Emergency commanding and dispatching method and system for sudden environmental events

Abstract

The invention relates to the technical field of emergency command and dispatch, and discloses an emergency command and dispatch method and system for sudden environmental events. The method comprises the following steps: firstly, establishing a post-disaster transportation network map according to a map after an emergent environment event occurs, then acquiring emergency commanding and dispatching personnel data, and generating an emergency commanding personnel dispatching scheme; secondly, establishing a GRU neural network material prediction model based on a time sequence, and outputting a post-disaster material consumption prediction value of a disaster-affected point; establishing a material scheduling objective function according to the post-disaster material consumption predicted value of the disaster-affected point, and solving the material scheduling objective function by using an olfactory optimization algorithm to obtain an optimal material scheduling path; and finally, generating an emergency commanding material scheduling scheme according to the optimal material scheduling path, and generating a final emergency commanding scheduling scheme in combination with the emergency commanding personnel scheduling scheme. According to the method, the emergency command and dispatch personnel and materials are dispatched, the purpose of emergency command and dispatch is achieved, and the method is accurate and objective.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency command and dispatch, and specifically relates to an emergency command and dispatch method and system for sudden environmental events. Background Art

[0002] Chinese Patent CN113919605B discloses an emergency command and dispatch management system and method based on big data. The method specifically includes: obtaining tunnel historical safety data, constructing a multi-source database, establishing a tunnel safety prediction model according to the multi-source data, and outputting the tunnel safety risk prediction value for the next cycle of the tunnel; secondly, using remote sensing information to obtain the topographic map of the tunnel area, constructing rescue points based on actual geographical factors, and storing rescue materials; then using the historical data of the traffic conditions in the tunnel area, constructing a big data information database, obtaining rescue points, establishing a traffic rescue path for the next cycle, and obtaining a preset traffic rescue path; finally, constructing an emergency command and dispatch simulation platform. When the tunnel safety risk prediction value for the next cycle exceeds the threshold, an emergency simulation rescue is carried out according to the preset traffic rescue path, and the results of the emergency plan are recorded. However, this invention does not optimize resource scheduling, and the scheduling cost is relatively high.

[0003] Traditional command and dispatch methods usually carry out scheduling of materials, information, etc. according to a specified plan, resulting in an ineffective scheduling plan in the face of sudden environmental events; at the same time, after a sudden environmental event occurs, traditional emergency command and dispatch methods are slow in emergency command and dispatch because they do not use technologies such as artificial intelligence, thus affecting the efficiency and effect of emergency disposal. Summary of the Invention

[0004] In view of the problems in the related art, the present invention provides an emergency command and dispatch method and system for sudden environmental events to overcome the above technical problems existing in the existing related technologies.

[0005] To solve the above technical problems, the present invention is realized through the following technical solutions:

[0006] The present invention provides an emergency command and dispatch method for sudden environmental events, including the following steps:

[0007] S1. Obtain the map after a sudden environmental event occurs, establish a post-disaster transportation network map according to the map after the sudden environmental event occurs, then obtain the emergency command and dispatch personnel data, and combine the emergency command and dispatch personnel data and the post-disaster personnel consumption for command and dispatch to generate an emergency command personnel scheduling plan;

[0008] S2. Obtain the post-disaster material consumption set, establish a GRU neural network material prediction model based on time series, and output the post-disaster material consumption prediction value of the affected points in the post-disaster transportation network map;

[0009] S3. Based on the predicted value of post-disaster material consumption at the disaster-stricken points, establish a material scheduling objective function based on the principles of material adaptation and minimum transportation time, and use the olfactory optimization algorithm to solve the material scheduling objective function to obtain the optimal material scheduling path;

[0010] S4. Generate an emergency command material scheduling plan according to the optimal material scheduling path, and combine the emergency command personnel scheduling plan and the emergency command material scheduling plan to generate a final emergency command scheduling plan to complete the emergency command and scheduling of sudden environmental incidents.

[0011] The invention establishes a post-disaster transportation network map based on the map after a sudden environmental incident occurs, and considers the degree of road damage, and then conducts command and scheduling to generate an emergency command personnel scheduling plan; this method simplifies the map to establish a topological map, converts the abstract problem into a mathematical model, which is convenient for the subsequent solution of the algorithm; secondly, based on time series, a GRU neural network material prediction model is established to output the predicted value of post-disaster material consumption at the disaster-stricken points in the post-disaster transportation network map; compared with other neural networks, the GRU neural network has fewer training parameters, lower computational complexity, solves the problem of gradient disappearance, and has a better prediction effect using time series, and the output predicted value is convenient for subsequent material scheduling processing; then, based on the principles of material adaptation and minimum transportation time, a material scheduling objective function is established, and the olfactory optimization algorithm is used to solve the material scheduling objective function to obtain the optimal material scheduling path, realizing the shortest transportation time and the least material underdelivery, and effectively dealing with the optimization problem of the material scheduling path; the olfactory optimization algorithm finds the optimal solution by simulating the release, diffusion and detection process of odor molecules, and this optimization algorithm has strong global search ability, effectively avoids falling into local optimal solutions, and is easy to implement and adjust, with good applicability; finally, generate an emergency command material scheduling plan, and then combine the emergency command material scheduling plan to generate a final emergency command scheduling plan to complete the emergency command and scheduling of sudden environmental incidents.

[0012] Preferably, the S1 includes the following steps:

[0013] S11. After a sudden environmental incident occurs, obtain the map after the sudden environmental incident in the target area. Consider each affected area in the map after the sudden environmental incident as an affected point, and the non-affected area as a non-affected point. Obtain the damaged data of each transportation road in the map after the sudden environmental incident, set a damage threshold, calculate the proportion of damaged transportation roads based on the damaged data of the transportation roads. When the proportion of damaged transportation roads is greater than the damage threshold, mark the corresponding transportation road as an invalid transportation road, and delete the invalid transportation road from the map after the sudden environmental incident. Otherwise, mark the corresponding transportation road as a valid transportation road, and mark the valid transportation road in the map after the sudden environmental incident to form a post-disaster transportation road network; combine the post-disaster transportation road network, affected points, and non-affected points to establish a post-disaster transportation network map;

[0014] S12. Obtain emergency command and dispatching personnel data, which includes emergency expert data, emergency personnel scheduling data, emergency command center data, etc. Obtain the post-disaster personnel consumption after the sudden environmental incident occurs. Conduct command and dispatching based on the post-disaster personnel consumption and emergency command and dispatching personnel data to generate an emergency command personnel scheduling plan. The specific steps are as follows:

[0015] S121. According to the post-disaster personnel consumption after the sudden environmental incident occurs, calculate the post-disaster personnel consumption of each affected point in the post-disaster transportation network map, and conduct quantification processing to obtain a set of post-disaster personnel consumption data; calculate the weights of the post-disaster personnel consumption data in the set of post-disaster personnel consumption data to obtain post-disaster personnel consumption weights;

[0016] S122. Allocate the emergency command and dispatching personnel data according to the post-disaster personnel consumption weights, and record and update the post-disaster personnel consumption in real time, dynamically adjust the emergency command and dispatching personnel data to obtain an emergency command personnel scheduling plan.

[0017] The invention establishes a post-disaster transportation network map based on the map after the sudden environmental incident occurs, and considering the degree of road damage, simplifies the map to establish a topological map, converts the abstract problem into a mathematical model, which is convenient for the subsequent algorithm to solve, and conducts command and dispatching to generate an emergency command personnel scheduling plan.

[0018] Preferably, the S2 includes the following steps:

[0019] S21. Statistically calculate the post-disaster material consumption of the affected points in the post-disaster transportation network map. The post-disaster material consumption includes drinking water, grain and oil, protective clothing, etc., to obtain a set of post-disaster material consumption A′={a1, a2, a3,..., a m}, where a mIt represents the consumption quantity of the m-th post-disaster material; then obtain the historical data of the post-disaster material consumption to get the historical data set A″ of the post-disaster material consumption = {b1, b2, b3,..., b n}, where b n represents the historical data of the n-th post-disaster material consumption; and count the post-disaster material consumption time points to form the material consumption time series A″′ = {A1, A2, A3,..., A p}, where A p represents the p-th post-disaster material consumption time point;

[0020] S22. According to the historical data set of the post-disaster material consumption and the material consumption time series, establish a GRU neural network material prediction model based on the time series. The specific steps are as follows:

[0021] S221. Divide the material consumption time series into the first time series B1 = {A1, A2, A3,..., A q} and the second time series B2 = {A q+1 , A q+2 ,..., A p}, where A q represents the q-th post-disaster material consumption time point. Obtain the historical data of the post-disaster material consumption corresponding to the first time series, denoted as the training sample data set, and obtain the historical data of the post-disaster material consumption corresponding to the second time series, denoted as the test sample data set; set the GRU neural network model to include an input layer, a hidden layer, and an output layer. The input layer includes an update gate and a reset gate. Input the training sample data set into the GRU neural network model and continuously iterate until the loss function in the GRU neural network model converges to obtain a trained GRU neural network model;

[0022] S222. Then input the test sample data set into the trained GRU neural network model, output the prediction result, set the accuracy threshold. When the accuracy of the prediction result is greater than the accuracy threshold, stop the iteration to obtain the GRU neural network material prediction model; otherwise, adjust the weights until the accuracy of the prediction result is greater than the accuracy threshold;

[0023] S23. Input the post-disaster material consumption set into the GRU neural network material prediction model, output the predicted value of each post-disaster material consumption, and then sequentially perform post-disaster material consumption predictions on other disaster-stricken points in the post-disaster transportation network map, and output the predicted values of the post-disaster material consumption at the disaster-stricken points in the post-disaster transportation network map.

[0024] The invention establishes a GRU neural network material prediction model based on time series, outputs the predicted value of post-disaster material consumption. Compared with other neural networks, the GRU neural network has fewer training parameters, lower computational complexity, solves the problem of gradient disappearance, and has a better prediction effect using time series. The output predicted value is convenient for subsequent material scheduling processing.

[0025] Preferably, S3 includes the following steps:

[0026] S31. According to the post-disaster transportation network map, regard the disaster-stricken points and non-disaster-stricken points as transportation nodes. According to the predicted value of post-disaster material consumption at the disaster-stricken points, obtain the post-disaster material demand and the shortage quantity at the disaster-stricken points, and set the shortage coefficient D b , when the shortage coefficient is equal to 1, the b-th disaster-stricken point needs post-disaster materials; when the shortage coefficient is equal to 0, the b-th disaster-stricken point does not need post-disaster materials, and establish a minimum material shortage target function based on the material matching principle where s represents the number of disaster-stricken points, and D b ′ represents the post-disaster material demand at the b-th disaster-stricken point;

[0027] Set up a material distribution center, calculate the transportation time from the material distribution center to each disaster-stricken point in the post-disaster transportation network map, record the material scheduling path from the material distribution center to each disaster-stricken point in the post-disaster transportation network map, and set the transportation time from the i-th transportation node to the j-th transportation node as C i,j , and the decision variable is α i,j , when the decision variable is equal to 1, select the path from the i-th transportation node to the j-th transportation node as the material scheduling path; when the decision variable is equal to 0, do not select the path from the i-th transportation node to the j-th transportation node as the material scheduling path, and establish a shortest transportation time target function based on the shortest transportation time principle where r represents the number of transportation nodes;

[0028] Combine the shortest transportation time target function and the minimum material shortage target function, and assign weights to establish a material scheduling target function;

[0029] S32. Use the material scheduling target function as the fitness function, use the fitness function value to measure the material scheduling path, and use the olfactory optimization algorithm to solve the material scheduling target function to obtain the optimal material scheduling path. The specific steps are as follows:

[0030] S321. Set up the search space. There are odor molecules in the search space, and each odor molecule represents a feasible material scheduling path. Encode the material distribution center and transportation nodes, initialize the odor molecules, and determine the initial positions and initial velocities of the odor molecules. During the sniffing phase of the odor molecules, it is set that the odor molecules are affected by temperature and mass. The temperature coefficient and mass coefficient are χ and δ respectively, the odor constant is c, d1 represents a random number between the interval (0, 1), then the update amount of the initial velocity of the odor molecules Set the current iteration number as t, and the velocity of the e-th odor molecule at the t-th iteration is v e (t). Then, at the (t + 1)-th iteration, the velocity v e (t + 1) = v e (t) + v′. At the t-th iteration, the position of the e-th odor molecule is w e (t). Then, at the (t + 1)-th iteration, the position w e (t + 1) = w e (t) + v e (t + 1);

[0031] S322. At this time, calculate the fitness function value corresponding to the position of the odor molecule, obtain the current best fitness function value and the current worst fitness function value, and record the odor molecule corresponding to the current best fitness function value as the current optimal material scheduling path. During the following phase of the odor molecules, set d2 and d3 to represent random numbers between the interval (0, 1), the olfactory coefficient is ε, the position of the odor molecule corresponding to the best fitness function value at the t-th iteration is w′(t), and the position of the odor molecule corresponding to the worst fitness function value at the t-th iteration is w″(t). Then w e (t + 1) = w e (t) + d2·ε·(w′(t) - w e (t)) - d3·ε·(w″(t) - w e (t)); During the random phase of the odor molecules, the concentration of the odor molecules changes. Set the random search step size as g, d4 represents a random number between the interval (0, 1), update the position of the e-th odor molecule at the (t + 1)-th iteration, and obtain w e (t + 1) = w e (t) + d4·g, generate new odor molecules for the next iteration; Move in the direction with the highest concentration of following odor molecules. Set the maximum number of iterations. When the current iteration number reaches the maximum number of iterations, stop the iteration to obtain the final position of the odor molecule;

[0032] S323. Obtain the best fitness function value at the final position of the odor molecule, and obtain the optimal material scheduling path according to the final position of the odor molecule.

[0033] The invention establishes a material scheduling objective function based on the principles of material adaptation and minimum transportation time, uses the olfactory optimization algorithm to solve the material scheduling objective function, and finds the optimal solution by simulating the release, diffusion, and detection processes of odor molecules, obtaining the optimal material scheduling path, achieving the shortest transportation time and the least material shortage. The optimization algorithm has strong global search ability, effectively avoids falling into local optimal solutions, and is easy to implement and adjust, with good applicability.

[0034] Preferably, S4 includes the following steps:

[0035] S41. Transport the post-disaster materials to the disaster-stricken areas according to the optimal material scheduling path, and achieve the least material shortage and the shortest transportation time, generating an emergency command material scheduling plan; combine the emergency command personnel scheduling plan and the emergency command material scheduling plan to obtain the final emergency command scheduling plan; after a sudden environmental event occurs, adjust the post-disaster materials and the distribution of post-disaster personnel in real time according to the final emergency command scheduling plan to complete the emergency command and dispatch of the sudden environmental event.

[0036] This embodiment also discloses a system for the emergency command and dispatch method of a sudden environmental event, specifically including: an emergency command personnel scheduling module, a post-disaster material consumption prediction module, a material scheduling path optimization module, and an emergency command scheduling plan generation module;

[0037] The emergency command personnel scheduling module is used to conduct command and dispatch according to the emergency command and dispatch personnel data and the post-disaster personnel consumption;

[0038] The post-disaster material consumption prediction module is used to establish a GRU neural network material prediction model and output the predicted value of the post-disaster material consumption at the disaster-stricken area;

[0039] The material scheduling path optimization module is used to use the olfactory optimization algorithm to solve the material scheduling objective function and establish the optimal material scheduling path;

[0040] The emergency command scheduling plan generation module is used to combine the emergency command personnel scheduling plan and the emergency command material scheduling plan to generate the final emergency command scheduling plan.

[0041] The present invention has the following beneficial effects:

[0042] 1. The invention establishes a post-disaster transportation network map based on the map after a sudden environmental event occurs, and considering the degree of road damage, simplifies the map to establish a topological map, converting the abstract problem into a mathematical model, which is convenient for the subsequent solution of the algorithm, and conducts command and dispatch to generate an emergency command personnel scheduling plan.

[0043] 2. The invention establishes a GRU neural network material prediction model based on time series, outputs the predicted value of post-disaster material consumption. Compared with other neural networks, the GRU neural network has fewer training parameters, lower computational complexity, solves the problem of gradient disappearance, and has better prediction effect using time series. The output predicted value is convenient for subsequent material scheduling processing.

[0044] 3. The invention establishes a material scheduling objective function based on the principles of material adaptation and minimum transportation time, uses the olfactory optimization algorithm to solve the material scheduling objective function, and searches for the optimal solution by simulating the release, diffusion, and detection processes of odor molecules, obtaining the optimal material scheduling path, achieving the shortest transportation time and minimum material underdelivery. The optimization algorithm has strong global search ability, effectively avoids falling into local optimal solutions, and is easy to implement and adjust, with good applicability.

[0045] Of course, it is not necessary for any product implementing the present invention to simultaneously achieve all the above-mentioned advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0047] Figure 1 The present invention provides a schematic flowchart of the emergency command and dispatch process of an emergency command and dispatch system for sudden environmental events. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0049] In the description of the present invention, it should be understood that the terms "opening", "upper", "lower", "top", "middle", "inner", etc. indicating the orientation or position relationship are only for the convenience of describing the invention and simplifying the description, rather than indicating or implying that the components or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention.

[0050] Embodiment 1

[0051] Please refer to Figure 1, this implementation discloses an emergency command and dispatch method for sudden environmental incidents, which specifically includes the following content:

[0052] S1. Obtain the map after the sudden environmental incident occurs, establish a post-disaster transportation network map based on the map after the sudden environmental incident occurs, then obtain the emergency command and dispatch personnel data, and conduct command and dispatch by combining the emergency command and dispatch personnel data and the post-disaster personnel consumption, so as to generate an emergency command personnel dispatch plan;

[0053] The S1 includes the following steps:

[0054] S11. After the sudden environmental incident occurs, obtain the map after the sudden environmental incident in the target area, regard each affected area in the map after the sudden environmental incident as an affected point, and regard the non-affected area as a non-affected point, and obtain the affected data of each transportation road in the map after the sudden environmental incident, set a damage threshold, calculate the proportion of damaged transportation roads according to the affected data of the transportation roads, and when the proportion of damaged transportation roads is greater than the damage threshold, mark the corresponding transportation road as an invalid transportation road and delete the invalid transportation road in the map after the sudden environmental incident, otherwise mark the corresponding transportation road as a valid transportation road, mark the valid transportation road in the map after the sudden environmental incident, and form a post-disaster transportation road network; Combine the post-disaster transportation road network, affected points and non-affected points to establish a post-disaster transportation network map;

[0055] S12. Obtain the emergency command and dispatch personnel data, where the emergency command and dispatch personnel data includes emergency expert data, emergency personnel scheduling data, emergency command center data, etc., obtain the post-disaster personnel consumption after the sudden environmental incident occurs, and conduct command and dispatch according to the post-disaster personnel consumption and the emergency command and dispatch personnel data to generate an emergency command personnel dispatch plan. The specific steps are as follows:

[0056] S121. According to the post-disaster personnel consumption after the sudden environmental incident occurs, calculate the post-disaster personnel consumption of each affected point in the post-disaster transportation network map, and conduct quantization processing to obtain a post-disaster personnel consumption data set; Calculate the weights of the post-disaster personnel consumption data in the post-disaster personnel consumption data set to obtain the post-disaster personnel consumption weights;

[0057] S122. Allocate the emergency command and dispatch personnel data according to the post-disaster personnel consumption weights, and record and update the post-disaster personnel consumption in real time, and dynamically adjust the emergency command and dispatch personnel data to obtain an emergency command personnel dispatch plan;

[0058] S2. Obtain the post-disaster material consumption set, establish a GRU neural network material prediction model based on the time series, and output the predicted value of the post-disaster material consumption of the affected points in the post-disaster transportation network map;

[0059] S2 includes the following steps:

[0060] S21. Statistically analyze the post-disaster material consumption at the disaster-stricken points in the post-disaster transportation network map. The post-disaster material consumption includes drinking water, grain and oil, protective clothing, etc., and obtain the post-disaster material consumption set A′ = {a1, a2, a3,..., a m}, where a m represents the consumption quantity of the m-th post-disaster material; then obtain the historical data of the post-disaster material consumption quantity to get the historical data set of the post-disaster material consumption quantity A″ = {b1, b2, b3,..., b n}, where b n represents the historical data of the n-th post-disaster material consumption quantity; and statistically analyze the post-disaster material consumption time points to form a material consumption time series A″′ = {A1, A2, A3,..., A p}, where A p represents the p-th post-disaster material consumption time point;

[0061] S22. Based on the historical data set of the post-disaster material consumption quantity and the material consumption time series, establish a GRU neural network material prediction model based on the time series. The specific steps are as follows:

[0062] S221. Divide the material consumption time series into a first time series B1 = {A1, A2, A3,..., A q} and a second time series B2 = {A q+1 , A q+2 ,..., A p}, where A q represents the q-th post-disaster material consumption time point. Obtain the historical data of the post-disaster material consumption quantity corresponding to the first time series, denoted as the training sample data set, and obtain the historical data of the post-disaster material consumption quantity corresponding to the second time series, denoted as the test sample data set; set that the GRU neural network model includes an input layer, a hidden layer, and an output layer. The input layer includes an update gate and a reset gate. Input the training sample data set into the GRU neural network model and continuously iterate until the loss function in the GRU neural network model converges to obtain a trained GRU neural network model;

[0063] S222. Then input the test sample data set into the trained GRU neural network model to output a prediction result. Set an accuracy threshold. When the accuracy of the prediction result is greater than the accuracy threshold, stop the iteration to obtain the GRU neural network material prediction model; otherwise, adjust the weights until the accuracy of the prediction result is greater than the accuracy threshold;

[0064] S23. Input the post-disaster material consumption set into the GRU neural network material prediction model to output the predicted value of each post-disaster material consumption. Then, sequentially conduct post-disaster material consumption predictions for other disaster-stricken points in the post-disaster transportation network map, and output the predicted values of post-disaster material consumption for the disaster-stricken points in the post-disaster transportation network map.

[0065] S3. Based on the predicted values of post-disaster material consumption at the disaster-stricken points, establish a material scheduling objective function based on the principles of material adaptation and minimum transportation time. Use the olfactory optimization algorithm to solve the material scheduling objective function to obtain the optimal material scheduling path.

[0066] The S3 includes the following steps:

[0067] S31. According to the post-disaster transportation network map, regard the disaster-stricken points and non-disaster-stricken points as transportation nodes. Based on the predicted values of post-disaster material consumption at the disaster-stricken points, obtain the post-disaster material demand and the under-delivery quantity at the disaster-stricken points, and set the under-delivery coefficient D b . When the under-delivery coefficient is equal to 1, the b-th disaster-stricken point needs post-disaster materials. When the under-delivery coefficient is equal to 0, the b-th disaster-stricken point does not need post-disaster materials. And establish a minimum material under-delivery objective function based on the principle of material adaptation where s represents the number of disaster-stricken points, and D b ′ represents the post-disaster material demand at the b-th disaster-stricken point;

[0068] Set the material distribution center, calculate the transportation time from the material distribution center to each disaster-stricken point in the post-disaster transportation network map, record the material scheduling paths from the material distribution center to each disaster-stricken point in the post-disaster transportation network map, and set the transportation time from the i-th transportation node to the j-th transportation node as C i,j , and the decision variable is α i,j . When the decision variable is equal to 1, select the i-th transportation node to the j-th transportation node as the material scheduling path. When the decision variable is equal to 0, do not select the i-th transportation node to the j-th transportation node as the material scheduling path. Establish a shortest transportation time objective function based on the principle of the shortest transportation time where r represents the number of transportation nodes;

[0069] Combine the shortest transportation time objective function and the minimum material under-delivery objective function, and assign weights to establish a material scheduling objective function;

[0070] S32. Take the material scheduling objective function as the fitness function, use the fitness function value to measure the material scheduling path, and use the olfactory optimization algorithm to solve the material scheduling objective function to obtain the optimal material scheduling path. The specific steps are as follows:

[0071] S321. Set the search space. There are odor molecules in the search space, and each odor molecule represents a feasible material scheduling path. Encode the material distribution center and transportation nodes, initialize the odor molecules, and determine the initial positions and initial velocities of the odor molecules. During the sniffing stage of the odor molecules, it is set that the odor molecules are affected by temperature and mass, with the temperature coefficient and mass coefficient being χ and δ respectively, the odor constant being c, and d1 representing a random number in the interval (0, 1). Then the updated amount of the initial velocity of the odor molecules Set the current iteration number as t, and the velocity of the e-th odor molecule at the t-th iteration is v e (t). Then the velocity of the e-th odor molecule at the (t + 1)-th iteration v e (t + 1) = v e (t) + v′. The position of the e-th odor molecule at the t-th iteration is w e (t). Then the position of the e-th odor molecule at the (t + 1)-th iteration w e (t + 1) = w e (t) + v e (t + 1);

[0072] S322. At this time, calculate the fitness function value corresponding to the position of the odor molecule, obtain the current best fitness function value and the current worst fitness function value, and record the odor molecule corresponding to the current best fitness function value as the current optimal material scheduling path. During the following stage of the odor molecules, it is set that d2 and d3 represent random numbers in the interval (0, 1), the olfactory coefficient is ε, the position of the odor molecule corresponding to the best fitness function value at the t-th iteration is w′(t), and the position of the odor molecule corresponding to the worst fitness function value at the t-th iteration is w″(t). Then w e (t + 1) = w e (t) + d2·ε·(w′(t) - w e (t)) - d3·ε·(w″(t) - w e (t)); During the random stage of the odor molecules, the concentration of the odor molecules changes. Set the random search step size as g, and d4 represents a random number in the interval (0, 1). Update the position of the e-th odor molecule at the (t + 1)-th iteration to obtain w e (t + 1) = w e (t) + d4·g, generate new odor molecules for the next iteration; Move in the direction with the highest concentration of following odor molecules. Set the maximum number of iterations. When the current iteration number reaches the maximum number of iterations, stop the iteration to obtain the final position of the odor molecule;

[0073] S323. Obtain the best fitness function value at the final position of the odor molecule, and obtain the optimal material scheduling path according to the final position of the odor molecule;

[0074] S4. Generate an emergency command material dispatching plan according to the optimal material dispatching path, and generate a final emergency command dispatching plan by combining the emergency command personnel dispatching plan and the emergency command material dispatching plan, thus completing the emergency command and dispatching for sudden environmental incidents.

[0075] The S4 includes the following steps:

[0076] S41. Transport the post-disaster materials to the disaster-stricken areas according to the optimal material dispatching path, and achieve the minimum material shortage and the shortest transportation time to generate an emergency command material dispatching plan; generate a final emergency command dispatching plan by combining the emergency command personnel dispatching plan and the emergency command material dispatching plan; after a sudden environmental incident occurs, adjust the post-disaster materials and the distribution of post-disaster personnel in real time according to the final emergency command dispatching plan, thus completing the emergency command and dispatching for sudden environmental incidents.

[0077] Embodiment 2

[0078] This embodiment also discloses a system for the emergency command and dispatching method of sudden environmental incidents, which specifically includes: an emergency command personnel dispatching module, a post-disaster material consumption prediction module, a material dispatching path optimization module, and an emergency command dispatching plan generation module;

[0079] The emergency command personnel dispatching module is used to conduct command and dispatching according to the emergency command dispatching personnel data and the post-disaster personnel consumption.

[0080] The post-disaster material consumption prediction module is used to establish a GRU neural network material prediction model and output the predicted value of the post-disaster material consumption at the disaster-stricken areas.

[0081] The material dispatching path optimization module is used to solve the material dispatching objective function using the olfactory optimization algorithm and establish the optimal material dispatching path.

[0082] The emergency command dispatching plan generation module is used to generate a final emergency command dispatching plan by combining the emergency command personnel dispatching plan and the emergency command material dispatching plan.

[0083] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0084] The preferred embodiments of the invention disclosed above are merely used to assist in the description of the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification in order to better explain the principles and practical applications of the invention, so that those skilled in the art can well understand and utilize the invention.

Claims

1. A method for emergency command and dispatch of sudden environmental incidents, characterized in that, It includes the following steps: S1. Obtain the map after the occurrence of a sudden environmental incident, establish a post-disaster transportation network map based on the map after the occurrence of the sudden environmental incident, then obtain the data of emergency command and dispatch personnel, and conduct command and dispatch in combination with the data of emergency command and dispatch personnel and post-disaster personnel consumption to generate an emergency command personnel dispatch plan; S2. Obtain the post-disaster material consumption set, establish a GRU neural network material prediction model based on the time series, and output the predicted value of post-disaster material consumption at the disaster-affected points in the post-disaster transportation network map; S3. According to the predicted value of post-disaster material consumption at the disaster-affected points, and based on the principles of material adaptation and minimum transportation time, establish a material dispatch objective function, solve the material dispatch objective function, and obtain the optimal material dispatch path; S4. Generate an emergency command material dispatch plan according to the optimal material dispatch path, and combine the emergency command personnel dispatch plan and the emergency command material dispatch plan to generate a final emergency command and dispatch plan to complete the emergency command and dispatch of sudden environmental incidents.

2. The emergency command and dispatch method for sudden environmental incidents according to claim 1, characterized in that The S1 includes the following steps: S11. After the occurrence of a sudden environmental incident, obtain the map after the occurrence of the sudden environmental incident in the target area to obtain the disaster-affected points and non-disaster-affected points; set a damage threshold, then calculate the proportion of damaged transportation roads to obtain the effective transportation roads, mark the effective transportation roads on the map after the occurrence of the sudden environmental incident, and establish a post-disaster transportation network map; S12. Obtain the data of emergency command and dispatch personnel and the post-disaster personnel consumption after the occurrence of the sudden environmental incident, and conduct command and dispatch according to the post-disaster personnel consumption and the data of emergency command and dispatch personnel to generate an emergency command personnel dispatch plan.

3. The emergency command and dispatch method for sudden environmental incidents according to claim 2, characterized in that, The S12 includes the following steps: S121. According to the post-disaster personnel consumption after the occurrence of the sudden environmental incident, obtain the post-disaster personnel consumption data set, calculate the weights of the post-disaster personnel consumption data in the post-disaster personnel consumption data set to obtain the post-disaster personnel consumption weights; S122. Distribute the data of emergency command and dispatch personnel according to the post-disaster personnel consumption weights, and record and update the post-disaster personnel consumption in real time, and dynamically adjust the data of emergency command and dispatch personnel to obtain an emergency command personnel dispatch plan.

4. The emergency command and dispatch method for sudden environmental incidents according to claim 3, characterized in that, The S2 includes the following steps: S21. Statistically analyze the post-disaster material consumption at the disaster-affected points in the post-disaster transportation network map to obtain the post-disaster material consumption set; then obtain the historical data of post-disaster material consumption to obtain the historical data set of post-disaster material consumption, and statistically analyze the post-disaster material consumption time points to form a material consumption time series; S22. Based on the historical data set of post-disaster material consumption and the material consumption time series, establish a GRU neural network material prediction model based on the time series; S23. Input the post-disaster material consumption set into the GRU neural network material prediction model, output the predicted value of each post-disaster material consumption, and obtain the predicted value of post-disaster material consumption at the disaster-affected points in the post-disaster transportation network map.

5. The emergency command and dispatch method for sudden environmental events according to claim 4, characterized in that, The S22 includes the following steps: S221. Divide the historical data set of post-disaster material consumption according to the material consumption time series to obtain a training sample data set and a test sample data set; set that the GRU neural network model includes an input layer, a hidden layer, and an output layer, and the input layer includes an update gate and a reset gate. Input the training sample data set into the GRU neural network model and iterate continuously until the loss function in the GRU neural network model converges to obtain a trained GRU neural network model. S222. Then input the test sample data set into the trained GRU neural network model to output a prediction result. Set an accuracy threshold. When the accuracy of the prediction result is greater than the accuracy threshold, stop the iteration to obtain the GRU neural network material prediction model; otherwise, adjust the weights until the accuracy of the prediction result is greater than the accuracy threshold.

6. The emergency command and dispatch method for sudden environmental incidents according to claim 5, wherein The S3 includes the following steps: S31. According to the post-disaster transportation network map, regard the disaster-stricken points and non-disaster-stricken points as transportation nodes. According to the predicted value of post-disaster material consumption at the disaster-stricken points, obtain the post-disaster material demand and the shortage quantity at the disaster-stricken points, and establish a minimum material shortage objective function based on the material adaptation principle. Set a material distribution center, calculate the transportation time from the material distribution center to each disaster-stricken point in the post-disaster transportation network map, record the material dispatching path from the material distribution center to each disaster-stricken point in the post-disaster transportation network map, and establish a shortest transportation time objective function based on the principle of the shortest transportation time. Combine the shortest transportation time objective function and the minimum material shortage objective function, and assign weights to establish a material dispatching objective function. S32. Use the material dispatching objective function as the fitness function, use the fitness function value to measure the material dispatching path, and use the olfactory optimization algorithm to solve the material dispatching objective function to obtain the optimal material dispatching path.

7. The emergency command and dispatch method for sudden environmental incidents according to claim 6, wherein The S32 includes the following steps: S321. Set the search space. There are odor molecules in the search space, and each odor molecule represents a feasible material scheduling path. Encode the material distribution center and transportation nodes, initialize the odor molecules, and determine the initial positions and initial velocities of the odor molecules. During the sniffing stage of the odor molecules, it is set that the odor molecules are affected by temperature and mass. The temperature coefficient and mass coefficient are χ and δ respectively, the odor constant is c, and d1 represents a random number between the interval (0, 1). Then the updated amount of the initial velocity of the odor molecule Set the current iteration number as t, and the velocity of the e-th odor molecule at the t-th iteration is v e (t). Then the velocity of the e-th odor molecule at the (t + 1)-th iteration is v e (t + 1) = v e (t) + v′. The position of the e-th odor molecule at the t-th iteration is w e (t). Then the position of the e-th odor molecule at the (t + 1)-th iteration is w e (t + 1) = w e (t) + v e (t + 1); S322. At this time, calculate the fitness function value corresponding to the position of the odor molecule to obtain the current best fitness function value and the current worst fitness function value. Denote the odor molecule corresponding to the current best fitness function value as the current optimal material scheduling path. During the following stage of the odor molecule, set d2 and d3 to represent random numbers in the interval (0, 1), the olfactory coefficient as ε, the position of the odor molecule corresponding to the best fitness function value at the t-th iteration as w′(t), and the position of the odor molecule corresponding to the worst fitness function value at the t-th iteration as w″(t). Then w e (t + 1) = w e (t) + d2·ε·(w′(t) - w e (t)) - d3·ε·(w″(t) - w e (t)); During the random stage of the odor molecule, the concentration of the odor molecule changes. Set the random search step size as g, d4 represents a random number in the interval (0, 1), update the position of the e-th odor molecule at the (t + 1)-th iteration to obtain w e (t + 1) = w e (t) + d4·g, generate a new odor molecule for the next iteration; Move in the direction with the highest concentration of the following odor molecule. Set the maximum number of iterations. When the current number of iterations reaches the maximum number of iterations, stop the iteration to obtain the final position of the odor molecule; S323. Obtain the best fitness function value at the position of the final odor molecule, and obtain the optimal material dispatching path according to the position of the final odor molecule.

8. The emergency command and dispatch method for sudden environmental incidents according to claim 7, wherein The S4 includes the following steps: S41. Transport the post-disaster materials to the disaster-stricken points according to the optimal material dispatching path, and achieve the minimum material shortage and the shortest transportation time to generate an emergency command material dispatching plan; according to the emergency command personnel dispatching plan and the emergency command material dispatching plan, combine them to obtain the final emergency command dispatching plan; after a sudden environmental event occurs, adjust the post-disaster materials and post-disaster personnel allocation in real time according to the final emergency command dispatching plan to complete the emergency command dispatching for the sudden environmental event.

9. A system for implementing the emergency command and dispatch method for sudden environmental incidents as described in any one of claims 1-8, characterized in that, Specifically include: An emergency command personnel dispatching module, a post-disaster material consumption prediction module, a material dispatching path optimization module, and an emergency command dispatching plan generation module; The emergency command personnel dispatching module is used for command and dispatching according to the emergency command dispatching personnel data and post-disaster personnel consumption. The post-disaster material consumption prediction module is used to establish a GRU neural network material prediction model and output the predicted value of post-disaster material consumption at the disaster-stricken points. The material dispatching path optimization module is used to use the olfactory optimization algorithm to solve the material dispatching objective function and establish the optimal material dispatching path. The emergency command and dispatch plan generation module is used to generate a final emergency command and dispatch plan by combining the emergency command personnel dispatch plan and the emergency command material dispatch plan.

Images