Early warning method for separation type water inrush and sand inrush disasters based on multivariate information fusion

Abstract

The invention provides an early warning method for separation type water inrush and sand inrush disasters based on multivariate information fusion, and relates to the technical field of mine water disaster prevention and control. The method comprises the steps that multiple pieces of risk index information of each evaluation unit are determined, and the multiple pieces of risk index information comprise the separation layer development rate, the water pressure change rate, the sand grain starting probability, the water guiding fracture zone height and the water-resisting layer instability probability; determining the weight of each piece of risk index information by adopting an entropy weight method; and based on the multiple pieces of risk index information and the corresponding weights of each evaluation unit, determining a comprehensive risk value of each evaluation unit, and further determining early warning areas with separation type water inrush and sand inrush disaster risks in the to-be-monitored working face and a risk level of each early warning area. According to the method, collaborative sensing and deep fusion of multiple parameters in the whole separation type water and sand inrush disaster inoculation process are achieved, a comprehensive risk fusion mechanism based on the entropy weight method is constructed, the early warning accuracy and advancement are remarkably improved, mining activities are responded in real time, and the catastrophe precursor recognition capacity is remarkably enhanced.

Description

Technical Field

[0001] This disclosure relates to the field of mine water hazard prevention and control technology, and in particular to an early warning method for delamination-type water inrush and sand collapse disasters based on multi-source information fusion. Background Technology

[0002] With the increasing depth and intensity of coal resource development, the frequency of delamination-type water inrush and sand collapse disasters is on the rise. In these coalfields, the main coal seams are generally buried at depths exceeding 400 meters, with the overlying strata consisting primarily of interbedded sandstone and mudstone. Under the influence of mining, delamination easily occurs at the interface between hard sandstone and weak mudstone. As the delamination space expands and the overlying aquifer continuously replenishes the water supply, the coupling effect of water pressure and mining stress causes water-conducting fracture zones to penetrate the delamination area. This allows accumulated water to flow along channels and through weakly cemented rock strata, leading to the deterioration and disintegration of the rock strata into sand particles under water pressure and seepage. Ultimately, the sand particles are carried into the mining face by the water flow, triggering a roof delamination-type water inrush and sand collapse disaster. The amount of sand in this type of disaster can reach several thousand cubic meters, easily causing roadway blockage, equipment burial, and casualties. It is characterized by insidious precursors, strong suddenness, high destructiveness, and high difficulty in prediction. Monitoring of such disasters in related technologies mainly relies on traditional roof water hazard monitoring methods, such as water level and water pressure monitoring, combined with geological modeling for risk assessment. However, the disaster-causing mechanism of delamination-type water inrush and sand burst is significantly different from that of conventional roof water inrush, and existing monitoring methods are difficult to accurately monitor delamination-type water inrush and sand burst disasters. Summary of the Invention

[0003] This disclosure aims to at least partially address one of the technical problems in the related art.

[0004] Therefore, the first aspect of this disclosure proposes an early warning method for delamination-type water inrush and sediment collapse disasters based on multi-source information fusion, comprising: The working face to be monitored is divided into multiple evaluation units; Multiple risk indicators for each evaluation unit are determined, including: delamination development rate, water pressure change rate, sand grain initiation probability, water-conducting fracture zone height, and aquitard instability probability. The weight of each risk indicator information is determined using the entropy weight method. Based on multiple risk indicator information of each evaluation unit and the weight of each risk indicator information, the comprehensive risk value of each evaluation unit is determined; Based on the comprehensive risk value of each evaluation unit, the warning areas for the risk of delamination-type water inrush and sand collapse disaster in the monitored working face and the risk level of each warning area are determined.

[0005] The second aspect of this disclosure proposes an early warning device for delamination-type water inrush and sediment collapse disasters based on multi-source information fusion, comprising: The partitioning module is used to divide the working face to be monitored into multiple evaluation units; The first determining module is used to determine multiple risk indicator information for each of the evaluation units. The risk indicator information includes: delamination development rate, water pressure change rate, sand grain initiation probability, water-conducting fracture zone height, and water-impermeable layer instability probability. The second determining module is used to determine the weight of each of the risk indicator information using the entropy weight method; The third determining module is used to determine the comprehensive risk value of each evaluation unit based on multiple risk indicator information of each evaluation unit and the weight of each risk indicator information; The fourth determination module is used to determine the early warning area and the risk level of each early warning area in the working face to be monitored based on the comprehensive risk value of each evaluation unit.

[0006] A third aspect of this disclosure provides an electronic device, including: a processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method described in the first aspect above.

[0007] This publicly disclosed method for early warning of delamination-type water inrush and sand collapse disasters based on multi-source information fusion integrates multi-source monitoring data from both static and dynamic perspectives. It achieves collaborative perception and deep fusion of multiple parameters throughout the entire process of delamination-type water inrush and sand collapse disaster development, significantly enhancing the ability to identify precursors of the disaster. Driven by real-time mining data and based on static geological and engineering parameters, it comprehensively identifies key risk indicators such as dynamic development of delamination, water accumulation and pressure, evolution of mining-induced fracture fields, deterioration of weakly cemented rock strata, and sand particle initiation. A comprehensive risk fusion mechanism based on the entropy weight method is constructed, significantly improving the accuracy and predictability of early warnings. It provides real-time responses to mining activities such as mining progress and dynamic expansion of mining-induced fractures, greatly reducing the false alarm and missed alarm rates.

[0008] Additional aspects and advantages of this disclosure will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this disclosure. Attached Figure Description

[0009] The above and / or additional aspects and advantages of this disclosure will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, in which: Figure 1 A schematic diagram of a process for early warning of delamination-type water inrush and sand collapse disaster based on multi-source information fusion, provided for embodiments of this disclosure; Figure 2This is a schematic diagram illustrating the process of generating a real-time dynamic risk cloud map of a working face, as provided in an embodiment of this disclosure. Figure 3 A schematic diagram of a monitoring and early warning system architecture for delamination-type water inrush and sand breaching provided in an embodiment of this disclosure; Figure 4 A flowchart illustrating the operation of a monitoring and early warning system provided in this embodiment of the present disclosure; Figure 5 This is a schematic diagram of an early warning device for delamination-type water inrush and sand collapse disaster based on multi-source information fusion, provided in an embodiment of this disclosure. Detailed Implementation

[0010] Embodiments of this disclosure are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this disclosure, and should not be construed as limiting this disclosure.

[0011] Specifically, the following describes an early warning method for delamination-type water inrush and sand collapse disasters based on multi-source information fusion, according to an embodiment of the present disclosure, with reference to the accompanying drawings.

[0012] Figure 1 This is a flowchart illustrating an early warning method for delamination-type water inrush and sediment collapse disasters based on multi-source information fusion, provided as an embodiment of this disclosure. Figure 1 As shown, the early warning method for delamination-type water inrush and sediment collapse disasters based on multi-source information fusion may include the following steps: Step 101: Divide the working surface to be monitored into multiple evaluation units.

[0013] Step 102: Determine multiple risk indicator information for each evaluation unit. The risk indicator information includes: delamination development rate, water pressure change rate, sand grain initiation probability, water-conducting fracture zone height, and aquitard instability probability.

[0014] This disclosed embodiment constructs risk indicators from multiple dimensions, including dynamic development of delamination (the generation and expansion of delamination space at the interface between hard and weak rock layers under mining action), water supply and water accumulation and pressure (the overlying aquifer continuously supplies water to the delamination cavity, leading to increased water pressure), water-conducting fracture penetration (mining fracture zones develop upward and penetrate the delamination area, forming water-conducting channels), and instability initiation of weakly cemented rock layers (weakly cemented rock layers disintegrate under the action of water flow scouring and seepage pressure, and sand particles are initiated and collapse into the well with the water flow). It comprehensively captures key information on the multi-physical field co-evolution during the incubation process of delamination-type water inrush and sand collapse, and integrates multi-dimensional dynamic monitoring data to accurately capture the precursor information of delamination-type water inrush and sand collapse disasters.

[0015] In some embodiments of this disclosure, the delamination development rate of the evaluation unit can be determined by the following steps S11-S14: S11, acquire the distributed fiber optic sensing data, number of microseismic events, and cumulative energy of the evaluation unit.

[0016] Optionally, in one implementation, distributed optical fibers can be deployed along the roof boreholes and key rock strata in the roadways to construct a distributed optical fiber sensor network. This network can monitor formation strain in real time and continuously, locate areas of abnormal formation deformation such as delamination development, rock strata bending, and accelerated subsidence zones, and quantify the delamination development rate. Additionally, a microseismic sensor network can be deployed in key rock strata around the working face to monitor the location, energy, and frequency of mining-induced microseismic events in real time. This information can be used to characterize the spatiotemporal evolution of the mining fracture field (water-conducting fracture zone) and quantify the degree of penetration of water-conducting channels.

[0017] S12, the strain abrupt change region of the evaluation unit based on distributed optical fiber sensing data.

[0018] As an example, a strain rate threshold can be set. For example, 0.05% / min, when the strain rate of a certain region in the evaluation unit is... When this occurs, it is marked as a strain abrupt change region (or it may also be called a potential delamination region).

[0019] S13, within the strain abrupt change region, if per unit time If the number of internal microseismic events is greater than or equal to the first threshold and the cumulative energy is greater than or equal to the second threshold, then the corresponding strain abrupt change region is identified as an active delamination region.

[0020] As an example, the first threshold It can be set to 3 times, the second threshold. Can be set to J.

[0021] S14, calculate the area of ​​the active abscission zone, and perform differential calculation on the rate of change of the area of ​​the active abscission zone over time to obtain the abscission development rate of the evaluation unit.

[0022] As an example, the area of ​​the active exoplane region can be calculated using integration or image processing algorithms. , The rate of change of the abscission area over time was calculated using differential calculations to obtain the abscission development rate. .

[0023] By collecting distributed optical fiber strain data, microseismic event location and energy data, and rock strata mechanical parameters, the dynamic development process of the delamination space is quantified and the delamination development rate is inverted.

[0024] In some embodiments of this disclosure, the water pressure change rate of the evaluation unit can be determined by the following steps S21-S23: S21, collect historical water pressure time series data and mining intensity index information of the evaluation unit.

[0025] Optionally, the mining intensity index information may include the advance rate. Distance between the nearest microseismic sensor and the microseismic event Among them, the microseismic sensor is a component of the microseismic sensor network pre-deployed underground.

[0026] S22, input historical water pressure time series data and mining intensity index information into a pre-trained LSTM neural network model to obtain water pressure prediction data for multiple future time steps. The LSTM neural network model has learned the ability to predict future water pressure based on historical water pressure time series data and mining intensity index information.

[0027] The LSTM time series forecasting algorithm is used to predict the trend of water pressure changes in the near future and to calculate the rate of change of water pressure. As an example, historical water pressure time series data can be used... Information on mining intensity indicators Both are used as input. The model output is the future. Predicted values ​​at each time step The model needs to be trained using historical data to minimize prediction error.

[0028] S23, determine the water pressure change rate of the evaluation unit based on water pressure prediction data.

[0029] Using the prediction results, the instantaneous rate of change of water pressure is calculated. In the formula The rate of change directly reflects the swiftness of water pressure changes and is a key indicator for early warning, as it monitors the time step.

[0030] In some embodiments of this disclosure, based on fluid dynamics theory and critical initiation conditions, and combined with the monitored hydraulic gradient, water flow velocity, and particle size distribution characteristics of the weakly cemented rock layer, the probability of the weakly cemented rock layer deteriorating under the action of water flow to form a sand source and initiate initiation is determined. An initiation function is constructed based on the Shields Criterion, and the sand grain initiation probability of the evaluation unit can be determined through the following steps S31-S33: S31, obtain the hydraulic gradient of the evaluation unit.

[0031] S32, calculates the shear force of the water flow based on the hydraulic gradient. Refer to the following formula:

[0032] in, The density of water, It is the acceleration due to gravity. Porosity For hydraulic gradient, is the permeability coefficient.

[0033] S33, determine the sand particle initiation probability of the evaluation unit based on the water flow shear force, critical shear force and preset initiation function.

[0034] The critical shear force is expressed as follows:

[0035] in, It is the critical Shields number (related to the particle Reynolds number, usually taken as 0.03-0.06). For sand particle density, The particle size is denoted as 'Sand Particle Size'.

[0036] Define the starting function and calculate the probability. , Start at the designated time. To handle uncertainty, [the system will]... If we consider it as a random variable, then the starting probability can be defined as: Alternatively, smoothing functions such as the Logistic function can be used for mapping: In the formula, Probability of sand particles being activated. The steepness coefficient controls the rate at which the function transitions from 0 to 1. This model combines mechanical principles with probabilistic assessment, outputting a continuous risk probability value to facilitate system-wide decision-making.

[0037] In some embodiments of this disclosure, a multivariate nonlinear regression model can be used to dynamically predict the upward development height of the water-conducting fracture zone under the influence of mining, combined with data such as microseismic energy, rock mechanics parameters, and working face advance speed. The height of the water-conducting fracture zone in the evaluation unit can be determined through the following steps S41-S42: S41, determine the working face advance speed and microseismic energy release rate of the evaluation unit.

[0038] S42 uses a multivariate nonlinear regression model, combined with the working face advance speed and microseismic energy release rate to dynamically predict the height of the upward development of the water-conducting fracture zone under the influence of mining, and obtains the height of the water-conducting fracture zone in the evaluation unit.

[0039] The prediction model for the development height of water-conducting fracture zones is dynamically corrected based on empirical formulas. In the formula, For regression coefficients, For mining, For lithological combination coefficient, To improve the working face advance speed, This represents the energy release rate of microseismic events.

[0040] Dynamic correction function Introducing real-time mining disturbance factors: The complete formula is: In the formula, The height of the water-conducting fracture zone development, These are the fitting coefficients determined through multiple nonlinear regression analysis based on the geological and mining conditions of the target mining area. This represents the lithological combination coefficient.

[0041] In some embodiments of this disclosure, a method combining limit state functions and reliability theory can be used to assess the instability probability of the aquitard in the risk assessment unit under mining stress and water pressure. This can be determined through the following steps S51-S53: S51, determine the water pressure below the aquitard, the water pressure above the aquitard, and the mining stress of the evaluation unit.

[0042] S52, based on the water pressure below the aquitard, the water pressure above the aquitard, the mining stress, and the preset limit state function, the stability state of the aquitard in the evaluation unit is determined.

[0043] Define the limit state function: In the formula, To enhance the resistance to damage of the waterproof layer, The total load is This is the stress coefficient caused by mining. The effective thickness of the waterproof layer, For the water pressure below the waterproof layer, For the water pressure above the waterproof layer, For rock mass compressive strength, This is the stress caused by mining.

[0044] S53, when the stable state of the aquitard is the instability of the aquitard, determine the probability of instability of the aquitard in the evaluation unit.

[0045] when This indicates that the waterproof layer is stable. The time indicates the instability of the impermeable layer. The instability probability is calculated, taking into account the uncertainty of the parameters. and Treating it as a random variable, the probability of the aquitard becoming unstable is: If we assume and If all follow a normal distribution, then In the formula, As a reliability indicator, The cumulative function of the standard normal distribution. and These are the mean and standard deviation, respectively.

[0046] Step 103: Use the entropy weight method to determine the weight of each risk indicator information.

[0047] The entropy weight method determines the weight of each indicator based on the amount of information contained in its observed values. The greater the dispersion of the indicator data sequence, the smaller its entropy value, and the greater the amount of information it contains; therefore, the higher the weight of that indicator. In one implementation, the weight of each risk indicator can be determined through the following steps: (1) Constructing the original evaluation matrix It has There are 1 evaluation unit, and each evaluation unit has 10 evaluation units. Several evaluation indicators (i.e., the delamination development rate, water pressure change rate, sand grain initiation probability, water-conducting fracture zone height, and aquitard instability probability in step 102) are used to form the original data matrix. : ,in, , , Representing the The first evaluation unit The observed values ​​of the indicators.

[0048] (2) Data standardization (normalization) To eliminate the influence of different dimensions and magnitudes of the indicators, the original matrix is ​​normalized. Since the delamination development rate, water pressure change rate, sand grain initiation probability, water-conducting fracture zone height, and aquitard instability probability are all positive indicators (the higher the value, the higher the risk), the following formula can be used to normalize the data: The standardized matrix is ​​obtained. To prevent undefined cases when calculating logarithms, all... Shift by a minimum value (like ).

[0049] (3) Calculate the first The first item under the indicator Contribution of each evaluation unit Treating the standardized values ​​as probabilities, calculate the weight of each indicator value in the indicator series. : ; Calculate the first Entropy value of the item index ,in, As a normalization constant, ensure .

[0050] (4) Calculate the first Difference coefficients and weights of indicators Coefficient of difference: , No. Entropy value of the item index The smaller the coefficient of variation, the higher the coefficient of variation. The larger the value, the more valuable information the indicator provides to decision-makers, and the higher its weight should be.

[0051] Weighting: Normalizing the difference coefficients yields the objective weights of each indicator. :

[0052] Step 104: Based on the multiple risk indicator information of each evaluation unit and the weight of each risk indicator information, determine the comprehensive risk value of each evaluation unit.

[0053] Using the obtained weights, calculate the evaluation unit for each evaluation unit. Comprehensive risk score In the formula, It is between The closer the scalar value is to 1, the higher the risk of that unit.

[0054] Step 105: Based on the comprehensive risk value of each evaluation unit, determine the early warning areas in the working face to be monitored that have the risk of delamination-type water inrush and sand collapse disasters, and the risk level of each early warning area.

[0055] In some embodiments of this disclosure, the overall risk value among multiple evaluation units may be greater than or equal to a first risk threshold. The evaluation unit was identified as the warning area.

[0056] Furthermore, when the comprehensive risk value of the evaluation unit Greater than or equal to the first risk threshold And less than the second risk threshold At that time, the corresponding evaluation unit is determined to be at a low-risk level (yellow warning). When the comprehensive risk value of the evaluation unit Greater than or equal to the second risk threshold And less than the third risk threshold At that time, the corresponding evaluation unit was determined to be at a medium-risk level (orange alert). When the comprehensive risk value of the evaluation unit Greater than or equal to the third risk threshold At that time, the corresponding evaluation unit is determined to be at a high-risk level (red alert).

[0057] Optionally, the first risk threshold Second risk threshold and the third risk threshold The risk value can be determined by clustering analysis of historical disaster data, engineering experience, or by using clustering algorithms (such as K-Means) for specific mining areas.

[0058] In one implementation, after determining the warning area and the risk level of each warning area, a real-time dynamic risk cloud map covering the entire evaluation unit can be automatically generated periodically to intuitively display the risk level and provide a basis for intelligent decision-making and automatic response. Figure 2 This is a schematic diagram illustrating the process of generating a real-time dynamic risk cloud map of a working face, as provided in an embodiment of this disclosure.

[0059] In some embodiments of this disclosure, the embodiments may further include a graded response mechanism linkage control. When there are low-risk evaluation units among multiple evaluation units, the downhole drainage system is automatically linked and controlled, such as increasing drainage capacity and activating backup drainage devices. When there are medium-risk evaluation units among multiple evaluation units, the downhole drainage system is automatically linked and controlled, the water-tight sand gate is closed, and personnel are notified to evacuate. When there are high-risk evaluation units among multiple evaluation units, personnel are urgently evacuated from the well, the downhole drainage system is automatically linked and controlled, the water-tight sand gate is closed, the grouting system is used to grout and plug water in the monitoring face and reinforce the risk area, and the emergency rescue plan is activated. Existing systems rely on manual judgment and decision-making after issuing warnings, lacking an efficient and automated linkage mechanism with engineering control facilities such as drainage systems and water-tight sand gates, resulting in significant response delays. This disclosure establishes an intelligent push system for early warning information and automatic matching of prevention and control measures, thereby transforming the response from passive to proactive intervention. It also establishes a closed-loop management system covering the entire chain of "perception-analysis-decision-response," which systematically improves the mine's ability to monitor, warn, dynamically assess, and proactively prevent and control delamination-type water inrush and sand collapse disasters. This provides reliable technical support for mine safety production and enhances the timeliness and reliability of disaster emergency response.

[0060] Figure 3 This diagram illustrates a monitoring and early warning system architecture for delamination-type water inrush and sand collapse, as provided in this embodiment. The intelligent decision-making layer, acting as the central hub of the entire early warning system, achieves real-time risk assessment and tiered early warning through a cloud-edge collaborative intelligent decision-making architecture. Combined with an automated engineering linkage mechanism, it significantly shortens the response time from early warning to prevention and control, enhancing the initiative and timeliness of disaster response. It possesses multi-dimensional information integration, real-time assessment, and visualized decision support functions. (1) Multi-source information integration and visualization: This layer gathers real-time monitoring data from the perception layer and various risk assessment results output by the analysis layer (such as risk level, water-conducting fracture zone range, overburden failure mode, water pressure field dynamics, etc.), and dynamically displays the mining-induced fracture initiation, expansion and penetration process, as well as the spatiotemporal evolution trend of the risk area through a three-dimensional visualization platform, providing managers with intuitive decision-making basis.

[0061] (2) Edge computing real-time assessment: Deploy edge computing units at key nodes downhole to receive and process real-time monitoring data streams and preliminary results of hierarchical early warning models, realize localized, low-latency risk assessment and early warning signal generation, and ensure that core decision-making functions can still be maintained in harsh network environments.

[0062] Optionally, embodiments of this disclosure may further include intelligent disaster avoidance and escape route planning and emergency rescue decision support, dynamically generating and pushing optimal disaster avoidance and escape routes based on real-time risk distribution maps, tunnel topology, and personnel location information. After a disaster occurs, it provides auxiliary decision-making information such as disaster impact range simulation, trapped personnel location prediction, and rescue channel safety assessment.

[0063] By implementing the embodiments of this disclosure, multi-dimensional monitoring data are integrated from both static and dynamic perspectives, enabling collaborative perception and deep fusion of multiple parameters throughout the entire process of delamination-type water inrush and sand collapse disasters, significantly enhancing the ability to identify disaster precursors. Driven by real-time mining data and based on static geological and engineering parameters, a comprehensive assessment of key risk indicators such as dynamic development of delamination, water accumulation and pressure, evolution of mining-induced fracture fields, deterioration of weakly cemented strata, and sand particle initiation is achieved. A comprehensive risk fusion mechanism based on the entropy weight method is constructed, significantly improving the accuracy and predictability of early warnings, enabling real-time response to mining activities such as mining progress and dynamic expansion of mining-induced fractures, and greatly reducing the false alarm and missed alarm rates for disasters.

[0064] To better understand the early warning method for delamination-type water inrush and sand collapse disasters based on multi-source information fusion proposed in this disclosure, a specific illustration is provided using a mining face as an example. Figure 4 This is a flowchart illustrating the operation of a monitoring and early warning system provided in this embodiment. The working face used for the experiment has a burial depth of approximately 500 m, a mining height of 10.16 m, and a measured fracture-to-mining ratio of 23.31. The maximum height of the water-conducting fracture zone reaches 236.83 m, locally affecting the thick sandstone and conglomerate aquifers of the Luohe and Yijun Formations. Below this aquifer lies a weakly cemented rock layer, which is highly susceptible to delamination at the sand-mudstone interface under mining conditions. Adjacent working faces using a traditional water hazard monitoring system (monitoring only aquifer water level and pressure) experienced multiple incidents of delamination-induced water inrush carrying mud and sand during mining operations, indicating a high risk of delamination-type water inrush and sand collapse in this area.

[0065] The monitoring and early warning system of this invention is deployed at the working face. In the track roadway, transport roadway, and cut-off sections of the working face, 32 high-precision microseismic sensors are deployed at 20-meter intervals, forming a three-dimensional monitoring network covering key areas of the roof. This network is used to capture microseismic signals in real time indicating the initiation, expansion, and penetration of mining-induced fractures. Distributed fiber optic strain sensors (BOTDA) are laid in boreholes along the roof strike of the return airway to continuously and in real-time monitor the strain of key layers during mining, accurately identifying the location and dynamics of delamination development. In the area expected to be affected by the roof aquifer, a cross-shaped electrode array is set up to monitor the spontaneous potential and induced polarization effect generated during rock fracture and seepage channel formation in real time. Eight sets of high-precision water level gauges and water pressure gauges are deployed in the Luohe Formation aquifer and the underlying delamination, forming a hydrological monitoring network to capture sudden drops in water level and abnormal fluctuations in water pressure in real time.

[0066] To illustrate how the entropy weight method integrates the outputs of the five core models, this disclosure provides a computational example based on simulated data. The state of a high-risk area (an assessment unit) on the working face roof is assessed at different time series, and four sets of key risk indicators output by the core models at different times are selected.

[0067] (1) Constructing the original evaluation matrix Table 1 shows the original values ​​of risk indicators calculated by each core prediction model based on real-time monitoring data at four different times, forming the original evaluation matrix X. These data are derived from the comprehensive and normalized processing of historical monitoring data, laboratory test data, and numerical simulation results from a typical working face in the western Huanglong mining area, aiming to simulate and demonstrate a complete disaster evolution process. Among these, a negative water pressure change rate indicates a sharp drop in water pressure, a precursor to water inrush; however, the larger its absolute value, the higher the risk, so it is treated as a positive indicator in the calculation.

[0068] Table 1. Parameters of the Original Evaluation Matrix

[0069] (2) Data standardization processing: The original evaluation matrix X is normalized so that all risk indicators are positive indicators, and the standardized matrix R is obtained.

[0071] (3) Calculate the contribution of each indicator at each time point, and then calculate the entropy value and the difference coefficient, as shown in Table 2.

[0072] Table 2. Risk Indicator Contribution, Entropy Value, and Difference Coefficient

[0073] (4) Calculate the weight of each indicator The difference coefficients are normalized to obtain the final weights. (5) Calculate the comprehensive risk value at each time point according to the formula. As shown in Table 3. Based on the mine conditions, the risk threshold is set as low risk (yellow warning), 0. <0.35; Medium risk (orange alert), 0.35 0.70; High risk (red alert), 0.70 1.00.

[0076] Table 3. Overall Risk Values ​​and Risk Levels at Each Time Point

[0077] At times T1 and T2, All values ​​are less than 0.35, triggering a yellow alert. The system then activates the drainage system to increase drainage volume. At time T3, = 0.466 At 0:35, the system triggered an orange alert, immediately activating the drainage system to increase drainage volume, activating backup drainage pumps, closing and reinforcing waterproof sand doors, alerting relevant personnel to take precautions and increase monitoring frequency, and preparing for evacuation at any time; at time T4, = 0.738>0.70, the edge computing node completes real-time analysis within seconds, determining the current risk level to be high, and the system immediately triggers a red alert. The system urgently activates the drainage system to increase drainage volume and activates the backup drainage pump; it also activates the waterproof sand gate to close and reinforce it; and it starts the grouting system to inject grout to block water and reinforce the risk area; based on the real-time risk distribution map, tunnel topology, and personnel location information, the system dynamically generates and pushes the optimal disaster avoidance and escape route, and activates the emergency rescue plan.

[0078] This monitoring and early warning system is based on multi-source monitoring data analysis, and its early warning time and accuracy are superior to traditional water hazard monitoring and early warning systems. The system uses real-time risk assessment to issue early warning levels and automatically link corresponding prevention and control measures. Compared with manual decision-making and prevention and control response, the disaster early warning-prevention and control linkage response time is significantly shortened, which significantly improves the safety of mining face.

[0079] Figure 5 This is a schematic diagram of an early warning device for delamination-type water inrush and sandstorm disasters based on multi-source information fusion, provided as an embodiment of this disclosure. Figure 5 As shown, the early warning device for delamination-type water inrush and sand collapse disaster based on multi-source information fusion may include: a division module 501, a first determination module 502, a second determination module 503, a third determination module 504, and a fourth determination module 505.

[0080] The division module 501 is used to divide the working surface to be monitored into multiple evaluation units.

[0081] The first determining module 502 is used to determine multiple risk indicator information for each evaluation unit. The risk indicator information includes: delamination development rate, water pressure change rate, sand grain initiation probability, water-conducting fracture zone height, and water-impermeable layer instability probability.

[0082] The second determining module 503 is used to determine the weight of each risk indicator information using the entropy weight method.

[0083] The third determination module 504 is used to determine the comprehensive risk value of each evaluation unit based on multiple risk indicator information of each evaluation unit and the weight of each risk indicator information.

[0084] The fourth determination module 505 is used to determine the early warning area and the risk level of each early warning area in the working face to be monitored based on the comprehensive risk value of each evaluation unit.

[0085] In some embodiments of this disclosure, the first determining module 502 is specifically used for: acquiring distributed optical fiber sensing data, the number of microseismic events, and the cumulative energy of the evaluation unit; identifying strain abrupt change regions of the evaluation unit based on the distributed optical fiber sensing data; within the strain abrupt change region, if the number of microseismic events per unit time is greater than or equal to a first threshold and the cumulative energy is greater than or equal to a second threshold, then the corresponding strain abrupt change region is determined as an active delamination region; calculating the area of ​​the active delamination region, and performing differential calculation on the rate of change of the area of ​​the active delamination region over time to obtain the delamination development rate of the evaluation unit.

[0086] In some embodiments of this disclosure, the first determining module 502 is specifically used for: collecting historical water pressure time-series data and mining intensity index information of the evaluation unit; inputting the historical water pressure time-series data and mining intensity index information into a pre-trained LSTM neural network model to obtain water pressure prediction data for multiple future time steps, wherein the LSTM neural network model has learned the ability to predict future water pressure based on historical water pressure time-series data and mining intensity index information; and determining the water pressure change rate of the evaluation unit based on the water pressure prediction data.

[0087] In some embodiments of this disclosure, the first determining module 502 is specifically used to: obtain the hydraulic gradient of the evaluation unit; The shear force of the water flow is calculated based on the hydraulic gradient; the sand particle initiation probability of the evaluation unit is determined based on the shear force of the water flow, the critical shear force, and the preset initiation function.

[0088] In some embodiments of this disclosure, the first determining module 502 is specifically used to: determine the working face advance speed and microseismic energy release rate of the evaluation unit; and use a multivariate nonlinear regression model to dynamically predict the height of the upward development of the water-conducting fracture zone under the influence of mining, in combination with the working face advance speed and microseismic energy release rate, to obtain the height of the water-conducting fracture zone of the evaluation unit.

[0089] In some embodiments of this disclosure, the first determining module 502 is specifically used to: determine the water pressure below the aquitard, the water pressure above the aquitard, and the mining stress of the evaluation unit; determine the stable state of the aquitard of the evaluation unit based on the water pressure below the aquitard, the water pressure above the aquitard, the mining stress, and a preset limit state function; and determine the probability of aquitard instability of the evaluation unit when the stable state of the aquitard is aquitard instability.

[0090] In some embodiments of this disclosure, the fourth determining module 505 is specifically used to: determine the evaluation units among multiple evaluation units whose comprehensive risk value is greater than or equal to a first risk threshold as warning areas; when the comprehensive risk value of an evaluation unit is greater than or equal to the first risk threshold and less than a second risk threshold, determine the corresponding evaluation unit as a low-risk level; when the comprehensive risk value of an evaluation unit is greater than or equal to the second risk threshold and less than a third risk threshold, determine the corresponding evaluation unit as a medium-risk level; when the comprehensive risk value of an evaluation unit is greater than or equal to the third risk threshold, determine the corresponding evaluation unit as a high-risk level.

[0091] In some embodiments of this disclosure, the fourth determining module 505 is further configured to: automatically control the downhole drainage system when there is a low-risk evaluation unit among the multiple evaluation units; automatically control the downhole drainage system and close the water-tight sand gate when there is a medium-risk evaluation unit among the multiple evaluation units; and automatically control the downhole drainage system and close the water-tight sand gate when there is a high-risk evaluation unit among the multiple evaluation units, grout the working face to be monitored through the grouting system to plug water and reinforce the risk area, and activate the emergency rescue plan.

[0092] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0093] To implement the above embodiments, this disclosure also proposes an electronic device, including: a processor and a memory communicatively connected to the processor; the memory stores computer execution instructions; the processor executes the computer execution instructions stored in the memory to implement the method provided in the foregoing embodiments. To implement the above embodiments, this disclosure also proposes a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the methods provided in the foregoing embodiments.

[0094] To implement the above embodiments, this disclosure also proposes a computer program product, including a computer program that, when executed by a processor, implements the methods provided in the foregoing embodiments.

[0095] In the foregoing descriptions of the embodiments, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0096] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this disclosure, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0097] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of this disclosure includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of this disclosure pertain.

[0098] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0099] It should be understood that various parts of this disclosure can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0100] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0101] Furthermore, the functional units in the various embodiments of this disclosure can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0102] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of the present disclosure have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present disclosure.

Claims

1. An early warning method for delamination-type water inrush and sediment outburst disasters based on multi-source information fusion, characterized in that, Includes the following steps: The working face to be monitored is divided into multiple evaluation units; Multiple risk indicators for each evaluation unit are determined, including: delamination development rate, water pressure change rate, sand grain initiation probability, water-conducting fracture zone height, and aquitard instability probability. The weight of each risk indicator information is determined using the entropy weight method. Based on multiple risk indicator information of each evaluation unit and the weight of each risk indicator information, the comprehensive risk value of each evaluation unit is determined; Based on the comprehensive risk value of each evaluation unit, the warning areas for the risk of delamination-type water inrush and sand collapse disaster in the monitored working face and the risk level of each warning area are determined.

2. The method according to claim 1, characterized in that, The delamination development rate of the evaluation unit was determined in the following way: Acquire distributed fiber optic sensing data, number of microseismic events, and cumulative energy of the evaluation unit; Based on the distributed optical fiber sensing data, the strain abrupt change region of the evaluation unit is identified; Within the strain abrupt change region, if the number of microseismic events per unit time is greater than or equal to a first threshold, and the cumulative energy is greater than or equal to a second threshold, then the corresponding strain abrupt change region is determined as an active delamination region. The area of ​​the active abscission zone is calculated, and the rate of change of the area of ​​the active abscission zone over time is differentially calculated to obtain the abscission development rate of the evaluation unit.

3. The method according to claim 1, characterized in that, The rate of change of water pressure in the evaluation unit was determined in the following way: Collect historical water pressure time-series data and mining intensity index information for the evaluation unit; The historical water pressure time series data and mining intensity index information are input into a pre-trained LSTM neural network model to obtain water pressure prediction data for multiple future time steps. The LSTM neural network model has learned the ability to predict future water pressure based on historical water pressure time series data and mining intensity index information. The water pressure change rate of the evaluation unit is determined based on the water pressure prediction data.

4. The method according to claim 1, characterized in that, The sand particle initiation probability of the evaluation unit is determined in the following way: Obtain the hydraulic gradient of the evaluation unit; Calculate the flow shear force based on the hydraulic gradient; The sand particle initiation probability of the evaluation unit is determined based on the water flow shear force, critical shear force, and preset initiation function.

5. The method according to claim 1, characterized in that, The height of the water-conducting fracture zone in the evaluation unit was determined in the following way: Determine the working face advance speed and microseismic energy release rate of the evaluation unit; A multivariate nonlinear regression model was used to dynamically predict the height of the upward development of the water-conducting fracture zone under the influence of mining, based on the working face advance speed and the microseismic energy release rate, thus obtaining the height of the water-conducting fracture zone in the evaluation unit.

6. The method according to claim 1, characterized in that, The probability of instability of the impermeable layer in the evaluation unit is determined in the following way: Determine the water pressure below the aquitard, the water pressure above the aquitard, and the mining stress in the evaluation unit; Based on the water pressure below the aquitard, the water pressure above the aquitard, the mining stress, and the preset limit state function, the stability state of the aquitard in the evaluation unit is determined. When the stable state of the waterproof layer is the instability of the waterproof layer, the probability of the waterproof layer instability of the evaluation unit is determined.

7. The method according to claim 1, characterized in that, The determination of the early warning area and the risk level of each early warning area based on the comprehensive risk value of each evaluation unit includes: The evaluation unit whose comprehensive risk value is greater than or equal to the first risk threshold among the multiple evaluation units is identified as the early warning area; When the comprehensive risk value of an evaluation unit is greater than or equal to the first risk threshold and less than the second risk threshold, the corresponding evaluation unit is determined to be of low risk level. When the comprehensive risk value of an evaluation unit is greater than or equal to the second risk threshold and less than the third risk threshold, the corresponding evaluation unit is determined to be of medium risk level. When the comprehensive risk value of an evaluation unit is greater than or equal to the third risk threshold, the corresponding evaluation unit is determined to be at a high risk level.

8. The method according to claim 7, characterized in that, Also includes: When there is a low-risk evaluation unit among the multiple evaluation units, the downhole drainage system is automatically linked and controlled. When there is a medium-risk evaluation unit among the multiple evaluation units, the downhole drainage system is automatically linked to control the closure of the water-tight sand gate; When a high-risk evaluation unit is found among the multiple evaluation units, the downhole drainage system is automatically activated, the water-tight sand gate is closed, the grouting system is used to grout and plug the water in the working face to be monitored and reinforce the risk area, and the emergency rescue plan is activated.

9. An early warning device for delamination-type water inrush and sediment collapse disasters based on multi-source information fusion, characterized in that, include: The partitioning module is used to divide the working face to be monitored into multiple evaluation units; The first determining module is used to determine multiple risk indicator information for each of the evaluation units. The risk indicator information includes: delamination development rate, water pressure change rate, sand grain initiation probability, water-conducting fracture zone height, and water-impermeable layer instability probability. The second determining module is used to determine the weight of each of the risk indicator information using the entropy weight method; The third determining module is used to determine the comprehensive risk value of each evaluation unit based on multiple risk indicator information of each evaluation unit and the weight of each risk indicator information; The fourth determination module is used to determine the early warning area and the risk level of each early warning area in the working face to be monitored based on the comprehensive risk value of each evaluation unit.

10. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-8.

Images