Geological disaster risk management and control system and method based on multi-source sensing and dynamic decision

Abstract

The invention discloses a geological disaster risk management and control system and method based on multi-source sensing and dynamic decision. The management and control method comprises the following steps: acquiring multi-source data and a standard space-time axis; according to the standard space-time axis and the multi-source data, aligning the space-time axis of each data in the multi-source data; determining characteristic parameters of the data according to the data after time-space axis alignment; determining a dynamic risk index according to each characteristic parameter; determining a preliminary execution decision according to the dynamic risk index and each characteristic parameter; and correcting the preliminary execution decision based on Nash equilibrium, and determining a final execution decision. On the basis of multi-source data combination and a standard time-space axis, time and space alignment of the multi-source data is achieved, then feature parameters of all the data are obtained through a series of means, a preliminary execution decision is determined, finally, a final execution scheme is determined through Nash equilibrium correction, casualties and losses caused by geological disasters are avoided to the maximum extent, and the method is suitable for popularization and application. Accurate management and control are realized, and rapid early warning and timely disaster avoidance can be ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster monitoring and prevention, and in particular to a geological disaster risk management and control system and method based on multi-source sensing and dynamic decision-making. Background Art

[0002] Geological disasters are natural or human-induced hazards related to geological processes, such as landslides, mudslides, debris flows, ground subsidence, ground fissures, and subsidence, that endanger people's lives and property. With the rapid development of society and the economy, various engineering activities have a greater impact on the geological environment. Due to both natural and human factors, geological disasters will continue to occur frequently, resulting in significant casualties and property losses each year.

[0003] Currently, existing geological disaster monitoring methods have certain limitations. For example, traditional systems rely on single sensors such as displacement meters and rain gauges, which capture limited data and lack intelligent and timely risk assessment and control. Furthermore, traditional control methods struggle to integrate multi-source data for accurate analysis. Static warning thresholds are flawed, resulting in high false alarm rates and an inability to adapt to the dynamic evolution of geological conditions. This results in an inability to provide effective early warning and control measures before geological disasters occur, failing to meet the needs of disaster prevention and mitigation in the new era. Summary of the Invention

[0004] The main purpose of this invention is to provide a geological disaster risk management and control system and method based on multi-source sensing and dynamic decision-making, aiming to solve the problem that effective early warning and control cannot be made before geological disasters occur, and cannot meet the needs of disaster prevention and mitigation in the new era.

[0005] To achieve the above objectives, the present invention proposes a geological disaster risk management method based on multi-source sensing and dynamic decision-making, including:

[0006] Acquire multi-source data and standard space-time axes;

[0007] Aligning the spatiotemporal axes of the data in the multi-source data according to the standard spatiotemporal axis and the multi-source data;

[0008] Determining characteristic parameters of each data according to the data after the time-space axis alignment;

[0009] Determining a dynamic risk index based on each of the characteristic parameters;

[0010] Determining a preliminary execution decision based on the dynamic risk index and each of the characteristic parameters;

[0011] The preliminary execution decision is revised based on Nash equilibrium to determine the final execution decision.

[0012] Preferably, the multi-source data includes at least one of microseismic data, stress data, displacement data and environmental data.

[0013] Preferably, the formula for the time-space axis alignment is as follows:

[0014]

[0015] Among them, t sync is the synchronization time, t sensor is the time of the sensor, d is the distance between sensors, v p is the rock mass longitudinal wave velocity, Δt clock is the transmission delay time.

[0016] Preferably, the step of determining characteristic parameters of each data according to each data after the time-space axis alignment includes:

[0017] Decomposing the data aligned according to the time and space axes in layers by wavelet packet decomposition technology;

[0018] Then, the data of the hierarchical decomposition grooves are denoised by PCA dimensionality reduction to determine the characteristic parameters.

[0019] Preferably, the formula of the dynamic risk index is as follows:

[0020]

[0021] Among them, F i (t) is the i-th dimension characteristic parameter, which includes at least one of the characteristic parameters of microseismic data, displacement data, stress data, and environmental data; w i (t) is the dynamic weight of the feature parameter of the i-th dimension; F i,base is the benchmark value of the characteristic parameter; R(t) is the dynamic risk index.

[0022] Preferably, the formula for the dynamic weight of the characteristic parameter is as follows:

[0023]

[0024] Among them, α i is the characteristic attenuation coefficient; t is the time variable; e is a natural constant.

[0025] Preferably, the correction formula of the Nash equilibrium is:

[0026] U=0.8Q-0.2C+0.1E;

[0027] Among them, U is the multi-objective weight; Q is the safety performance index; C is the cost index; and E is the energy efficiency index.

[0028] Preferably, Q is iterated by a deep reinforcement learning algorithm, and the convergence condition of Q is:

[0029] max|Q k+1 (s,a)-Q k (s,a)|≤0.01;

[0030] Among them, S is the state space; α is the action space; Q K is the current safety performance index; Q K+1 Next safety performance indicator.

[0031] Preferably, the step of determining a preliminary execution decision based on the dynamic risk index and each of the characteristic parameters includes:

[0032] Determine whether the dynamic risk index is within the preset range;

[0033] When the dynamic risk index is less than the predetermined interval, a preliminary implementation plan is determined based on the third response and each of the characteristic parameters;

[0034] When the dynamic risk index is within a preset interval, determining a preliminary implementation plan based on the second response and each of the characteristic parameters;

[0035] When the dynamic risk index is greater than a preset interval range, a preliminary execution plan is determined based on the first response and each of the characteristic parameters.

[0036] To achieve the above-mentioned object, the present invention proposes a geological disaster risk management and control system based on multi-source sensing and dynamic decision-making, which applies any of the above-mentioned geological disaster risk management and control methods based on multi-source sensing and dynamic decision-making, and is characterized in that it includes a sensing module, a processing module and an execution module, wherein the processing module is respectively connected to the sensing module and the execution module by signal;

[0037] The sensing module is used to obtain multi-source data and standard space-time axis;

[0038] The processing module is configured to align the spatiotemporal axes of each data in the multi-source data according to the standard spatiotemporal axis and the multi-source data; determine characteristic parameters of each data according to the data after the spatiotemporal axis alignment; determine a dynamic risk index according to each characteristic parameter; determine a preliminary execution decision according to the dynamic risk index and each characteristic parameter; and modify the preliminary execution decision based on a Nash equilibrium to determine a final execution decision.

[0039] The execution module is used to execute the final execution decision.

[0040] Compared with the prior art, the present invention has at least the following beneficial effects:

[0041] Based on the combination of multi-source data and standard space-time axis, the time and space alignment of multi-source data is achieved to facilitate unified processing. Then, through a series of means, the characteristic parameters of each data in the multi-source data are obtained, and the preliminary execution decision is determined through the dynamic risk index. Finally, the final execution plan is determined through Nash equilibrium correction to avoid casualties and losses caused by geological disasters to the greatest extent, achieve precise management and control, and ensure rapid warning and timely disaster avoidance. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0043] Figure 1 1 is a flow chart of an embodiment of a geological disaster risk management method based on multi-source sensing and dynamic decision-making according to the present invention;

[0044] Figure 2 This is a structural framework diagram of an embodiment of a geological disaster risk management and control system based on multi-source sensing and dynamic decision-making according to the present invention;

[0045] Figure 3 It is a three-dimensional surface diagram for dynamic game decision making;

[0046] Figure 4 A decision flow chart for preliminary implementation.

[0047] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0048] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be understood as limiting the present invention.

[0049] The following describes a geological disaster risk management and control system and method based on multi-source sensing and dynamic decision-making according to an embodiment of the present invention with reference to the accompanying drawings.

[0050] Figure 1 This is a flow chart of an embodiment of geological disaster risk management based on multi-source sensing and dynamic decision-making of the present invention.

[0051] See also Figure 1 and Figure 2 To achieve the above objectives, the first embodiment of the present invention provides geological disaster risk management and control based on multi-source sensing and dynamic decision-making, including:

[0052] Step S10, obtaining multi-source data and standard space-time axis;

[0053] Step S20, aligning the spatiotemporal axes of the multi-source data according to the standard spatiotemporal axis and the multi-source data;

[0054] Step S30, determining characteristic parameters of each data according to each data after the time-space axis alignment;

[0055] Step S40, determining a dynamic risk index based on each characteristic parameter;

[0056] Step S50, determining a preliminary execution decision based on the dynamic risk index and various characteristic parameters;

[0057] Step S60: Modify the preliminary execution decision based on the Nash equilibrium to determine the final execution decision.

[0058] Based on the combination of multi-source data and standard space-time axis, the time and space alignment of multi-source data is achieved to facilitate unified processing. Then, through a series of means, the characteristic parameters of each data in the multi-source data are obtained, and the preliminary execution decision is determined through the dynamic risk index. Finally, the final execution plan is determined through Nash equilibrium correction to avoid casualties and losses caused by geological disasters to the greatest extent, achieve precise management and control, and ensure rapid warning and timely disaster avoidance.

[0059] In a second embodiment of the present invention, based on the first embodiment, the multi-source data includes at least one of microseismic data, stress data, displacement data, and environmental data.

[0060] Specifically, microseismic data are collected through a microseismic array; stress data and displacement data are collected separately through a fiber optic Bragg grating stress sensing network; and environmental data are collected through a multispectral environmental perception camera.

[0061] The microseismic monitoring array consists of 6-12 nodes in a hexagonal topology. This layout offers numerous advantages. The hexagonal topology ensures uniform node distribution, effectively covers the monitoring area, reduces blind spots, and improves monitoring accuracy and comprehensiveness. Its frequency response range is 0.1-200Hz, enabling it to capture microseismic signals of varying frequencies, accurately sensing both low-frequency vibrations and subtle high-frequency fluctuations. With a sensitivity of 5mV / g, this high sensitivity ensures that even extremely weak microseismic signals can be detected, providing reliable data support for early warning and subsequent analysis.

[0062] The fiber Bragg grating (FBG) stress sensing network utilizes a double-helix layout. This double-helix design enhances the network's stability and reliability, better adapts to complex environmental conditions, and improves its ability to sense stress changes. Its wavelength resolution is 1 pm, enabling precise detection of even minute wavelength shifts caused by stress changes. With strain measurement accuracy reaching ±0.1 με, this high-precision strain measurement provides accurate data for geological disaster safety assessments and timely identification of potential safety hazards.

[0063] The multispectral environmental perception camera includes three channels: visible light (550nm), near-infrared (850nm), and short-wave infrared (1550nm). Different spectral channels provide different information. The visible light channel can capture the visual characteristics of the target object, the near-infrared channel is important for detecting objects' reflective properties and moisture content, and the short-wave infrared channel has unique advantages in penetrating clouds and fog and detecting specific substances. By working together, these multiple channels can achieve comprehensive environmental perception, providing richer and more accurate information for subsequent decision-making.

[0064] In the third embodiment of the present invention, based on the second embodiment, the formula for time-space axis alignment is as follows:

[0065]

[0066] Among them, t sync is the synchronization time, t sensor is the time of the sensor, d is the distance between sensors, v p is the rock mass longitudinal wave velocity, Δt clock is the transmission delay time.

[0067] Temporal alignment ensures that data from different sources and of different types are consistent in the time dimension, providing an accurate time benchmark for subsequent analysis, which demonstrates its unique value.

[0068] Specifically, step S30 includes:

[0069] Step S31, obtaining standard three-dimensional coordinates, and unifying the coordinates of each data after aligning the time-space axis according to the standard three-dimensional coordinates;

[0070] Step S32: determining characteristic parameters of each data according to the data after the time-space axis is aligned and the coordinates are unified.

[0071] Realize the unification of the location of each data and achieve consistency of each data in the spatial dimension.

[0072] In a fourth embodiment of the present invention, based on any one of the first to third embodiments, step S30 includes:

[0073] Step S33, performing hierarchical decomposition on the data aligned according to the time-space axis using wavelet packet decomposition technology;

[0074] In step S34, PCA dimensionality reduction is used to remove noise from the data of the hierarchical decomposition grooves to determine the characteristic parameters.

[0075] Using wavelet packet decomposition technology and performing layered decomposition operations can deeply explore detailed features in the data, breaking down complex data signals into sub-signals of different frequencies, thereby extracting more valuable information. PCA dimensionality reduction is then used to select principal components, remove redundant information from the data, and reduce the data dimension. This improves the efficiency and accuracy of subsequent analysis while retaining key features.

[0076] In a fifth embodiment of the present invention, based on any one of the first to third embodiments, the formula of the dynamic risk index is as follows:

[0077]

[0078] Among them, F i (t) is the i-th dimension characteristic parameter, which includes at least one of the characteristic parameters of microseismic data, displacement data, stress data, and environmental data; w i (t) is the dynamic weight of the feature parameter of the i-th dimension; F i,base is the benchmark value of the characteristic parameter; R(t) is the dynamic risk index.

[0079] In the sixth embodiment of the present invention, based on the fifth embodiment, the formula for the dynamic weight of the characteristic parameter is as follows:

[0080]

[0081] Among them, α i is the characteristic attenuation coefficient; t is the time variable; e is a natural constant.

[0082] Dynamic weighting of characteristic parameters can dynamically assign weights based on real-time data changes and the importance of different characteristics, making the weight allocation more scientific and reasonable. Based on the dynamic weighting of characteristic parameters, combined with the dynamic risk index, and taking into account various factors, a comprehensive and accurate assessment of geological hazard risks can be conducted.

[0083] After a series of rigorous processing and analysis, the system outputs dynamic risk indices and corresponding instructions. These dynamic risk indices intuitively reflect the current risk status of the monitored object, while the corresponding instructions provide relevant personnel with clear response measures and decision-making basis, helping to take timely and effective action to reduce losses caused by geological disaster risks.

[0084] In the seventh embodiment of the present invention, based on any one of the first to third embodiments, the modified formula of Nash equilibrium is:

[0085] U=0.8Q-0.2C+0.1E;

[0086] Among them, U is the multi-objective weight; Q is the safety performance index; C is the cost index; and E is the energy efficiency index.

[0087] Before determining the final implementation decision, it is necessary to combine the actual situation and highlight the decision-making orientation of prioritizing safety and benefits.

[0088] In the eighth embodiment of the present invention, based on the seventh embodiment, Q is iterated by a deep reinforcement learning algorithm, and the convergence condition of Q is:

[0089] max|Q k+1 (s,a)-Q k (s,a)|≤0.01;

[0090] Among them, S is the state space; α is the action space; Q K is the current safety performance index; Q K+1 Next safety performance indicator.

[0091] In the process of solving Nash equilibrium, a Q-value iteration mechanism is constructed through deep reinforcement learning algorithm. By designing the state space s (including dimensions such as device status, environmental parameters, and emergency resources) and the action space a (corresponding to the three-party strategy combination), Q is updated in each round of iteration. k (s,a)-valued function.

[0092] The convergence condition is that when the maximum fluctuation amplitude of the Q value of all state-to-action pairs in two consecutive iterations is lower than the 1% threshold, the strategy stability is determined to be achieved. At this time, the optimal response strategy constitutes an equilibrium solution that cannot be unilaterally improved, providing quantitative support for dynamic decision-making in complex emergency scenarios.

[0093] In a first embodiment of the present invention, based on any one of the first to third embodiments, step S50 includes:

[0094] Step S51, determining whether the dynamic risk index is within a preset range;

[0095] Step S52: When the dynamic risk index is less than the predetermined range, a preliminary execution plan is determined based on the third response and the characteristic parameters;

[0096] Step S53: When the dynamic risk index is within the preset range, a preliminary implementation plan is determined based on the second response and the characteristic parameters;

[0097] Step S54: When the dynamic risk index is greater than the preset interval, a preliminary execution plan is determined based on the first response and each characteristic parameter.

[0098] Specifically, when the dynamic risk index is smaller than the predetermined range and the dynamic risk index is smaller than the warning value, step S10 is executed after a preset time period.

[0099] Among them, the warning value is smaller than the preset range.

[0100] See also Figure 3 and Figure 4 , set the preset interval range to 0.5≤R(t)<1, and the warning value to 0.5.

[0101] When R(t)<0.5, the conventional monitoring mode is maintained and multi-source data are collected at a cycle of 5 minutes.

[0102] When 0.5≤R(t)<0.7, based on the three-level response and various characteristic parameters, the laser-induced cavitation drainage system will be turned on as a preliminary execution decision to achieve rapid drainage and prevent water erosion of the rock and soil.

[0103] When 0.7≤R(t)<0.9, based on the secondary response and various characteristic parameters, the laser-induced cavitation drainage system will be turned on and the grouting pressure will be increased to 8 MPa as the initial execution decision to suppress the deterioration of the seepage field.

[0104] When R(t)≥0.9, based on the first-level response and various characteristic parameters, the laser-induced cavitation drainage system, intelligent grouting system, shape memory alloy support structure and emergency broadcast system will be simultaneously activated as the preliminary execution decision to achieve all-round active prevention and control of disaster risks.

[0105] Through the above-mentioned grading mechanism, the dynamic game decision-making engine can optimize the response strategy in real time, ensuring the precise matching of prevention and control measures with risk levels, with the probability of false action less than 2%.

[0106] To achieve the above-mentioned object, the present invention further discloses a geological disaster risk management and control system based on multi-source sensing and dynamic decision-making. The management and control system applies any of the above-mentioned geological disaster risk management and control methods based on multi-source sensing and dynamic decision-making, and includes a sensing module, a processing module and an execution module. The processing module is signal-connected to the sensing module and the execution module respectively.

[0107] The sensing module is used to obtain multi-source data and standard space-time axes;

[0108] The processing module is used to align the time and space axes of each data in the multi-source data according to the standard time and space axes and the multi-source data; determine the characteristic parameters of each data according to the data after the time and space axes are aligned; determine the dynamic risk index according to each characteristic parameter; determine the preliminary execution decision according to the dynamic risk index and each characteristic parameter; and modify the preliminary execution decision based on the Nash equilibrium to determine the final execution decision;

[0109] The execution module is used to execute the final execution decision.

[0110] Specifically, the sensing module includes a microseismic array, a fiber Bragg grating stress sensor network, and a multispectral environmental perception camera. Serving as the control system's "sensing antennae," the sensing module is primarily responsible for collecting various information related to the target object, providing a data foundation for subsequent analysis and decision-making. It is composed of multiple advanced sensing devices, each with unique functions and characteristics.

[0111] Specifically, the processing module includes a geological-specific computing chip and a dynamic game decision engine.

[0112] The processing module is the system's "intelligent brain," responsible for preliminary processing and analysis of the data collected by the sensor module and making appropriate decisions. It consists of an advanced computing chip and a decision-making engine.

[0113] The dedicated geological computing chip (Geo-Edge V3) boasts 4TOPS of computing power. This powerful capability enables rapid processing of large amounts of sensor data, ensuring the real-time and efficient operation of the management and control system. It supports INT8 quantitative reasoning, which effectively reduces computational complexity and power consumption while ensuring accuracy, improving chip efficiency. The built-in LSTM time series processing unit effectively processes and analyzes time series data, uncovering underlying patterns and trends within the data and providing strong support for subsequent decision-making.

[0114] The dynamic game decision engine is based on a multi-objective optimization algorithm based on Nash equilibrium. In complex geological environments, there are often multiple interrelated objectives and decision-making factors, requiring comprehensive consideration of all factors to reach the optimal decision. The Nash equilibrium algorithm finds a stable strategy combination among multiple players in the game, preventing each player from unilaterally changing their strategy to gain greater benefits while the strategies of other players remain fixed. By leveraging this multi-objective optimization algorithm based on Nash equilibrium, the dynamic game decision engine is able to make optimal decisions in complex environments, improving the overall performance and efficiency of the system.

[0115] Specifically, the execution module includes a laser-induced cavitation drainage system, an intelligent grouting system, and a shape memory alloy support structure.

[0116] The execution module is the "action executor" of the management and control system, responsible for performing corresponding operations and controls on the target object based on the decision results of the processing module.

[0117] The laser-induced cavitation drainage system uses high-energy pulsed lasers to induce cavitation within the rock and soil. The system then utilizes the microjets and shock waves generated by the collapse of cavitation bubbles to enhance the permeability of pore water seepage channels, thereby achieving rapid drainage. The core of the system is to focus the laser on the aquifer, instantly vaporizing water to form cavitation bubbles. These bubbles collapse under external pressure, generating localized high pressure (up to 100 MPa) and high-speed microjets (speeds >100 m / s). These microjets scour soil particles, expand seepage channels, and improve drainage efficiency.

[0118] The intelligent grouting system has a pressure range of 0-20 MPa, allowing for flexible adjustment of grouting pressure based on varying needs and geological conditions. With a flow control accuracy of ±1.5%, this high-precision flow control ensures stability and accuracy during the grouting process, preventing the effects of excessive or insufficient flow. Based on the decisions made by the processing module, the intelligent grouting system precisely injects slurry into designated locations, reinforcing and repairing geological hazards.

[0119] The shape memory alloy support structure has a phase transition temperature of 45±2°C. When the ambient temperature reaches the phase transition temperature, the shape memory alloy undergoes a phase transition, generating a recovery stress of 600MPa. This property enables the shape memory alloy support structure to automatically function under specific conditions, providing effective support and protection against geological disasters.

[0120] In summary, the three-level distributed architecture of "sensing module-processing module-execution module" realizes comprehensive monitoring, intelligent analysis and effective control of target objects through the collaborative work of various levels, providing strong support for safety assurance and decision-making in complex geological environments.

[0121] Through the collaborative innovation of multi-source fusion sensor networks, intelligent computing and dynamic game decision-making, three major effects of geological disaster risk management have been achieved: first, the early warning accuracy has been significantly improved, and the false alarm rate has been reduced. The spatiotemporal alignment algorithm based on the fusion of microseismic-stress-displacement-environment four-dimensional data can identify instability precursors in advance; second, the response time has broken through the limit, and the edge computing nodes have achieved local decision-making in seconds. Laser drainage, grouting pressure regulation and support structure triggering form a closed-loop control in seconds; third, the energy efficiency ratio is comprehensively optimized. The dynamic game model reduces the energy consumption of prevention and control, reduces the loss of grouting materials, and improves the efficiency of laser-induced cavitation drainage.

[0122] The examples are as follows:

[0123] 1. Site Selection

[0124] Monitoring area: Select a typical landslide body (length 200m × width 150m × thickness 10m), with a sliding surface inclination of 30°±5° and sandy siltstone as the rock type.

[0125] 2. Preliminary Arrangements

[0126] Configuration of the sensor module:

[0127] Microseismic array: 12 nodes are arranged along the main slip direction, with a hexagonal topology (side length 8m±0.5m), a node burial depth of 5m, a sensitivity of 5mV / g, and a sampling rate of 1kHz;

[0128] Fiber Bragg grating stress sensing network: 8 sets of sensing units arranged in a double helix (wavelength 1528-1565nm), with an inclination angle of 45° and a spatial resolution of 5cm;

[0129] Multispectral environmental perception camera: installed at a high point in the monitoring area, with three channels of visible light (550nm) / near infrared (850nm) / shortwave infrared (1550nm), and a resolution of 0.5m@100m.

[0130] Processing module configuration:

[0131] Geo-Edge V3 chip: deployed in the middle section of the landslide, with a built-in LSTM acceleration unit (processing latency < 200ms) and a storage capacity of 64GB;

[0132] Decision parameters: safety weight 0.8, cost weight 0.2, energy efficiency weight 0.1, game learning rate α = 0.01.

[0133] Configuration of execution module:

[0134] Intelligent grouting system: 6 plunger pumps (single flow rate 50L / min), pressure sensor accuracy ±0.3%;

[0135] Laser-induced cavitation drainage system: 8 emission holes (aperture 10 mm) and laser (pulse energy 150 mJ, frequency 50 Hz) are arranged along the sliding belt.

[0136] 3. Data Collection and Processing

[0137] Microseismic data: 153 microseismic events were captured per day, with an energy range of 1×10 3 -5×10J, calculated energy entropy change rate ΔE=0.12;

[0138] Stress data: Maximum principal stress gradient (critical threshold 2.0MPa / m);

[0139] Displacement data: Maximum acceleration d 2 u / dt 2 =5.2mm / h2 (Threshold 6.0mm / h 2 ).

[0140] Calculation process:

[0141] 1. Space-time axis alignment: compensation for wave velocity differences (vp = 1500m / s), time synchronization error < 1ms;

[0142] 2. Fusion of characteristic parameters in each data: Generate a 16-dimensional characteristic vector (including microseismic b-value 0.85 and stress anisotropy coefficient 0.32);

[0143] 3. Dynamic weight of characteristic parameters: When α = 0.05, the weight distribution w microseismic = 0.52, w stress = 0.31, and w displacement = 0.17;

[0144] 4. Dynamic risk index calculation: R(t) = 0.78 (triggering a secondary response).

[0145] 4. Execution layer operations

[0146] Secondary response measures:

[0147] Grouting control: pressure increased to 12 MPa (PID parameters Kp = 0.8, Ki = 0.05, Kd = 0.1), grouting volume 38m 3 ;

[0148] Laser drainage: Start 50Hz pulse mode, single hole drainage rate 12.5L / min, water content dropped 4.7% within 30 minutes;

[0149] Data feedback: stress gradient drops to R(t) dropped back to 0.41, and the system switched to routine monitoring.

[0150] 5. Effect Verification

[0151] index The measured value of this system Traditional system comparison value Improvement Early warning accuracy 94.2％ 63.5％ +48.3％ Response delay 0.75s 14.2s -94.7％ Grouting material loss <![CDATA[0.38m 3 / MPa]]> <![CDATA[0.65m 3 / MPa]]> -41.5％ Drainage energy consumption <![CDATA[0.68kWh / m 3 ]]> <![CDATA[2.15kWh / m 3 ]]> -68.4％

[0152] 6. Summary of Technical Advantages

[0153] This example verifies the three core advantages of the invention:

[0154] 1. Deep coupling of multi-source data: Through the fusion of microseismic, stress, displacement and laser modes, the precursor recognition rate is increased to 94.2%;

[0155] 2. Real-time decision-making: Local computing latency is less than 0.8 seconds, which is 18 times more efficient than cloud processing.

[0156] 3. Dynamic and precise prevention and control: Grouting-drainage coordinated control based on the game model reduces energy consumption for prevention and control by 68.4% and material loss by 41.5%.

[0157] Throughout this specification, references to terms such as "one embodiment," "another embodiment," "other embodiments," or "first to Xth embodiments" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. Throughout this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, method steps, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0158] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0159] The serial numbers of the embodiments of the present invention are for description only and do not represent the advantages and disadvantages of the embodiments. Through the description of the above implementation modes, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, a magnetic disk, an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0160] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are protected by the present invention.

Claims

1. A geological disaster risk management and control method based on multi-source sensing and dynamic decision-making, characterized by: include: Acquire multi-source data and standard space-time axes; Aligning the spatiotemporal axes of the data in the multi-source data according to the standard spatiotemporal axis and the multi-source data; Determining characteristic parameters of each data according to the data after the time-space axis alignment; Determining a dynamic risk index based on each of the characteristic parameters; Determining a preliminary execution decision based on the dynamic risk index and each of the characteristic parameters; The preliminary execution decision is revised based on Nash equilibrium to determine the final execution decision.

2. The geological disaster risk management and control method based on multi-source sensing and dynamic decision-making according to claim 1 is characterized in that: The multi-source data includes at least one of microseismic data, stress data, displacement data and environmental data.

3. The geological disaster risk management and control method based on multi-source sensing and dynamic decision-making according to claim 2 is characterized in that: The formula for the space-time axis alignment is as follows: Among them, t sync is the synchronization time, t sensor is the time of the sensor, d is the distance between sensors, v p is the rock mass longitudinal wave velocity, Δt clock is the transmission delay time.

4. The geological disaster risk management method based on multi-source sensing and dynamic decision-making according to any one of claims 1 to 3, characterized in that: The step of determining characteristic parameters of each data according to each data after the time-space axis alignment includes: Decomposing the data aligned according to the time and space axes in layers by wavelet packet decomposition technology; Then, the data of the hierarchical decomposition grooves are denoised by PCA dimensionality reduction to determine the characteristic parameters.

5. The geological disaster risk management and control method based on multi-source sensing and dynamic decision-making according to any one of claims 1 to 3, characterized in that: The formula for the dynamic risk index is as follows: Among them, F i (t) is the i-th dimension characteristic parameter, which includes at least one of the characteristic parameters of microseismic data, displacement data, stress data, and environmental data; w i (t) is the dynamic weight of the feature parameter of the i-th dimension; F i,base is the benchmark value of the characteristic parameter; R(t) is the dynamic risk index.

6. The geological disaster risk management and control method based on multi-source sensing and dynamic decision-making according to claim 5 is characterized in that: The formula for the dynamic weight of the characteristic parameter is as follows: Among them, α i is the characteristic attenuation coefficient; t is the time variable; e is a natural constant.

7. The geological disaster risk management and control method based on multi-source sensing and dynamic decision-making according to any one of claims 1 to 3, characterized in that: The modified formula of the Nash equilibrium is: U=0.8Q-0.2C+0.1E; Among them, U is the multi-objective weight; Q is the safety performance index; C is the cost index; and E is the energy efficiency index.

8. The geological disaster risk management and control method based on multi-source sensing and dynamic decision-making according to claim 7 is characterized in that: Q is iterated through the deep reinforcement learning algorithm, and the convergence condition of Q is: max|Q k+1 (s,a)-Q k (s,a)|≤0.01; Among them, S is the state space; α is the action space; Q K is the current safety performance index; Q K+1 Next safety performance indicator.

9. The geological disaster risk management and control method based on multi-source sensing and dynamic decision-making according to any one of claims 1 to 3, characterized in that: The step of determining a preliminary execution decision based on the dynamic risk index and each of the characteristic parameters includes: Determine whether the dynamic risk index is within the preset range; When the dynamic risk index is less than the predetermined interval, a preliminary implementation plan is determined based on the third response and each of the characteristic parameters; When the dynamic risk index is within a preset interval, determining a preliminary implementation plan based on the second response and each of the characteristic parameters; When the dynamic risk index is greater than a preset interval range, a preliminary execution plan is determined based on the first response and each of the characteristic parameters.

10. A geological disaster risk management and control system based on multi-source sensing and dynamic decision-making, wherein the geological disaster risk management and control system applies the geological disaster risk management and control method based on multi-source sensing and dynamic decision-making according to any one of claims 1 to 9, characterized in that: It includes a sensing module, a processing module and an execution module, wherein the processing module is respectively connected to the sensing module and the execution module by signals; The sensing module is used to obtain multi-source data and standard space-time axis; The processing module is configured to align the spatiotemporal axes of each data in the multi-source data according to the standard spatiotemporal axis and the multi-source data; determine characteristic parameters of each data according to the data after the spatiotemporal axis alignment; determine a dynamic risk index according to each characteristic parameter; and determine a preliminary execution decision according to the dynamic risk index and each characteristic parameter; Modify the preliminary execution decision based on Nash equilibrium to determine the final execution decision; The execution module is used to execute the final execution decision.

Images