Scheduling information management method and system based on fire fighting truck fire fighting

Abstract

The invention discloses a scheduling information management method and system based on fire fighting of a fire fighting truck, and relates to the technical field of intelligent fire fighting scheduling, and the method comprises the steps: receiving fire alarm information, and collecting real-time fire scene data based on a preliminary fire behavior evaluation result; combining the collected real-time fire scene data with historical fire data, and analyzing a fire spreading trend through an AI fire prediction model; according to the fire behavior spreading trend and the equipment state of the fire fighting truck, dynamically allocating fire fighting truck resources, and planning an optimal driving route; the driving state of the fire fighting truck is monitored in real time, dynamic adjustment is performed based on road conditions and equipment information, and task allocation of the fire fighting truck is updated according to the real-time fire scene situation; in the fire extinguishing process, resources are dispatched in real time for supply according to the material consumption condition of the fire fighting truck. By monitoring the material consumption condition of the fire fighting truck in real time, the supply resources are intelligently dispatched, and the continuity and high efficiency of the fire extinguishing action are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent fire dispatching, and in particular to a dispatching information management method and system based on fire fighting by fire trucks. Background Art

[0002] With the acceleration of urbanization and the increase in building density, the frequency and complexity of fire accidents are increasing, especially in high-rise buildings and densely populated urban areas. Firefighting and rescue work is facing unprecedented challenges. In traditional fire dispatch, fire trucks are the main firefighting force, and their dispatch usually relies on manual decision-making and experience judgment of the command center. Due to the lack of real-time dynamic fire scene information, the resource allocation efficiency of fire trucks is often low, and it is difficult to make flexible adjustments according to the actual situation of the fire scene. In addition, the planning of fire truck driving routes mostly relies on predefined static paths, which cannot respond to complex and changeable traffic conditions and the spread of fire in a timely manner. This leads to the fact that in actual fire rescue, the arrival time, resource allocation and action command of fire trucks cannot reach the optimal state, affecting the fire extinguishing efficiency and rescue effect.

[0003] Although the introduction of intelligent technology in recent years has brought certain improvements to fire dispatch, such as the use of Internet of Things technology to monitor the status of fire trucks in real time and drones to collect fire scene information in real time, the existing technology still faces many shortcomings. First, the existing dispatch often lacks the ability to dynamically dispatch fire truck resources based on the fire scene situation, and fails to fully utilize the fusion analysis of real-time fire scene data and historical fire data. Secondly, when the fire truck is driving, the existing technology fails to effectively combine the real-time traffic information and the status of the fire truck equipment to dynamically optimize the driving route. In addition, the existing consumption and supply management of fire scene materials is also relatively lagging, making it difficult to achieve timely and accurate supply dispatch during the fire fighting process, thus affecting the stability and efficiency of continuous rescue. Summary of the invention

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

[0005] Therefore, the present invention provides a dispatch information management method based on fire truck firefighting to solve the problem that the fire truck resource dispatch is not intelligent and real-time enough.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for managing dispatch information based on firefighting by fire trucks, comprising receiving fire alarm information and collecting real-time fire scene data based on preliminary fire assessment results;

[0008] Combine the collected real-time fire scene data with historical fire data and analyze the fire spread trend through the AI ​​fire prediction model;

[0009] Dynamically allocate fire truck resources and plan the optimal driving route based on the fire spread trend and the equipment status of the fire truck;

[0010] Monitor the driving status of fire trucks in real time, make dynamic adjustments based on road conditions and equipment information, and update the task allocation of fire trucks according to the real-time fire situation;

[0011] During the fire-fighting process, resources are dispatched in real time for replenishment based on the material consumption of the fire truck.

[0012] As a preferred solution of the dispatch information management method based on fire truck firefighting of the present invention, wherein: the receiving of fire alarm information and collecting real-time fire scene data based on the preliminary fire situation assessment result, the specific steps are as follows:

[0013] Receive fire alarm information and format it;

[0014] Kalman filtering is used to remove fire alarm information for noise filtering;

[0015] Set a preset fire threshold, use the Bayesian inference model to calculate the probability of fire occurrence, and conduct a preliminary fire assessment when the probability of fire occurrence exceeds the preset threshold;

[0016] Obtain real-time environmental data of the fire site, and output preliminary fire assessment results through a fire prediction model that combines a convolutional neural network with a long short-term memory network;

[0017] Based on the initial fire assessment results, the flight mission is assigned to the drone;

[0018] The flight path of the drone is optimized through the ant colony algorithm, and it arrives at the fire scene to collect real-time fire data.

[0019] As a preferred solution of the dispatch information management method based on fire fighting by fire trucks described in the present invention, the real-time fire scene data collected is combined with historical fire data, and the fire spread trend is analyzed by an AI fire prediction model. The specific steps are as follows:

[0020] The real-time fire scene data and historical data of the fire are represented in quantum states, and multi-dimensional data are processed using quantum superposition and entanglement states;

[0021] Based on multidimensional data, high-dimensional data is mapped to the frequency domain through quantum Fourier transform, and the expression is:

[0022]

[0023] Among them, Ψ(t) represents the quantum state fusion model at time t, n represents the number of dimensions involved in fire prediction, i represents the weight coefficient of dimensional data, α iThe weight of the i-th group of data, |D i > re represents the quantum state of the real-time fire scene data in the i-th dimension, re indicates that the data is collected in real time, |H i > hl represents the quantum state of the historical data of the i-th dimension, hl indicates that the data comes from historical records, represents tensor product;

[0024] On the basis of quantum data fusion, nonlinear dynamic fuzzy logic is introduced to fuzzify the real-time fire scene data;

[0025] By analyzing the fuzzy sets of various parameters and combining the nonlinear differential equations, the change of fire intensity over time is dynamically described. The expression is:

[0026]

[0027] in, represents the first-order derivative of the fire spread trend F(t) with time t, m represents the number of real-time fire scene data, j represents the index variable of real-time fire scene data, μ j (x) is the fuzzy membership function of each environmental parameter x, Represents the jth real-time fire scene data P j (t) is the rate of change over time t, v(t′) represents the weight function of the historical data at the past time point t′, e -β(t-t′) represents an exponential decay function, β represents a parameter that controls the decay rate, β(tt′) represents an exponential decay coefficient that controls the effect of historical data on the current prediction, dt represents a small increment of time t, and dt′ represents a small increment of past time t′;

[0028] By analyzing the first-order derivative of the fire spread trend F(t) over time t, the fire spread speed, spread direction and spread range are output.

[0029] As a preferred solution of the dispatch information management method based on fire truck firefighting described in the present invention, wherein: on the basis of quantum data fusion, nonlinear dynamic fuzzy logic is introduced to fuzzify the real-time fire scene data, and the specific steps are as follows:

[0030] For the fused quantum data, define a fuzzy set and set a membership function for the fuzzy set;

[0031] The real-time fire scene data collected in real time are converted into fuzzy values ​​through membership functions;

[0032] Through the fuzzy reasoning mechanism and combined with the fuzzy rule base, the activation degree of each rule is calculated, and the fuzzy output of the fire spread trend is derived.

[0033] As a preferred solution of the dispatch information management method based on fire truck firefighting described in the present invention, wherein: the fire truck resources are dynamically allocated according to the fire spread trend and the equipment status of the fire truck, and the optimal driving route is planned. The specific steps are:

[0034] Monitor the status of each fire truck in real time through the Internet of Vehicles;

[0035] Combined with the fire spread speed, direction and range, deep reinforcement learning is used to intelligently dispatch fire truck resources;

[0036] Use spatiotemporal graph convolutional networks to combine real-time traffic data and geographic information to plan the optimal driving route for fire trucks.

[0037] As a preferred solution of the dispatch information management method based on fire fighting of fire trucks described in the present invention, the real-time monitoring of the driving status of fire trucks, dynamic adjustment based on road conditions and equipment information, and updating of task allocation of fire trucks according to the real-time fire situation are specifically performed as follows:

[0038] Combined with the fire truck status information obtained by the Internet of Vehicles and the real-time road conditions provided by the traffic management platform, the road traffic conditions are predicted through the spatiotemporal graph convolutional network;

[0039] If abnormal road conditions occur, the fire truck route will be dynamically adjusted through the adaptive dynamic programming model;

[0040] Obtain the fire spread speed, direction and range in real time through drones and monitoring equipment;

[0041] The multi-objective optimization method is used in combination with the fire truck status and fire situation to adjust the task allocation of each fire truck.

[0042] As a preferred solution of the dispatch information management method based on fire fighting of fire trucks described in the present invention, in which: during the fire extinguishing process, resources are dispatched in real time for replenishment according to the material consumption of the fire trucks, and the specific steps are:

[0043] Monitor the remaining quantity of firefighting materials in fire trucks in real time, and transmit the remaining quantity data of firefighting materials to the central control platform via wireless network;

[0044] The central control platform pre-processes the received data on the remaining quantity of fire-fighting material reserves;

[0045] Based on the preprocessed remaining data of fire-fighting material reserves, a material consumption prediction model is constructed using deep learning to predict the consumption rate of materials in the future. The expression is:

[0046]

[0047] Among them, k(t) is the material consumption rate of the fire truck at time t, ρ is the density of the fire extinguishing agent, u is the fire extinguishing agent injection velocity vector, is the divergence of the velocity field, v is the velocity field, V is the volume of the injection area, g is the gravity acceleration vector, θ is the injection angle, α represents the time attenuation coefficient, and dV represents a small increment in the volume of the injection area;

[0048] According to the consumption rate obtained by the material consumption prediction model, combined with the location of fire trucks, fire conditions and material reserves of supply stations, the supply path and timing are dynamically found through particle swarm optimization.

[0049] In a second aspect, the present invention provides a dispatch information management system based on fire fighting by fire trucks, including a fire assessment module, a trend prediction module, a resource allocation module, a path planning module and a material supply module;

[0050] The fire situation assessment module is used to receive fire alarm information and collect real-time fire scene data based on preliminary fire situation assessment results;

[0051] The trend prediction module is used to combine the collected real-time fire scene data with historical fire data and analyze the fire spread trend through the AI ​​fire prediction model;

[0052] The resource allocation module is used to dynamically allocate fire truck resources and plan the optimal driving route according to the fire spread trend and the equipment status of the fire truck;

[0053] The path planning module is used to monitor the driving status of the fire truck in real time, make dynamic adjustments based on road conditions and equipment information, and update the task allocation of the fire truck according to the real-time fire situation;

[0054] The material supply module is used to dispatch resources for supply in real time according to the material consumption of the fire truck during the fire-fighting process.

[0055] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for managing dispatch information based on fire truck firefighting as described in the first aspect of the present invention is implemented.

[0056] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the method for managing dispatch information based on fire truck firefighting as described in the first aspect of the present invention is implemented.

[0057] The beneficial effects of the present invention are as follows: the present invention accurately predicts the trend of fire spread by introducing drones to collect fire scene data in real time, combining AI fire prediction models with historical fire data, as well as methods such as quantum data fusion and dynamic fuzzy logic, and dynamically optimizes the resource scheduling and driving routes of fire trucks based on the equipment status of fire trucks and real-time road conditions; at the same time, during the fire-fighting process, it can monitor the material consumption of fire trucks in real time, intelligently schedule supply resources, and ensure the continuity and efficiency of fire-fighting operations; overall, it significantly improves the intelligence level of fire truck scheduling, shortens the rescue response time, and improves the coordination efficiency of fire rescue. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0059] Figure 1 This is a flow chart of the dispatch information management method based on fire truck firefighting in Example 1.

[0060] Figure 2 This is a module diagram of the dispatch information management system based on fire truck firefighting in Example 1. DETAILED DESCRIPTION

[0061] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0062] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0063] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0064] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a method for managing dispatch information based on firefighting by fire trucks, comprising the following steps:

[0065] S1. Receive fire alarm information and collect real-time fire scene data based on preliminary fire assessment results.

[0066] Furthermore, fire alarm information is received from multiple data sources and formatted;

[0067] Among them, multiple data sources, including data from fire detectors, manual alarms, satellite thermal imaging and other different sources, unify their key information such as timestamps and geographic coordinates.

[0068] Kalman filtering is used to remove fire alarm information for noise filtering, remove random errors and noise in the data, and improve data reliability;

[0069] Set a preset fire threshold and use the Bayesian inference model to calculate the probability of fire. When the probability of fire exceeds the preset threshold, the fire alarm is confirmed to be valid and a preliminary fire assessment is conducted.

[0070] Obtain real-time environmental data (including temperature, humidity, wind speed, wind direction, etc.) at the fire site, and output preliminary fire assessment results (including the predicted range, direction, and speed of fire spread) through a fire prediction model that combines a convolutional neural network (CNN) with a long short-term memory network (LSTM);

[0071] Based on the initial fire assessment results, the flight mission is assigned to the drone;

[0072] The flight path of the drone is optimized through the ant colony algorithm, and it arrives at the fire scene to collect real-time fire data;

[0073] Among them, the real-time fire scene data, including fire temperature distribution, fire spread, wind speed and other information, is pre-processed through edge computing nodes and transmitted to the command center in real time using the 5G network to assist further decision-making and fire assessment.

[0074] S2. Combine the collected real-time fire scene data with historical fire data and analyze the fire spread trend through the AI ​​fire prediction model.

[0075] Furthermore, the real-time fire scene data and historical data of the fire are represented in quantum states, and quantum superposition and entanglement are used to process multi-dimensional data;

[0076] It should be noted that traditional fire prediction models (such as CNN, LSTM, etc.) rely on classical computing architectures, and the computing time will increase exponentially with the increase in data dimensions. Quantum computing can process large amounts of data in parallel, greatly improving computing efficiency and solving the computing bottleneck in traditional prediction models. Especially when processing the fusion of real-time fire scene data and historical data, the super-strong parallelism and entangled state of quantum computing enable it to find the best data fusion strategy in a very short time.

[0077] Based on multidimensional data, high-dimensional data is mapped to the frequency domain through quantum Fourier transform (QFT), and the expression is:

[0078]

[0079] Among them, Ψ(t) represents the quantum state fusion model at time t, n represents the number of dimensions involved in fire prediction, i represents the weight coefficient of dimensional data, which is used to control the influence of real-time fire scene data and historical data of the i-th dimension on the overall quantum state fusion, α i The weight of the i-th group of data is used to control the influence of different data sources. i > re represents the quantum state of the real-time fire scene data in the i-th dimension, re indicates that the data is collected in real time, |H i > hl represents the quantum state of the historical data of the i-th dimension, hl indicates that the data comes from historical records, represents tensor product;

[0080] On the basis of quantum data fusion, nonlinear dynamic fuzzy logic is introduced to fuzzify real-time fire scene data (such as wind speed, temperature, and humidity). The specific steps are as follows:

[0081] For the fused quantum data, define a fuzzy set (such as low wind speed, medium wind speed and high wind speed) and set a membership function for the fuzzy set;

[0082] The real-time fire scene data collected in real time are converted into fuzzy values ​​through membership functions to represent the membership degree of each parameter in different fuzzy sets;

[0083] Through the fuzzy reasoning mechanism and combined with the fuzzy rule base, the activation degree of each rule is calculated, and the fuzzy output of the fire spread trend is derived.

[0084] It should be noted that fire spread is a highly nonlinear and dynamically changing process, and traditional linear models are difficult to capture this complexity. We use Dynamic Nonlinear Fuzzy Logic System (DNFLS) to deal with the uncertainty and ambiguity in the fire spread process; by fuzzifying different environmental parameters (such as wind speed, humidity, temperature, etc.), we can effectively deal with the uncertainty in the fire spread trend.

[0085] By analyzing the fuzzy sets of various parameters and combining the nonlinear differential equations, the change of fire intensity over time is dynamically described. The expression is:

[0086]

[0087] in, represents the first-order derivative of the fire spread trend F(t) with time t, that is, it describes the rate at which the fire changes with time. m represents the number of real-time fire data (such as wind speed, humidity, etc.), j represents the index variable of real-time fire data, and μ j (x) is the fuzzy membership function of each environmental parameter x, which is used to indicate the influence of this parameter on the fire. Represents the jth real-time fire scene data P j (t) is the rate of change over time t, v(t′) represents the weight function of the historical data at the past time point t′, e -β(t-t′) represents an exponential decay function, β represents a parameter that controls the decay rate, β(tt′) represents an exponential decay coefficient that controls the effect of historical data on the current prediction, dt represents a small increment of time t, and dt′ represents a small increment of past time t′;

[0088] Preferably, by introducing a set of nonlinear differential equations, the conversion process between fuzzy sets can be dynamically described. For example, when humidity decreases and wind speed increases, the tendency of fire spread will intensify nonlinearly. Through nonlinear dynamic modeling, this complex change process can be accurately captured.

[0089] By analyzing the first-order derivative of the fire spread trend F(t) over time t, the fire spread speed, spread direction and spread range are output.

[0090] S3. Dynamically allocate fire truck resources and plan the optimal driving route based on the fire spread trend and the equipment status of the fire truck.

[0091] Furthermore, the status information of each fire truck can be monitored in real time through the Internet of Vehicles, including location, speed, equipment integrity, material reserves (such as water, fuel, etc.) and power level;

[0092] Combined with the fire spread speed, direction and range, deep reinforcement learning is used to intelligently dispatch fire truck resources;

[0093] Specifically, the decision takes into account the direction of fire spread, the status of fire trucks, and geographic and traffic information, and finds the optimal resource allocation plan by maximizing the cumulative reward function.

[0094] Use the spatiotemporal graph convolutional network (ST-GCN) to combine real-time traffic data and geographic information to plan the optimal driving route of the fire truck;

[0095] Specifically, by representing the road network as a graph structure, ST-GCN predicts future traffic congestion and calculates the shortest driving path in combination with the Dijkstra or A* algorithm.

[0096] Preferably, the spatiotemporal graph convolutional network (ST-GCN) is a deep learning model for processing spatiotemporal data, which can combine real-time traffic data (such as congestion, road closures, etc.) and geographic information (such as road networks, terrain, etc.) to plan the optimal path for each fire truck, ensuring that the fire truck can reach the fire scene as quickly as possible.

[0097] S4. Monitor the driving status of fire trucks in real time, make dynamic adjustments based on road conditions and equipment information, and update the task allocation of fire trucks according to the real-time fire situation.

[0098] Furthermore, the fire truck status information obtained by the Internet of Vehicles, including location, speed, water tank capacity, equipment status, and fuel / power, and the real-time road conditions provided by the traffic management platform are combined to predict road traffic conditions through the spatiotemporal graph convolutional network (ST-GCN);

[0099] It should be noted that the spatiotemporal graph convolutional network (ST-GCN) is a deep learning model for modeling spatiotemporal data, which is particularly suitable for dealing with complex systems with spatiotemporal dependencies such as traffic networks. It represents the road network as a graph structure, extracts spatial features through convolution operations on the graph, and combines time series data to predict future traffic conditions.

[0100] Furthermore, ST-GCN uses a graph to represent the connection between each road segment and road in the traffic network. The graph consists of nodes and edges:

[0101] Nodes: Nodes in the graph represent road segments, intersections, or other physical locations related to traffic in the transportation network.

[0102] Edges: The edges between nodes represent the transmission path of traffic flow, that is, the connection relationship between roads. Edges can be directed (indicating the direction of traffic flow) or undirected (indicating a bidirectional connection between road sections).

[0103] Specifically in the traffic scenario, the road network can be represented by a graph G = (V, E):

[0104] V: Node set, representing all road sections or traffic key points in the road network.

[0105] E: Edge set, representing the connections between road segments (such as intersections, traffic flow directions).

[0106] Each node (road section) has different information such as traffic flow, speed, vehicle density, etc. at different time points. This information can be represented as a time series data, that is, at each time step, the state of the node is the traffic characteristics of the node (such as traffic flow, vehicle speed, etc.).

[0107] Therefore, the input data can be represented as a spatiotemporal feature matrix, where each row represents the state of a node and the column represents the feature value at the time step;

[0108] In ST-GCN, graph convolution is used to capture dependencies in the spatial dimension, that is, the mutual influence between different road sections. Compared with traditional CNN, which performs convolution on two-dimensional plane pixels, graph convolution is performed on the nodes of the graph, and the features of each node are updated through the information of its neighboring nodes.

[0109] In each layer, the node features are weighted summed with the features of its neighboring nodes through the adjacency matrix A to update the node features. This process captures the spatial traffic flow relationship.

[0110] In addition to the spatial dimension, traffic data also has dependencies in the temporal dimension. For example, the traffic conditions at the current time point are often closely related to the traffic conditions in the past few minutes or hours. Therefore, ST-GCN also captures the temporal dependency of traffic flow through temporal convolution.

[0111] Temporal convolution usually uses one-dimensional convolution (1D-CNN) to process time series data. For each node, a one-dimensional convolution is used to perform a sliding window operation on the time dimension to extract the temporal change pattern:

[0112] ST-GCN gradually extracts the spatial dependencies and temporal dynamic dependencies between nodes in the traffic network by alternating the hierarchical structure of graph convolution and temporal convolution. After multiple layers of convolution operations, ST-GCN can learn complex spatiotemporal features and ultimately be used to predict traffic status.

[0113] The last layer is usually a fully connected layer or a one-dimensional convolution, which is used to output future traffic status predictions, such as traffic volume and vehicle speed in the next few minutes.

[0114] If abnormal road conditions occur, the fire truck route is dynamically adjusted through an adaptive dynamic programming model, with the goal of minimizing travel time and resource consumption;

[0115] The fire spread speed, direction and range are obtained in real time through drones and monitoring equipment. The multi-objective optimization method is used in combination with the status of the fire truck and the fire situation to adjust the task allocation of each fire truck to ensure that the fire truck performs the optimal task according to the latest fire situation.

[0116] For example, when the fire in a certain area weakens, nearby fire trucks are adjusted and transferred to other more urgent areas for support.

[0117] S5. During the fire-fighting process, resources are dispatched in real time for replenishment based on the material consumption of the fire truck.

[0118] Furthermore, with the support of Internet of Things (IoT) technology, sensors installed on fire trucks can monitor the remaining amount of firefighting materials (such as water, foam, etc.) in real time and send data to the central control platform;

[0119] The central control platform pre-processes the received data on the remaining quantity of fire-fighting materials reserves, including data cleaning and formatting, to ensure the quality and availability of the data;

[0120] Based on the preprocessed remaining data of fire-fighting material reserves, a material consumption prediction model is constructed using deep learning to predict the consumption rate of materials in the future. The expression is:

[0121]

[0122] Among them, k(t) is the material consumption rate of the fire truck at time t, which represents the mass of fire extinguishing agent consumed per unit time, ρ is the density of the fire extinguishing agent, which represents the mass of the fire extinguishing agent per unit volume, and u is the fire extinguishing agent injection velocity vector, which represents the speed and direction of the fire extinguishing agent injection. is the divergence of the velocity field, which indicates the degree of diffusion of the fluid flow, that is, the net outflow of the fluid per unit volume; v is the velocity field, which indicates the velocity distribution of the fluid in space; V is the volume of the spraying area, which indicates the spatial range covered by the spraying of the fire extinguishing agent; g is the gravity acceleration vector, which indicates the influence of gravity on the spraying of the fire extinguishing agent; θ is the spraying angle, which indicates the angle between the spraying direction of the fire extinguishing agent and the horizontal plane; α indicates the time attenuation coefficient, which indicates the natural downward trend of the material consumption rate with the increase of time; dV indicates the small increment of the volume of the spraying area;

[0123] According to the consumption rate obtained by the material consumption prediction model, combined with the location of the fire truck, the fire scene and the material reserves of the supply station, the particle swarm optimization is used to dynamically find the supply path and timing;

[0124] Specifically, the steps for dynamically finding supply paths and opportunities are as follows:

[0125] First, the current location of each fire truck is obtained in real time through the Internet of Vehicles, and the real-time situation of the fire scene is collected simultaneously, including the spread speed, direction and dangerous areas of the fire scene. In addition, the location, material storage and supply capacity of each supply station need to be obtained. At the same time, combined with the material consumption prediction model, the material consumption rate and remaining material quantity of each fire truck are dynamically calculated.

[0126] Secondly, determine the optimization goal. The core goal of the optimization is to ensure that the fire truck can be resupplied in time before the supplies are exhausted and find the best path to complete the resupply in the shortest time. This means finding a reasonable time for resupply and planning the optimal path from the current location of the fire truck to the resupply station. In this process, the dangerous areas of the fire scene, real-time traffic conditions, and the urgency of the firefighting task should be fully considered.

[0127] Furthermore, the particle swarm optimization (PSO) algorithm is used to dynamically optimize the replenishment plan. The specific steps are as follows:

[0128] Initialize the particle swarm: Each particle represents a possible supply plan, which includes two key elements: supply timing (when to supply) and supply path (the optimal route from the current location of the fire truck to the supply station).

[0129] Fitness evaluation: The fitness function is used to evaluate the pros and cons of each solution. The fitness function takes the following factors into consideration:

[0130] Resupply time: The round-trip time from the fire truck’s current location to the resupply station, including traffic conditions inside and outside the fire scene.

[0131] Supply depletion risk: Whether the rate of supply consumption will cause the fire truck to run out of supplies before being resupplied.

[0132] Impact on fire situation: Whether the fire truck leaving the fire scene for resupply will affect the firefighting operation or cause the fire to get out of control.

[0133] Particle swarm iterative update: Through the iterative update of the particle swarm, the replenishment timing and path of each particle are dynamically adjusted to gradually approach the optimal solution.

[0134] Finally, the particle swarm optimization algorithm outputs the optimal supply timing to ensure that the fire truck is replenished in time before the supplies are exhausted. At the same time, the optimal supply path is also given to ensure that the fire truck completes the supply task in the shortest time and reduce the interruption and impact on the firefighting operation.

[0135] The present embodiment also provides a dispatch information management system based on fire fighting of fire trucks, including: a fire assessment module, a trend prediction module, a resource allocation module, a path planning module and a material supply module; the fire assessment module is used to receive fire alarm information and collect real-time fire scene data based on the preliminary fire assessment results; the trend prediction module is used to combine the collected real-time fire scene data with historical fire data, and analyze the fire spread trend through an AI fire prediction model; the resource allocation module is used to dynamically allocate fire truck resources and plan the optimal driving route according to the fire spread trend and the equipment status of the fire truck; the path planning module is used to monitor the driving status of the fire truck in real time, make dynamic adjustments based on road conditions and equipment information, and update the task allocation of the fire truck according to the real-time fire situation; the material supply module is used to dispatch resources for replenishment in real time according to the material consumption of the fire truck during the fire extinguishing process.

[0136] This embodiment also provides a computer device, which is suitable for the dispatch information management method based on fire truck firefighting, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the dispatch information management method based on fire truck firefighting as proposed in the above embodiment.

[0137] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0138] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for implementing the dispatch information management method based on fire fighting of fire trucks as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0139] In summary, the present invention accurately predicts the trend of fire spread by introducing drones to collect fire scene data in real time, combining AI fire prediction models with historical fire data, as well as methods such as quantum data fusion and dynamic fuzzy logic, and dynamically optimizes the resource scheduling and driving routes of fire trucks based on the equipment status and real-time road conditions of fire trucks; at the same time, during the fire-fighting process, it can monitor the material consumption of fire trucks in real time, intelligently schedule supply resources, and ensure the continuity and efficiency of fire-fighting operations; overall, it significantly improves the intelligence level of fire truck scheduling, shortens the rescue response time, and improves the coordination efficiency of fire rescue.

[0140] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of a dispatch information management method based on fire truck firefighting is provided.

[0141] The experiment simulates a fire scene to evaluate whether the method can show significant performance advantages in multi-source data access and drone dispatching. The experiment compares the existing traditional fire monitoring methods with the method proposed in this paper, focusing on key indicators such as the initial assessment of the fire, drone dispatch efficiency, response time and accuracy of fire scene data collection and analysis.

[0142] In the experiment, the traditional fire monitoring method used a fire detection and response method based on a single data source. It usually relies on fixed fire detectors (such as smoke sensors, temperature sensors, etc.) to alarm and judge the occurrence of fire through simple logical conditions (such as temperature or smoke concentration exceeding the set threshold). The fire assessment of the traditional method relies on the data of these fixed sensors, which cannot dynamically capture the complex changes of the fire scene, and lacks efficient fusion and processing mechanisms when facing multi-source data (such as manual alarms, satellite thermal imaging, etc.). In addition, the drone dispatch in the traditional method usually adopts a preset flight path and is not optimized according to the actual situation of the fire, resulting in a long response time.

[0143] The experimental group uses the multi-source data fusion fire monitoring method proposed in this paper, which can receive multi-source data from fire detectors, manual alarms, satellite thermal imaging, etc. In the experimental group, these data will be de-noised through Kalman filtering, and the timestamps and geographic coordinates will be unified. After the data is formatted, the Bayesian inference model is used to calculate the probability of fire occurrence. When the probability exceeds the preset threshold, the fire alarm is confirmed to be valid, and a preliminary fire assessment is performed.

[0144] Next, the experiment simulated the real-time environmental data of the fire scene (including temperature, humidity, wind speed, wind direction, etc.). The experimental group method outputs the range, direction and speed of fire spread through a fire prediction model combining a convolutional neural network (CNN) and a long short-term memory network (LSTM). At the same time, based on the fire assessment results, the experimental group method dispatched drones with sufficient power, optimized the flight path of the drones through the ant colony algorithm, and collected real-time fire data at the fire scene. After the drone arrived at the fire scene, it collected information such as the temperature distribution and fire spread of the fire scene, and preprocessed the data through edge computing nodes, and transmitted it to the command center in real time using the 5G network.

[0145] In the experiment, the experimental group method used quantum states to represent multidimensional data, combined real-time fire data with historical fire data, and mapped the data to the frequency domain through quantum Fourier transform (QFT) to predict the trend of fire spread. Compared with traditional fire monitoring methods, the experimental group method demonstrated extremely high parallel processing efficiency, especially when processing multi-source data with complex dimensions.

[0146] After the experiment, the performance of the experimental group was compared with the traditional fire monitoring method in terms of data processing efficiency, drone dispatch response time, and fire scene data collection accuracy.

[0147] The details are shown in Table 1 below:

[0148] Table 1 Experimental data comparison table

[0149]

[0150] Through the above experimental data, it can be clearly seen that the multi-source data fusion fire monitoring method proposed in this paper has significant advantages over traditional fire monitoring methods in multiple key indicators. First, in terms of data processing efficiency, the method of the experimental group reached 90MB / s, which is almost twice that of the traditional method. This is due to the noise elimination of the Kalman filter and the preprocessing of multi-source data, which avoids the interference of redundant and noisy data and greatly improves the processing speed of the method. However, the traditional method has low data processing efficiency due to its reliance on a single data source and lack of effective processing capabilities for multi-source data.

[0151] Secondly, in terms of drone dispatch response time, the experimental group method performed well, with a response time of only 5 seconds, while the traditional method required 12 seconds. The traditional method of drone dispatch is often based on a preset flight path and cannot be dynamically adjusted according to changes in the fire situation, so the response speed is slow. The experimental group method combines the fire assessment results and the ant colony algorithm to dynamically optimize the flight path of the drone, ensuring that the drone can be reasonably dispatched to the fire scene in the shortest time.

[0152] In terms of the accuracy of fire scene data collection, the experimental group's drone real-time data collection is consistent with the actual fire scene situation by up to 95%, which is significantly better than the 78% of the traditional method. Traditional methods are usually unable to effectively combine multi-source data (such as satellite thermal imaging, environmental sensors, etc.), and can only rely on a single data source to judge the fire situation, resulting in low accuracy of data collection. The experimental group method pre-processes the data collected by the drone through the edge computing node, and deeply integrates it with historical fire data, making the collected data more accurate and reliable.

[0153] In terms of fire spread prediction error, the experimental group's error was 6%, much lower than the 15% of traditional methods. Traditional methods usually rely on simple linear models or rules based on historical experience to predict fire spread, which cannot cope with complex fire environments. The experimental group method effectively improved the prediction accuracy of fire spread through quantum state data fusion and quantum Fourier transform (QFT). The parallel processing capability of quantum computing shows great advantages when facing high-dimensional data, while traditional methods have difficulty processing such complex multi-dimensional data.

[0154] Finally, in terms of real-time fire scene data transmission delay, the experimental group method has a delay of 2 seconds, which is significantly lower than the 8 seconds of the traditional method. The traditional method uses a lower-bandwidth network transmission technology and lacks edge computing support, resulting in slow data transmission speed. The experimental group method, through the support of 5G network and edge computing nodes, ensures that the fire scene data can be quickly transmitted to the command center to assist further decision-making.

[0155] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A dispatch information management method based on fire fighting by fire trucks, characterized in that: include, Receive fire alarm information and collect real-time fire scene data based on preliminary fire assessment results; Combine the collected real-time fire scene data with historical fire data and analyze the fire spread trend through the AI ​​fire prediction model; Dynamically allocate fire truck resources and plan the optimal driving route based on the fire spread trend and the equipment status of the fire truck; Monitor the driving status of fire trucks in real time, make dynamic adjustments based on road conditions and equipment information, and update the task allocation of fire trucks according to the real-time fire situation; During the fire-fighting process, resources are dispatched in real time for replenishment based on the material consumption of the fire truck.

2. The method for managing dispatch information based on firefighting by fire trucks according to claim 1, characterized in that: The receiving of fire alarm information and collecting real-time fire scene data based on preliminary fire situation assessment results are specifically performed as follows: Receive fire alarm information and format it; Kalman filtering is used to remove fire alarm information for noise filtering; Set a preset fire threshold, use the Bayesian inference model to calculate the probability of fire occurrence, and conduct a preliminary fire assessment when the probability of fire occurrence exceeds the preset threshold; Obtain real-time environmental data of the fire site, and output preliminary fire assessment results through a fire prediction model that combines a convolutional neural network with a long short-term memory network; Based on the initial fire assessment results, assign flight missions to drones; The flight path of the drone is optimized through the ant colony algorithm, and it arrives at the fire scene to collect real-time fire data.

3. The method for managing dispatch information based on firefighting by fire trucks according to claim 2, characterized in that: The real-time fire scene data collected is combined with historical fire data to analyze the fire spread trend through the AI ​​fire prediction model. The specific steps are: The real-time fire scene data and historical data of the fire are represented in quantum states, and multi-dimensional data are processed using quantum superposition and entanglement states; Based on multidimensional data, high-dimensional data is mapped to the frequency domain through quantum Fourier transform, and the expression is: Among them, Ψ(t) represents the quantum state fusion model at time t, n represents the number of dimensions involved in fire prediction, i represents the weight coefficient of dimensional data, α i The weight of the i-th group of data, |D i > re represents the quantum state of the real-time fire scene data in the i-th dimension, re indicates that the data is collected in real time, |H i > hl represents the quantum state of the historical data of the i-th dimension, hl indicates that the data comes from historical records, represents tensor product; On the basis of quantum data fusion, nonlinear dynamic fuzzy logic is introduced to fuzzify the real-time fire scene data; By analyzing the fuzzy sets of various parameters and combining the nonlinear differential equations, the change of fire intensity over time is dynamically described. The expression is: in, represents the first-order derivative of the fire spread trend F(t) with time t, m represents the number of real-time fire scene data, j represents the index variable of real-time fire scene data, u j (x) is the fuzzy membership function of each environmental parameter x, Represents the jth real-time fire scene data P j (t) is the rate of change over time t, ν(t′) represents the weight function of the historical data at the past time point t′, e -β(t-t′) represents an exponential decay function, β represents a parameter that controls the decay rate, β(tt′) represents an exponential decay coefficient that controls the effect of historical data on the current prediction, dt represents a small increment of time t, and dt′ represents a small increment of past time t′; By analyzing the first-order derivative of the fire spread trend F(t) over time t, the fire spread speed, spread direction and spread range are output.

4. The method for managing dispatch information based on firefighting by fire trucks according to claim 3, characterized in that: On the basis of quantum data fusion, nonlinear dynamic fuzzy logic is introduced to fuzzify the real-time fire scene data. The specific steps are as follows: For the fused quantum data, define a fuzzy set and set a membership function for the fuzzy set; The real-time fire scene data collected in real time are converted into fuzzy values ​​through membership functions; Through the fuzzy reasoning mechanism and combined with the fuzzy rule base, the activation degree of each rule is calculated, and the fuzzy output of the fire spread trend is derived.

5. The method for managing dispatch information based on firefighting by fire trucks according to claim 4, characterized in that: The specific steps of dynamically allocating fire truck resources and planning the optimal driving route according to the fire spreading trend and the equipment status of the fire truck are as follows: Monitor the status of each fire truck in real time through the Internet of Vehicles; Combined with the fire spread speed, direction and range, deep reinforcement learning is used to intelligently dispatch fire truck resources; Use spatiotemporal graph convolutional networks to combine real-time traffic data and geographic information to plan the optimal driving route for fire trucks.

6. The method for managing dispatch information based on firefighting by fire trucks according to claim 5, characterized in that: The real-time monitoring of the driving status of the fire truck, dynamic adjustment based on road conditions and equipment information, and updating the task allocation of the fire truck according to the real-time fire situation are specifically as follows: Combined with the fire truck status information obtained by the Internet of Vehicles and the real-time road conditions provided by the traffic management platform, the road traffic conditions are predicted through the spatiotemporal graph convolutional network; If abnormal road conditions occur, the fire truck route will be dynamically adjusted through the adaptive dynamic programming model; Obtain the fire spread speed, direction and range in real time through drones and monitoring equipment; The multi-objective optimization method is used in combination with the fire truck status and fire situation to adjust the task allocation of each fire truck.

7. The method for managing dispatch information based on firefighting by fire trucks according to claim 6, characterized in that: In the fire-fighting process, resources are dispatched in real time for replenishment according to the material consumption of the fire truck. The specific steps are: Monitor the remaining quantity of firefighting materials in fire trucks in real time, and transmit the remaining quantity data of firefighting materials to the central control platform via wireless network; The central control platform pre-processes the received data on the remaining quantity of fire-fighting material reserves; Based on the preprocessed remaining data of fire-fighting material reserves, a material consumption prediction model is constructed using deep learning to predict the consumption rate of materials in the future. The expression is: Among them, k(t) is the material consumption rate of the fire truck at time t, ρ is the density of the fire extinguishing agent, u is the fire extinguishing agent injection velocity vector, is the divergence of the velocity field, v is the velocity field, V is the volume of the injection area, g is the gravity acceleration vector, θ is the injection angle, α represents the time attenuation coefficient, and dV represents a small increment in the volume of the injection area; According to the consumption rate obtained by the material consumption prediction model, combined with the location of fire trucks, fire conditions and material reserves of supply stations, the supply path and timing are dynamically found through particle swarm optimization.

8. A dispatch information management system based on fire truck firefighting, based on the dispatch information management method based on fire truck firefighting according to any one of claims 1 to 7, characterized in that: Including fire assessment module, trend prediction module, resource allocation module, path planning module and material supply module; The fire situation assessment module is used to receive fire alarm information and collect real-time fire scene data based on preliminary fire situation assessment results; The trend prediction module is used to combine the collected real-time fire scene data with historical fire data and analyze the fire spread trend through the AI ​​fire prediction model; The resource allocation module is used to dynamically allocate fire truck resources and plan the optimal driving route according to the fire spread trend and the equipment status of the fire truck; The path planning module is used to monitor the driving status of the fire truck in real time, make dynamic adjustments based on road conditions and equipment information, and update the task allocation of the fire truck according to the real-time fire situation; The material supply module is used to dispatch resources for supply in real time according to the material consumption of the fire truck during the fire fighting process.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the dispatch information management method based on fire fighting by fire trucks are implemented as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for managing dispatch information based on firefighting by fire trucks according to any one of claims 1 to 7 are implemented.

Images