Intelligent fire-fighting command auxiliary decision-making system

Abstract

The invention provides an intelligent fire-fighting command auxiliary decision-making system, which belongs to the field of fire rescue and comprises a data acquisition module, a data processing module, an intelligent decision-making module, a communication module, a positioning and navigation module and a storage module. According to the system, fire scene environment data and fireman state data are collected through various sensor devices, and then data cleaning and recognition are conducted through the data processing module. The intelligent decision-making module generates a fire extinguishing strategy and personnel scheduling suggestions and assists commanders in making decisions. The system also uses GPS or other positioning technologies to monitor the positions of firefighters and fire fighting trucks in real time, and generates a fire extinguishing path and a personnel evacuation path. The system further comprises a storage module used for storing historical data and decision-making processes. The design can better assist fire control commanders to make decisions, and the efficiency and effect of fire rescue are improved.

Description

Technical Field

[0001] The invention belongs to the field of firefighting and rescue, and in particular relates to an intelligent firefighting combat command auxiliary decision-making system. Background Art

[0002] In the current fire command decision-making, the decision mainly relies on the fire commander's judgment and experience of the fire scene. This method is not only slow in decision-making, but also easily affected by personal experience and subjective judgment, which can easily lead to inaccurate and non-optimal decisions. In addition, in the case of complex fire environment, large amount of information, and frequent changes, it is difficult for fire commanders to quickly and accurately grasp the overall information and provide effective reference for command decision-making. Therefore, how to improve the efficiency and accuracy of fire operation command decision-making through scientific and technological means has become an important topic at present.

[0003] In the past, some researchers have tried to establish intelligent fire command systems by introducing computers and information technology to assist fire command decision-making. However, most of these systems rely on preset rules and processes, and their flexibility and breadth of application are limited to a certain extent.

[0004] In recent years, with the development of artificial intelligence and big data technology, applying them to fire command decision-making has the potential to greatly improve the efficiency and quality of decision-making. For example, by collecting fire scene environment data and firefighter status data in real time, combined with advanced data processing and decision-making algorithms, fire-fighting strategies and personnel scheduling suggestions can be generated to assist commanders in making decisions. However, how to build such a system is still a technical problem to be solved. Summary of the invention

[0005] The main technical problem solved by the present invention is how to build an intelligent firefighting command and decision-making support system, combine big data and artificial intelligence technology, realize real-time and accurate collection of fire scene environment and firefighter status, effectively process these data, further identify fire sources, predict fire intensity, identify safe passages, and generate fire extinguishing strategies and personnel scheduling suggestions, so as to enhance the efficiency, accuracy and effectiveness of firefighting command decision-making and improve the success rate of firefighting operations.

[0006] In order to achieve the above-mentioned object, the present invention is implemented by adopting the following technical solutions: a data acquisition module, connected to the data processing module, includes various sensor devices, and is used to collect fire scene environment data and firefighter status data, including the firefighter's position, heart rate, body temperature, blood oxygen saturation, remaining capacity of oxygen cylinders, and pressure resistance level of firefighter suits, and transmit them to the subsequent data processing module in real time; Data processing module: After cleaning the collected fire scene environment data and firefighter status data, fire source identification and safety channel identification are performed; The intelligent decision-making module is connected to the data processing module and is used to generate fire-fighting strategies and personnel dispatch suggestions based on the fire source identification results and the safety channel identification results, assist the commander in making decisions, and generate fire-fighting paths and personnel evacuation paths; A communication module, connected to the intelligent decision-making module, used for data transmission between different modules and / or data exchange with external devices; The positioning and navigation module is connected to the intelligent decision-making module to monitor the location of firefighters and fire trucks in real time based on positioning technology; The storage module is connected with the data processing module and the intelligent decision-making module and is used to store historical data and decision-making processes.

[0007] In one embodiment, the data acquisition module collects data from sensors in the fire environment, including sensor devices fixed in buildings, or mobile sensor devices carried by drones or intelligent robots.

[0008] In one embodiment, the data processing module includes: Data cleaning: Clean the collected raw fire scene environment data and firefighter status data, and delete invalid, abnormal or isolated data; Fire source identification: Use the support vector machine (SVM) model to identify the fire source location; Fire prediction: Use long short-term memory neural network LSTM to predict the development of fire based on historical data; Safe passage identification: Use path planning algorithms, combined with current environmental data and building map data, to plan the safest and shortest evacuation route.

[0009] In one embodiment, the intelligent decision module adopts an intelligent fire fighting decision method of hierarchical-grayscale correlation analysis method, including: Establish mathematical models for intelligent decision-making mechanisms; The hierarchical-grayscale correlation analysis method is used to solve the model.

[0010] In one embodiment, the mathematical model of the intelligent decision-making mechanism is specifically: Record the fire station as an alternative: , represents n optional fire stations; The set of alternative options refers to all fire stations; gather Every fire station in The factors affecting the selection of the plan, such as the geographical location of the disaster, the nature of the burning material, the water source of the accident site, the degree of road congestion, the geographical location of the fire station, and the status of the fire trucks in the fire station, are used as the decision criteria and recorded as: , The set of decision criteria that affect the selection of fire stations is represented by a set of m decision indicators; gather Each decision criterion in represents the first criterion, the second criterion, and so on, until the mth criterion; the ultimate goal is to select the appropriate fire station to arrive at the disaster site as soon as possible; It is called the evaluation sequence. Fire Station In decision-making criteria The assessed value below.

[0011] In one embodiment, the hierarchical-grayscale correlation analysis method includes: The weight of each evaluation factor is calculated, and three judgment matrices are obtained according to different levels of evaluation parameters: a global comparison judgment matrix of the fire station's ability to handle the disaster, the firefighting forces that can be dispatched by the fire station, the water source conditions at the accident site, the geographical characteristics of the fire site, and the number of fire-fighting resources reserved by the fire station, as well as a local judgment matrix of the number of personnel that can be dispatched in the fire station and the number of vehicles that can be dispatched, and a local judgment matrix of the distance from the accident site to the fire station and the degree of road congestion between the fire stations. Use grey correlation analysis to select the best fire station.

[0012] In one embodiment, the communication module includes: Internal communication: Within the system, data transmission between different modules is realized through network protocols TCP / IP and HTTP; External communication: Data exchange between the system and external devices is carried out through wireless communication 4G / 5G, LoRa, and ZigBee.

[0013] In one solution, the storage module uses a relational database to store structured data, a NoSQL database to store large-scale, semi-structured data, and a time series database to store sensor data; Read data from the database for decision analysis, report generation, and system monitoring.

[0014] Beneficial effects of the present invention: The intelligent firefighting command and decision-making support system of the present invention combines multiple modules such as data collection, data processing, intelligent decision-making, communication, positioning and navigation. It can realize real-time and accurate monitoring and analysis of the fire environment and the status of firefighters, and generate fire-fighting strategies and personnel dispatch suggestions based on the analysis results, thereby improving the efficiency and safety of firefighting operations.

[0015] 1) Realize intelligent decision-making: By collecting, processing and analyzing real-time data on the fire environment and the status of firefighters, the system can generate effective fire-fighting strategies and personnel dispatch suggestions, thereby improving the efficiency and accuracy of decision-making, reducing the possibility of misjudgment caused by human factors, and effectively improving the fire-fighting effect and ensuring the personal safety of firefighters.

[0016] 2) Realize real-time monitoring: The data acquisition module can monitor the fire environment and the status of firefighters in real time, detect abnormalities in time, and transmit them to the subsequent processing module in real time through the communication module, so that the command personnel can make quick judgments and decisions.

[0017] 3) Achieve precise positioning and navigation: The positioning and navigation module can monitor the location of firefighters and fire trucks in real time, generate fire-fighting paths and personnel evacuation paths, which helps to improve the efficiency of fire-fighting operations and ensure personnel safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a system block diagram of the present invention; Figure 2 It is the decision-making flow chart of the present invention; Figure 3 This is the decision factor hierarchy model of the present invention. DETAILED DESCRIPTION

[0019] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Typical embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described in the present invention. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0020] Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0021] like Figure 1 As shown, the intelligent fire fighting command decision-making auxiliary system comprises: Data acquisition module: collects fire scene environment data and firefighter status data through various sensor devices, and transmits them to the subsequent data processing module in real time. The data acquisition module collects environmental data from various sensors installed in the fire scene environment, including sensor devices fixed in buildings, or mobile sensor devices carried by drones or intelligent robots.

[0022] Firefighter status data generally refers to various types of information related to the physical and environmental status of firefighters during their missions. This includes but is not limited to the firefighters’ location, physiological status such as heart rate, body temperature, blood oxygen saturation, etc., mission status such as ambient temperature, humidity, presence of harmful gases, etc., and equipment status such as the remaining capacity of oxygen cylinders and the pressure rating of fire suits. These data can be collected in real time or at regular intervals through various sensor devices and transmitted to subsequent processing modules or equipment for analysis and processing, which are used to monitor the safety of firefighters, estimate possible dangers, and make corresponding emergency responses in real time.

[0023] Environmental data is collected through sensor devices fixed in the building. These devices include smoke detectors, temperature sensors, humidity sensors, etc. These devices are installed in fixed locations in the building and immediately send signals to the data acquisition module once a fire source or environmental change is detected. In addition to real-time data collection, these devices can also provide historical data to help understand the development of the fire source.

[0024] Drones and smart robots can carry a variety of sensor devices, such as high-definition cameras, infrared detectors, thermal imaging cameras, etc. They can quickly move to different locations of the fire scene according to actual needs and collect data from multiple perspectives in the air and on the ground. For example, drones can take a full picture of the fire scene from the air, and smart robots can enter narrow areas or areas that are inaccessible to people to collect data.

[0025] During the data collection process, all collected data will be transmitted to the data processing module in real time. The collected data will be sent to the data processing center using 4G / 5G wireless communication technology. The data processing center will perform operations such as fire source identification, fire prediction, and safe channel identification based on the received data. The data collection module should be able to quickly and in real time obtain detailed information about the fire scene environment, providing an important basis for subsequent decision-making.

[0026] Data processing module: cleans the collected data, including fire source identification and safety channel identification.

[0027] The data processing module comprises: Data cleaning: Clean the collected raw data and delete invalid, abnormal or isolated data.

[0028] Fire source identification: Use the support vector machine (SVM) model to identify the location of the fire source. First, train an SVM identification model. The training data includes various scenarios of fire sources and non-fire sources, such as the presence or absence of burning objects in the room. Each scenario corresponds to a label to indicate whether there is a fire source. After the model training is completed, the collected fire environment data is input into the trained SVM model, and the fire source location is identified through the model's prediction results.

[0029] SVM (Support Vector Machines) is a supervised learning model, mainly used for classification and regression analysis. In fire source identification, the SVM model can be trained based on the characteristics of the fire source (such as temperature, smoke density, etc.) and category (whether it is a fire source).

[0030] The main idea of ​​SVM is to find a hyperplane that maximizes the distance between the plane and the closest point of two types of samples. Assume a two-dimensional classification problem, that is, the feature space is . The following function is used for classification: ; Where w is the weight vector, x is the feature vector, and b is the bias term. SVM determines the optimal hyperplane by optimizing the following problem: .

[0031] ; in, is the square of the norm of w, and yi is the class label of each sample.

[0032] Fire prediction: Use the long short-term memory neural network LSTM to predict the development of the fire based on historical data. In this process, we first collect historical fire data, including the location of the fire source, flame size, temperature, wind speed, wind direction and other information. Then use this data to train an LSTM model to let it learn how the fire changes according to historical status. After the model is trained, the real-time collected environmental data is input into the model, and the model can predict the development of the fire in the future.

[0033] LSTM (Long Short-Term Memory) is a special recurrent neural network (RNN) that is suitable for time series prediction problems. It can handle long-term dependency problems in time series and is suitable for predicting the development of fires. The characteristic of LSTM is the introduction of a "gate" structure. Each LSTM unit has a forget gate, an input gate, and an output gate, which can learn the degree of retention or forgetting of information at each time point.

[0034] Safe passage identification: Use the path planning algorithm Dijkstra algorithm, combined with the current environmental data and building map data, to plan the safest and shortest evacuation path. First, the building map needs to be converted into a graph. Each node represents a location on the map, and the edge between the nodes indicates that the two locations can pass directly. The weight of the edge represents the difficulty of passing this path, such as distance, fire intensity, etc. Then, we use the Dijkstra algorithm to find a path to the safe zone from the fire exit (or the current location of the firefighter) so that the sum of the weights of this path is minimized, that is, to find the safest and shortest evacuation path.

[0035] The Dijkstra algorithm is an algorithm for solving the shortest path problem and can be used for secure channel identification. Its main steps are as follows: a) Initialization: Take the starting point as the root of the shortest path tree and initialize the distance from the starting point to all other points. If they are directly connected, it is the weight of the edge, otherwise it is infinite.

[0036] b) Selection and update: Select the node with the shortest distance from the current point to the starting point among the candidate nodes, remove it from the candidate nodes, and include it in the shortest path tree. Then update the distance from this node to the candidate nodes directly connected to it.

[0037] c) Repeat the second step until all points are included in the shortest path tree or the candidate node is empty.

[0038] Intelligent decision-making module: Generates fire-fighting strategies and personnel dispatch suggestions to assist commanders in making decisions.

[0039] The intelligent decision-making module adopts an intelligent fire fighting decision-making method of hierarchical-grayscale correlation analysis method, including: like Figure 2 As shown, S1. Establish a mathematical model of intelligent decision-making mechanism; when a disaster occurs, the fire command and dispatch system uses the grayscale association algorithm to comprehensively analyze the factors affecting the command and dispatch, such as the nature of the burning material, the water source at the accident site, the status and characteristics of the fire trucks in the fire station, and finally assist the command center to quickly and reasonably dispatch fire resources. The specific process of applying the grayscale association algorithm is as follows:. S101. First, determine the selection model and record the fire station as an alternative: Indicates n optional fire station, The set of alternative options refers to all fire stations; gather Each fire station in the fire station. The factors affecting the selection of the plan, such as the geographical location of the disaster, the nature of the burning material, the water source of the accident site, the degree of road congestion, the geographical location of the fire station, and the status of the fire truck in the fire station, are used as the decision criteria and recorded as: represents a set of m decision indicators, The set of decision criteria that affect the selection of fire stations is represented by a set of m decision indicators; gather Each decision criterion in represents the first criterion, the second criterion, and so on, until the mth criterion; the ultimate goal is to select the appropriate fire station to arrive at the disaster site as soon as possible; It is called the evaluation sequence. Fire Station In decision-making criteria The hierarchical relationship between the decision factors is shown in Figure 3 .

[0040] S102, then determine the weight of the selection factor to the overall goal, analyze the relationship between the basic elements in the evaluation system, and establish the hierarchical structure of the system. Compare the importance of each element at the same level to a certain criterion at the previous level, and construct a judgment matrix; then use the judgment matrix to calculate the relative weight of the compared elements to the criterion, and perform a consistency test; finally, calculate the weight of each layer of elements relative to the overall goal, and get the sequence .

[0041] S103, then the reference series and the comparison series are dimensionless. Due to the different physical meanings of the factors in the system, the dimensions of the data are not necessarily the same, which makes it difficult to compare or to get the correct conclusion when comparing. Therefore, dimensionless data processing is generally required when performing grey correlation analysis. The next step is to obtain the grey correlation coefficient, that is, first determine is an ideal sequence, and then the correlation coefficient of each evaluation sequence is calculated according to the relevant formula , thus obtaining the correlation coefficient matrix: Finally, the correlation is calculated based on Calculate the correlation degree G. Select the best fire station based on the correlation degree.

[0042] S2. Use hierarchical-grayscale correlation analysis method to solve the model.

[0043] The hierarchical-grayscale correlation analysis method comprises: S201, calculate the weight of each evaluation factor, assuming m parameters Affects the selection of fire stations, two parameters are selected each time and ,by express and The ratio of the influence on the selection of fire stations, thus obtaining the pairwise comparison judgment matrix: , the elements in A should satisfy the following conditions: ; The method of the present invention adopts the 1-9 scale method as The scaling method. and When the ratio is equal Take 1, when the ratio of the two is strong Take 3, when the ratio of the two is strong, Take 5, when the ratio between the two is very strong Take 7, when the ratio of the two is absolutely strong Take 9.

[0044] According to the different levels of evaluation parameters, three judgment matrices are obtained. They are the global comparison judgment matrix of the fire station's ability to handle the disaster accident (a), the firefighting force that can be dispatched by the fire station (b), the water source situation at the accident site (e), the geographical characteristics of the fire site (f), and the number of fire-fighting resources reserved by the fire station (g), as well as the local judgment matrix of the number of personnel that can be called up in the fire station (j), the number of vehicles that can be dispatched (h), and the local judgment matrix of the distance from the accident site to the fire station (i), and the degree of road congestion between fire stations (k). The weights of the evaluation factors are calculated using the characteristic root method for the obtained judgment matrix. The process is as follows: First, calculate the characteristic root of the judgment matrix A: ; In the formula —The maximum eigenvalue Z of A— The corresponding feature vector W is then normalized to obtain the weight vector and the global weight is obtained by normalization calculation. . Local weight and .

[0045] Because the judgment matrix is ​​determined subjectively, it is necessary to check the consistency of the judgment matrix. The average random consistency index of the same order The ratio of , to determine whether the consistency of matrix A is acceptable: ; Where n is the order of the judgment matrix A. When , the consistency of A is acceptable, otherwise adjust A. It can be obtained by checking correction table 1.

[0046] Table 1 Value comparison table: ;

[0047] After the above preparation, the evaluation parameter weights can now be determined, namely:

[0048] Where t is the parameter in the evaluation target tree. Number of ancestors; —Evaluation parameters The weights obtained at the bottom layer; -parameter The weight obtained by the i-th ancestor in the hierarchical single sorting.

[0049] The weight of each evaluation parameter is: .

[0050] S202. Use grey correlation analysis method to select the best fire station.

[0051] Now there are 10 fire stations in the city as the fire stations to be evaluated, and the values ​​of the m indicators in the 10 fire stations are listed in a matrix, then: ; Fire Station In decision-making criteria The assessed value below.

[0052] The above 10 matrices are called matrices to be evaluated.

[0053] Constructing an ideal fire station sequence based on the relative optimization principle , that is, if the maximum value of a certain indicator is good, then take the maximum value of the indicator in each solution; if the minimum value of a certain indicator is good, then take the minimum value of the indicator in each solution, and the ideal sequence is: ; Because the dimensions of each parameter are different, in order to eliminate this effect, different dimensionless treatments are used for different types of parameters.

[0054] For very large parameters: ; For very small parameters: ; in , .

[0055] Now we start to calculate the correlation coefficient. Assume that the dimensionless values ​​of the ideal sequence and the evaluation sequence are and , To evaluate the sequence Evaluation indicators With ideal sequence Evaluation indicators The grey correlation coefficient between . Then we get: ; in, is the resolution coefficient, , usually taken as 0.5.

[0056] Because this model only considers the maximum and minimum cases, ,Pick , the correlation matrix is ​​obtained by calculation: ; Next, according to the formula The grey correlation degree is obtained, where Z is the weight vector of the evaluation parameters obtained above relative to the overall target. According to the grey correlation analysis, the fire station with the largest correlation degree is the best fire station.

[0057] In the present invention, the process of realizing intelligent decision-making involves the joint work of a data acquisition module, a data processing module and an intelligent decision-making module.

[0058] First, the data acquisition module collects data on the fire scene environment and the status of firefighters (such as location, heart rate, body temperature, blood oxygen saturation, remaining capacity of oxygen cylinders, pressure resistance level of fire suits, etc.) in real time through various sensor devices and transmits these data to the data processing module in real time through the communication module. These sensor devices can be fixed in buildings or carried by drones or intelligent robots to achieve comprehensive monitoring of the fire scene environment.

[0059] Then, after receiving the data, the data processing module will first clean the data to remove invalid, abnormal or isolated data to avoid irrelevant interference. After that, the module will use the cleaned data to identify the fire source and safe passage. Among them, fire source identification is achieved by using the support vector machine (SVM) model, while safe passage identification uses a path planning algorithm, combined with current environmental data and building map data, to plan the safest and shortest evacuation path. In addition, the data processing module also uses a long short-term memory neural network (LSTM) to predict the fire.

[0060] The intelligent decision-making module adopts the intelligent fire decision-making method of hierarchical-grayscale correlation analysis, which specifically includes establishing a mathematical model of the intelligent decision-making mechanism and then solving the model using hierarchical-grayscale correlation analysis.

[0061] The established mathematical model will take the fire station as an alternative option, and the factors affecting the option selection, such as the geographical location of the disaster, the nature of the burning materials, the water source conditions at the accident site, the degree of road congestion, the geographical location of the fire station, and the status of the fire trucks in the fire station, will be used as the decision criteria. The ultimate goal is to select a suitable fire station that can reach the disaster site in the shortest time.

[0062] The system will also generate firefighting strategies and personnel dispatch suggestions based on the firefighters' status data and fire scene environment data. If the firefighters' status data shows that their physiological indicators such as heart rate, body temperature or blood oxygen saturation are abnormal, the system will suggest that they temporarily leave the fire scene for a rest. If the fire scene environment data shows that the location of the fire source has changed or a new fire source has been discovered, the system will readjust the firefighting strategy, such as changing the firefighting route, mobilizing more firefighting personnel or equipment, etc.

[0063] The above is the relevant process of generating fire-fighting strategies and personnel dispatch suggestions. Through this method, flexible and accurate fire-fighting decisions can be made based on the on-site environment and the specific conditions of the firefighters, greatly improving the efficiency and safety of fire-fighting.

[0064] The system uses a special path planning algorithm (such as the Dijkstra algorithm) to plan the safest path in and out of the fire zone and evacuate based on the real-time fire source location, firefighter location, fire truck location, and environmental conditions. The process is as follows: First, the fire scene environment is abstracted into a network or graph model, where each node represents a possible location and each edge represents the connection between two locations. This model divides the entire environment into many small areas. The fire source conditions, traffic conditions and other information of each area are obtained through the data acquisition module and then marked on the corresponding nodes. The current position of the firefighter or fire truck and the fire source location are two nodes in the graph. The goal of path planning is to find the optimal path from the current location node to the fire source location node. Then, use the path planning algorithm to search in this graph model to find a path with the lowest fire source intensity and the least difficulty of passage. This path is the safest path to enter the fire or evacuate.

[0065] After finding the optimal path, the system will transmit the information of this path to the navigation equipment of firefighters and fire trucks through the communication module. The navigation equipment provides real-time travel guidance for firefighters or fire trucks based on this path information, instructing them on how to act in order to reach the destination as quickly and safely as possible.

[0066] Communication module: used for data transmission between different modules or data exchange with external devices.

[0067] Internal communication: Within the system, data transmission between different modules will follow certain network protocols. These protocols mainly include TCP / IP and HTTP. TCP / IP is the basic protocol for network communication. It is responsible for dividing data into several small blocks for transmission and ensuring that all data is accurate during transmission. The HTTP protocol is an application layer protocol that is responsible for determining the format and content of the data, as well as the method of data transmission. In actual applications, data exchange between the server and the client is usually carried out through the HTTP protocol, and the HTTP protocol itself works based on the TCP / IP protocol. For example, the data acquisition module sends the collected data to the server through the HTTP protocol. After the server receives the data, it distributes the data to other modules through the TCP / IP protocol.

[0068] External communication: Data exchange between the system and external devices is mainly completed through wireless communication technologies, such as 4G / 5G, LoRa, ZigBee, etc. 4G / 5G is the current mainstream mobile communication technology. They have the characteristics of high-speed transmission and large coverage area, and can realize real-time and remote data transmission. For example, through the 4G / 5G network, drones can transmit the captured fire scene pictures back to the data center in real time. LoRa and ZigBee are very power-saving wireless communication technologies designed specifically for IoT applications. They are suitable for communication between short-distance and low-power devices. For example, in a fire scene, the equipment carried by firefighters can be networked through these two technologies to realize data exchange and communication.

[0069] Positioning and navigation module: Through GPS or other positioning technologies, the location of firefighters and fire trucks is monitored in real time, and fire-fighting paths and personnel evacuation paths are generated.

[0070] The positioning and navigation module is an important part of the intelligent fire fighting command and decision-making support system. It is mainly responsible for real-time monitoring of the positions of firefighters and fire trucks, and generating fire extinguishing paths and personnel retreat paths. The specific implementation process is as follows: Firefighters and fire trucks are equipped with positioning devices, such as GPS receivers. When they are working at a fire scene, these devices continuously transmit radio wave signals, and the receivers on the devices can calculate the accurate location information of firefighters or fire trucks. This location information is transmitted to the data processing module in real time for real-time monitoring and decision-making of fire situations.

[0071] Storage module: used to store historical data and decision-making process.

[0072] Structured data, such as fire truck information, firefighter information, or fire station information, is generally stored in a relational database, such as MySQL or Oracle. When new data is generated, it is written into the database through SQL statements, such as INSERT INTOXXX. Then, when this information is needed for decision analysis, report generation, and system monitoring, the data is obtained by writing and executing SQL query statements, such as SELECT * FROMXXX.

[0073] Large-scale, semi-structured data processing, such data may come from camera videos at the fire scene, pictures taken by drones, etc., which are usually stored in NoSQL databases such as MongoDB or Cassandra. These data can be stored in the database in binary form. When these data need to be used, they can be queried and processed through programming interfaces such as Java's MongoDBdriver and Python's pymongo library.

[0074] Processing of sensor data. Sensor data is usually a set of time-related data series, such as temperature, humidity, smoke concentration, etc. These data need to be stored through a time series database, such as InfluxDB. Sort the acquired data by timestamp and store it in the time series database. When using it, you can query it in combination with time, such as SELECT*FROMXXXWHEREtime>='2021-01-01T00:00:00Z' in InfluxDB.

[0075] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0076] It should be understood that the detailed description of the technical solutions of the present invention by means of the preferred embodiments is illustrative rather than restrictive. A person skilled in the art may modify the technical solutions described in the embodiments, or replace some of the technical features by equivalents, based on reading the specification of the present invention; and these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent fire fighting command decision-making support system, characterized by: The system comprises: The data acquisition module is connected to the data processing module and includes various sensor devices for collecting fire scene environment data and firefighter status data, including the firefighters' location, heart rate, body temperature, blood oxygen saturation, remaining capacity of oxygen cylinders, and pressure resistance level of firefighter suits, and transmits them to the subsequent data processing module in real time; The data processing module cleans the collected fire scene environment data and firefighter status data, and then identifies the fire source and safe passage; The intelligent decision-making module is connected to the data processing module and is used to generate fire-fighting strategies and personnel dispatch suggestions based on the fire source identification results and the safety channel identification results, assist the commander in making decisions, and generate fire-fighting paths and personnel evacuation paths; The communication module is connected with the intelligent decision-making module, the data processing module, the data acquisition module, the field sensor, and the drone, and is used for data transmission between different modules and / or data exchange with external devices; The positioning and navigation module is connected to the communication module to monitor the location of firefighters and fire trucks in real time based on positioning technology; The storage module is connected with the data processing module, the intelligent decision-making module, the field sensor, and the drone to store historical data and decision-making process.

2. The intelligent fire fighting command decision-making auxiliary system according to claim 1 is characterized in that: The data acquisition module collects data from sensors in the fire environment, including sensor devices fixed in buildings, or mobile sensor devices carried by drones or intelligent robots.

3. The intelligent fire fighting command decision support system according to claim 1 is characterized in that: The data processing module comprises: Data cleaning: Clean the collected raw fire scene environment data and firefighter status data, and delete invalid, abnormal or isolated data; Fire source identification: Use the support vector machine (SVM) model to identify the fire source location; Fire prediction: Use long short-term memory neural network LSTM to predict the development of fire based on historical data; Safe passage identification: Use path planning algorithms, combined with current environmental data and building map data, to plan the safest and shortest evacuation route.

4. The intelligent fire fighting command decision-making support system according to claim 1 is characterized in that: The intelligent decision-making module adopts an intelligent fire fighting decision-making method of hierarchical-grayscale correlation analysis method, including: Establish mathematical models for intelligent decision-making mechanisms; The hierarchical-grayscale correlation analysis method is used to solve the model.

5. The intelligent fire fighting command decision-making support system according to claim 4 is characterized in that: The mathematical model of the intelligent decision-making mechanism is specifically: Record the fire station as an alternative: , represents n optional fire stations; The set of alternative options refers to all fire stations; gather Every fire station in The factors affecting the selection of the plan, such as the geographical location of the disaster, the nature of the burning material, the water source of the accident site, the degree of road congestion, the geographical location of the fire station, and the status of the fire trucks in the fire station, are used as the decision criteria and recorded as: , The set of decision criteria that affect the selection of fire stations is represented by a set of m decision indicators; gather Each decision criterion in represents the first criterion, the second criterion, and so on, until the mth criterion; the ultimate goal is to select the appropriate fire station to arrive at the disaster site as soon as possible; It is called the evaluation sequence. Fire Station In decision-making criteria The assessed value below.

6. The intelligent fire fighting command decision-making support system according to claim 4 is characterized by: The hierarchical-grayscale correlation analysis method comprises: The weight of each evaluation factor is calculated, and three judgment matrices are obtained according to different levels of evaluation parameters: a global comparison judgment matrix of the fire station's ability to handle the disaster, the firefighting forces that can be dispatched by the fire station, the water source conditions at the accident site, the geographical characteristics of the fire site, and the number of fire-fighting resources reserved by the fire station, as well as a local judgment matrix of the number of personnel that can be dispatched in the fire station and the number of vehicles that can be dispatched, and a local judgment matrix of the distance from the accident site to the fire station and the degree of road congestion between the fire stations. Use grey correlation analysis to select the best fire station.

7. The intelligent fire fighting command decision-making auxiliary system according to claim 1 is characterized by: The communication module comprises: Internal communication: Within the system, data transmission between different modules is realized through network protocols TCP / IP and HTTP; External communication: Data exchange between the system and external devices is carried out through wireless communication 4G / 5G, LoRa, and ZigBee.

8. The intelligent fire fighting command decision-making support system according to claim 1 is characterized by: The storage module uses a relational database to store structured data, a NoSQL database to store large-scale, semi-structured data, and a time series database to store sensor data; Read data from the database for decision analysis, report generation, and system monitoring.

Images