A method and system for multi-dimensional fire safety monitoring in smart campuses

Abstract

This application relates to the field of fire safety technology, providing a method and system for multi-dimensional fire safety monitoring in smart campuses. The method includes: collecting multi-source safety data from campus buildings; extracting features based on preset risk dimensions to generate fire risk feature vectors; constructing a risk model based on historical fire cases, scoring the feature vectors to obtain real-time risk scores; triggering tiered early warning commands when the score exceeds a dynamic threshold; generating emergency plans based on the early warnings, and coordinating the execution of operations with fire-fighting equipment and personnel evacuation systems. This application solves the technical problem of existing campus fire safety monitoring systems lacking multi-dimensional, real-time, and comprehensive risk assessment methods, making it difficult to accurately determine different fire risks and respond in real time. It achieves the technical effect of improving the level of campus fire safety management by realizing accurate fire risk early warning and efficient emergency response through multi-source data fusion and intelligent analysis.

Description

Technical Field

[0001] This application relates to the field of fire safety technology, specifically to a multi-dimensional fire safety monitoring method and system for smart campuses. Background Technology

[0002] With the continuous advancement of smart campus construction, campus building structures are becoming increasingly complex, personnel mobility is increasing, and functional areas are becoming more diversified, posing new challenges to fire safety. Traditional campus fire safety management methods largely rely on manual inspections and fixed sensor deployments, lacking the ability to comprehensively perceive multi-source heterogeneous data, making it difficult to identify potential fire risks in a timely and accurate manner. In practical applications, factors such as inconsistent building fire resistance, opaque operating status of fire-fighting equipment, drastic fluctuations in environmental parameters, and dense population distribution can all contribute to fire hazards. However, existing technologies generally lack a comprehensive risk analysis mechanism based on building structural characteristics, equipment status, environmental changes, and personnel dynamics, making it difficult to achieve real-time assessment and dynamic response to fire risks, resulting in delayed early warnings and inefficient emergency response. To improve the overall fire safety level of campuses, it is urgent to introduce intelligent, multi-dimensional, and dynamically adaptive monitoring methods and systems that integrate multi-source data analysis, machine learning modeling, and emergency linkage control technologies to achieve accurate identification, scientific early warning, and efficient handling of campus fire risks. Summary of the Invention

[0003] This application provides a multi-dimensional fire safety monitoring method and system for smart campuses, aiming to solve the technical problem that existing campus fire safety monitoring systems lack multi-dimensional, real-time and comprehensive risk assessment means, making it difficult to accurately judge different fire risks and respond in real time.

[0004] The first aspect disclosed in this application provides a method for multi-dimensional fire safety monitoring in a smart campus. The method includes: collecting multi-source safety data from campus buildings, including building structure data, fire equipment status data, environmental sensor data, and personnel distribution data; extracting features from the multi-source safety data based on a preset risk dimension to generate a fire risk feature vector; constructing a multi-dimensional risk probability model based on historical fire case data, calculating a risk score on the fire risk feature vector, and obtaining a real-time risk score; triggering a tiered early warning instruction when the real-time risk score exceeds a dynamic risk threshold; and generating an emergency response plan based on the tiered early warning instruction, linking the fire equipment control system and the personnel evacuation guidance system to execute emergency operations.

[0005] Another aspect of this application discloses a smart campus multi-dimensional fire safety monitoring system, the system comprising: a data acquisition module for acquiring multi-source safety data of campus buildings, including building structure data, fire equipment status data, environmental sensor data, and personnel distribution data; a feature extraction module for extracting features from the multi-source safety data based on preset risk dimensions to generate a fire risk feature vector; a scoring calculation module for constructing a multi-dimensional risk probability model based on historical fire case data, calculating a risk score on the fire risk feature vector, and obtaining a real-time risk score; an instruction triggering module for triggering a graded early warning instruction when the real-time risk score exceeds a dynamic risk threshold; and an emergency operation module for generating an emergency response plan based on the graded early warning instruction, and linking the fire equipment control system and the personnel evacuation guidance system to execute emergency operations.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The aforementioned multi-dimensional fire safety monitoring method for smart campuses comprehensively assesses the campus's internal safety situation by collecting various safety data from campus buildings, including building structure, fire equipment operation status, environmental sensor information, and personnel distribution. Subsequently, based on preset risk dimensions, this data undergoes in-depth analysis and feature extraction to construct a feature vector reflecting the current fire risk status. Combined with numerous historical fire cases, a multi-dimensional risk probability model is established to score the current feature vector, deriving a real-time fire risk level. When the risk score exceeds a system-set dynamic threshold, the system automatically triggers different levels of warning commands. Following this, corresponding emergency response plans are generated based on the warning level, and fire equipment and personnel evacuation guidance systems are activated to achieve rapid response and intelligent handling in the early stages of a fire. This method achieves closed-loop management from risk identification to emergency response, enhancing the intelligence level of campus fire safety.

[0008] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used 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 those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating a multi-dimensional fire safety monitoring method for smart campuses in one embodiment.

[0011] Figure 2 This is an architecture diagram of a multi-dimensional fire safety monitoring system for a smart campus, as shown in one embodiment.

[0012] Figure labeling: Data acquisition module 11, feature extraction module 12, scoring calculation module 13, instruction triggering module 14, emergency operation module 15. Detailed Implementation

[0013] This application provides a smart campus multi-dimensional fire safety monitoring method and system to solve the technical problem that existing campus fire safety monitoring systems lack multi-dimensional, real-time and comprehensive risk assessment means, making it difficult to accurately judge different fire risks and respond in real time.

[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0015] It should be noted that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product, or device.

[0016] Example 1, as Figure 1 As shown, this application provides a multi-dimensional fire safety monitoring method for smart campuses, the method comprising:

[0017] Collect and acquire multi-source safety data of campus buildings, including building structure data, fire equipment status data, environmental sensor data, and personnel distribution data.

[0018] In this embodiment, various safety information of campus buildings is collected in real time through multiple sensors and data sources. This data includes building structure data (such as fire resistance rating, structural design, and distribution of safety exits), fire equipment status data (such as fire extinguisher pressure, sprinkler system operation status, and smoke alarm operation status), environmental sensor data (such as temperature, humidity, smoke concentration, and carbon dioxide detection data), and personnel distribution data (such as the number of people on different floors and in different areas, personnel flow paths, and personnel density). This multi-source data provides a foundation for a comprehensive assessment of fire risks, enabling the system to more accurately perceive and analyze the fire safety status within the campus.

[0019] Based on a preset risk dimension, features are extracted from the multi-source safety data to generate a fire risk feature vector.

[0020] In one embodiment, collected multi-source safety data is analyzed and processed according to pre-defined risk dimensions. First, data on building structure, fire-fighting equipment, environmental sensors, and personnel distribution are categorized according to the preset risk dimensions. Each category of data is then extracted based on fire risk-related characteristics. For example, the building's fire resistance rating, equipment operating status, ambient temperature and smoke concentration, and personnel density are all converted into digitized feature values. For non-numerical data, one-heat encoding can be used for quantification. Subsequently, these feature values ​​are integrated to form a comprehensive fire risk feature vector, which includes all factors that may affect the occurrence and spread of a fire. In this way, the current fire risk status on campus can be quantified and represented, providing foundational data for subsequent risk assessment and decision-making.

[0021] Furthermore, this application provides a method for feature extraction from the multi-source security data based on a preset risk dimension, comprising:

[0022] The building structure data is divided into a building risk dimension, extracting building fire resistance rating, safety exit density, and fire compartment integrity index; the fire equipment status data is divided into an equipment availability dimension, extracting equipment failure rate, fire extinguisher pressure value, and smoke alarm response time; the environmental sensor data is divided into an environmental hazard dimension, extracting temperature change rate, smoke concentration gradient, and CO concentration peak; the personnel distribution data is divided into a personnel density dimension, extracting the number of people on each floor, escape route congestion index, and number of people remaining in key areas; the data from the building risk dimension, equipment availability dimension, environmental hazard dimension, and personnel density dimension are normalized to generate the fire risk feature vector.

[0023] Preferably, the system analyzes the preset risk dimensions to determine the required data types of these dimensions, and extracts specific features related to fire risk from multi-source safety data based on these data types. Specifically, it extracts building risk dimension features from the building structure data of multi-source safety data to obtain building fire resistance rating, safety exit density, and fire compartment integrity index. The building fire resistance rating assesses the fire resistance of building materials and structures, and is categorized into A, B, and C, representing the fire resistance of building materials. The safety exit density is the ratio of the number of available safety exits per floor to the total building area, reflecting emergency evacuation capabilities. The fire compartment integrity index assesses the integrity of each fire compartment in the building, such as the integrity of firewalls and the sealing of doors and windows. It also extracts equipment availability dimension features from the fire equipment status data of multi-source safety data to obtain equipment failure rate, fire extinguisher pressure value, and smoke alarm response time. The equipment failure rate reflects equipment reliability; the fire extinguisher pressure value determines whether the fire extinguisher pressure is within the normal range, with low pressure potentially indicating a malfunction; and the smoke alarm response time reflects the sensitivity and response speed of the equipment. Environmental hazard features were extracted from multi-source safety data, including environmental sensor data, to obtain temperature change rate, smoke concentration gradient, and CO concentration peak. Temperature change rate refers to the rate of temperature change in the environment; a rapid temperature rise may be an indicator of fire spread. Smoke concentration gradient reflects differences in smoke concentration across different areas; areas with larger gradients may be the source or spread zone of a fire. CO concentration peak refers to the maximum concentration of carbon monoxide in the air; an increase in carbon monoxide concentration is usually an indicator of a fire. Personnel density features were extracted from multi-source safety data, including the number of people per floor, escape route congestion index, and number of people remaining in key areas. The number of people per floor refers to the real-time number of people on each floor; the escape route congestion index indicates the density of people in escape routes, with excessive congestion reducing evacuation efficiency; and the number of people remaining in key areas refers to the number of people remaining in key areas (such as laboratories and classrooms), as excessive numbers of people remaining may affect evacuation efficiency. After obtaining the feature data for these dimensions, the data for each dimension is normalized to convert different types of data into standard comparable values. Common normalization methods include maximum value normalization and Z-score normalization. This step ensures that data from different dimensions are compared on the same scale, and features with different dimensions can be directly combined. Finally, the normalized feature values ​​of each dimension are concatenated to form a comprehensive fire risk feature vector. This feature vector integrates risk information from various aspects such as building structure, fire equipment status, environmental hazards, and personnel density, comprehensively reflecting the overall fire risk status on campus and serving as the basis for subsequent risk assessment and response decisions.

[0024] A multi-dimensional risk probability model is constructed based on historical fire case data, and a risk score is calculated on the fire risk feature vector to obtain a real-time risk score.

[0025] In one embodiment, historical fire cases in campuses or similar buildings are collected and organized via a network, including information such as the time and location of the fire, the fire's development process, the extent of damage, and the fire response. By feature-labeling and recording each historical fire case, key data such as building structural characteristics, equipment status, environmental parameters, and personnel distribution can be obtained. Subsequently, based on this data, supervised learning is used to construct building risk detection units, equipment risk detection units, environmental risk detection units, and personnel risk detection units. These units can perform scoring and prediction based on the received data, and obtain a comprehensive risk score through weighted averaging. By integrating these units, a multi-dimensional risk probability model can be obtained. When the multi-dimensional risk probability model receives a fire risk feature vector, it decomposes the fire risk feature vector into multiple sub-vectors, each corresponding to a type of data, such as a building structure vector or an equipment status vector. Then, the multi-dimensional risk probability model uses its internal units to calculate risk scores for these sub-vectors, thereby predicting multiple scores. After weighted summation by the model's output layer, a real-time risk score is obtained. This score reflects the level of fire risk in the current building environment; the higher the score, the greater the fire risk. This real-time score is a dynamic indicator that updates as real-time data changes. The system continuously updates the risk score based on various data collected in real time (such as ambient temperature, smoke concentration, equipment status, etc.), providing decision support for subsequent early warning and emergency response.

[0026] Furthermore, this application provides that the aforementioned multi-dimensional risk probability model is constructed in the following manner:

[0027] Historical fire case data is acquired, and the corresponding building structure characteristics, equipment status characteristics, environmental characteristics, and personnel characteristics for each event are labeled. The building structure characteristics and their building structure scores are used as input to train a building risk detection unit; the equipment status characteristics and their equipment status scores are used as input to train an equipment risk detection unit; the environmental characteristics and their environmental scores are used as input to train an environmental risk detection unit; and the personnel characteristics and their personnel scores are used as input to train a personnel risk detection unit. The building risk detection unit, equipment risk detection unit, environmental risk detection unit, and personnel risk detection unit are integrated to generate the multi-dimensional risk probability model.

[0028] Preferably, historical fire case data of campuses or similar buildings are collected, and the building structure characteristics, equipment status characteristics, environmental characteristics, and personnel characteristics in these case data are labeled. These characteristics include corresponding dimensional data (such as building fire resistance rating, equipment failure rate, temperature change rate, etc.) and corresponding feature scores. For example, for building structure characteristics, this includes building fire resistance rating, safety exit density, fire compartment integrity index, and building structure score. Subsequently, the building structure characteristics and their corresponding building structure scores are used as input to train a building risk detection unit using supervised learning algorithms (such as decision trees, support vector machines, neural networks, etc.). This unit will be able to assess the risk status of the building based on the input building structure characteristics and predict whether the building structure is likely to exacerbate the spread of fire or cause greater damage in the event of a fire. Similarly, the equipment status characteristics and their corresponding equipment status scores are used as input to train an equipment risk detection unit. This unit assesses whether the fire-fighting equipment is in good condition based on the working status of the equipment, such as whether fire extinguishers are effective and whether smoke detectors can respond in a timely manner, thereby assessing whether the equipment can effectively cope with fire risks. Similarly, environmental characteristics and their corresponding environmental scores are used as input to train an environmental risk detection unit. This unit assesses the likelihood of a fire by analyzing environmental conditions (such as temperature, humidity, and gas concentration). For example, changes in high temperature, dense smoke, or CO concentration may be precursors to a fire, and the environmental risk unit can predict the environmental risk status based on these factors. Likewise, personnel characteristics and their corresponding personnel scores are used as input to train a personnel risk detection unit. This unit assesses the impact of personnel distribution on fire, including evacuation capacity during a fire, passageway congestion, and the number of people remaining in key areas. Personnel density and the unobstructedness of evacuation routes have a significant impact on fires, and this unit can assess the ease of evacuation and potential risks based on these factors. Finally, the building risk detection unit, equipment risk detection unit, environmental risk detection unit, and personnel risk detection unit are integrated in parallel to form a comprehensive multi-dimensional risk probability model. This model can comprehensively assess the severity of a fire based on multiple factors such as the input building structure, equipment status, environmental data, and personnel distribution, and generate a real-time risk score based on the assessment results. This training process enables accurate assessment of current fire risks based on historical fire data, real-time data input, and intelligent models, and allows for risk prediction and management based on multi-dimensional assessment results.

[0029] Furthermore, this application provides a training building risk detection unit, including:

[0030] The building structure features are paired with the corresponding building structure scores to construct a training sample set; the training sample set is then normalized for features and standardized for labels to generate standardized training data; the standardized training data is input into the initial building risk detection unit and trained using a supervised learning algorithm to form a building risk detection unit capable of assessing building structure risk.

[0031] Optionally, building structural features can be extracted from historical fire case data. These features may include the building's fire resistance rating, number of safety exits, fire compartment integrity, and building structural rating. By pairing each building's fire resistance rating, number of safety exits, and fire compartment integrity with its building structural rating, a training sample set is constructed. Each sample consists of a building structural feature and its corresponding rating, forming an input-output pair. Subsequently, since different building structural features (such as fire resistance rating and number of safety exits) may have different dimensions or ranges, these features need to be normalized. Common methods include maximum value normalization (scaling each feature to between 0 and 1) or Z-score standardization (converting the data to a standard normal distribution with a mean of 0 and a variance of 1). By standardizing this data and then concatenating the processed data, multiple sample feature vectors can be obtained. The building structural rating is used as the label for each sample feature vector, thus generating standardized training data. Through normalization and standardization, all features and labels will be unified within the same dimension and range, avoiding excessive influence of features with different dimensions on model training. Next, the standardized training dataset is input into the initial building risk detection unit, which is a model built based on machine learning algorithms (such as support vector machines, decision trees, random forests, neural networks, etc.). Taking a multilayer perceptron (MLP) as an example, an MLP is used to construct a building risk detection unit, including an input layer, hidden layers, and an output layer. The weights of the building risk detection unit are initialized using random numbers, and the standardized training data is input into the initialized building risk detection unit for forward propagation. The data is passed layer by layer through the input layer, hidden layer, and output layer to extract features from the building structure data and generate a risk score for the building structure. Then, the mean squared error loss function is used to calculate the loss value between the predicted result and the actual label, and the gradient of the loss with respect to the weights of each layer is calculated layer by layer through backpropagation. The Adam optimizer is then used to optimize the unit parameters, adjusting the weights to minimize the value of the loss function. This process is repeated until the maximum number of iterations is reached. After training, the unit performance is tested using a validation set to evaluate the accuracy of the unit on the building structure risk assessment task. If the accuracy meets expectations, the current building risk detection unit is output. Conversely, hyperparameters such as the learning rate and the number of training batches can be adjusted to further improve the effectiveness of the unit in building structure risk assessment.

[0032] When the real-time risk score exceeds the dynamic risk threshold, a tiered early warning instruction is triggered.

[0033] In one embodiment, when the calculated real-time fire risk score exceeds a preset dynamic risk threshold, it means that the current fire risk has reached a certain dangerous level and may pose a threat to campus safety. At this point, a tiered early warning mechanism is automatically triggered, issuing corresponding warning instructions based on different risk levels (such as yellow, orange, and red alerts). Each warning level corresponds to different response measures, helping relevant personnel to take timely preventative or emergency measures to mitigate potential fire risks. Through this tiered early warning system, different situations can be dynamically addressed, ensuring campus fire safety.

[0034] Furthermore, this application provides that the determination of the dynamic risk threshold includes:

[0035] The system obtains the campus activity scenario types for the current time period, including class time, assembly time, and nighttime rest time. It matches preset basic thresholds based on the scenario type, where the basic thresholds correspond to class time, the assembly time thresholds are dynamically adjusted based on the risk of personnel gathering, and the nighttime rest time thresholds are dynamically adjusted based on the risk of response delay. It also overlays a real-time weather condition impact assessment; when meteorological conditions are detected to exacerbate the risk of fire spread, the basic thresholds are adaptively adjusted to obtain a dynamic risk threshold.

[0036] Preferably, the first step is to identify and obtain the campus activity scenario type for the current time period. Depending on the time of day, campus activities and behaviors differ. Common scenario types include class time, assembly time, and nighttime rest time. Class time is generally during the day, with people relatively dispersed in classrooms, laboratories, etc., making evacuation relatively easier. Assembly time includes campus activities, lectures, or gatherings, where people are concentrated in a certain area, leading to greater evacuation pressure. Nighttime rest time sees most people in dormitories or rest areas with less movement, typically resulting in a longer response time after a fire. Then, a preset base threshold is matched based on the scenario type. For class time, due to the dispersed nature of people and regular use of building facilities, the fire risk is relatively low; therefore, a lower base threshold is set to reflect the fire risk level during this period. For assembly time, considering the concentrated number of people, the difficulty of evacuation and the potential risk are greater in the event of a fire. In this case, the base threshold is lowered to improve early warning sensitivity, typically by 20%, but can be adjusted according to actual needs. During nighttime rest periods, the response to a fire is delayed due to the resting state of personnel, requiring more sensitive early warnings. In this case, the base threshold is lowered further, typically by 30%. Subsequently, real-time monitoring of weather changes, particularly temperature, humidity, and wind speed, is conducted. When weather conditions exacerbate the risk of fire spread—that is, when one or more of the temperature, humidity, and wind speed exceed the meteorological impact threshold—the base threshold is adaptively adjusted based on weather factors. For example, higher wind speeds or excessively high temperatures may accelerate fire spread and increase fire risk. In this case, the difference between these abnormal parameters and the threshold is calculated, and the ratio is calculated to the threshold. The results are then weighted and summed to obtain an impact coefficient. This impact coefficient is then multiplied by the base threshold, and the product is subtracted from the base threshold to obtain a dynamic risk threshold. This dynamic risk threshold reflects the real-time fire risk level, providing a more accurate reference for subsequent early warnings and emergency responses, ensuring timely and appropriate responses in different scenarios and environments, and maximizing campus safety.

[0037] Furthermore, this application provides that the triggering of the tiered early warning instruction includes:

[0038] Set three dynamic risk thresholds: yellow, orange, and red. When the real-time risk score exceeds the yellow warning threshold for the first time, initiate equipment self-check and warning broadcast. When the score continues to exceed the orange warning threshold for a preset duration, trigger the sprinkler system to pre-start and unlock the escape route access control. When the score instantaneously exceeds the red warning threshold, implement the building-wide power outage protection and activate emergency lighting and evacuation guidance.

[0039] Optionally, based on historical experience and expert decisions, a three-tiered dynamic risk threshold system—yellow, orange, and red—can be set. The yellow threshold indicates a low risk but a potential fire hazard; when the real-time risk score first exceeds the yellow threshold, the fire risk is considered to have entered the warning stage, requiring initial response. The orange threshold indicates a high fire risk; if the score continues to exceed this threshold, the risk of a fire increases, requiring more preventative and emergency measures. The red threshold indicates an extremely high fire risk, with a serious threat of a fire outbreak; in this case, the maximum possible emergency response should be implemented immediately to ensure personnel safety and property protection. The system first compares the real-time fire risk score with the yellow threshold. Once the score first exceeds the yellow threshold, it indicates an increased fire risk, requiring the initiation of initial preventative measures. At this point, the system automatically initiates a self-check program to inspect the operational status of relevant fire-fighting equipment (such as fire extinguishers, smoke detectors, sprinkler systems, etc.) to ensure they are in normal working order. Simultaneously, an early warning broadcast is activated, broadcasting fire warning information to all personnel on campus, reminding everyone to be aware of fire hazards and prepare accordingly. When the fire risk score remains consistently above the orange alert threshold and exceeds a set time threshold (e.g., for more than one minute), the fire risk is considered to have increased, triggering a higher-level emergency response phase. At this point, the sprinkler system is pre-activated, with some valves opened in advance to ensure rapid fire suppression in the event of a fire. Simultaneously, escape route access is unlocked to ensure smooth evacuation and prevent people from being trapped inside. When the real-time fire risk score instantaneously exceeds the red alert threshold, indicating an extremely dangerous fire situation, the largest-scale emergency measures must be implemented immediately. In this case, automatic power outage protection is activated, cutting off power to the building to prevent electrical equipment from causing the fire to spread or exacerbate the disaster. Emergency lighting is activated to ensure that emergency exits, stairwells, and exits remain well-lit after the power outage, preventing people from becoming disoriented during evacuation. Furthermore, the evacuation guidance system is activated, using electronic displays or broadcasts to guide people to safe areas, ensuring an orderly evacuation process. Through the above process, emergency response measures can be dynamically adjusted according to real-time changes in fire risk, and corresponding fire-fighting equipment and personnel evacuation mechanisms can be automatically triggered according to different levels of warning, so as to maximize the safety of campus personnel.

[0040] An emergency response plan is generated based on the tiered early warning instructions, and the fire equipment control system and personnel evacuation guidance system are linked to execute emergency operations.

[0041] In one embodiment, based on real-time fire risk scores and warning levels, an appropriate emergency response plan is automatically selected. For example, a yellow warning may only require initiating equipment self-checks and warning broadcasts, while a red warning requires a comprehensive emergency response, involving power outage protection and sprinkler system activation. The emergency response plan includes not only the control of fire-fighting equipment but also specific arrangements for personnel evacuation. For instance, under an orange warning, the sprinkler system may need to be activated in advance, while under a red warning, all safety exits will be unlocked and evacuation routes will be unobstructed. According to the emergency response plan, the fire-fighting equipment control system will be activated automatically to initiate relevant fire-fighting equipment for fire prevention and suppression. For example, sprinkler systems, fire extinguishers, and smoke detectors will automatically activate under different warning levels. Under a red warning, the entire building's power outage protection will be fully activated to cut off power and prevent the spread of electrical fires. The sprinkler system will automatically adjust its spray area according to the fire's spread to quickly extinguish the fire. Simultaneously, the personnel evacuation guidance system will be activated, initiating emergency evacuation procedures. Based on the fire risk assessment, all escape routes will be automatically opened, safety exits will be unlocked, and personnel will be guided to evacuate quickly via emergency broadcasts or signs. During a red alert, emergency lighting will be activated to ensure that evacuation routes within the building remain clearly visible even in the event of a power outage, preventing people from becoming disoriented due to insufficient lighting. Evacuation guidance information will also be updated in real time via electronic displays and audio broadcasts to ensure orderly evacuation to safe areas. Through these coordinated operations, various emergency measures can be quickly and effectively activated in the event of a fire, ensuring rapid evacuation of personnel and timely response of fire-fighting equipment, thereby minimizing fire damage and protecting the lives of teachers and students.

[0042] Furthermore, this application provides an emergency response plan generated based on the aforementioned tiered early warning instructions, including:

[0043] Real-time acquisition of personnel location data and fire spread data; acquisition of three-dimensional evacuation routes based on building BIM model and fire spread data; calculation of optimal evacuation routes based on personnel location data and three-dimensional evacuation routes using Dijkstra's algorithm, and dynamic generation of the evacuation guidance.

[0044] Optionally, sensors, cameras, or other positioning devices installed within the building can be used to track and acquire the location information of each person inside the building in real time. This data can be transmitted to the system via wireless networks or other positioning technologies. Environmental sensors (such as smoke detectors and temperature sensors) are used to acquire information about the spread of the fire. This data reflects the location of the fire, its spread speed, and the areas that may be affected, helping to understand which areas have been occupied by the fire and which areas remain safe. Subsequently, a building BIM model is constructed. This BIM model includes structural information, floor plans, location data of key facilities such as emergency exits, escape routes, and stairwells. Through the BIM model, the layout of each floor and the building can be clearly understood. Combined with the fire spread data, it is possible to update in real time which areas are no longer suitable for passage and which areas are still safe. Based on the building BIM model and fire spread data, a three-dimensional evacuation route is calculated. This route not only considers the physical structure of the building but also avoids areas affected by the fire, ensuring that people can avoid the fire source and evacuate smoothly to a safe area. Next, the Dijkstra algorithm is activated. Dijkstra's algorithm is used to calculate the shortest path in a graph. Here, each node in the building (such as rooms, corridors, stairwells, etc.) is considered a node in the graph, and each edge on the path (such as corridors, doors, stairs) has a weight. The weight can be adjusted based on factors such as path length and the risk of fire spread. The system uses personnel location data as a starting point and combines it with a 3D evacuation path graph model to calculate the shortest and safest evacuation path from each employee's current location to a safe exit. This process takes into account the impact of fire spread, avoiding passage through fire-prone areas. Through the Dijkstra algorithm, the optimal evacuation path for each person is dynamically calculated, ensuring that people at each location can reach the nearest safe exit or safe area in the shortest possible time. Finally, based on the calculated optimal evacuation paths, evacuation guidance is generated in real time. This guidance can be disseminated to personnel through electronic displays, broadcast systems, etc., helping them quickly find the safest evacuation route and ensuring the safe evacuation of personnel.

[0045] In summary, the embodiments of this application have at least the following technical effects:

[0046] This application embodiment first collects multi-source safety data of campus buildings, including building structure data, fire equipment status data, environmental sensor data, and personnel distribution data. Based on a preset risk dimension, features are extracted from the multi-source safety data to generate a fire risk feature vector. A multi-dimensional risk probability model is constructed based on historical fire case data, and a risk score is calculated on the fire risk feature vector to obtain a real-time risk score. When the real-time risk score exceeds a dynamic risk threshold, a tiered early warning instruction is triggered. An emergency response plan is generated based on the tiered early warning instruction, and the fire equipment control system and personnel evacuation guidance system are linked to execute emergency operations. These technical effects collectively solve the technical problem that existing campus fire safety monitoring systems lack multi-dimensional, real-time, and comprehensive risk assessment methods, making it difficult to accurately determine different fire risks and respond in real time. This achieves the technical effect of realizing accurate fire risk early warning and efficient emergency response through multi-source data fusion and intelligent analysis, thereby improving the level of campus fire safety management.

[0047] Example 2, based on the same inventive concept as the multi-dimensional fire safety monitoring method for smart campuses in the foregoing examples, such as... Figure 2 As shown, this application provides a smart campus multi-dimensional fire safety monitoring system, the system comprising: a data acquisition module 11: acquiring multi-source safety data of campus buildings, including building structure data, fire equipment status data, environmental sensor data, and personnel distribution data; a feature extraction module 12: extracting features from the multi-source safety data based on a preset risk dimension to generate a fire risk feature vector; a scoring calculation module 13: constructing a multi-dimensional risk probability model based on historical fire case data, calculating a risk score on the fire risk feature vector, and obtaining a real-time risk score; an instruction triggering module 14: triggering a graded early warning instruction when the real-time risk score exceeds a dynamic risk threshold; and an emergency operation module 15: generating an emergency response plan based on the graded early warning instruction, and linking the fire equipment control system and the personnel evacuation guidance system to execute emergency operations.

[0048] Furthermore, the feature extraction module 12 is also used to perform the following method:

[0049] The building structure data is divided into a building risk dimension, extracting building fire resistance rating, safety exit density, and fire compartment integrity index; the fire equipment status data is divided into an equipment availability dimension, extracting equipment failure rate, fire extinguisher pressure value, and smoke alarm response time; the environmental sensor data is divided into an environmental hazard dimension, extracting temperature change rate, smoke concentration gradient, and CO concentration peak; the personnel distribution data is divided into a personnel density dimension, extracting the number of people on each floor, escape route congestion index, and number of people remaining in key areas; the data from the building risk dimension, equipment availability dimension, environmental hazard dimension, and personnel density dimension are normalized to generate the fire risk feature vector.

[0050] Furthermore, the scoring calculation module 13 is also used to perform the following method:

[0051] Historical fire case data is acquired, and the corresponding building structure characteristics, equipment status characteristics, environmental characteristics, and personnel characteristics for each event are labeled. The building structure characteristics and their building structure scores are used as input to train a building risk detection unit; the equipment status characteristics and their equipment status scores are used as input to train an equipment risk detection unit; the environmental characteristics and their environmental scores are used as input to train an environmental risk detection unit; and the personnel characteristics and their personnel scores are used as input to train a personnel risk detection unit. The building risk detection unit, equipment risk detection unit, environmental risk detection unit, and personnel risk detection unit are integrated to generate the multi-dimensional risk probability model.

[0052] Furthermore, the scoring calculation module 13 is also used to perform the following method:

[0053] The building structure features are paired with the corresponding building structure scores to construct a training sample set; the training sample set is then normalized for features and standardized for labels to generate standardized training data; the standardized training data is input into the initial building risk detection unit and trained using a supervised learning algorithm to form a building risk detection unit capable of assessing building structure risk.

[0054] Furthermore, the instruction triggering module 14 is also used to execute the following method:

[0055] The system obtains the campus activity scenario types for the current time period, including class time, assembly time, and nighttime rest time. It matches preset basic thresholds based on the scenario type, where the basic thresholds correspond to class time, the assembly time thresholds are dynamically adjusted based on the risk of personnel gathering, and the nighttime rest time thresholds are dynamically adjusted based on the risk of response delay. It also overlays a real-time weather condition impact assessment; when meteorological conditions are detected to exacerbate the risk of fire spread, the basic thresholds are adaptively adjusted to obtain a dynamic risk threshold.

[0056] Furthermore, the instruction triggering module 14 is also used to execute the following method:

[0057] Set three dynamic risk thresholds: yellow, orange, and red. When the real-time risk score exceeds the yellow warning threshold for the first time, initiate equipment self-check and warning broadcast. When the score continues to exceed the orange warning threshold for a preset duration, trigger the sprinkler system to pre-start and unlock the escape route access control. When the score instantaneously exceeds the red warning threshold, implement the building-wide power outage protection and activate emergency lighting and evacuation guidance.

[0058] Furthermore, the emergency operation module 15 is also used to perform the following methods:

[0059] Real-time acquisition of personnel location data and fire spread data; acquisition of three-dimensional evacuation routes based on building BIM model and fire spread data; calculation of optimal evacuation routes based on personnel location data and three-dimensional evacuation routes using Dijkstra's algorithm, and dynamic generation of the evacuation guidance.

[0060] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0061] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0062] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for multi-dimensional fire safety monitoring in a smart campus, characterized in that, The method comprises: Collecting multi-source safety data of the campus building, the multi-source safety data comprising building structure data, fire-fighting equipment state data, environmental sensor data and personnel distribution data; Performing feature extraction on the multi-source safety data based on preset risk dimensions to generate a fire risk feature vector; According to historical fire case data, a multi-dimensional risk probability model is constructed, and risk score calculation is performed on the fire risk feature vector to obtain a real-time risk score; When the real-time risk score exceeds a dynamic risk threshold, a hierarchical early warning instruction is triggered; An emergency handling scheme is generated according to the hierarchical early warning instruction, and a fire-fighting equipment control system and a personnel evacuation guidance system are linked to perform emergency operations; The determination of the dynamic risk threshold comprises: Obtaining a campus activity scene type of a current period, the scene type comprising a class period, an assembly period and a night sleep period; According to the scene type, a preset basic threshold is matched, wherein the basic threshold corresponds to the class period, the assembly period dynamically adjusts the basic threshold based on personnel gathering risk, and the night sleep period dynamically adjusts the basic threshold based on response delay risk; Superimposing real-time weather condition influence evaluation, when it is monitored that the meteorological environment aggravates the fire spread risk, the basic threshold is adaptively adjusted to obtain the dynamic risk threshold; According to the hierarchical early warning instruction, an emergency handling scheme is generated, comprising: Real-time acquisition of personnel position data and fire spread data; Based on the building BIM model and the fire spread data, a three-dimensional evacuation path is acquired; Based on the personnel position data and the three-dimensional evacuation path, the Dijkstra algorithm is used to calculate the optimal evacuation path, and an evacuation guide is dynamically generated.

2. The method of claim 1, wherein, The method for performing feature extraction on the multi-source safety data based on preset risk dimensions comprises: The building structure data is divided into a building risk dimension, and the building fire resistance rating, safety exit density and fire-fighting partition integrity index are extracted; The fire-fighting equipment state data is divided into an equipment availability dimension, and the equipment failure rate, fire extinguisher pressure value and smoke alarm response time are extracted; The environmental sensor data is divided into an environmental hazard dimension, and the temperature change rate, smoke concentration gradient and CO concentration peak value are extracted; The personnel distribution data is divided into a personnel density dimension, and the floor personnel quantity, escape passage congestion index and key area retention number are extracted; The data of the building risk dimension, equipment availability dimension, environmental hazard dimension and personnel density dimension are normalized to generate the fire risk feature vector.

3. The method of claim 1, wherein, The multi-dimensional risk probability model is constructed by the following way: Historical fire case data is acquired, and the building structure features, equipment state features, environmental features and personnel features corresponding to each event are labeled; The building structure features and building structure scores thereof are taken as inputs to train a building risk detection unit; The equipment state features and equipment state scores thereof are taken as inputs to train an equipment risk detection unit; The environmental features and environmental scores thereof are taken as inputs to train an environmental risk detection unit; The personnel features and personnel scores thereof are taken as inputs to train a personnel risk detection unit; The building risk detection unit, the equipment risk detection unit, the environment risk detection unit and the personnel risk detection unit are integrated to generate the multi-dimensional risk probability model.

4. The method of claim 1, wherein, The triggering of the hierarchical early warning instruction comprises: Setting yellow early warning, orange early warning and red early warning three-level dynamic risk thresholds; When the real-time risk score exceeds the yellow early warning threshold for the first time, starting equipment self-checking and early warning broadcast; When the score continuously exceeds the orange early warning threshold for a preset time length, triggering the pre-starting of the sprinkler system and unlocking the escape passage access control; When the score momentarily exceeds the red early warning threshold, performing full building power-off protection and activating emergency lighting and evacuation guidance.

5. The method of claim 3, wherein, Training the building risk detection unit comprises: Data pairing the building structure features and corresponding building structure scores to construct a training sample set; Performing feature normalization and label standardization processing on the training sample set to generate standardized training data; Inputting the standardized training data into the initial building risk detection unit and training by using a supervised learning algorithm to form a building risk detection unit capable of evaluating building structure risks.

6. A multi-dimensional fire safety monitoring system for a smart campus, characterized in that, The system is used to execute the intelligent campus multi-dimensional fire safety monitoring method of any one of claims 1-5, comprising: A data acquisition module: acquiring multi-source safety data of campus buildings, the multi-source safety data comprising building structure data, fire equipment state data, environment sensor data and personnel distribution data; A feature extraction module: extracting features from the multi-source safety data based on preset risk dimensions to generate a fire risk feature vector; A score calculation module: constructing a multi-dimensional risk probability model according to historical fire case data, calculating the risk score of the fire risk feature vector to obtain a real-time risk score; An instruction triggering module: triggering a hierarchical early warning instruction when the real-time risk score exceeds a dynamic risk threshold; An emergency operation module: generating an emergency treatment scheme according to the hierarchical early warning instruction, and executing emergency operations by linking the fire equipment control system and the personnel evacuation guidance system.

Images