uses the correction results of the hydrogeological model as the model basis for the channel inversion module to generate the inference results of the inferred leakage channel inversion next time.

[0211] In this embodiment, when a new water level disturbance event occurs and a new multi-source anomaly consistency judgment result is generated, the channel inversion module first reads the candidate seepage unit status table and channel record table in the hydrogeological model correction result to determine whether the new anomaly spatial location falls into a candidate seepage unit that has been marked as presumed to be in a conducting state or its adjacent candidate seepage unit.

[0212] The risk level determination unit determines the leakage risk level based on the hydrogeological model correction results, the candidate leakage unit where the leakage channel is located, the estimated development direction of the leakage channel, the number of abnormal response types, and the number of channel units.

[0213] Specifically, the number of abnormal response types is determined by the number of types that participate in generating the multi-source anomaly consistency judgment results from the temperature abnormal response results, acoustic abnormal response results, and osmotic pressure abnormal response results. The number of channel units is determined by the number of candidate leakage units after deduplication in the candidate leakage units where the presumed leakage channel is located. The risk level determination unit determines the leakage risk level according to the combination relationship between the number of abnormal response types and the number of channel units.

[0214] When there are two abnormal response types and one channel unit, the risk level determination unit will determine the leakage risk level as the leakage warning level.

[0215] It should be noted that two abnormal response types indicate that two types of monitoring responses are pointing to the same presumed leakage process, and one channel unit indicates that the presumed leakage location is still concentrated in one candidate leakage unit. Therefore, the leakage risk level is determined as the leakage warning level.

[0216] When there are two abnormal response types and at least two channel units, the risk level determination unit will determine the leakage risk level as a continuous leakage level.

[0217] It should be noted that having at least two channel units indicates that the presumed leakage process has crossed multiple candidate leakage units, and the risk level determination unit will raise the leakage risk level from the leakage warning level to the continuous leakage level accordingly.

[0218] When there are three abnormal response types and one channel unit, the risk level determination unit will determine the leakage risk level as a continuous leakage level.

[0219] It should be noted that the number of abnormal response types is three, indicating that the abnormal temperature response, abnormal sound wave response, and abnormal seepage pressure response all participate in the generation of multi-source anomaly consistency judgment results. Even if the number of channel units is one, it means that multiple monitoring responses appear in the same candidate leakage unit and all point to the same presumed leakage process. Therefore, the leakage risk level is determined to be the continuous leakage level.

[0220] When there are three abnormal response types and at least two channel units, and the leakage channel is presumed to be developing towards the outward boundary space, the risk level determination unit determines the leakage risk level as the outward risk level.

[0221] It should be noted that the extravasation risk level is used to indicate that the inversion results of the presumed leakage channel have simultaneously met the following conditions: the three types of abnormal responses point in the same direction, the candidate leakage unit where the presumed leakage channel is located spans at least two candidate leakage units, and the development direction of the presumed leakage channel points to the extravasation boundary space. During data processing, whether the development direction of the presumed leakage channel points to the extravasation boundary space is determined based on the fact that the extravasation boundary distance of the direction termination unit is less than the extravasation boundary distance of the presumed leakage initiation unit. Therefore, the extravasation risk level is not triggered solely by the number of abnormal response types and the number of channel units, but is jointly triggered by the multi-source abnormal responses, the spatial extension range, and the directional relationship towards the extravasation boundary space.

[0222] The assessment result generation unit generates leakage risk assessment results based on the leakage risk level, the candidate leakage unit where the estimated leakage channel is located, and the estimated development direction of the leakage channel.

[0223] Specifically, the assessment result generation unit binds the leakage risk level, the candidate leakage unit where the inferred leakage channel is located, the inferred leakage channel development direction and the corresponding water level disturbance event results to generate leakage risk assessment results that can be used for interface display, inspection positioning or operation record storage.

[0224] The leakage risk assessment results include leakage risk level, estimated leakage location, and estimated leakage path development direction. The estimated leakage location is determined by the candidate leakage unit where the estimated leakage path is located.

[0225] In this embodiment, by writing the inversion results of the presumed seepage channel into the basic hydrogeological model and updating the connectivity state of the candidate seepage unit corresponding to the candidate seepage unit where the presumed seepage channel is located to the presumed conduction state, it is possible to directly call the corrected spatial connectivity record when the water level disturbance event occurs again, avoiding the need to re-determine the seepage channel with the initial model each time.

[0226] It should be noted that, in this embodiment, real-time monitoring and assessment refers to the data processing process from the acquisition of reservoir water level monitoring information to the generation of leakage risk assessment results, which meets the requirements of continuous operation. Specifically, the system acquires reservoir water level monitoring information according to a preset acquisition interval. When the amplitude of the change in reservoir water level is detected to reach the preset water level disturbance threshold, the system immediately calls up the temperature monitoring information, sound wave monitoring information, and seepage pressure monitoring information within the corresponding time period, and completes the abnormal response judgment according to a unified response time window.

[0227] When generating the inversion results of the presumed leakage channels, it is only necessary to query the corresponding candidate leakage units, the connectivity relationship of the candidate leakage units, and the spatial distance relationship based on the consistency judgment results of the multi-source anomalies, and determine the presumed leakage starting unit, the set of presumed leakage extension units, and the development direction of the presumed leakage channels according to the preset judgment rules. It is not necessary to repeatedly perform groundwater flow field calculations or fissure seepage numerical simulation calculations, so it can meet the real-time risk update requirements during the operation of coal mine underground water reservoirs. By incorporating the presumed seepage channel development direction and the connectivity of candidate seepage units into the hydrogeological basic model and generating hydrogeological model correction results, the channel direction obtained from a single monitoring inversion can be used as the basis for subsequent inversions. For example, when adjacent boundary responses occur twice consecutively near the coal pillar dam, candidate seepage units already in a presumed conductive state and their adjacent sections can be prioritized for verification. By determining the seepage risk level by combining the number of abnormal response types and the number of channel units, the co-occurrence of multiple monitoring responses and whether the presumed seepage channel crosses multiple candidate seepage units can be included in the risk assessment. This allows two types of anomalies within a single spatial unit to correspond to a seepage warning level, two types of anomalies within multiple spatial units or three types of anomalies within a single spatial unit to correspond to a continuous seepage level, and three types of anomalies within multiple spatial units to correspond to an external risk level. By simultaneously outputting the seepage risk level, presumed seepage location, and presumed seepage channel development direction through the seepage risk assessment results, on-site personnel can directly determine the candidate seepage unit where the presumed seepage channel needs to be inspected, the priority investigation direction, and whether water level control, grouting sealing, or boundary verification measures are needed.

[0228] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention.

Claims

1. A real-time monitoring and assessment system for leakage risk in underground water reservoirs of coal mines, characterized by: It includes a water level triggering module, a response evidence collection module, a collaborative discrimination module, a channel inversion module, and a feedback evaluation module; The water level triggering module is used to collect water level monitoring information in underground water reservoirs in coal mines, and generate water level disturbance event results as the triggering basis for leakage identification based on the water level monitoring information and preset water level disturbance conditions. The response evidence collection module is used to construct a unified response time window based on the results of water level disturbance events, and extract temperature response information, acoustic response information, and seepage pressure response information from temperature monitoring information, acoustic wave monitoring information, and seepage pressure monitoring information, respectively, within the unified response time window. The collaborative discrimination module is used to perform spatiotemporal consistency discrimination on temperature response information, acoustic response information and seepage pressure response information based on a unified response time window, and generate multi-source anomaly consistency discrimination results to indicate whether temperature response information, acoustic response information and seepage pressure response information jointly point to the same presumed leakage process. The channel inversion module is used to divide the coal pillar dam body, roof fracture zone, floor water-proof boundary and external boundary into candidate seepage units based on the hydrogeological basic model, and to map the multi-source anomaly consistency discrimination results to the candidate seepage units, generating inversion results of the inferred seepage channel that represent the candidate seepage unit where the inferred seepage channel is located, the development direction of the inferred seepage channel, the number of anomaly response types and the number of channel units; The feedback assessment module is used to correct the hydrogeological basic model based on the inversion results of the inferred leakage channels, generate the corrected hydrogeological model results, and feed the corrected hydrogeological model results back to the channel inversion module as the model basis for subsequent channel inversion. At the same time, it generates leakage risk assessment results based on the corrected hydrogeological model results and the inversion results of the inferred leakage channels.

2. The real-time monitoring and assessment system for leakage risk of underground water reservoirs in coal mines according to claim 1, characterized in that: The water level triggering module includes a water level acquisition unit, a water level change acquisition unit, and a disturbance event generation unit; The water level acquisition unit acquires the current water level value and the reference water level value in the reservoir from the reservoir water level monitoring information according to a preset acquisition interval; The benchmark water level in the reservoir is the average value of the reservoir water level monitoring information within a preset water level benchmark time period prior to the current water level value. The water level change acquisition unit obtains the reservoir water level change amplitude based on the absolute value of the difference between the current reservoir water level and the benchmark reservoir water level. The disturbance event generation unit compares the magnitude of the water level change in the reservoir with the preset water level disturbance threshold in the preset water level disturbance conditions; When the magnitude of the water level change in the reservoir reaches the preset water level disturbance threshold, the disturbance event generation unit determines the collection time when the preset water level disturbance threshold is first reached as the water level disturbance start time, and generates the water level disturbance event result based on the water level disturbance start time and the magnitude of the water level change in the reservoir.

3. The real-time monitoring and assessment system for leakage risk of underground water reservoirs in coal mines according to claim 2, characterized in that: The response evidence collection module includes a time window determination unit, a response information extraction unit, and a response set generation unit; The time window determination unit uses the start time of the water level disturbance as the starting point of the unified response time window. The time window determination unit takes the time after the preset response duration following the start time of the water level disturbance as the end point of the unified response time window. The time window determination unit determines the unified response time window based on the start and end points of the unified response time window. The response information extraction unit extracts temperature response information, acoustic response information, and osmotic pressure response information from temperature monitoring information, acoustic wave monitoring information, and osmotic pressure monitoring information respectively within a unified response time window; The response set generation unit consists of a disturbance-related response set composed of temperature response information, acoustic response information, and osmotic pressure response information.

4. The real-time monitoring and assessment system for leakage risk of underground water reservoirs in coal mines according to claim 3, characterized in that: The channel inversion module includes a spatial modeling unit, a channel positioning unit, and a channel result generation unit; The spatial modeling unit obtains the spatial range of the water storage area, the spatial range of the coal pillar dam, the spatial range of the roof fracture zone, the spatial range of the floor water-proof boundary, and the spatial range of the outward boundary from the hydrogeological basic model. The spatial modeling unit divides candidate leakage units according to the spatial range of the coal pillar dam body, the spatial range of the roof fissure zone, the spatial range of the floor water-proof boundary, the spatial range of the outward boundary, and the preset spatial division scale. The spatial modeling unit determines the location of each candidate seepage unit based on the basic hydrogeological model. The spatial modeling unit defines the relationship between two candidate leakage units as having a shared boundary or the shortest distance between two candidate leakage units not exceeding a preset spatial division scale as the relationship between adjacent candidate leakage units. The spatial modeling unit determines the relationship between two candidate seepage units that have adjacent candidate seepage unit relationships as candidate seepage unit connectivity relationships if there is a water-conducting fracture connectivity record and no water-impeding boundary isolation record in the hydrogeological basic model. The candidate leakage unit locations and the subsequently acquired abnormal spatial locations use the same spatial coordinate reference.

5. The real-time monitoring and assessment system for leakage risk of underground water reservoirs in coal mines according to claim 4, characterized in that: The collaborative discrimination module includes an abnormal response generation unit, an abnormal result selection unit, and a consistency result generation unit; The abnormal response generation unit obtains the temperature residual value, acoustic energy value, and osmotic pressure change value based on the disturbance-related response set. The temperature baseline information is the average value of temperature monitoring information at the same monitoring location within a preset baseline time period; The seepage pressure baseline information is the average value of seepage pressure monitoring information at the same monitoring location within a preset baseline time period; The preset benchmark time period is located before the start of the unified response time window; The temperature residual value is the absolute value of the difference between the temperature monitoring information and the temperature reference information at the same monitoring location within a unified response time window. The acoustic energy value is the energy value of acoustic monitoring information at the same monitoring location within a preset acoustic frequency band within a unified response time window. The seepage pressure change value is the absolute value of the difference between the seepage pressure monitoring information and the seepage pressure reference information at the same monitoring location within a unified response time window; The abnormal response generation unit compares the temperature residual value, the sound wave energy value, and the osmotic pressure change value with the preset temperature residual threshold, the preset sound wave energy threshold, and the preset osmotic pressure change threshold, respectively. The abnormal response generation unit generates a temperature abnormal response result when the temperature residual value reaches a preset temperature residual threshold, generates an acoustic abnormal response result when the acoustic energy value reaches a preset acoustic energy threshold, and generates an osmotic pressure abnormal response result when the osmotic pressure change value reaches a preset osmotic pressure change threshold. The results of the temperature anomaly response, acoustic anomaly response, and osmotic pressure anomaly response all include the time of anomaly occurrence and the spatial location of the anomaly. The anomaly occurs when the temperature residual value, acoustic energy value, or osmotic pressure change value first reaches the corresponding threshold during data collection. The abnormal spatial location is the location of the monitoring point where the temperature residual value, sound wave energy value, or osmotic pressure change value first reaches the corresponding threshold under the same spatial coordinate reference.

6. The real-time monitoring and assessment system for leakage risk of underground water reservoirs in coal mines according to claim 5, characterized in that: The abnormal result selection unit selects at least two types of abnormal response results from the temperature abnormal response results, the sound wave abnormal response results, and the seepage pressure abnormal response results; The consistency result generation unit determines whether the occurrence times of the anomalies in at least two types of abnormal response results are both within a unified response time window. The consistency result generation unit determines, based on the abnormal spatial location and the candidate leakage unit location, whether the abnormal spatial locations of at least two types of abnormal response results are mapped to the same candidate leakage unit, or whether they are mapped to two candidate leakage units that have an adjacent candidate leakage unit relationship. When the occurrence times of at least two types of abnormal response results are all within a unified response time window, and the spatial locations of the abnormalities of at least two types of abnormal response results are mapped to the same candidate leakage unit or to two candidate leakage units that have an adjacent candidate leakage unit relationship, the consistency result generation unit generates a multi-source abnormality consistency discrimination result. The multi-source anomaly consistency discrimination result includes at least two types of anomaly response results, consistency candidate leakage units corresponding to the at least two types of anomaly response results, and the number of consistency candidate leakage units.

7. The real-time monitoring and assessment system for leakage risk of underground water reservoirs in coal mines according to claim 6, characterized in that: The channel positioning unit determines the set of presumed leakage initiation units and presumed leakage extension units based on the consistency candidate leakage units, the number of consistency candidate leakage units, and the connectivity of candidate leakage units in the multi-source anomaly consistency discrimination results. The channel positioning unit determines the shortest distance from the location of each consistent candidate leakage unit to the spatial range of the water storage area as the water storage area distance. The channel positioning unit determines the shortest distance from the candidate leakage unit position of each consistent candidate leakage unit to the outer boundary space range as the outer boundary distance; When the number of consistent candidate leakage units is one, the channel positioning unit determines the consistent candidate leakage unit as the presumed leakage initiation unit and determines the presumed leakage extension unit set as an empty set. When there are at least two consistent candidate leakage units, the channel positioning unit determines the consistent candidate leakage unit with the smallest distance from the water storage area as the presumed leakage initiation unit, and determines the consistent candidate leakage units other than the presumed leakage initiation unit that have a candidate leakage unit connection relationship with the presumed leakage initiation unit as the presumed leakage extension unit set. The channel positioning unit generates the inversion result of the estimated leakage channel based on the estimated leakage initiation unit, the estimated leakage extension unit set, and the connectivity of the candidate leakage units.

8. The real-time monitoring and assessment system for leakage risk of underground water reservoirs in coal mines according to claim 7, characterized in that: The channel result generation unit combines the estimated leakage initiation unit and the estimated leakage extension unit to form the candidate leakage unit where the estimated leakage channel is located. When the set of estimated leakage extension units is empty, the channel result generation unit determines the direction from the position of the candidate leakage unit of the estimated leakage initiation unit to the position in the outer boundary space range that is closest to the position of the candidate leakage unit of the estimated leakage initiation unit as the estimated leakage channel development direction. When the set of estimated leakage extension units is not empty, the channel result generation unit determines the estimated leakage extension unit with the smallest outward boundary distance in the set of estimated leakage extension units as the direction termination unit, and determines the direction from the estimated leakage start unit to the direction termination unit as the estimated leakage channel development direction. The channel result generation unit determines the number of abnormal response types based on the number of at least two types of abnormal response results involved in generating the multi-source abnormality consistency discrimination result. The channel result generation unit determines the number of channel units based on the number of candidate leakage units where the estimated leakage channel is located. The inversion results of the presumed leakage channels include the candidate leakage unit where the presumed leakage channel is located, the development direction of the presumed leakage channel, the number of abnormal response types, and the number of channel units.

9. The real-time monitoring and assessment system for leakage risk of underground water reservoirs in coal mines according to claim 8, characterized in that: The feedback assessment module includes a model feedback unit, a risk level determination unit, and an assessment result generation unit; The model feedback unit updates the connectivity state of the candidate seepage unit corresponding to the candidate seepage unit where the estimated seepage channel is located in the hydrogeological basic model to the estimated conduction state based on the inversion results of the estimated seepage channel. The presumed conduction state is used to indicate that the corresponding candidate leakage unit has been included in the presumed leakage channel inversion result; The model feedback unit writes the inferred development direction of the seepage channel and the connectivity of the candidate seepage units in the candidate seepage unit where the inferred seepage channel is located into the basic hydrogeological model. The model feedback unit generates hydrogeological model correction results based on the updated connectivity status of candidate seepage units, the presumed development direction of seepage channels after writing, and the connectivity relationship of candidate seepage units after writing. The model feedback unit uses the correction results of the hydrogeological model as the model basis for the channel inversion module to generate the inversion results of the inferred leakage channel inversion next time.

10. The real-time monitoring and assessment system for leakage risk of underground water reservoirs in coal mines according to claim 9, characterized in that: The risk level determination unit determines the leakage risk level based on the hydrogeological model correction results, the candidate leakage unit where the leakage channel is located, the estimated development direction of the leakage channel, the number of abnormal response types, and the number of channel units. When there are two abnormal response types and one channel unit, the risk level determination unit will determine the leakage risk level as the leakage warning level. When there are two abnormal response types and at least two channel units, the risk level determination unit will determine the leakage risk level as a continuous leakage level. When there are three abnormal response types and one channel unit, the risk level determination unit will determine the leakage risk level as a continuous leakage level. When there are three abnormal response types and at least two channel units, and the leakage channel is presumed to be developing towards the outward boundary space, the risk level determination unit determines the leakage risk level as the outward risk level. The assessment result generation unit generates leakage risk assessment results based on the leakage risk level, the candidate leakage unit where the estimated leakage channel is located, and the estimated development direction of the leakage channel. The leakage risk assessment results include leakage risk level, estimated leakage location, and estimated leakage path development direction. The estimated leakage location is determined by the candidate leakage unit where the estimated leakage path is located.

Images