Potential disaster intelligent sensing and emergency data engineering system driven by multi-modal data

Abstract

The invention discloses a multi-modal data-driven potential disaster intelligent sensing and emergency data engineering system, which relates to the technical field of intelligent information processing, and comprises a metasurface communication module for calculating a phase gradient between communication nodes by using a disaster risk probability graph and terrain height data, generating a beam forming matrix, performing low-power-consumption non-line-of-sight data transmission, and transmitting data to a data processing module; monitoring link quality and feeding back communication energy consumption, and generating an encrypted data packet; the cloud collaboration module is used for decrypting the encrypted data packet and carrying out three-dimensional disaster deduction to generate a disaster influence thermodynamic diagram; and the rescue decision module is used for carrying out path optimization on the disaster influence thermodynamic diagram by utilizing an improved path planning algorithm, generating an optimized rescue path and generating updated communication power and redundancy configuration parameters in combination with communication energy consumption data and returned topographic parameters. According to the invention, the reliability of emergency communication and the energy utilization efficiency are greatly improved, and the stability and redundancy of a communication link in a key area are ensured.

Description

Technical Field

[0001] The present invention relates to the field of intelligent information processing technology, and in particular to a multimodal data-driven potential disaster intelligent perception and emergency data engineering system. Background Art

[0002] In recent years, with the development of sensing technology, communication technology, and high-performance computing, the real-time monitoring and emergency response capabilities for natural disasters have been significantly improved. The emergence of multimodal heterogeneous data acquisition and fusion technology has enabled the integration of multi-source data from satellite remote sensing, ground sensor networks, unmanned aerial vehicle platforms, and meteorological radars in the same spatiotemporal benchmark, thus providing a richer information basis for disaster prediction and risk assessment. At the same time, continuous progress in geomechanical modeling and high-performance numerical solution methods has made it possible to simulate the evolution of disasters based on physical mechanisms; and emerging metasurface communication and beamforming technologies have provided new means for achieving highly reliable and low-power data transmission in complex terrain and harsh environments. In addition, the development of cloud-based collaborative computing and intelligent decision-making algorithms has provided a higher level of automation and intelligence for the full-link information processing and command and dispatch of disaster emergency response.

[0003] However, in existing technology systems, the spatiotemporal alignment of multimodal data, the energy-constrained calculation of multi-source physical models, and the adaptive optimization of communication links in dynamic disaster environments are often implemented independently, lacking real-time closed-loop collaboration within the same architecture. For example, in disaster simulation calculations, physical modeling and numerical solutions lack the ability to dynamically adapt to energy consumption, which can lead to reduced computing efficiency in energy-constrained environments. In communications, although beamforming technology can optimize link quality, parameter adjustments often fail to fully integrate the spatial distribution of disaster risks and energy status for comprehensive optimization. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a multimodal data-driven potential disaster intelligent perception and emergency data engineering system to solve the problems of insufficient coordination among multiple links and dynamic optimization of energy consumption.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: The present invention provides a multimodal data-driven potential disaster intelligent perception and emergency data engineering system, which includes a data processing module that collects multimodal heterogeneous data and performs spatiotemporal alignment processing to generate a standardized spatiotemporal matrix; an energy management module that monitors the energy state in real time according to the standardized spatiotemporal matrix, selects the calculation mode and drives the physical neural operator accelerator to output energy control instructions; a physical calculation module that uses three-dimensional geomechanical equations to perform analytical modeling, gridding, equation discretization and accelerated solution on the standardized spatiotemporal matrix under the constraints of energy control instructions, extracts the mechanical field distribution and calculates the instability probability to generate Disaster risk probability map; metasurface communication module, which uses the disaster risk probability map and terrain height data to calculate the phase gradient between communication nodes, generate a beamforming matrix, perform low-power non-line-of-sight data transmission, monitor link quality and feedback communication energy consumption, and generate encrypted data packets; cloud collaboration module, which decrypts the encrypted data packets and performs three-dimensional disaster simulation to generate a disaster impact heat map; rescue decision-making module, which uses an improved path planning algorithm to optimize the path of the disaster impact heat map, generate an optimized rescue path, and combine communication energy consumption data and return terrain parameters to generate updated communication power and redundancy configuration parameters.

[0007] As a preferred solution of the multimodal data driven potential disaster intelligent perception and emergency data engineering system of the present invention, wherein: the generating of the standardized spatiotemporal matrix is ​​carried out in the following specific steps: The dynamic time warping algorithm is used to synchronize the multimodal heterogeneous data and output a time-aligned data sequence. The time-aligned data sequence is converted into spatial coordinates in combination with RTK positioning information to generate spatiotemporal aligned data. The format of the spatiotemporal alignment data is standardized to form a standardized spatiotemporal matrix.

[0008] As a preferred solution of the multimodal data driven potential disaster intelligent perception and emergency data engineering system of the present invention, the output energy control instruction has the following specific steps: Based on the time and position synchronization information of the standardized space-time matrix, the output power and supercapacitor power are collected in real time through the energy monitoring sensor, and pre-processed to form energy state parameters; According to the energy state parameters, the current energy sufficiency and the task computing requirements are judged, the mode selection logic is performed to determine the matching computing mode, and the selected computing mode is encoded into a computing mode signal; The computing mode signal is converted into the corresponding energy control instruction, sent to the physical neural operator accelerator to drive the operation, and the energy control instruction is output.

[0009] As a preferred solution of the multimodal data driven potential disaster intelligent perception and emergency data engineering system of the present invention, the specific steps of generating the disaster risk probability map are as follows: Parse the standardized space-time matrix and energy control instructions, extract space-time field variables and solution configurations, map and update the three-dimensional computational grid, and generate a discrete grid model through spatial interpolation assignment; Apply initial mechanical conditions and boundary conditions to the discrete grid model, set the coupling equations, time step and solver parameters according to the solution configuration, and form a discretized set of equations; The discretized equations are loaded into the physical neural operator accelerator for numerical solution to extract the mechanical field distribution; The instability probability of the mechanical field distribution is calculated and spatially mapped to generate a disaster risk probability map.

[0010] As a preferred solution of the multimodal data-driven potential disaster intelligent perception and emergency data engineering system described in the present invention, the following specific steps are performed: parsing the standardized space-time matrix and energy control instructions, extracting space-time field variables and solution configurations, mapping and updating the three-dimensional computational grid, and generating a discrete grid model through spatial interpolation assignment. Parse the standardized space-time matrix and energy control instructions, extract space-time field variables and calculation requirement parameters, and output space-time field variables and solution configuration; Generate the initial grid structure based on the 3D calculation area and solution configuration, and output the 3D calculation grid; Map the space-time field variables to a three-dimensional computational grid, adjust the grid density based on the disaster risk probability map, mechanical field distribution, and energy state parameters of the previous computational cycle, and output the updated three-dimensional computational grid; Perform spatial interpolation on the updated three-dimensional computational grid, assign the space-time field variables to the grid nodes, and generate a discrete grid model.

[0011] As a preferred solution of the multimodal data driven potential disaster intelligent perception and emergency data engineering system of the present invention, the specific steps of generating the beamforming matrix are as follows: The terrain height data in the existing digital terrain model database is used in combination with the disaster risk probability map to form a terrain risk dataset; Using the terrain risk dataset, we calculate the spatial coordinate differences and height differences between communication nodes to obtain the distance and height difference matrix between nodes. The phase gradient of the communication node is calculated using the distance between nodes and the height difference matrix, and is mapped to the array elements to generate phase modulation values, which are combined to form a beamforming matrix.

[0012] As a preferred solution of the multimodal data driven potential disaster intelligent perception and emergency data engineering system of the present invention, the specific steps of generating the encrypted data packet are as follows: The beamforming matrix and the disaster risk probability map are used to transmit data and obtain the transmission status; According to the transmission status, the communication link quality is evaluated and the communication energy consumption is calculated to generate communication energy consumption data; The communication energy consumption data is used to encrypt the disaster risk probability map and generate an encrypted data packet.

[0013] As a preferred solution of the multimodal data driven potential disaster intelligent perception and emergency data engineering system of the present invention, the specific steps of generating the disaster impact heat map are as follows: Decrypt the encrypted data packet and output the decrypted disaster risk probability map; Map the decrypted disaster risk probability map to the discrete grid model nodes, initialize the initial values ​​of the mechanical field and spatial boundary constraints, perform three-dimensional disaster simulation, and generate preliminary disaster simulation data; The preliminary disaster simulation data is integrated with the meteorological and hydrological observation data collected by on-site sensors, and three-dimensional disaster simulation is performed again to generate a disaster impact heat map.

[0014] As a preferred solution of the multimodal data driven potential disaster intelligent perception and emergency data engineering system of the present invention, the specific steps of generating the optimized rescue path are as follows: Extract the risk level and obstacle information of the path planning area from the disaster impact heat map to generate path planning input data; Based on the path planning input data, the initial parameters and heuristic functions of the improved path planning algorithm are initialized, path search is performed, the cost of candidate paths is evaluated, the path selection is optimized, and a preliminary rescue path is generated; The preliminary rescue path is smoothed and risk adjusted to generate an optimized rescue path.

[0015] As a preferred solution of the multimodal data-driven potential disaster intelligent perception and emergency data engineering system of the present invention, wherein: the generating of the updated communication power and redundancy configuration parameters is carried out in the following specific steps: Extract the coordinates of high-risk areas from the optimized rescue path, trigger the correction of beamforming matrix control parameters, and generate corrected communication parameters; Monitor and collect communication link quality indicators, calculate link power consumption and energy consumption based on the corrected communication parameters, generate energy consumption feedback information, and perform fusion processing to generate comprehensive energy status data; The adaptive optimization algorithm is used to analyze the comprehensive energy status data, dynamically adjust the communication power allocation ratio and the redundancy setting of the beamforming matrix, and generate updated communication power and redundancy configuration parameters.

[0016] The beneficial effects of the present invention are: by real-time triggering of beamforming matrix control parameter correction, and combining link quality monitoring with adaptive optimization algorithms to dynamically adjust communication power and redundancy, a closed-loop optimization mechanism for disaster situation awareness, task requirements and communication resource management is formed, which greatly improves the reliability and energy utilization efficiency of emergency communications, ensures that communication links in key areas are stable and redundant, and saves resources in low-risk areas. It is suitable for emergency rescue scenarios with limited energy and complex communication environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 these drawings without paying any creative work.

[0018] Figure 1 Schematic diagram of the multimodal data-driven potential disaster intelligent perception and emergency data engineering system in the present invention.

[0019] Figure 2 Flowchart for generating disaster risk probability map in the present invention.

[0020] Figure 3 This is a flow chart of beamforming matrix generation in the present invention.

[0021] Figure 4 This is a flow chart for generating a disaster impact heat map in the present invention. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0025] Reference Figures 1 to 4, is an embodiment of the present invention, which provides a multimodal data-driven potential disaster intelligent perception and emergency data engineering system, including the following steps: The data processing module collects multimodal heterogeneous data, performs spatiotemporal alignment processing, and generates a standardized spatiotemporal matrix.

[0026] The dynamic time warping algorithm is used to synchronize the multimodal heterogeneous data and output the time-aligned data sequence. Combined with the RTK positioning information, the time-aligned data sequence is converted into spatial coordinates to generate spatiotemporal aligned data.

[0027] Specifically, multimodal heterogeneous data include meteorological sensor measurement data, hydrological sensor measurement data, geological sensor measurement data, and satellite remote sensing data; Using the dynamic time warping algorithm, the collected multimodal heterogeneous data are compared sequence by sequence according to the timestamp, and the time series of different data sources are nonlinearly stretched or compressed to align the features corresponding to each time point, and the time-aligned data series are output; The time-aligned data sequence is matched with the latitude, longitude, and elevation coordinates in the RTK positioning information, and the data of each time step is mapped to the corresponding spatial coordinates to form time-space aligned data, which contains time information and spatial coordinate information. For example, the temperature, humidity, and vibration data collected by different sensors in the same second are mapped to the same geographic location coordinates to achieve a unified representation of time and space.

[0028] The format of the spatiotemporal alignment data is standardized to form a standardized spatiotemporal matrix.

[0029] Specifically, the format of the spatiotemporal alignment data is standardized, including unit unification, numerical normalization, and missing value filling of the feature data of each time step, arranging the features of different data types in column order, and generating a unified data table structure; The time information, spatial coordinates, and various eigenvalues ​​are organized into a matrix form, so that the rows of the matrix correspond to the time steps and the columns correspond to the spatial positions and feature channels. For example, the temperature, humidity, vibration, and displacement data are arranged in a fixed order in the matrix columns to form a standardized space-time matrix.

[0030] The energy management module monitors the energy status in real time according to the standardized space-time matrix, selects the computing mode and drives the physical neural operator accelerator to output energy control instructions.

[0031] According to the time and position synchronization information of the standardized space-time matrix, the output power and supercapacitor power are collected in real time through the energy monitoring sensor, and preprocessed to form energy state parameters.

[0032] Specifically, based on the time and position synchronization information of the standardized space-time matrix, the output power and supercapacitor power are collected in real time through energy monitoring sensors, and the collected voltage, current and power signals are subjected to noise filtering and signal smoothing. For example, a moving average method or a weighted filtering method is used to process the continuous sampling data. The power and power at each moment are normalized and calculated according to the sampling time, and the units and numerical ranges are unified. The processed power values ​​and power values ​​are arranged according to the time step and spatial position to form energy state parameters containing the output power and supercapacitor power at each time step.

[0033] The current energy sufficiency and task computing requirements are judged according to the energy state parameters, the mode selection logic is performed to determine the matching computing mode, and the selected computing mode is encoded into a computing mode signal.

[0034] Specifically, the energy state parameters of each time step are compared with the preset output power threshold and supercapacitor power threshold to determine whether the current energy meets the load and energy storage requirements, thereby generating an energy sufficiency classification result; According to the task computing requirements, the computing task type is matched with the corresponding computing load. For example, high-load tasks are matched with high-energy states, and low-load tasks are matched with medium- and low-energy states. The computing mode matched for each time step is determined through the mode selection logic; the selected computing mode is converted into a computing mode signal according to preset coding rules, such as binary or multi-bit coding, and the computing mode signal sequence corresponding to each time step is output.

[0035] The specific steps for the output power threshold and supercapacitor power threshold should also be explained: based on the task computing requirements and energy supply capabilities, preset output power thresholds corresponding to the three energy states of high, medium and low, for example, the high energy state threshold is 80%, and the low energy state threshold is 20%; at the same time, based on the supercapacitor charging and discharging characteristics and continuous energy supply time requirements, preset supercapacitor power thresholds corresponding to the three energy states of high, medium and low, for example, the high energy state threshold is 80%, and the low energy state threshold is 20%; at each time step, the real-time collected output power and supercapacitor power are compared with the preset supercapacitor power threshold to determine whether the current energy state belongs to high, medium or low energy state, thereby providing a basis for subsequent calculation mode selection.

[0036] The preset coding rules establish a mapping relationship between each computing mode and the corresponding coding value based on the computing mode classification and task computing requirements in the standardized space-time matrix; the coding value can be expressed in binary or decimal, for example, the high-performance computing mode is encoded as "11", the energy-saving mode is encoded as "01", and the ordinary computing mode is encoded as "10"; the coding rules are derived from the analysis of the functional characteristics of the computing mode and the energy management strategy to ensure that the coding can uniquely identify each computing mode and adapt to the energy control instruction transmission and parsing requirements.

[0037] The calculation mode signal is converted into a corresponding energy control instruction, which is sent to the physical neural operator accelerator to drive operation, and the energy control instruction is output.

[0038] Specifically, the calculation mode signal is parsed according to a preset encoding rule and mapped to a parameter value corresponding to the energy control instruction, for example, the code "11" is converted into a high-performance calculation mode instruction, and the code "01" is converted into an energy-saving mode instruction; the mapped energy control instruction is sent to the physical neural operator accelerator through the control interface of the physical neural operator accelerator to drive operation, and the physical neural operator adjusts the calculation resource allocation and energy use strategy according to the received energy control instruction, completes the operation control, and outputs the energy control instruction.

[0039] It should be noted that the physical neural operator accelerator uses existing high-performance computing hardware and programmable accelerator technology, combined with neural network computing optimization scheme, to realize fast processing of complex numerical operation and equation solving through hardware-level parallel computing and instruction set optimization, relying on existing GPU, FPGA or ASIC accelerator platform, using mature hardware interface and control protocol to drive operation, without the need to build non-public computing devices, ensuring that the energy control instruction can be efficiently executed and the calculation result can be output.

[0040] The physical computing module uses three-dimensional geomechanics equations to analyze modeling, gridding, equation discretization and accelerated solving of the standardized space-time matrix under the constraint of the energy control instruction, extracts the mechanical field distribution and calculates the instability probability, and generates a disaster risk probability map.

[0041] It should be noted that the standardized space-time matrix of multi-modal heterogeneous data is closely combined with the energy control instruction, and based on the three-dimensional geomechanics equation, the analytical modeling, gridding and equation discretization solving are realized, and the mechanical field distribution and instability probability are efficiently extracted through accelerated calculation, so as to generate a refined disaster risk probability map, realizing the real-time closed-loop combination of data-driven physical modeling and disaster risk quantification, and significantly improving the accuracy and calculation efficiency of disaster risk prediction.

[0042] The beneficial effects are that by coupling the standardized space-time matrix of multi-modal heterogeneous data with the energy control instruction, using three-dimensional geomechanics equations for analytical modeling, gridding and discretization solving, the mechanical field distribution can be extracted in real time and high precision, and the instability probability can be calculated, so as to generate a disaster risk probability map. Unlike the existing technology which usually relies on a single data source or static empirical model for disaster risk assessment, the deep integration of multi-source data-driven, physical modeling and accelerated calculation is realized, which not only improves the accuracy and spatial resolution of disaster risk prediction, but also significantly reduces the calculation delay, realizes the real-time and dynamic adjustability of disaster risk quantification, and embodies the innovation in disaster intelligent perception and emergency data processing.

[0043] The standardized space-time matrix and energy control instructions are parsed, the space-time field variables and solution configurations are extracted, the three-dimensional computational grid is mapped and updated, and a discrete grid model is generated through spatial interpolation assignment.

[0044] Furthermore, the standardized space-time matrix and energy control instructions are parsed, the space-time field variables and calculation requirement parameters are extracted, and the space-time field variables and solution configuration are output.

[0045] Specifically, the time information, position coordinate information, meteorological sensor measurement data, hydrological sensor measurement data and geological sensor measurement data in the standardized space-time matrix are read, and the time information, position coordinate information, meteorological sensor measurement data, hydrological sensor measurement data and geological sensor measurement data are indexed and matched according to the corresponding relationship between the time and space coordinates; the energy control instructions are parsed, the calculation mode code and energy constraint information are extracted, and the information is combined with the task calculation requirement data in the standardized space-time matrix to calculate the computing resources and energy requirements required for each spatial node; according to the calculation requirement parameters and the node position, a solution configuration is generated, including the time step, equation type, boundary conditions and grid division scheme; and the space-time field variables and solution configuration corresponding to each spatial node are output.

[0046] Generates the initial mesh structure based on the 3D computational domain and solution configuration, and outputs the 3D computational mesh.

[0047] Specifically, the voxel grid of the three-dimensional calculation area is divided according to the spatial boundary information of the three-dimensional calculation area, the grid size, the time step in the solution configuration, the equation coupling parameters and the solver settings, and the grid nodes uniformly distributed along the X-axis, Y-axis and Z-axis are generated. The connection relationship between the nodes is established to form an initial grid structure, and the initial grid structure is output as the three-dimensional calculation grid.

[0048] The space-time field variables are mapped to a three-dimensional computational grid, and the grid density is adjusted according to the disaster risk probability map, mechanical field distribution, and energy state parameters of the previous computational cycle, and the updated three-dimensional computational grid is output.

[0049] Specifically, trilinear interpolation is performed spatially on the space-time field variables in the standardized space-time matrix according to the node coordinates of the three-dimensional computational grid, and the interpolation results are written into the grid node attributes of the three-dimensional computational grid; For each grid node of the three-dimensional computational grid, the risk value at the center of the disaster risk probability map of the previous computational cycle is obtained; the stress or displacement value of the mechanical field distribution at the grid node is obtained and the mechanical field gradient amplitude is calculated by differential calculation of adjacent grid nodes; and the output power and supercapacitor charge of the energy state parameter corresponding to the time step are read and the normalized value of the energy state parameter is calculated; Determine the local refinement or coarsening operation based on the risk value, the mechanical field gradient amplitude, and the normalized value of the energy state parameter. For example, when the risk value exceeds an example high risk standard or the mechanical field gradient amplitude exceeds an example high gradient standard and the energy state parameter normalized value exceeds an example high energy standard, mark it as refinement; when the risk value is lower than an example low risk standard and the mechanical field gradient amplitude is lower than an example low gradient standard or the energy state parameter normalized value is lower than an example low energy standard, mark it as coarsening; Perform local eight-division or two-division refinement on the cubic grid marked as refinement, insert new grid nodes and update the node connectivity relationship, merge adjacent cubic grids of the same level that are marked as coarsening and meet the merging conditions, remove redundant nodes, and adjust the level consistency of adjacent cubic grids to maintain a smooth transition of grid resolution; finally, output the updated three-dimensional computational grid.

[0050] Perform spatial interpolation on the updated three-dimensional computational grid, assign the space-time field variables to the grid nodes, and generate a discrete grid model.

[0051] Specifically, according to the node coordinates of the updated three-dimensional computational grid, spatial interpolation processing is performed on the space-time field variables in the standardized space-time matrix, and the sensor measurement data at each time step and position are mapped to the corresponding grid nodes; Existing three-dimensional interpolation methods, such as trilinear interpolation or Kriging interpolation, are used to calculate the field variable values ​​at the grid nodes, and the interpolated displacement, stress, temperature or measurement parameters are assigned to each grid node to generate a discrete grid model.

[0052] Apply initial mechanical and boundary conditions to the discrete mesh model, set the coupling equations, time step, and solver parameters according to the solution configuration, and form a discretized set of equations.

[0053] Specifically, according to the grid node information of the discrete grid model, the initial mechanical condition parameters and boundary condition parameters in the standardized space-time matrix are read, the initial displacement, velocity, stress and temperature are assigned to the grid nodes of the discrete grid model, and displacement constraints, force or pressure boundary conditions are applied to the boundary nodes of the discrete grid model; Based on the equation type, time step, and solver parameters in the solution configuration, the discretized three-dimensional geomechanical partial differential equations are assembled according to the grid node numbers. Combined with the initial conditions and boundary conditions, the mechanical coupling relationship between the nodes is established, forming a combination of stiffness matrix, mass matrix, and force vectors to obtain the coupled equations, and the discretized equation group is output for numerical solution.

[0054] It should also be explained that the solution configuration is derived from the standard numerical calculation method and engineering experience parameters for solving three-dimensional geomechanical equations, including the type of discretization method adopted, time step setting, solver selection, boundary conditions and initial condition parameters of the coupled equations, as well as the calculation accuracy and stability requirements obtained through experimental verification in similar geomechanical calculation tasks, to ensure that the numerical solution of the discretized equation group can meet the accuracy and efficiency of disaster risk probability calculation.

[0055] The discretized set of equations is loaded into the physical neural operator accelerator for numerical solution to extract the mechanical field distribution.

[0056] Specifically, the discretized equations are converted into a numerical input format recognizable by the physical neural operator accelerator according to the solver parameters and time step requirements, the discretized equations are loaded into the calculation program of the physical neural operator accelerator, and the numerical iterative solution is performed in the physical neural operator accelerator according to the calculation mode and parallel computing strategy. The space-time field variables of the grid nodes are periodically updated until the solution convergence conditions are met, and the mechanical field distribution of each grid node, such as node displacement, stress and strain values, is output.

[0057] It should also be explained that when iteratively solving the discretized set of equations in the physical neural operator accelerator, the changes in the mechanical field distribution of each grid node in the three-dimensional computational grid are continuously calculated to determine whether the changes have reached the preset convergence threshold. If the changes in the mechanical field distribution of all grid nodes meet the convergence threshold, the solution is considered complete; if the changes in the mechanical field distribution of any grid node do not meet the convergence threshold, the mechanical field distribution of the grid node is updated according to the time step and the next round of iteration is performed. The process is repeated until the changes in the mechanical field distribution of all grid nodes meet the convergence requirements, and finally the converged extracted mechanical field distribution is output; The preset convergence threshold is determined based on the accuracy of the mechanical solution, the stability of the discretized equation and the reliability of the disaster risk prediction. For example, the maximum allowable difference in the displacement or stress change of the grid node is set. Specifically, the accuracy requirements are determined based on the three-dimensional calculation grid resolution and time step, and the convergence threshold is set for sensitive areas in combination with the disaster risk probability map. In the iterative solution process, the change in the mechanical field distribution of each grid node is compared with the convergence threshold in real time to determine the convergence.

[0058] The instability probability of the mechanical field distribution is calculated and spatially mapped to generate a disaster risk probability map.

[0059] Specifically, the mechanical field distribution obtained from the discretized equations is analyzed node by node, and the node mechanical characteristic vector is formed based on the stress, strain and displacement data of the grid nodes. The node mechanical characteristic vector is compared with the preset instability judgment conditions one by one, and the instability probability is calculated node by node using the transfinite mapping method. The instability probability of the grid nodes is then arranged according to the node number to form a node instability probability matrix. The node instability probability matrix is ​​mapped to the spatial position coordinates of the three-dimensional computational grid, and a continuously distributed disaster risk probability map is generated through interpolation processing. The disaster risk probability map is then spatially smoothed and visualized to form a disaster risk probability map.

[0060] It should also be explained that the preset instability judgment conditions are determined based on existing geomechanics theory, experience in solving three-dimensional geomechanics equations and historical disaster observation data, including the strength limit, friction angle, cohesion and parameters of typical instability modes of rock and soil materials, combined with the mechanical field distribution characteristics and stress and strain critical values, obtained through on-site surveys and experimental measurement methods, to form a set of judgment conditions that can be used to calculate the instability probability of grid nodes.

[0061] The metasurface communication module uses disaster risk probability maps and terrain height data to calculate the phase gradient between communication nodes, generate a beamforming matrix, perform low-power non-line-of-sight data transmission, monitor link quality and provide feedback on communication energy consumption, and generate encrypted data packets.

[0062] The terrain height data in the existing digital terrain model database is used in combination with the disaster risk probability map to form a terrain risk dataset.

[0063] Specifically, terrain height data is extracted from the existing digital terrain model database, and the terrain height information of the corresponding area is screened according to the spatial range of the disaster risk probability map. The terrain height of each grid node is matched with the disaster risk probability of the corresponding node, and a terrain risk dataset is generated according to the grid node position. For example, the terrain height values ​​and disaster risk probability values ​​are formed into a two-dimensional table or a three-dimensional array for storage, and finally a terrain risk dataset is formed that can be used for subsequent communication node phase gradient calculation.

[0064] Using the terrain risk dataset, the spatial coordinate differences and height differences between communication nodes are calculated to obtain the distance and height difference matrix between nodes.

[0065] Specifically, according to the spatial coordinates and terrain height of each communication node in the terrain risk dataset, any two communication nodes are taken in turn. and , calculate the horizontal distance and height difference between nodes. The horizontal distance calculation expression is:

[0066] in, Represents a communication node Communication nodes The horizontal distance between Represents a communication node The spatial coordinates of The value of the axis direction, Represents a communication node The spatial coordinates of The value of the axis direction, Represents a communication node The spatial coordinates of The value of the axis direction, Represents a communication node The spatial coordinates of The value of the axis direction, Indicates the number of the first communication node participating in the calculation, Indicates the number of the second communication node participating in the calculation, and The value range of ,in is the total number of communication nodes; The height difference calculation expression is:

[0067] in, Represents a communication node Communication nodes The height difference between Represents a communication node The terrain height value, Represents a communication node The terrain height value; All communication nodes are combined to calculate the distance and height difference matrix between nodes .

[0068] The phase gradient of the communication node is calculated using the distance between nodes and the height difference matrix, and is mapped to the array elements to generate phase modulation values, which are combined to form a beamforming matrix.

[0069] Specifically, target communication link pairs are selected one by one according to the link identifier, and a link geometry information sequence is generated based on the horizontal distance and height difference of each link pair. Using the array reference element as the zero phase reference, the phase gradient increments along the horizontal and vertical directions of the array are calculated for each link according to the link geometry information sequence to obtain a phase gradient sequence. Mapping the phase gradient sequence to each array element according to the array element number, accumulating the phase increment element by element, and performing phase normalization and quantization (e.g., discretization according to a preset quantization bit width) to obtain a phase modulation value sequence of the array element; Finally, according to the two-dimensional index corresponding to the array element number and the link number, the array element phase modulation values ​​corresponding to each link are combined column by column to output the beamforming matrix.

[0070] The beamforming matrix and the disaster risk probability map are used for data transmission to obtain the transmission status.

[0071] Specifically, the disaster risk probability map is sliced ​​and packaged according to its grid index to generate disaster risk probability map data fragments; lossless compression and message encapsulation are performed on the disaster risk probability map data fragments, and timestamps and spatial coordinate tags are added to form a data stream to be sent; The beamforming matrix is ​​loaded at the transmitting end, and the baseband symbol stream corresponding to the data stream to be transmitted is precoded according to the beamforming matrix, and mapped according to the RF channel of the transmitting array; Channel coding and modulation (e.g., LDPC coding and QPSK modulation) are used to generate radio frequency signals, which are then transmitted via a wireless link. Timing and carrier synchronization, demodulation, and decoding are performed at the receiving end. The disaster risk probability map data slices are restored based on their sequence numbers and CRC checks, and the reception results are recorded. The valid frame count, frame loss count, number of retransmissions, average round-trip delay, and instantaneous throughput of the transmitter and receiver are counted, and the transmission status is summarized and generated by combining the beam number and transmit power records of the beamforming matrix.

[0072] According to the transmission status, the communication link quality is evaluated and the communication energy consumption is calculated to generate communication energy consumption data.

[0073] Specifically, data such as valid frame count, frame loss count, number of retransmissions, average round-trip delay, instantaneous throughput, beam number, and transmit power in the transmission status are read and analyzed. The frame success rate is calculated based on the valid frame count and frame loss count, and the communication link quality level is evaluated based on the average round-trip delay and instantaneous throughput. During the evaluation process, link performance differences are recorded and categorized by beam number. The total transmission energy is calculated using the transmission power, number of retransmissions, and transmission duration of each frame. The reception energy is calculated by combining the working duration of the receiving end and the reception power consumption. The transmission energy and the reception energy are accumulated to obtain the communication energy consumption. The communication link quality level is associated with the communication energy consumption to generate communication energy consumption data containing timestamp, spatial coordinates, link quality level, and communication energy consumption value.

[0074] The communication energy consumption data is used to encrypt the disaster risk probability map and generate an encrypted data packet.

[0075] Specifically, it reads communication energy consumption data and disaster risk probability maps, evaluates available encryption overhead based on the communication energy consumption data, and selects a symmetric encryption algorithm and key length; The session key is derived using an existing key derivation function (e.g., HKDF) based on the preset master key and the timestamp and spatial coordinate information of the disaster risk probability map, and an initialization vector is generated using a true random number generator. The disaster risk probability map is divided into shards and each shard is encrypted using the selected symmetric encryption algorithm, the derived session key and the initialization vector, while a message authentication code is generated and a timestamp and spatial coordinate tag are attached; Encapsulate the encrypted fragment, encryption algorithm identifier, derived session key parameters, timestamp and spatial coordinate label in a message format and output the encrypted data packet.

[0076] It should also be explained that the preset master key is derived from a high-strength random number sequence generated by a key management center that has passed the national commercial cryptography testing and certification. A 256-bit binary key is generated using a hardware true random number generator. This key is securely distributed by the key management center and written into an encrypted secure storage area through a secure channel that complies with the national SM4 encryption algorithm. It is loaded and solidified before deployment to ensure the uniqueness and unpredictability of the preset master key throughout its entire usage cycle.

[0077] The cloud-based collaborative module decrypts encrypted data packets and conducts three-dimensional disaster simulation to generate a heat map of disaster impact.

[0078] Decrypt the encrypted data packet and output the decrypted disaster risk probability map.

[0079] Specifically, read the header and payload of the encrypted data packet, and parse out the encryption algorithm identifier, session key derivation parameters, initialization vector, message authentication code, data fragmentation, timestamp and spatial coordinate label; Based on the pre-set master key and derived session key parameters, the session key is derived using HKDF. The parsed initialization vector and the derived session key are then used to perform message authentication code verification and decryption on each data fragment using the encryption algorithm indicated in the message header (e.g., AES-GCM or ChaCha20-Poly1305). The integrity and sequence number of the decrypted fragment are then verified (e.g., by CRC check or sequence number comparison). All verified decrypted slices are reassembled according to timestamps and spatial coordinate labels and merged into a complete disaster risk probability map data stream in the original slice order; Losslessly decompress the merged disaster risk probability map data stream (if compressed), and output the decrypted disaster risk probability map.

[0080] The decrypted disaster risk probability map is mapped to the discrete grid model nodes, the initial values ​​of the mechanical field and the spatial boundary constraints are initialized, three-dimensional disaster simulation is performed, and preliminary disaster simulation data is generated.

[0081] Specifically, the decrypted disaster risk probability map is aligned with the spatial coordinates of the discrete grid model nodes, and the risk value is calculated for each discrete grid model node based on trilinear spatial interpolation and written into the node attributes; Assign initial values ​​of the mechanical field to each discrete grid model node based on node attributes, including initial values ​​of node displacement, stress, and velocity; Identify the boundary nodes of the discrete grid model and impose spatial boundary constraints, such as fixed displacement boundaries, force / pressure boundaries, or seepage boundary parameters, based on the solution configuration. Perform three-dimensional disaster simulations on the discretized system of equations using explicit or implicit time integration schemes based on the solution configuration, update the mechanical fields at the nodes of the discrete grid model at each time step, and perform necessary numerical stability and convergence checks. The node field variables of the discrete grid model at each time step are organized and exported according to spatial grid points and time series to generate preliminary disaster simulation data.

[0082] The preliminary disaster simulation data is integrated with the meteorological and hydrological observation data collected by on-site sensors, and three-dimensional disaster simulation is performed again to generate a disaster impact heat map.

[0083] Specifically, the preliminary disaster simulation data and the meteorological and hydrological observation data collected by on-site sensor measurements are read, and the meteorological and hydrological observation data collected by on-site sensor measurements are temporally and spatially interpolated and aligned to the discrete grid model nodes according to the timestamps and spatial coordinates; Calculate the node-by-node deviation between the observed value and the preliminary disaster simulation data at the registered nodes, apply the deviation as a correction term to the initial mechanical field values ​​and boundary conditions of the discrete grid model nodes using existing data assimilation methods (such as least squares incremental correction or ensemble Kalman filtering), and output the corrected discrete grid initial values ​​and boundary conditions; Based on the calibrated discrete grid initial values ​​and solution configuration, a three-dimensional disaster simulation is performed according to the time integration scheme specified by the solution configuration. The grid node field variables are updated time-step by time-step until the simulation period is completed, and the time series field data of the three-dimensional disaster simulation is derived. According to the calculation rules of impact indicators, the time series field data output by the three-dimensional disaster simulation are subjected to spatiotemporal aggregation and rasterization processing, and the rasterized impact values ​​are spatially interpolated and smoothed as necessary to finally generate a disaster impact heat map.

[0084] It should also be explained that the sources of the impact indicator calculation rules are publicly published disaster assessment industry standards, meteorological and hydrological monitoring technical specifications, and statistical analysis results of disasters in previous years, and the calculation factors and weight distribution are determined in combination with the typical physical characteristics of the target disaster type. For example, rainfall intensity, runoff, slope stability coefficient, etc. are determined as evaluation factors based on national or industry disaster risk assessment standards, and the weight and scope of each evaluation factor are determined by referring to the impact scope and loss degree caused by similar disasters in the historical disaster database, so as to form quantifiable and repeatable impact indicator calculation rules.

[0085] The rescue decision-making module uses an improved path planning algorithm to optimize the disaster impact heat map and generate an optimized rescue path. It also combines communication energy consumption data and returned terrain parameters to generate updated communication power and redundancy configuration parameters.

[0086] It should be noted that the disaster impact heat map, communication energy consumption data and multi-dimensional information of returned terrain parameters are integrated, and the improved path planning algorithm is used to realize dynamic optimization of the rescue path. At the same time, the communication power allocation and beamforming redundancy are dynamically adjusted to achieve closed-loop collaborative optimization of rescue operations and communication support.

[0087] Beneficial effect: By integrating the disaster impact heat map, communication energy consumption data and return terrain parameters into the rescue decision-making module, unified optimization of rescue path optimization and coordinated scheduling of communication resources is achieved. Unlike the existing technology that usually relies only on single geographic information or disaster risk information for path planning, it can dynamically adjust the communication power allocation and beamforming redundancy to keep the communication links in key areas stable in high-risk areas, while saving energy in low-risk areas, thus forming a closed-loop optimization mechanism for disaster situation awareness, path planning and communication management. It significantly improves the efficiency of emergency rescue and the reliability of communication, is suitable for disaster-stricken areas with limited energy and complex communication environments, and embodies the innovation and practicality of the integration of path planning and communication assurance.

[0088] The risk level and obstacle information of the path planning area are extracted from the disaster impact heat map to generate path planning input data.

[0089] Specifically, the risk value of each grid in the disaster impact heat map is read, and the risk value is classified according to the preset risk level interval to generate the risk level of the path planning area. At the same time, the grid where the obstacle exists is determined based on the comparison between the risk value and the obstacle judgment threshold, and the spatial coordinates and attribute information of the obstacle are recorded. The risk level and obstacle information are organized in grid order to form the path planning input data.

[0090] It should also be noted that the preset risk level is divided into different levels, such as low, medium and high, according to the statistical results of geological disasters, meteorological and hydrological historical data and the calculation rules of disaster impact indicators, to reflect different degrees of potential danger; the obstacle determination threshold is determined according to the measurement data of the on-site sensor and the historical obstacle distribution characteristics, for example, by analyzing the obstacle probability distribution, the obstacle determination threshold is set to an example value of 0.7 to ensure that the path planning input data can accurately reflect the actual obstacle distribution situation.

[0091] According to the path planning input data, the initial parameters and heuristic functions of the improved path planning algorithm are initialized, the path search is performed and the candidate path cost is evaluated, the path selection is optimized, and the preliminary rescue path is generated.

[0092] Specifically, the risk level and obstacle information of each grid in the path planning input data are read, the improved path planning algorithm is initialized according to the heuristic function and initial parameter, the start point and end point grids are added to the search queue, the path cost of the candidate path in the queue is calculated in turn, including path length, risk level cumulative value and obstacle avoidance cost, and the optimal candidate path is selected according to the cost sorting, while the search queue and path state are updated, the search is repeated until the end point is reachable, and finally the preliminary rescue path sorted by cost optimization is output.

[0093] For example, the example path length weight in the heuristic function is 0.5, the risk level weight is 0.3, and the obstacle avoidance weight is 0.2. The heuristic function is derived from the classical heuristic search method in the field of artificial intelligence path planning, such as Manhattan distance, Euclidean distance or weighted distance function used in A* algorithm, which is improved in combination with the risk level and obstacle information in the path planning input data to guide the search to preferentially explore areas with high safety and low path cost, and to realize effective path search.

[0094] The preliminary rescue path is smoothed and risk adjusted to generate an optimized rescue path.

[0095] Specifically, the node coordinates of the preliminary rescue path are read, the line segments between adjacent nodes are smoothed using curve fitting method, the nodes in the curve located in high risk areas or close to obstacles are adjusted according to the path risk level and obstacle determination threshold in the path planning input data, the node coordinates are recalculated and the path order is updated to form the optimized rescue path.

[0096] The high-risk area coordinates are extracted from the optimized rescue path to trigger the correction of the beamforming matrix control parameters to generate the corrected communication parameters.

[0097] Specifically, the node coordinates of the optimized rescue path are read, the nodes located in the high-risk area are screened and the spatial coordinates are extracted, the phase modulation value and amplitude distribution of the beamforming matrix are adjusted according to the coordinates of the high-risk area, the control parameters of the array elements are recalculated, and the corrected communication parameters are generated.

[0098] Monitor and collect communication link quality indicators, calculate link power consumption and energy consumption based on the corrected communication parameters, generate energy consumption feedback information, and perform fusion processing to generate comprehensive energy status data.

[0099] Specifically, the communication link quality indicators are collected and parameters such as signal strength, signal-to-noise ratio and bit error rate are recorded. The instantaneous power consumption and cumulative energy consumption of each communication link are calculated based on the corrected communication parameters. The power consumption and energy consumption of each link are integrated to generate energy consumption feedback information. The energy consumption feedback information is fused with the communication link quality indicators to obtain comprehensive energy status data.

[0100] The adaptive optimization algorithm is used to analyze the comprehensive energy status data, dynamically adjust the communication power allocation ratio and the redundancy setting of the beamforming matrix, and generate updated communication power and redundancy configuration parameters.

[0101] Specifically, the comprehensive energy status data is read and normalized, and the communication link power consumption, energy consumption and signal quality indicators are used as input features. The adaptive optimization algorithm is used to calculate the adjustment amount of the communication power allocation ratio and the optimized value of the beamforming matrix redundancy. The communication power allocation ratio and the beamforming matrix redundancy settings are updated according to the optimization results, and finally the updated communication power and redundancy configuration parameters are generated for subsequent communication control.

[0102] It should be noted that the optimized rescue path and updated communication power and redundancy configuration parameters realize intelligent perception and emergency data engineering operations in potential disaster environments through multi-modal fusion processing of disaster impact heat maps, communication energy consumption data and return terrain parameters. Among them, the communication power and redundancy parameters are used as feedback signals to dynamically adjust the rescue path planning and communication guarantee strategies.

[0103] In summary, the present invention achieves this by: triggering beamforming matrix control parameter correction in real time, and dynamically adjusting communication power and redundancy in combination with link quality monitoring and adaptive optimization algorithms, thereby forming a closed-loop optimization mechanism for disaster situation awareness, mission requirements, and communication resource management. This significantly improves the reliability and energy efficiency of emergency communications, ensures that communication links in key areas are stable and redundant, and saves resources in low-risk areas. It is suitable for emergency rescue scenarios with limited energy and complex communication environments.

[0104] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A multimodal data-driven potential disaster intelligent perception and emergency data engineering system, characterized by: include, The data processing module collects multimodal heterogeneous data and performs spatiotemporal alignment processing to generate a standardized spatiotemporal matrix; The energy management module monitors the energy status in real time based on the standardized space-time matrix, selects the computing mode, drives the physical neural operator accelerator, and outputs energy control instructions; The physical calculation module uses three-dimensional geomechanical equations to perform analytical modeling, gridding, equation discretization, and accelerated solution of the standardized space-time matrix under the constraints of energy control instructions. It extracts the mechanical field distribution and calculates the instability probability to generate a disaster risk probability map. The metasurface communication module uses disaster risk probability maps and terrain height data to calculate the phase gradient between communication nodes, generate a beamforming matrix, perform low-power non-line-of-sight data transmission, monitor link quality, provide feedback on communication energy consumption, and generate encrypted data packets. The cloud-based collaborative module decrypts encrypted data packets and performs three-dimensional disaster simulation to generate a disaster impact heat map. The rescue decision-making module uses an improved path planning algorithm to optimize the disaster impact heat map and generate an optimized rescue path. It also combines communication energy consumption data and returned terrain parameters to generate updated communication power and redundancy configuration parameters.

2. The multimodal data-driven potential disaster intelligent perception and emergency data engineering system according to claim 1, characterized in that: The specific steps of generating the standardized space-time matrix are as follows: The dynamic time warping algorithm is used to synchronize the multimodal heterogeneous data and output a time-aligned data sequence. The time-aligned data sequence is converted into spatial coordinates in combination with RTK positioning information to generate spatiotemporal aligned data. The format of the spatiotemporal alignment data is standardized to form a standardized spatiotemporal matrix.

3. The multimodal data-driven potential disaster intelligent perception and emergency data engineering system according to claim 1, characterized in that: The output energy control instruction, the specific steps are as follows: Based on the time and position synchronization information of the standardized space-time matrix, the output power and supercapacitor power are collected in real time through the energy monitoring sensor, and pre-processed to form energy state parameters; According to the energy state parameters, the current energy sufficiency and the task computing requirements are judged, the mode selection logic is performed to determine the matching computing mode, and the selected computing mode is encoded into a computing mode signal; The computing mode signal is converted into the corresponding energy control instruction, sent to the physical neural operator accelerator to drive the operation, and the energy control instruction is output.

4. The multimodal data-driven potential disaster intelligent perception and emergency data engineering system according to claim 1, characterized in that: The specific steps of generating the disaster risk probability map are as follows: Parse the standardized space-time matrix and energy control instructions, extract space-time field variables and solution configurations, map and update the three-dimensional computational grid, and generate a discrete grid model through spatial interpolation assignment; Apply initial mechanical conditions and boundary conditions to the discrete grid model, set the coupling equations, time step and solver parameters according to the solution configuration, and form a discretized set of equations; The discretized equations are loaded into the physical neural operator accelerator for numerical solution to extract the mechanical field distribution; The instability probability of the mechanical field distribution is calculated and spatially mapped to generate a disaster risk probability map.

5. The multimodal data-driven potential disaster intelligent perception and emergency data engineering system according to claim 4, characterized in that: Parse the standardized space-time matrix and energy control instructions, extract space-time field variables and solution configuration, map and update the three-dimensional computational grid, and generate a discrete grid model through spatial interpolation assignment. The specific steps are as follows: Parse the standardized space-time matrix and energy control instructions, extract space-time field variables and calculation requirement parameters, and output space-time field variables and solution configuration; Generate the initial grid structure based on the 3D calculation area and solution configuration, and output the 3D calculation grid; Map the space-time field variables to a three-dimensional computational grid, adjust the grid density based on the disaster risk probability map, mechanical field distribution, and energy state parameters of the previous computational cycle, and output the updated three-dimensional computational grid; Perform spatial interpolation on the updated three-dimensional computational grid, assign the space-time field variables to the grid nodes, and generate a discrete grid model.

6. The multimodal data-driven potential disaster intelligent perception and emergency data engineering system according to claim 1, characterized in that: The specific steps of generating the beamforming matrix are as follows: The terrain height data in the existing digital terrain model database is used in combination with the disaster risk probability map to form a terrain risk dataset; Using the terrain risk dataset, we calculate the spatial coordinate differences and height differences between communication nodes to obtain the distance and height difference matrix between nodes. The phase gradient of the communication node is calculated using the distance between nodes and the height difference matrix, and is mapped to the array elements to generate phase modulation values, which are combined to form a beamforming matrix.

7. The multimodal data-driven potential disaster intelligent perception and emergency data engineering system according to claim 1, characterized in that: The specific steps of generating an encrypted data packet are as follows: The beamforming matrix and the disaster risk probability map are used to transmit data and obtain the transmission status; According to the transmission status, the communication link quality is evaluated and the communication energy consumption is calculated to generate communication energy consumption data; The communication energy consumption data is used to encrypt the disaster risk probability map and generate an encrypted data packet.

8. The multimodal data-driven potential disaster intelligent perception and emergency data engineering system according to claim 1, characterized in that: The specific steps for generating the disaster impact heat map are as follows: Decrypt the encrypted data packet and output the decrypted disaster risk probability map; Map the decrypted disaster risk probability map to the discrete grid model nodes, initialize the initial values ​​of the mechanical field and spatial boundary constraints, perform three-dimensional disaster simulation, and generate preliminary disaster simulation data; The preliminary disaster simulation data is integrated with the meteorological and hydrological observation data collected by on-site sensors, and three-dimensional disaster simulation is performed again to generate a disaster impact heat map.

9. The multimodal data-driven potential disaster intelligent perception and emergency data engineering system according to claim 1, characterized in that: The specific steps of generating the optimized rescue path are as follows: Extract the risk level and obstacle information of the path planning area from the disaster impact heat map to generate path planning input data; Based on the path planning input data, the initial parameters and heuristic functions of the improved path planning algorithm are initialized, path search is performed, the cost of candidate paths is evaluated, the path selection is optimized, and a preliminary rescue path is generated; The preliminary rescue path is smoothed and risk adjusted to generate an optimized rescue path.

10. The multimodal data-driven potential disaster intelligent perception and emergency data engineering system according to claim 1, characterized in that: The specific steps of generating the updated communication power and redundancy configuration parameters are as follows: Extract the coordinates of high-risk areas from the optimized rescue path, trigger the correction of beamforming matrix control parameters, and generate corrected communication parameters; Monitor and collect communication link quality indicators, calculate link power consumption and energy consumption based on the corrected communication parameters, generate energy consumption feedback information, and perform fusion processing to generate comprehensive energy status data; The adaptive optimization algorithm is used to analyze the comprehensive energy status data, dynamically adjust the communication power allocation ratio and the redundancy setting of the beamforming matrix, and generate updated communication power and redundancy configuration parameters.

Images