An intelligent strategy generation method for emergency rescue and related equipment

Abstract

The application provides an intelligent strategy generation method for emergency rescue and related equipment, and the method comprises the following steps: obtaining a rescue type and a rescue level to determine a basic treatment framework; obtaining meteorological data and road condition data of a rescue location, loading the environment correction parameters into the basic treatment framework after the environment correction parameters are generated, to generate an environment adaptation framework; obtaining a current disaster situation, performing similar case matching and treatment effect score extraction on the current disaster situation, to generate an optimized treatment step sequence, and performing fusion verification with the environment adaptation framework, resource matching and path calculation under the condition that the treatment constraint condition is met, to obtain an emergency equipment strategy, an emergency personnel deployment strategy and a support path strategy. Through the above scheme, the adaptability and executability of the emergency rescue strategy generation are improved, the response accuracy and implementation efficiency under complex environmental conditions are enhanced, and the problem of insufficient adaptability of the treatment process structure to the actual environmental conditions in the prior art is overcome.

Description

Technical Field

[0001] This application relates to the technical field of emergency rescue and disaster relief, and in particular to a method for generating intelligent strategies for emergency rescue and disaster relief, and related equipment. Background Technology

[0002] With the acceleration of urbanization and the frequent occurrence of extreme weather events, geological disasters, and major accidents, emergency rescue work is gradually developing towards informatization, digitalization, and intelligentization. Emergency management departments at all levels have gradually established comprehensive emergency management systems covering disaster monitoring, resource management, dispatch and command, and on-site response. These systems integrate multi-source data, including meteorological, transportation, communication, and material reserves, through information platforms to improve the speed of emergency response and coordination efficiency.

[0003] Among the relevant technical means, a rule matching method based on a preset emergency plan database is used to generate a response plan. This means that the corresponding standard response process is called from the plan database according to the disaster type and level, and the steps are adjusted in combination with human experience. This shortens the decision-making time, improves the efficiency of emergency dispatch, and reduces the uncertainty caused by relying entirely on human experience.

[0004] Although emergency response strategies can be quickly generated through pre-plan matching, when the actual emergency response environment changes dynamically (such as sudden changes in weather conditions or a decrease in road traffic capacity), the generated response process structure often lacks consistency verification with the actual on-site traffic conditions. This can easily lead to situations where the response steps are logically feasible but limited in terms of spatial execution, thus affecting the overall efficiency of emergency response. Summary of the Invention

[0005] The technical problem to be solved by the present invention is the insufficient adaptability of the existing process structure to actual environmental conditions, thereby providing an intelligent strategy generation method and related equipment for emergency rescue.

[0006] In view of this, a first aspect of the present invention provides an intelligent strategy generation method for emergency rescue, comprising: acquiring the rescue type and rescue level to determine the corresponding set of response rules and handling constraints, and generating a basic handling framework corresponding to the set of response rules; acquiring current meteorological data and road condition data at the rescue location, calculating a meteorological risk coefficient based on the meteorological data, calculating a road segment passability coefficient based on the road condition data, correcting the road segment passability coefficient using the meteorological risk coefficient, and generating an environmental correction parameter; loading the environmental correction parameter into the basic handling framework to generate an environmental adaptation framework; acquiring the current disaster situation, performing similar case matching and handling effect score extraction on the current disaster situation to generate an optimized handling step sequence; mapping each step item in the optimized handling step sequence to the corresponding handling stage node in the environmental adaptation framework, and removing step items whose resource requirement parameters conflict with the environmental correction parameters to obtain a fusion verification step set; performing resource matching and path calculation when the fusion verification step set satisfies the handling constraints to obtain an emergency equipment strategy, an emergency personnel deployment strategy, and a support path strategy.

[0007] Preferably, the step of obtaining the disaster relief type and disaster relief level to determine the corresponding set of response rules and handling constraints, and generating a basic handling framework corresponding to the set of response rules, includes: obtaining an identifier for the disaster relief type, matching the identifier with a pre-stored type-rule mapping table, and extracting the corresponding rule subset; obtaining the disaster relief level, matching the disaster relief level with a pre-stored level-weight mapping table, and obtaining the activation weight and constraint threshold corresponding to each rule subset; filtering and weighting the rule entries in the rule subset using the activation weight to obtain the set of response rules, and constructing handling constraints using the constraint threshold; and constructing a basic handling framework based on the set of response rules and the handling constraints.

[0008] Preferably, the step of constructing a basic handling framework based on the response rule set and the handling constraints includes: mapping each rule entry to a handling stage node according to the category attribute of each rule entry in the response rule set, to obtain a framework node set; establishing logical directed connections between each handling stage node in the framework node set according to the triggering dependency relationship between each rule entry, to generate a directed handling flowchart; and performing compliance verification between the directed handling flowchart and the handling constraints, removing nodes and connections that violate the handling constraints, to obtain the basic handling framework.

[0009] Preferably, the steps of acquiring current meteorological and road condition data at the disaster relief location, calculating a meteorological risk coefficient based on the meteorological data, calculating a road segment passability coefficient based on the road condition data, and correcting the road segment passability coefficient using the meteorological risk coefficient to generate environmental correction parameters include: acquiring real-time meteorological data at the disaster relief location from a meteorological data interface, matching the meteorological data item by item with a pre-stored meteorological-risk mapping table to obtain a meteorological risk value, and weighting and summing the meteorological risk values ​​according to the risk weights of each meteorological element to obtain the meteorological risk. The system obtains real-time road condition data of the road network surrounding the disaster relief location from the road condition data interface. It then combines this data with a pre-stored road condition-capacity mapping table to calculate the road segment capacity coefficient for each road segment in the surrounding road network. A road network capacity matrix is ​​constructed using these road segment capacity coefficients. The system then corrects the road segment capacity coefficients in the road network capacity matrix using the meteorological risk coefficient to obtain a corrected capacity matrix. Finally, it normalizes the difference between the meteorological risk coefficient and the corrected road segment capacity coefficients in the corrected capacity matrix to obtain environmental correction parameters.

[0010] Preferably, the step of loading the environmental correction parameters into the basic response framework to generate an environment adaptation framework includes: identifying nodes in the basic response framework whose response stage nodes include action space attributes or resource allocation attributes; loading the correction components in the environmental correction parameters corresponding to the action space attributes or resource allocation attributes of the nodes into the corresponding nodes to obtain an environmental correction node set; constructing an environmental accessibility assessment structure around the disaster relief location based on the corrected accessibility matrix; extracting accessibility indicators for the corresponding locations from the environmental accessibility assessment structure according to the execution locations corresponding to each response stage node; marking execution locations with accessibility indicators lower than a preset accessibility threshold as restricted areas; generating alternative execution nodes for response stage nodes involving restricted areas in the basic response framework; replacing the original nodes in the basic response framework with the environmental correction node set; and inserting the alternative execution nodes into the directed response flowchart at the positions corresponding to the original response stage nodes to generate an environment adaptation framework.

[0011] Preferably, the step of obtaining the current disaster situation, performing similar case matching and extracting the handling effect score of the current disaster situation to generate an optimized handling step sequence includes: obtaining the current disaster situation, converting the current disaster situation into a disaster feature vector according to a preset feature extraction rule; extracting historical case records from a historical emergency disaster relief database, each historical case record including a historical disaster feature vector, a corresponding handling step sequence, and a handling effect score; calculating the cosine similarity between the disaster feature vector and the historical disaster feature vector in each historical case record, and selecting historical case records with a cosine similarity higher than a preset similarity threshold as a matching case set; extracting the handling step sequence of each historical case record from the matching case set, deduplicating and merging the step items in the handling step sequence to obtain a candidate handling step set, and extracting the handling effect score corresponding to each step item in the matching case set; calculating the weighted product of the handling effect score of each step item and the cosine similarity of its corresponding historical case record to obtain a comprehensive priority score for each step item, and sorting the step items in the candidate handling step set in descending order according to the comprehensive priority score to obtain an optimized handling step sequence.

[0012] Preferably, the step of sequentially mapping each step item in the optimized disposal step sequence to the corresponding disposal stage node in the environment adaptation framework, and eliminating step items whose resource requirement parameters conflict with the environmental correction parameters, to obtain a fusion verification step set, and performing resource matching and path calculation to obtain the emergency equipment strategy, emergency personnel deployment strategy, and support path strategy when the fusion verification step set satisfies the disposal constraints, includes: sequentially mapping each step item in the optimized disposal step sequence to the corresponding disposal stage node in the environment adaptation framework, and comparing and verifying the resource requirement parameters of each step item with the environmental correction parameters of the corresponding disposal stage node item item by item, eliminating those with conflicting resource requirement parameters and environmental correction parameters. The steps are defined to obtain a fusion verification step set. The execution order and resource consumption of each step in the fusion verification step set are checked against the disposal constraints. Steps exceeding the constraint threshold are split or delayed to obtain a feasible step set that satisfies the disposal constraints. The equipment requirement list of each step in the feasible step set is matched with a pre-stored emergency equipment resource library to obtain an emergency equipment strategy. The personnel requirement list of each step in the feasible step set is matched with a pre-stored emergency personnel resource library to obtain an emergency personnel deployment strategy. Using the modified traffic capacity matrix as the road network weight input, with the disaster relief location as the endpoint and the current location of each emergency device and personnel as the starting point, the optimal path from each starting point to the endpoint is calculated to obtain a support path strategy.

[0013] A second aspect of this invention provides an intelligent dispatch system for emergency rescue, comprising: a framework generation module, used to acquire the type and level of the emergency rescue to determine the corresponding set of response rules and handling constraints, and to generate a basic handling framework corresponding to the set of response rules; an environment correction module, used to acquire current meteorological data and road condition data at the emergency rescue location, calculate a meteorological risk coefficient based on the meteorological data, calculate a road segment passability coefficient based on the road condition data, correct the road segment passability coefficient using the meteorological risk coefficient, and generate environment correction parameters; and a data loading module, used to load the environment correction parameters into the basic handling framework. An environment adaptation framework is generated; a case matching module is used to obtain the current disaster situation, perform similar case matching and treatment effect score extraction on the current disaster situation to generate an optimized treatment step sequence; a strategy generation module is used to map each step item in the optimized treatment step sequence to the corresponding treatment stage node in the environment adaptation framework, and remove step items whose resource requirement parameters and environmental correction parameters conflict, to obtain a fusion verification step set. If the fusion verification step set meets the treatment constraints, resource matching and path calculation are performed to obtain emergency equipment strategy, emergency personnel deployment strategy and support path strategy.

[0014] A third aspect of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the intelligent strategy generation method for emergency rescue as described above.

[0015] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is run by a processor, the processor causes the processor to execute the intelligent strategy generation method for emergency rescue as described in any one of the above embodiments.

[0016] The technical solution of this invention has the following advantages: by obtaining the disaster relief type and level, a set of response rules and disposal constraints are determined, and a basic disposal framework is generated. Environmental correction parameters are generated by combining current meteorological and road condition data at the disaster relief location. Environmental loading and node replacement are performed on the basic disposal framework to form an environmental adaptation framework. Simultaneously, a disaster feature vector is constructed based on the current disaster situation. Similar case matching and disposal effect scoring are extracted using a historical emergency disaster relief database to form an optimized disposal step sequence. This optimized disposal step sequence is then fused and verified with the environmental adaptation framework. Under the premise of meeting disposal constraints, resource matching and path calculation are completed, generating emergency equipment strategies, emergency personnel deployment strategies, and support path strategies. This achieves the organic combination of dynamic environmental factors and historical disposal experience, enabling collaborative optimization of disposal steps, resource allocation, and support paths under unified disposal constraints. This improves the adaptability and executability of emergency disaster relief strategy generation, enhances response accuracy and implementation efficiency under complex environmental conditions, and overcomes the problem of insufficient adaptability between the disposal process structure and actual environmental conditions in existing technologies. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating an intelligent strategy generation method for emergency rescue provided in an embodiment of the present invention; Figure 2 A schematic block diagram illustrating the construction of the basic processing framework provided in the embodiments of the present invention; Figure 3 A schematic block diagram illustrating the calculation structure of meteorological risk coefficient and road network capacity provided in an embodiment of the present invention; Figure 4 A schematic block diagram illustrating the structure for matching similar cases provided in embodiments of the present invention; Figure 5 A schematic block diagram of the structure generated for the strategy provided in the embodiments of the present invention; Figure 6 A schematic block diagram of an intelligent strategy generation system for emergency rescue provided in an embodiment of the present invention; Figure 7 A schematic block diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0019] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] In the description of this invention, it should be noted that the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0021] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0022] like Figure 1 As shown, this application provides an intelligent strategy generation method for emergency rescue, which can construct a handling process with logical constraints based on the type and level of rescue after a disaster occurs. The intelligent strategy generation method specifically includes the following steps: Step S1: Obtain the type and level of emergency response to determine the corresponding set of response rules and handling constraints, and generate a basic handling framework corresponding to the set of response rules.

[0023] The type of emergency response is input in the form of a predefined type code, and the level of emergency response is input in the form of a level value or level code. Based on the type and level of emergency response, a set of rule entries that meet the conditions are selected from the pre-stored rule dataset, and disposal constraints are constructed by combining the resource consumption limit, time limit threshold and safety constraint parameters corresponding to the level of emergency response. The selected rule entries are structured and organized to form a basic disposal framework with nodes and connections.

[0024] For example, if the emergency response type is "urban flooding" and the emergency response level is "Level II", then the rule entries corresponding to "urban flooding", such as drainage disposal rules, traffic management rules, and power protection rules, are selected from the rule dataset. At the same time, the constraint parameters corresponding to "Level II", such as the maximum number of personnel to be deployed is 200, the maximum number of equipment to be called is 50, and the maximum execution time of a single step is 4 hours, are read and the disposal constraints are constructed accordingly.

[0025] Specifically, such as Figure 2 As shown, step S1 can also preferably be implemented in the following manner: Obtain the identifier of the disaster relief type, match the identifier with the pre-stored type-rule mapping table, and extract the corresponding rule subset.

[0026] The type-rule mapping table uses a key-value pair data structure for storage. The key is the emergency rescue type identifier, and the value is a set of rule entry numbers. The rule entry number corresponds to a specific rule entry record in the rule data table. Each rule entry includes a rule number, rule category attribute, rule preconditions, rule triggering conditions, rule execution action description, and resource requirement parameters. By searching the mapping table for a key value that exactly matches the emergency rescue type identifier, the corresponding set of rule entry numbers is read, and a complete rule record is extracted from the rule data table based on the rule entry number to form a rule subset.

[0027] For example, the disaster relief type identifier is defined as a three-digit string code, where "FLD" represents urban flooding and "EQK" represents earthquake disaster. When the input identifier is "FLD", the record with the key "FLD" is read from the type-rule mapping table, and its value is {R101, R102, R205, R310}. Then, according to the rule entry numbers R101, R102, R205, and R310, the corresponding rule entries are extracted from the rule data table to form a subset of rules corresponding to urban flooding.

[0028] Obtain the emergency response level, match the emergency response level with the pre-stored level-weight mapping table to obtain the activation weight and constraint threshold corresponding to each rule subset.

[0029] The level-weight mapping table is stored in a two-dimensional table structure. The row index is the emergency response level, and the column fields include the rule category weight coefficient, resource upper limit threshold, time upper limit threshold, and risk tolerance threshold. The emergency response level is used as the row index to read the corresponding row data, the rule category weight coefficient is assigned to each rule entry in the rule subset, and the resource upper limit threshold, time upper limit threshold, and risk tolerance threshold are read as the parameter basis for the disposal constraints.

[0030] For example, when the emergency response level is "Level II", the corresponding row of data is read from the level-weight mapping table, where the weight coefficient for on-site drainage is 1.2, the weight coefficient for traffic diversion is 1.0, the weight coefficient for material support is 0.9, the upper limit threshold for resources is 50 pieces of equipment and 200 people, and the upper limit threshold for time is 4 hours. The above weight coefficients are assigned to the rule entries of the corresponding categories in the rule subset, and the upper limit thresholds for resources and time are written into the disposal constraint parameter set.

[0031] The rule entries in the rule subset are filtered and weighted by activation weights to obtain the response rule set, and disposal constraints are constructed using constraint thresholds.

[0032] For each rule entry in the rule subset, an activation value is calculated. The activation value is the product of the rule category weight coefficient and the basic priority score of the rule entry. The basic priority score of the rule entry is an integer score preset in the rule data table. Rule entries with activation values ​​greater than the preset activation threshold are selected as valid rule entries. The valid rule entries are deduplicated by rule number to form a response rule set. At the same time, the resource upper limit threshold, time upper limit threshold, and risk tolerance threshold are written into the structured constraint data table to form disposal constraints.

[0033] For example, if the basic priority score of rule entry R101 is 80 and the corresponding category weight coefficient is 1.2, then the activation value is 96. If the preset activation threshold is 70, then R101 is retained. If the basic priority score of rule entry R310 is 50 and the corresponding category weight coefficient is 0.9, then the activation value is 45, which is less than 70, so it is removed. The set of retained rule entries constitutes the response rule set, and at the same time, 50 devices, 200 personnel, and a maximum single-step execution time of 4 hours are written into the disposal constraint structure.

[0034] A basic handling framework is constructed based on the set of response rules and handling constraints.

[0035] The rule entries in the response rule set are structured and modeled according to the logical dependencies between the rules, generating a flowchart structure containing nodes and logical directed connections. The flowchart structure is then checked for constraint consistency, and nodes and connections that satisfy the disposal constraints are retained to form a basic disposal framework.

[0036] For example, a logical dependency is set between the rule entry "Start the drainage pump group" and the rule entry "Complete the site survey", that is, "Complete the site survey" is a prerequisite and "Start the drainage pump group" is a subsequent action, and a directed connection is established in the flowchart.

[0037] Furthermore, the step of constructing the basic handling framework based on the set of response rules and handling constraints can preferably be carried out in the following manner: Each rule entry in the response rule set is mapped to a disposal stage node according to its category attribute, thus obtaining a frame node set.

[0038] The nodes in the disposal stage are represented in a structured data format. Each disposal stage node includes a node number, node category, set of preconditions, resource requirement parameters, and a description of the execution action. According to the rule category attribute in the rule entry, rule entries of the same category are mapped to the same disposal stage node or different sub-nodes in the same stage. The mapping results are stored as a node array structure to form a frame node set.

[0039] For example, if the rule category attribute of rule entries R101 and R102 is "on-site drainage", then R101 and R102 will be mapped to the disposal stage node N1. Node N1 contains resource requirement parameters such as the number of water pumps (10), power support capacity (200kW), and number of personnel (20).

[0040] Based on the triggering dependencies between each rule entry, logical directed connections are established between each processing stage node in the framework node set. These logical directed connections are used to represent the execution order and triggering relationships between processing stages to generate a directed processing flowchart.

[0041] Dependency triggers are stored in an adjacency list structure. Each record contains the starting node number, the target node number, and the trigger condition expression. The precondition fields of the rule entries in the response rule set are traversed. The node corresponding to the precondition is used as the starting node, and the node corresponding to the current rule entry is used as the target node. Directed edge records are added to the adjacency list. After construction, a directed processing flowchart is formed.

[0042] For example, if the precondition for rule entry R205 is that R101 is completed, then add a record (N1, N3, condition = R101 completed) to the adjacency list, indicating that node N3 will be triggered to execute after node N1 is completed.

[0043] The directed disposal flowchart and disposal constraints are verified for compliance. Nodes and connections that violate the disposal constraints are removed to obtain the basic disposal framework.

[0044] A topology sorting algorithm is performed on the directed processing flowchart. The topology sorting algorithm is a directed acyclic graph sorting algorithm used to detect whether there are circular dependencies in the flowchart. If a circular dependency is detected, the newly added directed connection is deleted. At the same time, the resource requirement parameters of each node are accumulated and summed, and compared with the resource upper limit threshold in the processing constraints. If the accumulated resource requirement of a path containing a node exceeds the resource upper limit threshold, the node and its related connections are deleted. The nodes and logical directed connections retained after verification constitute the basic processing framework.

[0045] For example, if a closed loop is detected between node N4 and node N5 during topology sorting, the last directed connection established is deleted; if the total demand for a certain path is found to be 60 units during resource accumulation, which is greater than the upper limit threshold of 50 units in the disposal constraint, the node with the highest resource consumption in that path is deleted, and the calculation is recalculated until all paths meet the upper limit threshold requirement for resources.

[0046] Step S2: Obtain the current meteorological and road condition data at the disaster relief location, calculate the meteorological risk coefficient based on the meteorological data, calculate the road section passability coefficient based on the road condition data, correct the road section passability coefficient using the meteorological risk coefficient, and generate environmental correction parameters.

[0047] The disaster relief location is entered in the form of latitude and longitude coordinates or administrative division codes; the spatial query range is determined based on the disaster relief location, which is defined as a circular area with a radius of 5 to 20 kilometers centered on the disaster relief location; within this spatial query range, meteorological data interfaces and road condition data interfaces are called to obtain real-time meteorological data and real-time road condition data under the corresponding timestamps; the meteorological data and road condition data are subjected to structured parsing and numerical standardization processing, and environmental correction parameters are generated according to predefined risk calculation rules and traffic capacity calculation rules.

[0048] For example, the emergency location is latitude and longitude (31.2304, 121.4737), and the spatial query range is set to an area with a radius of 10 kilometers; the meteorological data interface returns the precipitation intensity of 25 mm / h, wind speed of 12 m / s, visibility of 800 m, and temperature of 5℃ at the timestamp of 10:00 on March 3, 2026; the road condition data interface returns the road congestion level of each road section in this area as level 3, the road surface damage status as minor damage, and no road closure information; after parsing the above data, the risk and traffic capacity calculation process begins.

[0049] Specifically, such as Figure 3 As shown, step S2 can also preferably be implemented in the following manner: Real-time meteorological data of the disaster relief location is obtained from the meteorological data interface. The meteorological data includes precipitation intensity, wind speed and direction, visibility, temperature and humidity. The meteorological data is matched item by item with the pre-stored meteorological-risk mapping table to obtain the meteorological risk value corresponding to each meteorological element. The meteorological risk values ​​are then weighted and summed according to the risk weight of each meteorological element to obtain the meteorological risk coefficient.

[0050] The meteorological-risk mapping table uses a segmented interval mapping structure for storage. Each meteorological element corresponds to multiple numerical intervals and interval risk values. Precipitation intensity intervals are divided into 0~5 mm / h, 5~20 mm / h, 20~50 mm / h, and greater than 50 mm / h, corresponding to risk values ​​of 0.1, 0.4, 0.7, and 0.9, respectively. Wind speed intervals are divided into 0~5 m / s, 5~10 m / s, 10~20 m / s, and greater than 20 m / s, corresponding to risk values ​​of 0.1, 0.3, 0.6, and 0.85, respectively. Visibility intervals are divided into... The risk values ​​are 0.1, 0.3, 0.6, and 0.9 for 2000m, 1000~2000m, 500~1000m, and less than 500m, respectively. Temperature and humidity are assigned an additional risk value of 0.1 when the temperature is below 0℃ or above 35℃. Risk weights are predefined as follows: precipitation intensity 0.4, wind speed and direction 0.3, visibility 0.2, and temperature and humidity 0.1. The meteorological risk coefficient is calculated using the weighted summation formula: meteorological risk coefficient = ∑(meteorological risk value × risk weight), and the range of the meteorological risk coefficient is limited to between 0 and 1.

[0051] For example, a precipitation intensity of 25 mm / h corresponds to a risk value of 0.7, a wind speed of 12 m / s corresponds to a risk value of 0.6, a visibility of 800 m corresponds to a risk value of 0.6, and a temperature of 5℃ corresponds to an additional risk value of 0; substituting these values ​​into the calculation, the meteorological risk coefficient is: 0.7 × 0.4 + 0.6 × 0.3 + 0.6 × 0.2 + 0 × 0.1 = 0.28 + 0.18 + 0.12 = 0.58.

[0052] Real-time traffic data of the road network surrounding the disaster relief location is obtained from the traffic data interface. The traffic data includes road congestion level, road surface damage status and road closure information. Based on the road congestion level, road surface damage status and road closure information, the traffic data is combined with the pre-stored traffic capacity mapping table to calculate the traffic coefficient of each road segment in the road network surrounding the disaster relief location, and a road network traffic capacity matrix is ​​constructed based on the traffic coefficient of each road segment.

[0053] The road network is represented using a graph structure, where nodes represent intersections and edges represent road segments. Each edge includes the segment number, length, design speed, and current status fields. The road condition-capacity mapping table maps road congestion levels 1-5 to basic capacity reduction coefficients of 1.0, 0.8, 0.6, 0.4, and 0.2, respectively. Road surface damage is categorized as intact, slightly damaged, moderately damaged, and severely damaged, corresponding to reduction coefficients of 1.0, 0.9, 0.7, and 0.4, respectively. When a road is closed, the capacity coefficient is set to 0. The formula for calculating the road segment capacity coefficient is: design speed × congestion reduction coefficient × damage reduction coefficient. If the road is closed, the road segment capacity coefficient is directly assigned to 0. After calculating the road segment capacity coefficient for all road segments, a road network capacity matrix is ​​constructed in the form of an adjacency matrix, where the matrix elements represent the capacity coefficient of the road segment corresponding to node i to node j.

[0054] For example, if a road segment has a design speed of 60 km / h, a congestion level of 3 with a reduction factor of 0.6, and a road surface damage condition of minor damage with a reduction factor of 0.9, then the road segment's capacity coefficient = 60 × 0.6 × 0.9 = 32.4. If the road segment is closed, then the road segment's capacity coefficient is 0. After all road segments are calculated, a road network capacity matrix is ​​formed.

[0055] The capacity coefficients of each road segment in the road network capacity matrix are corrected using the meteorological risk coefficient to obtain the corrected capacity matrix. The difference between the corrected meteorological risk coefficient and the corrected capacity coefficients of each road segment in the corrected capacity matrix is ​​normalized to obtain the environmental correction parameters.

[0056] The correction formula is defined as: Corrected traffic coefficient = Original road segment traffic coefficient × (1 - Meteorological risk coefficient); The above correction operation is performed on all non-zero elements in the road network capacity matrix to form the corrected traffic capacity matrix; The correction difference for each road segment is calculated as the original road segment traffic coefficient minus the corrected coefficient; Normalization is performed using the minimum-maximum normalization method, and the normalization formula is: Standard value = (Corrected difference - Minimum difference) / (Maximum difference - Minimum difference); All standard values ​​and meteorological risk coefficients are combined to form an environmental correction parameter set, which includes the meteorological risk coefficient and the standardized correction difference vector corresponding to each road segment.

[0057] For example, if the original traffic coefficient of a certain road segment is 32.4 and the meteorological risk coefficient is 0.58, then the corrected traffic coefficient = 32.4 × (1 - 0.58) = 13.608, and the corrected difference is 18.792. After performing minimum-maximum normalization on the differences of all road segments, a standard value between 0 and 1 is obtained. This standard value and the meteorological risk coefficient 0.58 are stored together as environmental correction parameters.

[0058] Step S3: Load the environment correction parameters into the basic processing framework to generate the environment adaptation framework.

[0059] Traverse all the nodes in the basic handling framework, identify the execution location field and resource allocation parameter field of each node; find the corresponding road segment traffic capacity index in the modified traffic capacity matrix according to the execution location field, and modify the node execution conditions and resource requirement parameters in combination with environmental modification parameters; after the modification is completed, update the node parameters and adjust the logical directed connection relationship to form an environment adaptation framework.

[0060] For example, if the execution location of a certain handling stage node corresponds to road segment number L25, and the traffic capacity index found in the corrected traffic capacity matrix is ​​0.3, which is lower than the preset threshold of 0.4, then the execution time parameter of that node will be multiplied by a correction factor of 1.5, and it will be marked as a restricted execution node.

[0061] Specifically, step S3 can also preferably be implemented in the following manner: Identify nodes in each stage of the basic response framework that include nodes with action space attributes or resource allocation attributes. Load the correction components in the environmental correction parameters corresponding to the action space attributes or resource allocation attributes of the nodes into the corresponding nodes. Perform product correction on the original parameters of each node in the basic response framework to obtain the environmental correction node set.

[0062] The action space attribute is defined as the set of road segment numbers or area numbers on which the node execution depends, and the resource allocation attribute is defined as the number of equipment and personnel required by the node. The standardized correction difference in the environmental correction parameters is mapped to the corresponding node according to the road segment number. The original execution time parameter and resource requirement quantity parameter of the node are corrected by multiplying by the correction coefficient = 1 + standardized correction difference. The corrected node is stored as a new node structure, forming an environmental correction node set.

[0063] For example, if the original execution time of a node is 2 hours and the standardized correction difference is 0.5, then the correction factor is 1.5, and the corrected execution time is 3 hours; if the original equipment requirement is 10 units, then the corrected equipment requirement is 15 units.

[0064] Execution locations where the traffic capacity index is lower than the preset traffic threshold are marked as restricted areas, and alternative execution nodes are generated for the disposal stage nodes involving restricted areas in the basic disposal framework.

[0065] The preset passage threshold is defined as 0.4 of the passage capability index in the modified passage capability matrix; the passage capability index corresponding to the execution position of all nodes is traversed, and those less than 0.4 are marked as restricted areas; for the handling stage nodes involving restricted areas, alternative execution nodes are generated according to the predefined alternative execution scheme field in the original rule entries. The alternative execution nodes have new execution position fields and resource allocation attributes; a new node number is assigned to the alternative execution nodes in the node data structure.

[0066] For example, if the traffic capacity index of a node at execution location L30 is 0.2, which is less than 0.4, then L30 is marked as a restricted area; if the alternative execution location is defined as L45 in the original rule entry, then an alternative execution node N8 is generated, with execution location L45.

[0067] The environment correction node set replaces the original nodes in the basic handling framework, and the replacement execution nodes are inserted into the positions corresponding to the original handling stage nodes in the directed handling flowchart to generate the environment adaptation framework.

[0068] In the adjacency list of the directed processing flowchart, the original node number is replaced with the environment correction node number; for cases where there is an alternative execution node, a logical directed connection is established between the predecessor node of the original node and the alternative execution node, and a logical directed connection is established between the alternative execution node and the successor node of the original node; the direct connection between the original node and the successor node is deleted; after all node replacements and connection adjustments are completed, a data structure representation of the environment adaptation framework is formed.

[0069] For example, if the original node N5 has a replacement execution node N8, then the (N5, N6) connection is deleted from the adjacency list, and the (N5, N8) and (N8, N6) connections are added; after all nodes are replaced, an environment adaptation framework containing environment correction nodes and replacement execution nodes is obtained.

[0070] Step S4: Obtain the current disaster situation, perform similar case matching and treatment effect score extraction on the current disaster situation, and generate an optimized treatment step sequence.

[0071] The current disaster situation is analyzed in a structured manner and converted into a numerical vector form with a unified dimension. Historical case records similar to the current disaster situation are retrieved from the historical emergency relief database, and the corresponding sequence of disposal steps and disposal effect scores are extracted. The disposal steps are sorted according to similarity and disposal effect scores to generate an optimized disposal step sequence.

[0072] The current disaster situation is "urban flooding" type, 3 square kilometers affected area, moderate severity, and 2 casualties. Several historical cases related to urban flooding were retrieved from the historical emergency relief database. Historical cases with similarity scores higher than the threshold were selected by similarity calculation, and their handling steps were comprehensively sorted.

[0073] Specifically, such as Figure 4 As shown, step S4 can also preferably be implemented in the following manner: Obtain the current disaster situation, including the type of disaster, the scope of the disaster, the degree of the disaster, and the number of casualties. Transform the current disaster situation into a disaster feature vector using preset feature extraction rules.

[0074] The preset feature extraction rules are defined as mapping rules that convert different data types into fixed-length numerical vectors; the disaster type is represented by one-hot encoding, for example, the complete set of disaster types is set as {urban flooding, earthquake disaster, fire accident, hazardous material leakage}, corresponding to a four-dimensional binary vector; the disaster area is normalized in square kilometers, and the normalization formula is (current value - minimum value) / (maximum value - minimum value), where the minimum and maximum values ​​are derived from the statistical range of the historical emergency disaster relief database; the degree of disaster is divided into mild, moderate, and severe, mapped to values ​​of 0.3, 0.6, and 0.9 respectively; the number of casualties is mapped to values ​​of 0, 0.4, 0.7, and 1 according to the number range of 0, 1-5, 6-20, and greater than 20; the above-mentioned values ​​are concatenated in a fixed order to form the disaster feature vector.

[0075] The current disaster situation is urban flooding, with an affected area of ​​3 square kilometers, a moderate degree of disaster, and 2 casualties. Assuming that the historical minimum affected area is 0 square kilometers and the maximum is 10 square kilometers, the normalized value of the affected area is 0.3. The unique hot code corresponding to urban flooding is (1,0,0,0); the moderate degree of disaster corresponds to 0.6; the number of casualties corresponds to 0.4; and the final disaster feature vector is (1,0,0,0,0.3,0.6,0.4).

[0076] Historical case records are extracted from the historical emergency disaster relief database. Each historical case record includes a historical disaster feature vector, a corresponding sequence of handling steps, and a handling effect score.

[0077] The historical emergency disaster relief database is stored using a relational data table structure, including a case number field, a historical disaster feature vector field, a response step sequence field, and a response effect score field. The response step sequence field is a set of step item numbers arranged in chronological order. The response effect score field is numerical data with a value range of 0 to 100, and the scoring criteria include response time, resource utilization efficiency, and the degree of secondary loss control. By querying all historical case records, the historical disaster feature vector and the corresponding response step sequence and response effect score are extracted.

[0078] For example, the historical disaster feature vector of a certain historical case is (1,0,0,0,0.25,0.6,0.3), the sequence of handling steps is {S101,S205,S310}, and the handling effect score is 85 points.

[0079] Calculate the cosine similarity between the disaster-affected feature vector and the historical disaster-affected feature vector in each historical case record, and select historical case records with a cosine similarity higher than a preset similarity threshold as the matching case set.

[0080] The cosine similarity algorithm is used for calculation. The cosine similarity formula is: similarity = (vector A · vector B) / (|A| × |B|); where vector A is the disaster-affected feature vector and vector B is the historical disaster-affected feature vector. The similarity value of all historical case records is calculated. The preset similarity threshold is set to 0.75. Historical case records with a similarity greater than or equal to 0.75 are selected as the matching case set.

[0081] If the cosine similarity between the current disaster-affected feature vector and a historical disaster-affected feature vector is 0.82, which is greater than 0.75, then the historical case record is included in the matching case set; if the similarity is 0.63, it is excluded.

[0082] Extract the sequence of handling steps from each historical case record in the matched case set, remove duplicate steps from the sequence and merge them to obtain a candidate set of handling steps, and extract the handling effect score corresponding to each step in the matched case set.

[0083] The system iterates through all the treatment step sequences in the matching case set, stores all step item numbers in a set structure to remove duplicates, and establishes a mapping table between step items and their source historical case records. For each step item, it records the treatment effect score in different historical case records, thus forming a candidate treatment step set and a corresponding score record table.

[0084] For example, if the matching case set contains two historical case records with disposition step sequences {S101,S205,S310} and {S101,S220,S330}, then the candidate disposition step set is {S101,S205,S310,S220,S330}; the disposition effect scores of step item S101 in the two historical case records are 85 and 90 points, respectively.

[0085] The comprehensive priority score of each step item is calculated by weighted product of the treatment effect score and the cosine similarity of the historical case record. The steps in the candidate treatment step set are then sorted in descending order according to the comprehensive priority score to obtain the optimized treatment step order.

[0086] Calculate the overall priority score for each step item using the formula: Overall Priority Score = ∑ (Disposal Effect Score × Cosine Similarity of Corresponding Historical Case Records). After calculating the overall priority score for all steps items, sort them in descending order according to the overall priority score using a quicksort algorithm. The sorted result is the optimized disposal step order.

[0087] For example, if the treatment effect scores of step item S101 in two historical case records are 85 and 90 respectively, with corresponding cosine similarity scores of 0.82 and 0.78, then the comprehensive priority score = 85 × 0.82 + 90 × 0.78 = 69.7 + 70.2 = 139.9; after calculating all step items, the optimized treatment step order is obtained by sorting them according to their numerical values.

[0088] Step S5: Map each step in the optimized disposal step sequence to the corresponding disposal stage node in the environment adaptation framework, and remove step items whose resource requirement parameters conflict with the environment correction parameters to obtain the fusion verification step set. If the fusion verification step set meets the disposal constraints, perform resource matching and path calculation to obtain the emergency equipment strategy, emergency personnel deployment strategy, and support path strategy.

[0089] The optimized response sequence is mapped one by one to the corresponding response stage node in the environment adaptation framework, and the consistency of resource requirement parameters and execution conditions is verified. Under the premise of meeting the response constraints, resources are matched according to the set of feasible steps, and support paths are calculated based on the modified accessibility matrix, outputting emergency equipment strategies, emergency personnel deployment strategies, and support path strategies.

[0090] For example, the step entry S101 in the optimized disposal step sequence is mapped to node N1 in the environment adaptation framework. The modified equipment requirement quantity and execution time parameters of this node are checked to see if they meet the disposal constraints. After the verification is passed, resource matching and path calculation are performed.

[0091] Specifically, such as Figure 5 As shown, step S5 can also preferably be implemented in the following manner: Each step in the optimized treatment sequence is mapped sequentially to the corresponding treatment stage node in the environment adaptation framework. The resource requirement parameters of each step are compared and verified with the environmental correction parameters of the corresponding treatment stage node. Steps with conflicting resource requirement parameters and environmental correction parameters are removed to obtain the fusion verification step set.

[0092] Resource requirement parameters include the number of equipment, the number of personnel, and the execution time; environmental correction parameters include the corrected upper limit of the number of equipment, the upper limit of the number of personnel, and the execution time adjustment coefficient; the comparison rule is that the resource requirement parameters must not exceed the corresponding values ​​of the environmental correction parameters; if they exceed the limit, the step item is marked as a conflicting step and removed; the remaining step items constitute the fusion verification step set.

[0093] For example, if a step requires 20 units of equipment, but the maximum number of units after the corresponding processing stage node is adjusted is 15, then that step is removed.

[0094] The execution order and resource consumption of each step in the fusion verification step set are constrained by the handling constraints. Steps that exceed the constraint threshold are split or delayed to obtain a feasible set of steps that meet the handling constraints.

[0095] The set of fusion verification steps is traversed according to the optimized handling step order, and the resource consumption of the currently arranged steps is calculated cumulatively. If the cumulative resource consumption exceeds the upper limit threshold in the handling constraints, the current step item is split into multiple sub-step items or adjusted to the next time window. The time window is divided in 1-hour units. After the processing is completed, a set of feasible steps that meets the requirements of the upper limit threshold of resources and the upper limit threshold of time is formed.

[0096] For example, if the cumulative number of devices occupied within a certain time window is 48, and the next step requires 10 devices, exceeding the upper limit of 50 devices, then the step is split into two sub-steps, each requiring 5 devices, and these are arranged in adjacent time windows.

[0097] The equipment requirement list for each step in the feasible steps set is matched with the pre-stored emergency equipment resource library. Equipment combinations that meet the requirements are selected according to equipment type, quantity and availability to obtain the emergency equipment strategy.

[0098] The emergency equipment resource database is stored in a data table structure, with fields including equipment number, equipment type, equipment quantity, current status, and current location. It iterates through the equipment requirement list for each step in the feasible steps set, filters equipment records in the emergency equipment resource database that match the equipment type and are currently available, allocates equipment according to quantity requirements, and updates the equipment status to occupied, thus forming an emergency equipment strategy.

[0099] For example, if the step requires 10 water pumps, filter the emergency equipment resource library for equipment type water pumps that are available, select equipment numbered E01 to E10 for allocation, and update their status to occupied.

[0100] The personnel requirement list for each step in the feasible steps is matched with the pre-stored emergency personnel resource library. Based on the personnel skill categories, quantities, and current on-duty status, personnel combinations that meet the requirements are selected to obtain the emergency personnel deployment strategy.

[0101] The emergency personnel resource database includes fields for personnel number, skill category, level, current on-duty status, and current location; it filters personnel records with matching skill categories and current on-duty status based on the personnel demand list for each step item; it selects personnel numbers according to the required quantity and records allocation information to form an emergency personnel allocation strategy.

[0102] For example, if a step requires 5 power maintenance personnel, then the emergency personnel resource pool will be used to select personnel whose skill category is power maintenance and whose current on-duty status is on-duty, and personnel numbered P11 to P15 will be selected for deployment.

[0103] Using the modified traffic capacity matrix as the road network weight input, with the disaster relief location as the endpoint and the current location of each emergency equipment and personnel as the starting point, the shortest weighted path algorithm is run to calculate the optimal path from each starting point to the endpoint, thus obtaining the support path strategy.

[0104] The shortest weighted path algorithm is Dijkstra's algorithm, with weights taken from the reciprocal of the modified traffic coefficient of each road segment in the modified traffic capacity matrix. Dijkstra's algorithm is run separately for each starting point to calculate the shortest path to the corresponding node at the disaster relief location. The calculated path node sequence is recorded as the support path. The support path strategy is formed after calculating the shortest path for all emergency equipment and personnel.

[0105] For example, the current location of the emergency equipment corresponds to node A1, and the location of the rescue operation corresponds to node B5. After converting the modified accessibility matrix into a weight matrix, Dijkstra's algorithm is run to obtain the path A1→A3→B2→B5, which has the minimum total weight. This path is the support path for the emergency equipment. After calculating for all equipment and personnel, the complete support path strategy is obtained.

[0106] In this embodiment, the disaster relief type and level are obtained, and the corresponding set of response rules and handling constraints are determined from a preset rule base based on the disaster relief type and level. On this basis, a basic handling framework corresponding to the set of response rules is constructed according to the triggering dependency relationship between the sets of response rules. Then, the current meteorological data and road condition data of the disaster relief location are obtained. The meteorological risk coefficient is calculated based on the meteorological data, and the road section passability coefficient is calculated based on the road condition data. The road section passability coefficient is corrected using the meteorological risk coefficient to generate environmental correction parameters. Then, the environmental correction parameters are loaded into the basic handling framework, and environmental adaptation processing is performed on nodes involving execution space or resource allocation to form an environmental adaptation framework. Next, the current disaster situation is obtained, and similar case matching is performed after feature extraction of the current disaster situation. The handling effect score is extracted from the historical handling results obtained by matching, thereby generating an optimized handling step sequence. Finally, the optimized handling step sequence is fused and verified with the environmental adaptation framework. Under the condition of meeting the handling constraints, resource matching and path calculation are carried out, and emergency equipment strategy, emergency personnel deployment strategy, and support path strategy are output, thereby forming a complete disaster relief execution plan for the current disaster situation.

[0107] By introducing environmental correction parameters to dynamically modify the basic response framework based on the generated response process structure, the environmental adaptation framework maintains consistency with the current meteorological and road condition data at the disaster site at the logical structure level. Simultaneously, by matching similar cases and extracting response effectiveness scores based on the current disaster situation, an optimized response step sequence is generated, providing historical data support for process optimization. In the final stage, the optimized response step sequence and the environmental adaptation framework are integrated and verified. Resource matching and path calculation are performed under the premise of meeting response constraints, thereby avoiding situations where response steps are logically feasible but spatially limited. This improves the adaptability between emergency equipment strategies, emergency personnel deployment strategies, and support path strategies and actual environmental conditions, enhancing the executability and stability of emergency rescue decisions and overcoming the problem of insufficient adaptability between the response process structure and actual environmental conditions in existing technologies.

[0108] like Figure 6 As shown in the example, this application also provides an intelligent dispatching system for emergency rescue, which specifically includes a framework generation module 10, an environment correction module 11, a data loading module 12, a case matching module 13, and a strategy generation module 14.

[0109] The framework generation module 10 is used to obtain the type and level of emergency response to determine the corresponding set of response rules and handling constraints, and to generate a basic handling framework corresponding to the set of response rules.

[0110] The environmental correction module 11 is used to acquire the current meteorological data and road condition data at the disaster relief location, calculate the meteorological risk coefficient based on the meteorological data, calculate the road section passability coefficient based on the road condition data, correct the road section passability coefficient using the meteorological risk coefficient, and generate environmental correction parameters.

[0111] The data loading module 12 is used to load environmental correction parameters into the basic processing framework to generate an environment adaptation framework.

[0112] The case matching module 13 is used to obtain the current disaster situation, perform similar case matching and extract the treatment effect score of the current disaster situation, so as to generate an optimized treatment step sequence.

[0113] The strategy generation module 14 is used to map each step item in the optimized disposal step sequence to the corresponding disposal stage node in the environment adaptation framework, and remove step items whose resource requirement parameters and environmental correction parameters conflict, to obtain a fusion verification step set. Under the condition that the fusion verification step set meets the disposal constraints, resource matching and path calculation are performed to obtain emergency equipment strategy, emergency personnel deployment strategy and support path strategy.

[0114] In this application example, by setting up a framework generation module 10, an environment correction module 11, a data loading module 12, a case matching module 13, and a strategy generation module 14, a modular division of labor and sequential connection of data processing links are formed, which include the determination of rules for disaster relief type and level, environmental quantification processing of meteorological and road condition data, structured loading of environmental correction parameters, matching of similar cases of the current disaster situation, generation of optimized disposal step sequences, and final resource matching and path calculation. This enables the transmission and constraint verification between functional units using parameters and step sequences as carriers, thereby completing the collaborative generation of emergency equipment strategies, emergency personnel deployment strategies, and support path strategies at the system level, improving the structural clarity and functional coupling of system operation, and enhancing the stability and scalability of overall scheduling.

[0115] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the system and each module described above can be referred to the corresponding process in the aforementioned Embodiment 1, and will not be repeated here.

[0116] like Figure 7 As shown, this application example also provides an electronic device 20, including a memory 21 and a processor 22. The memory 21 stores a computer program that can run on the processor 22. When the processor 22 executes the computer program, it implements the above-mentioned intelligent strategy generation method for emergency rescue.

[0117] In this application example, by storing a computer program for generating intelligent strategies for emergency rescue in the memory 21 of the electronic device 20, and having the processor 22 call and execute the computer program, the electronic device 20 is equipped with the ability to comprehensively process the type of rescue, the level of rescue, meteorological data, road condition data and the current disaster situation. The calculation of environmental correction parameters, the construction of environmental adaptation framework, the generation of optimized disposal step sequence, and the resource matching and path calculation process are integrated into an integrated operation process, realizing the automated generation of emergency equipment strategies, emergency personnel deployment strategies and support path strategies, improving strategy generation efficiency and reducing the uncertainty caused by human intervention.

[0118] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when run by a processor, causes the processor to execute the intelligent strategy generation method for emergency response described above.

[0119] In this application example, by storing a computer program for generating an intelligent strategy for emergency rescue in a computer-readable storage medium, the processor can perform steps such as identifying the type and level of emergency rescue, generating environmental correction parameters, constructing an environmental adaptation framework, matching similar cases and extracting the treatment effect score, and performing fusion verification and resource path calculation when reading and running the computer program. This allows the complete strategy generation process to be reproduced on different hardware platforms, enabling the programmatic deployment and cross-device application of the method, and improving the portability and application value of the technical solution.

[0120] The above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for generating intelligent strategies for emergency rescue and disaster relief, characterized in that, include: Obtain the identifier of the disaster relief type, match the identifier with the pre-stored type-rule mapping table, and extract the corresponding rule subset; Obtain the emergency response level, match the emergency response level with a pre-stored level-weight mapping table to obtain the activation weight and constraint threshold corresponding to each rule subset; The rule entries in the rule subset are filtered and weighted using the activation weight to obtain a response rule set, and disposal constraints are constructed using the constraint threshold. Each rule entry in the response rule set is mapped to a disposal stage node according to its category attribute, resulting in a framework node set. Logical directed connections are established between the disposal stage nodes in the framework node set based on the triggering dependencies between the rule entries to generate a directed disposal flowchart. The directed disposal flowchart and the disposal constraints are then subjected to compliance verification, and nodes and connections that violate the disposal constraints are removed to obtain the basic disposal framework. Obtain current meteorological and road condition data at the disaster relief location; calculate the meteorological risk coefficient based on the meteorological data; calculate the road segment passability coefficient based on the road condition data; construct a road network capacity matrix using the passability coefficients of each road segment; correct the road network capacity matrix using the meteorological risk coefficient to obtain a corrected capacity matrix; and generate environmental correction parameters based on the meteorological risk coefficient and the road segment passability coefficients in the corrected capacity matrix. The process involves identifying nodes in the basic response framework that include action space attributes or resource allocation attributes at each response stage. Correction components corresponding to the action space attributes or resource allocation attributes of the nodes in the environmental correction parameters are then loaded onto the corresponding nodes to obtain an environmental correction node set. An environmental accessibility assessment structure is constructed around the disaster relief location based on the corrected accessibility matrix. Accessibility indicators for each response stage node are extracted from the environmental accessibility assessment structure according to the execution location. Execution locations with accessibility indicators below a preset accessibility threshold are marked as restricted areas, and alternative execution nodes are generated in the basic response framework for response stage nodes involving restricted areas. The environmental correction node set replaces the original nodes in the basic response framework, and the alternative execution nodes are inserted into the directed response flowchart at positions corresponding to the original response stage nodes to generate an environmental adaptation framework. The current disaster situation is obtained, and similar cases are matched and the treatment effect score is extracted based on the current disaster situation to generate an optimized treatment step sequence; Each step in the optimized disposal sequence is sequentially mapped to the corresponding disposal stage node in the environment adaptation framework. Steps whose resource requirement parameters conflict with environmental correction parameters are removed to obtain a fusion verification step set. Resource matching and path calculation are performed when the fusion verification step set meets the disposal constraints to obtain emergency equipment strategy, emergency personnel deployment strategy, and support path strategy.

2. The intelligent strategy generation method for emergency rescue according to claim 1, characterized in that, The steps of acquiring current meteorological and road condition data at the disaster relief location, calculating a meteorological risk coefficient based on the meteorological data, calculating road segment traffic coefficients based on the road condition data, constructing a road network traffic capacity matrix using the traffic coefficients of each road segment, correcting the road network traffic capacity matrix using the meteorological risk coefficient to obtain a corrected traffic capacity matrix, and generating environmental correction parameters based on the meteorological risk coefficient and the traffic coefficients of each road segment in the corrected traffic capacity matrix include: Real-time meteorological data of the disaster relief location is obtained from the meteorological data interface. The meteorological data is matched item by item with the pre-stored meteorological-risk mapping table to obtain the meteorological risk value. The meteorological risk value is then weighted and summed according to the risk weight of each meteorological element to obtain the meteorological risk coefficient. The real-time road condition data of the road network around the disaster relief location is obtained from the road condition data interface. The road condition data is combined with the pre-stored road condition-traffic capacity mapping table to calculate the traffic coefficient of each road segment in the road network around the disaster relief location. The road network traffic capacity matrix is ​​constructed based on the traffic coefficient of each road segment. The traffic coefficients of each road segment in the road network capacity matrix are corrected using the meteorological risk coefficient to obtain a corrected traffic capacity matrix. The difference between the corrected meteorological risk coefficient and the corrected traffic coefficients of each road segment in the corrected traffic capacity matrix is ​​normalized to obtain environmental correction parameters.

3. The intelligent strategy generation method for emergency rescue according to claim 1, characterized in that, The steps of obtaining the current disaster situation, performing similar case matching and handling effect score extraction on the current disaster situation, and generating an optimized handling step sequence include: Obtain the current disaster situation and convert the current disaster situation into a disaster feature vector according to a preset feature extraction rule; Historical case records are extracted from the historical emergency disaster relief database. Each historical case record includes a historical disaster feature vector, a corresponding sequence of response steps, and a response effect score. Calculate the cosine similarity between the disaster-affected feature vector and the historical disaster-affected feature vector in each historical case record, and select historical case records with a cosine similarity higher than a preset similarity threshold as the matching case set; Extract the sequence of handling steps for each historical case record from the matched case set, remove duplicate steps from the sequence of handling steps and merge them to obtain a candidate set of handling steps, and extract the handling effect score corresponding to each step in the matched case set. The comprehensive priority score of each step item is calculated by weighted multiplication of the treatment effect score with the cosine similarity of the historical case record in which it is located. The step items in the candidate treatment step set are then sorted in descending order according to the comprehensive priority score to obtain the optimized treatment step sequence.

4. The intelligent strategy generation method for emergency rescue according to claim 2, characterized in that, The steps of sequentially mapping each step in the optimized disposal step sequence to the corresponding disposal stage node in the environment adaptation framework, and removing step entries whose resource requirement parameters conflict with environmental correction parameters, to obtain a fusion verification step set, and performing resource matching and path calculation to obtain the emergency equipment strategy, emergency personnel deployment strategy, and support path strategy when the fusion verification step set satisfies the disposal constraints, include: Each step in the optimized treatment sequence is mapped sequentially to the corresponding treatment stage node in the environment adaptation framework. The resource requirement parameters of each step are compared and verified with the environmental correction parameters of the corresponding treatment stage node. Steps with conflicting resource requirement parameters and environmental correction parameters are removed to obtain the fusion verification step set. The execution order and resource consumption of each step in the fusion verification step set are constrained by the aforementioned processing constraints. Steps that exceed the constraint threshold are split or delayed to obtain a feasible set of steps that meet the processing constraints. The equipment requirement list for each step in the feasible steps set is matched with the pre-stored emergency equipment resource library to obtain the emergency equipment strategy. The personnel demand list for each step in the feasible steps set is matched with the pre-stored emergency personnel resource database to obtain an emergency personnel deployment strategy. Using the modified traffic capacity matrix as the road network weight input, with the disaster relief location as the endpoint and the current location of each emergency equipment and personnel as the starting point, the optimal path from each starting point to the endpoint is calculated to obtain the support path strategy.

5. An intelligent dispatch system for emergency rescue and disaster relief, characterized in that, include: The framework generation module is used to obtain the identifier of the disaster relief type, match the identifier with the pre-stored type-rule mapping table, and extract the corresponding rule subset; Obtain the emergency response level, match the emergency response level with a pre-stored level-weight mapping table to obtain the activation weight and constraint threshold corresponding to each rule subset; The rule entries in the rule subset are filtered and weighted using the activation weight to obtain a response rule set, and disposal constraints are constructed using the constraint threshold. Each rule entry in the response rule set is mapped to a disposal stage node according to its category attribute, resulting in a framework node set. Logical directed connections are established between the disposal stage nodes in the framework node set based on the triggering dependencies between the rule entries to generate a directed disposal flowchart. The directed disposal flowchart and the disposal constraints are then subjected to compliance verification, and nodes and connections that violate the disposal constraints are removed to obtain the basic disposal framework. The environmental correction module is used to acquire current meteorological and road condition data at the disaster relief location, calculate a meteorological risk coefficient based on the meteorological data, calculate a road segment passability coefficient based on the road condition data, construct a road network capacity matrix using the passability coefficients of each road segment, correct the road network capacity matrix using the meteorological risk coefficient to obtain a corrected capacity matrix, and generate environmental correction parameters based on the meteorological risk coefficient and the road segment passability coefficients in the corrected capacity matrix. The data loading module is used to identify nodes in the basic response framework that include action space attributes or resource allocation attributes at each response stage. It loads the correction components in the environmental correction parameters corresponding to the action space attributes or resource allocation attributes of the nodes into the corresponding nodes, obtaining an environmental correction node set. Based on the corrected accessibility matrix, it constructs an environmental accessibility assessment structure around the disaster relief location. According to the execution location corresponding to each response stage node, it extracts the accessibility index of the corresponding location from the environmental accessibility assessment structure. Execution locations with accessibility indices lower than a preset accessibility threshold are marked as restricted areas, and alternative execution nodes are generated in the basic response framework for response stage nodes involving restricted areas. The environmental correction node set replaces the original nodes in the basic response framework, and the alternative execution nodes are inserted into the directed response flowchart at positions corresponding to the original response stage nodes to generate an environmental adaptation framework. The case matching module is used to obtain the current disaster situation, perform similar case matching and extract the handling effect score of the current disaster situation, so as to generate an optimized handling step sequence; The strategy generation module is used to sequentially map each step item in the optimized disposal step sequence to the corresponding disposal stage node in the environment adaptation framework, and remove step items whose resource requirement parameters and environmental correction parameters conflict, to obtain a fusion verification step set. If the fusion verification step set satisfies the disposal constraints, resource matching and path calculation are performed to obtain emergency equipment strategy, emergency personnel deployment strategy, and support path strategy.

6. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the intelligent strategy generation method for emergency rescue as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, It stores a computer program, which, when run by a processor, causes the processor to execute the intelligent strategy generation method for emergency rescue as described in any one of claims 1 to 4.

Images