Intelligent construction fire safety monitoring method, system, equipment and medium
Through intelligent construction fire safety monitoring methods, combined with environmental risk assessment and virtual environment simulation, dynamically adjusting the safety radius and automated monitoring, the problem of lack of accuracy and immediacy of fire safety monitoring in the existing technology is solved, and the construction site is improved safety management and emergency response capabilities.
Patent Information
- Application Number
- CN202411136828.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-19
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Due to the fire and explosion risks in construction, the existing safety monitoring methods are based on experience or fixed standards, neglecting in-depth analysis and dynamic adjustment of environmental risks, resulting in a lack of accuracy and immediacy of monitoring management, which reduces the speed of emergency response and disposal. efficiency.
Using intelligent construction fire safety monitoring method, by obtaining fire operation elements information, calculating environmental risk index, dynamically adjusting safety radius, monitoring personnel and equipment positions in real time, automatic control authority, combining virtual environment simulation and abnormal identification models, identifying potential risks And generate an early warning signal.
It improves the safety management level of the construction site, reduces the probability of fire and explosion accidents, ensures the response speed and disposal efficiency of emergency situations, and realizes accurate risk assessment of fire operations and dynamic adjustment of safety areas.
Smart Images

Figure CN119067444A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of hot work monitoring, and in particular to an intelligent construction hot work safety monitoring method, system, equipment and medium. Background Art
[0002] In the construction industry, hot work has always been a critical and challenging part of safety management due to its characteristics of directly or indirectly producing open flames. Specifically, hot work covers a wide range of unconventional operations, such as electric welding, gas welding (including thermal cutting processes such as oxyacetylene cutting and plasma cutting), use of blowtorches, grinding wheels, hot melt operations, and other operations that may cause hot surfaces. While these operations provide necessary technical support for building structure connections, metal component processing, equipment installation, etc., they also place extremely high demands on safety monitoring of construction sites due to their potential fire and explosion risks.
[0003] At present, the construction environment is complex and changeable. Hot work is often carried out in special environments containing a large amount of flammable materials, confined spaces, and near existing buildings or pipelines, which undoubtedly increases the difficulty of safety monitoring. Common hot work safety monitoring is mostly based on experience or fixed standards to define safe areas, ignoring the in-depth analysis of environmental risks and dynamic adjustment mechanisms, which leads to the lack of accuracy and timeliness in the monitoring and management of personnel and key equipment related to hot work, reducing the response speed and disposal efficiency when emergencies occur, and the safety management level of construction sites needs to be improved. Summary of the invention
[0004] In order to improve the safety management level of construction sites, the present application provides an intelligent construction fire safety monitoring method, system, equipment and medium.
[0005] In the first aspect, the present application provides an intelligent construction fire safety monitoring method, which adopts the following technical solutions:
[0006] An intelligent construction fire safety monitoring method, the monitoring method comprising:
[0007] According to the current hot work task, the corresponding hot work element information is obtained; wherein the hot work element information includes the hot work location information, the hot work level, the hot work person information, the fire watcher information and the fire extinguishing equipment information;
[0008] Based on the first preset mapping table, determining the corresponding initial safety radius according to the hot work level;
[0009] Obtaining the fire hazard level and real-time environmental parameters of the hot work location;
[0010] Calculate the environmental risk index corresponding to the hot work location according to the fire hazard level and real-time environmental parameters;
[0011] Dynamically adjusting the initial safety radius according to the environmental risk index to obtain a hot work safety radius;
[0012] According to the hot work location information and the hot work safety radius, determine a safe working area centered on the hot work location and send a safety isolation instruction;
[0013] Monitor the real-time location information of the fire starter, fire watcher and fire extinguishing equipment;
[0014] According to the real-time location information, it is determined in real time whether the person starting the fire, the person watching the fire, and the fire-fighting equipment are all in the safe working area;
[0015] If so, the user of the hot work is allowed to use the hot work equipment;
[0016] If not, the power supply of the hot-fire equipment is cut off and an early warning signal is sent to the management terminal.
[0017] By adopting the above technical solutions, a highly intelligent and flexible hot work safety management solution is constructed by integrating technical means such as environmental risk assessment, dynamic adjustment of safety radius, real-time location monitoring and automatic authority control. By combining environmental risks and the specific circumstances of hot work, the safety radius is dynamically adjusted to facilitate the precise definition and implementation of safe working areas, effectively isolating the potential risks of hot work. At the same time, combined with the dynamically adjusted safe working areas, through the automated monitoring and early warning mechanism of personnel and key equipment related to hot work, the probability of fire and explosion accidents is effectively reduced, ensuring the response speed and handling efficiency in the event of an emergency, and improving the safety management level of the construction site.
[0018] Optionally, the step of calculating the environmental risk index corresponding to the hot work location according to the fire hazard level and the real-time environmental parameters includes:
[0019] Based on the second preset mapping table, determining the corresponding regional hazard index according to the fire hazard level;
[0020] Standardizing real-time environmental parameters, wherein the real-time environmental parameters include temperature, humidity and wind speed;
[0021] Assign preset weights according to environmental parameter categories, multiply and sum the standardized environmental parameter values with the corresponding preset weights to obtain an environmental parameter index;
[0022] The environmental risk index is obtained according to the sum of the regional hazard index and the environmental parameter index.
[0023] By adopting the above technical solutions, the quantification of fire hazard levels is combined with the real-time monitoring of environmental parameters to establish a comprehensive environmental risk assessment system. By mapping the fire hazard levels, standardizing environmental parameters, reasonably allocating preset weights and calculating the final environmental risk index, an accurate assessment of fire risks in specific areas is achieved.
[0024] Optionally, the step of dynamically adjusting the initial safety radius according to the environmental risk index to obtain a hot work safety radius includes:
[0025] Based on a third preset mapping table, determining an index interval according to the environmental risk index, and determining an environmental risk level corresponding to the environmental risk index according to the index interval;
[0026] Based on a fourth preset mapping table, determining a corresponding adjustment factor according to the environmental risk level;
[0027] The hot work safety radius is calculated based on the initial safety radius and the adjustment factor.
[0028] By adopting the above technical solution, the environmental risk index is classified into a specific risk level according to the measured risk index, and then the adjustment factor is determined. Finally, the safety radius adapted to the current environmental risk is calculated by multiplying the adjustment factor by the initial safety radius. This technical solution ensures the safety of hot work operations, reduces potential safety hazards, and has good scalability and flexibility, and can be adjusted and optimized according to different application scenarios and needs.
[0029] Optionally, after the step of dynamically adjusting the initial safety radius according to the environmental risk index to obtain the hot work safety radius, the method further includes:
[0030] Determining whether the environmental risk index exceeds a risk index threshold;
[0031] If yes, then obtain hot work operation related parameters according to the current hot work operation task;
[0032] According to the hot work related parameters, a virtual environment model is constructed, and initial conditions and boundary conditions of the virtual environment model are configured, wherein the initial conditions include the environmental state at the beginning of the hot work, and the boundary conditions include a safe operation area determined according to a hot work safety radius;
[0033] Based on the configured virtual environment model, simulating the hot work process; the simulated hot work process includes the stages of ignition and combustion spread;
[0034] Recording environmental status data at various time points during the simulated hot work process; wherein the environmental status data includes temperature and smoke concentration;
[0035] Performing abnormal threshold analysis based on the environmental status data to identify potential risk points and corresponding risk assessment levels within the safe operation area;
[0036] According to the risk assessment level corresponding to each potential risk point, a risk warning signal is generated and sent to the management terminal.
[0037] By adopting the above technical solutions and comprehensively using technical means such as environmental risk index analysis, virtual environment modeling and simulation, and abnormal threshold analysis, it is possible to effectively identify and evaluate the potential risks in the process of hot work, and generate corresponding early warning signals according to the risk level, which not only improves the safety of hot work, but also reduces the possibility of safety accidents caused by human errors. Through real-time monitoring and early warning mechanisms, the safety management level of the work site can be significantly improved to ensure the safety of life and property of personnel.
[0038] Optionally, after the step of granting the hot work person the right to use the hot work equipment, the following step is further included:
[0039] Acquire monitoring image data and real-time environmental data of the safe operation area;
[0040] Performing data preprocessing on the monitoring image data and real-time environmental data;
[0041] Performing target detection on the preprocessed monitoring image data, identifying the target objects in the safe operation area and marking the target frames; wherein the target objects include personnel and equipment;
[0042] Perform feature extraction based on the surveillance image data after the target frame annotation to obtain the visual features of the target object;
[0043] Extracting environmental parameter features based on preprocessed real-time environmental data;
[0044] Fusing the visual feature with the environmental parameter feature to obtain a multimodal feature vector;
[0045] Based on a machine learning algorithm, the multimodal feature vector is input into a pre-trained anomaly recognition model to obtain an anomaly recognition result;
[0046] An abnormal warning signal is sent to the management terminal according to the abnormal identification result.
[0047] By adopting the above technical solutions, after the hot work begins, a comprehensive safety monitoring system is built using advanced image processing technology and environmental monitoring methods, which can not only accurately identify the status of personnel and equipment in the hot work area, but also timely discover potential risks in environmental changes. Through the application of multimodal feature fusion and anomaly recognition models, effective early warning of safety accidents is achieved, which greatly improves the safety management level of the work site, reduces the probability of accidents, and protects the life and property safety of workers.
[0048] Optionally, the safety monitoring method further includes a training step of the abnormality recognition model, and the training step includes:
[0049] Acquire a historical sample data set and perform preprocessing; wherein the historical sample data set includes historical sample image data of normal or abnormal behavior, historical environmental data, and pre-annotated sample labels;
[0050] Perform feature extraction and feature fusion on the preprocessed historical sample data set to obtain a multimodal sample feature vector set;
[0051] Dividing the multimodal sample feature vector set into a training set, a validation set, and a test set;
[0052] Inputting the training set into a pre-built machine learning model for initial training;
[0053] The machine learning model after initial training is verified and model parameters are optimized based on the verification set and the test set to obtain the trained anomaly recognition model.
[0054] By adopting the above technical solution, an anomaly recognition model is constructed to realize the automatic detection of abnormal behaviors at the hot work site. By combining multimodal sample features for training, it is ensured that the model can distinguish normal and abnormal patterns from the multimodal sample feature vectors, and then identify violations and abnormal environmental changes in actual monitoring, so as to quickly identify potential safety hazards and improve the safety management level of hot work.
[0055] Optionally, the training set is input into a pre-built machine learning model for initial training, and the steps of validating the machine learning model after initial training and optimizing model parameters based on the validation set and the test set include:
[0056] Inputting the training set into a pre-built machine learning model for iterative training, optimizing the model parameters and calculating the loss function of the model until the loss function meets the preset conditions, thereby obtaining a trained anomaly recognition model;
[0057] Validating the anomaly recognition model based on the validation set, evaluating the performance of the anomaly recognition model and adjusting the hyperparameters of the model according to the validation results;
[0058] The ability of the adjusted anomaly recognition model is tested based on the test set.
[0059] By adopting the above technical solution, the training set is used for iterative training and the model parameters are optimized to ensure that the model can learn the patterns in the data. The validation set is used to adjust the model hyperparameters to improve the generalization ability and prediction accuracy of the model. Finally, the model is comprehensively evaluated through the test set to ensure the effectiveness of the model in practical applications. The overall technical solution of this application can not only accurately identify potential safety hazards, but also send abnormal warning signals in a timely manner, thereby improving the safety management level of hot work operations and reducing the occurrence of accidents.
[0060] In the second aspect, the present application provides an intelligent construction fire safety monitoring system, which adopts the following technical solutions:
[0061] An intelligent construction fire safety monitoring system, the monitoring system comprising:
[0062] The first acquisition module is used to acquire corresponding hot work element information according to the current hot work task; wherein the hot work element information includes hot work location information, hot work level, hot work person information, fire watcher information and fire extinguishing equipment information;
[0063] An initial safety radius determination module, configured to determine a corresponding initial safety radius according to the hot work level based on a first preset mapping table;
[0064] A second acquisition module is used to obtain the fire hazard level and real-time environmental parameters of the hot work location;
[0065] An environmental risk index calculation module, used to calculate the environmental risk index corresponding to the hot work location according to the fire hazard level and real-time environmental parameters;
[0066] A dynamic adjustment module, used to dynamically adjust the initial safety radius according to the environmental risk index to obtain a hot work safety radius;
[0067] A safe operation area determination module, used to determine a safe operation area centered on the hot work position according to the hot work position information and the hot work safety radius;
[0068] A safety isolation module, used for sending a safety isolation instruction according to the safe operation area;
[0069] A location monitoring module, used to monitor the real-time location information of the fire starter, fire watcher and fire extinguishing equipment;
[0070] A judgment module, used to judge in real time whether the person starting the fire, the person watching the fire and the fire extinguishing equipment are all in the safe operation area according to the real-time position information; if so, output a first judgment result; if not, output a second judgment result;
[0071] A hot work equipment authority management module, used for granting the hot work person the authority to use the hot work equipment in response to the first judgment result;
[0072] A power-off warning module is used to, in response to the second judgment result, perform power-off control on the hot work equipment and send a warning signal to the management terminal.
[0073] In a third aspect, the present application provides a computer device, which adopts the following technical solution:
[0074] A computer device comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method as described in the first aspect.
[0075] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:
[0076] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any one of the methods in the first aspect.
[0077] To summarize, the present application includes at least one of the following beneficial technical effects: it integrates technical means such as environmental risk assessment, dynamic adjustment of safety radius, real-time location monitoring and automatic authority control to build a highly intelligent and flexible fire operation safety management plan; by combining environmental risks and the specific circumstances of fire operations, the safety radius is dynamically adjusted to facilitate the accurate definition and implementation of safe operating areas, thereby effectively isolating the potential risks of fire operations; at the same time, combined with the dynamically adjusted safe operating areas, through the automated monitoring and early warning mechanism of personnel and key equipment related to fire operations, the probability of fire and explosion accidents is effectively reduced, ensuring the response speed and disposal efficiency in emergencies, and improving the safety management level of construction sites. BRIEF DESCRIPTION OF THE DRAWINGS
[0078] Figure 1 It is a first flow chart of an intelligent construction fire safety monitoring method according to one of the embodiments of the present application.
[0079] Figure 2 It is a second flow chart of the intelligent construction fire safety monitoring method of one of the embodiments of the present application.
[0080] Figure 3 It is a third flow chart of the intelligent construction fire safety monitoring method of one of the embodiments of the present application.
[0081] Figure 4 It is a fourth flow chart of the intelligent construction fire safety monitoring method of one of the embodiments of the present application.
[0082] Figure 5 It is the fifth flow chart of the intelligent construction fire safety monitoring method of one of the embodiments of the present application.
[0083] Figure 6 It is the sixth flow chart of the intelligent construction fire safety monitoring method of one of the embodiments of the present application.
[0084] Figure 7 It is the seventh flow chart of the intelligent construction fire safety monitoring method of one of the embodiments of the present application. DETAILED DESCRIPTION
[0085] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-7 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0086] First, the terms related to this embodiment are explained.
[0087] Hot work refers to temporary work activities such as welding, cutting, grinding, heating, baking, burning, using open flames or generating sparks in the production or construction area. Such work is considered high-risk because it may directly or indirectly generate flames, sparks, hot surfaces and other fire sources, which may easily cause fire or explosion accidents. Hot work usually needs to be carried out under specific conditions, such as ensuring that there are no flammable and explosive substances in the surrounding environment, taking appropriate protective measures, and going through a strict approval process.
[0088] A hot work person is a person who directly participates in hot work, that is, a worker who actually performs welding, cutting, using a blowtorch, or other operations that generate flames or high temperatures. He is the person directly responsible for the hot work and must have the corresponding professional skills and safety knowledge, and understand the risks and preventive measures of the operation. A hot work person must hold a corresponding professional qualification certificate and receive safety training before the operation to ensure that the operation is carried out in accordance with the operating procedures and safety requirements.
[0089] Fire watchers, also known as fire guardians, are specialized personnel responsible for safety supervision and emergency preparedness at the site of hot work. Their main responsibilities include but are not limited to: checking the safety conditions of the site before and after the hot work, confirming that there are no fire hazards in and around the hot work area; supervising whether the hot work complies with safety regulations; continuously observing the work site during the hot work, and taking immediate measures to deal with the initial fire or evacuate personnel in case of abnormal conditions; being equipped with necessary firefighting equipment to ensure rapid response in emergency situations. Fire watchers also need to receive professional fire safety training, be familiar with emergency plans, and be able to effectively direct on-site personnel to rescue themselves and each other in emergency situations.
[0090] The embodiment of the present application discloses an intelligent construction fire safety monitoring method.
[0091] Reference Figure 1 , an intelligent construction fire safety monitoring method, the monitoring method comprises:
[0092] Step S101, obtaining corresponding hot work element information according to the current hot work task;
[0093] Among them, the hot work element information includes the hot work location information, hot work level, hot work person information, fire watcher information and fire extinguishing equipment information;
[0094] Specifically, the system collects all key information based on the upcoming hot work task, including the exact location of the hot work, the level of the hot work (such as level one hot work, level two hot work, etc.), the identity information of the hot work personnel, the identity information of the fire watch personnel and the specific information of the required fire-fighting equipment, to ensure sufficient preparation before the operation and provide accurate input data for subsequent risk assessment and formulation of safety measures.
[0095] In some embodiments, the level of hot work, that is, the danger level of hot work, can be determined based on the specific content of the hot work, including the tools used (such as welding torches, cutting machines, etc.), the work object (such as whether it is flammable materials or pipes containing flammable gases), the intensity of sparks or heat energy that may be generated during the operation, and the duration of the operation, etc., which are all key factors in determining the level of hot work.
[0096] Step S102, based on the first preset mapping table, according to the level of the hot work, determine the corresponding initial safety radius;
[0097] The first preset mapping table includes initial safety radiuses corresponding to multiple different hot work levels;
[0098] It is understandable that the higher the level of hot work, the higher the danger level of the hot work, and the larger the corresponding initial safety radius, so as to set up a larger safe working area and reduce the risk of fire spreading to other areas, thereby effectively reducing the probability of accidents and reducing the degree of damage to the surrounding environment. At the same time, it also provides more time and space for on-site personnel to identify dangers and evacuate in an orderly manner.
[0099] Step S103, obtaining the fire hazard level and real-time environmental parameters of the hot work location;
[0100] Among them, the fire hazard level of the location of the hot work operation can be predetermined according to the type, quantity, distribution, etc. of the combustible materials at the location; for example, the hot work performed near the flammable liquid storage tank area will be regarded as a high-hazard level due to the extremely high potential fire risk; in one of the embodiments of the present application, it can also be divided into Class A, B, and C fire hazard areas according to common fire hazard classification standards, among which Class A fire hazard level is the highest and Class C fire hazard level is the lowest.
[0101] In some embodiments, real-time environmental parameters (such as wind speed, temperature, humidity, etc.) can be collected through a field sensor network, and these data reflect the immediate safety status of the working environment.
[0102] It is understandable that real-time environmental parameters provide real-time data support for risk assessment, enabling safety measures to dynamically adapt to on-site changes and improving the timeliness and accuracy of safety monitoring.
[0103] Step S104, calculating the environmental risk index corresponding to the hot work location according to the fire hazard level and the real-time environmental parameters;
[0104] Among them, algorithms can be used to conduct a comprehensive assessment of the collected fire hazard levels and environmental parameters and convert them into a quantitative environmental risk index, which reflects the potential fire hazard level of operations under current environmental conditions. By quantifying risks, safety decisions are made more objective and scientific, which facilitates system automation processing and early warning triggering.
[0105] Step S105, dynamically adjusting the initial safety radius according to the environmental risk index to obtain a hot work safety radius;
[0106] Among them, the initial safety radius value is dynamically adjusted according to the environmental risk index. The larger the environmental risk index, the higher the degree of danger and the larger the adjustment factor to ensure that the safety distance matches the actual risk level.
[0107] It is understandable that dynamic adjustments based on real-time environmental risks enable personalized customization of safety distances, which neither overly restricts operations nor ignores actual risks, thereby improving the flexibility and safety of operations.
[0108] Step S106, determining a safe operation area centered on the hot work location according to the hot work location information and the hot work safety radius, and sending a safety isolation instruction;
[0109] Among them, the safe operating area of the hot work can be determined by using the location of the hot work and the dynamically adjusted safety radius. This area revolves around the center of the hot work point, and its radius is dynamically adjusted according to the environmental risk index and environmental parameters, thereby providing a clear boundary for the positioning and monitoring of personnel and equipment, and at the same time defining clear and reasonable safety boundaries for the hot work area, which is convenient for management and implementation.
[0110] Specifically, safety isolation instructions include not only physical isolation of hot work areas (such as setting up warning tapes, fences, etc.), but also restrictions on activities in and around the work area, such as restricting the entry and exit of unrelated personnel and vehicles, to ensure the purity and safety of the hot work area.
[0111] In some embodiments, the safety isolation instruction can be sent to on-site managers, safety supervisors, and automated control systems via wireless communication technology (such as Wi-Fi, Bluetooth, or a dedicated communication network). After receiving the instruction, the manager organizes personnel to implement physical isolation measures; at the same time, automated systems such as access control systems and surveillance cameras will also adjust permissions according to the instruction to ensure that only authorized personnel and equipment can enter the safe operation area.
[0112] Step S107, monitoring the real-time location information of the person starting the fire, the person watching the fire, and the fire extinguishing equipment;
[0113] In some embodiments, the location information of the person who starts the fire, the person who watches the fire, and the fire-fighting equipment can be tracked in real time through PFID electronic tag technology, UWB technology, ZigBee technology, Bluetooth technology, Wifi probe technology, etc.; for example, for the monitoring of the person who starts the fire and the person who watches the fire, an active or passive electronic monitoring module (PFID electronic tag, UWB module, ZigBee module, Bluetooth module, etc.) can be used, and by integrating it on wearable items (such as safety helmets, smart watches), the personnel location monitoring information can be fed back in real time; for the fire-fighting equipment, an active or passive electronic monitoring module (PFID electronic tag, UWB module, ZigBee module, Bluetooth module, etc.) can also be used, and by attaching it to the surface of the fire-fighting equipment bottle, the equipment location monitoring information can be fed back in real time; by monitoring the real-time location information of the three, strong control over various safety factors in the fire-starting area is achieved.
[0114] Step S108, judging in real time whether the person starting the fire, the person watching the fire, and the fire extinguishing equipment are all in the safe working area according to the real-time location information; if so, jump to step S109; if not, jump to step S110;
[0115] Specifically, the real-time location information obtained from monitoring will be compared with the safe working area to determine whether the person who starts the fire is on duty, whether the person who watches the fire is on duty, and whether the fire-fighting equipment is in place. Through real-time monitoring of the work site, violations can be discovered and corrected immediately to prevent safety accidents.
[0116] Step S109, granting the hot work person the right to use the hot work equipment;
[0117] Among them, according to the position monitoring results of personnel and equipment, the safety factors of hot work, namely the person who starts the fire, the person who watches the fire, and the fire extinguisher, are taken as monitoring targets. When the person who starts the fire, the person who watches the fire, and the fire extinguisher are all within the specified safe operation area, the current scene can be determined to be in a safe state, and the person who starts the fire can be controlled to have access to the hot work equipment, allowing the hot work to be carried out;
[0118] Step S110, power off the hot work equipment and send an early warning signal to the management terminal.
[0119] When it is identified that any of the safety conditions are not met, such as the person who started the fire is not on duty, the person watching the fire is not on duty, or the fire-fighting equipment is not in place, the current scene is determined to be unsafe. At this time, the power supply to the fire-starting equipment will be cut off, and warnings and reminders will be issued. When it is re-identified that all safety factors are within the safe range, the fire-starting operation authority will be restored;
[0120] It is understandable that by ensuring that hot work is carried out under absolutely safe conditions, the risk of fire caused by improper configuration of personnel or equipment is effectively prevented.
[0121] In the above implementation, technical means such as environmental risk assessment, dynamic adjustment of safety radius, real-time location monitoring and automatic authority control are integrated to construct a highly intelligent and flexible hot work safety management plan; by combining environmental risks and the specific circumstances of hot work, the safety radius is dynamically adjusted to facilitate the accurate definition and implementation of the safe working area, effectively isolating the potential risks of hot work; at the same time, combined with the dynamically adjusted safe working area, through the automated monitoring and early warning mechanism of personnel and key equipment related to hot work, the probability of fire and explosion accidents is effectively reduced, ensuring the response speed and disposal efficiency in the event of an emergency, and improving the safety management level of the construction site.
[0122] Reference Figure 2As an implementation method of step S104, the step of calculating the environmental risk index corresponding to the hot work location according to the fire hazard level and the real-time environmental parameters includes:
[0123] Step S201, based on the second preset mapping table, determine the corresponding regional hazard index according to the fire hazard level;
[0124] The second preset mapping table includes a plurality of regional hazard indexes corresponding to different fire hazard levels;
[0125] It can be understood that the higher the fire hazard level, the higher the danger level of the area, and the greater the corresponding regional hazard index; for example, for the specific standards of Class A, B, and C fire hazard areas, Class A represents extremely high hazard (the regional hazard index can be assigned a value of 3), Class B represents high hazard (the regional hazard index can be assigned a value of 2), and Class C represents general hazard (the regional hazard index can be assigned a value of 1).
[0126] Step S202, standardizing the real-time environmental parameters;
[0127] Among them, the real-time environmental parameters include temperature, humidity and wind speed; specifically, the original parameter data is standardized to normalize it to the [0, 1] interval to facilitate subsequent calculations.
[0128] Step S203, assigning preset weights according to environmental parameter categories, multiplying and summing the standardized environmental parameter values with the corresponding preset weights to obtain an environmental parameter index;
[0129] Specifically, the standardized environmental parameter values are multiplied by their corresponding weights, and then all the results are added together to obtain the environmental parameter index. In one embodiment of the present application, the environmental parameter index EPI = (T × W T )+(H×W H )+(V×W V ), where T is the temperature parameter value, H is the humidity parameter value, V is the wind speed parameter value, and W T , W H , W V They are the preset weights for temperature, humidity and wind speed respectively.
[0130] The preset weights reflect the relative importance of different environmental parameter categories to the possibility of fire. For example, under high temperature conditions, substances are more likely to ignite, so the weight of temperature is higher and can be set to 0.4; while humidity and wind speed affect the speed and spread of combustion respectively, so they can be given appropriate weights, both set to 0.3.
[0131] Step S204, obtaining an environmental risk index according to the sum of the regional hazard index and the environmental parameter index.
[0132] Among them, the environmental risk index combines fixed risk factors (such as regional hazard index) and dynamic risk factors (such as environmental parameter index) to provide a comprehensive risk assessment indicator.
[0133] In the above implementation mode, the quantification of fire hazard levels is combined with the real-time monitoring of environmental parameters to establish a comprehensive environmental risk assessment system; through the mapping of fire hazard levels, standardization of environmental parameters, reasonable allocation of preset weights and calculation of the final environmental risk index, an accurate assessment of fire risks in specific areas is achieved.
[0134] Reference Figure 3 As an implementation method of step S105, the step of dynamically adjusting the initial safety radius according to the environmental risk index to obtain the hot work safety radius includes:
[0135] Step S301, based on a third preset mapping table, determining an index interval according to the environmental risk index, and determining an environmental risk level corresponding to the environmental risk index according to the index interval;
[0136] Among them, the third preset mapping table includes a plurality of groups of correspondences between index intervals and environmental risk levels.
[0137] In some embodiments, the environmental risk level may be configured as low risk, medium risk, and high risk, respectively. The larger the environmental risk index, the higher the corresponding environmental risk level.
[0138] Step S302, based on a fourth preset mapping table, determining a corresponding adjustment factor according to the environmental risk level;
[0139] Among them, the fourth preset mapping table includes a plurality of sets of correspondences between environmental risk levels and adjustment factors.
[0140] Specifically, the higher the environmental risk level, the higher the degree of danger, and the larger the adjustment factor, to ensure that the safety distance matches the actual risk level. For example, in the fourth preset mapping table, the adjustment factor corresponding to low risk can be 1.0, the adjustment factor corresponding to medium risk can be 1.2, and the adjustment factor corresponding to high risk can be 1.5.
[0141] Step S303, calculating the hot work safety radius according to the initial safety radius and the adjustment factor.
[0142] Among them, the final hot work safety radius can be obtained by multiplying the initial safety radius with the adjustment factor; the safety radius is dynamically updated according to the current environmental risk index to ensure that the size of the hot work safety area matches the actual environmental risk and reduce the possibility of accidents.
[0143] For example, if the initial safety radius R0 is 20 meters and the adjustment factor f is 1.2, then the safety radius for hot work is R=R0×f=24 meters.
[0144] In the above implementation, the environmental risk index is measured and classified into a specific risk level, and then the adjustment factor is determined, and finally the safety radius adapted to the current environmental risk is calculated by multiplying the adjustment factor by the initial safety radius. This technical solution ensures the safety of hot work operations, reduces potential safety hazards, and has good scalability and flexibility, and can be adjusted and optimized according to different application scenarios and needs.
[0145] Reference Figure 4 As a further implementation method of the safety monitoring method, after dynamically adjusting the initial safety radius according to the environmental risk index to obtain the hot work safety radius, it also includes:
[0146] Step S401, determine whether the environmental risk index exceeds the risk index threshold; if so, jump to step S402; if not, do not perform any operation;
[0147] Among them, a predefined risk index threshold is set to determine whether the current environmental risk index has reached the level where additional safety measures need to be taken; if the environmental risk index exceeds this threshold, it means that there is a high fire risk under the current environmental conditions and further simulation evaluation is required.
[0148] Step S402, obtaining hot work operation related parameters according to the current hot work operation task;
[0149] Among them, the parameters related to hot work may include the type of hot work (welding, cutting, etc.), the type of fuel used, the specific coordinates of the work site, and the weather forecast data during the hot work (wind speed, temperature and humidity, etc.).
[0150] Step S403, constructing a virtual environment model according to hot work related parameters, and configuring initial conditions and boundary conditions of the virtual environment model;
[0151] The initial conditions include the environmental conditions at the beginning of the hot work, and the boundary conditions include the safe working area determined according to the hot work safety radius;
[0152] In one of the embodiments of the present application, based on the collected parameters related to hot work, a physical engine and environmental simulation software can be used to create a virtual environment model, which can reflect the characteristics of the actual working environment; the initial conditions of the model include the environmental status at the beginning of the hot work, which can be obtained based on the weather forecast data in the parameters related to the hot work, and the boundary conditions define the scope of the safe working area, which can be based on the safe working area determined by the safety radius of the hot work.
[0153] For example, assuming that the hot work will be carried out in an open space, the initial conditions may be the average temperature of the day at 20°C, relative humidity at 50%, and wind speed at 5 m / s; the boundary condition is a circular area with a safety radius of 20 meters centered on the location of the hot work; by establishing an accurate virtual environment model, the actual hot work process can be accurately simulated, thereby improving the accuracy of risk assessment.
[0154] Step S404, simulating the hot work process based on the configured virtual environment model;
[0155] The simulated hot work process includes the stages of ignition and combustion spread;
[0156] Specifically, in a configured virtual environment model, physical simulation software is used to simulate the hot work process, for example, the entire process from ignition to the flame spreading to the surrounding environment, to help identify potential safety hazards, such as the flame spreading too fast and the smoke spreading over a large range.
[0157] Step S405, recording the environmental status data at each time point during the simulated hot work process;
[0158] Among them, the environmental status data include temperature and smoke concentration;
[0159] Specifically, during the simulation of hot work, record the environmental status data at key time points, such as temperature, smoke concentration, etc., as a basis for subsequent risk assessment and to provide a basis for taking corresponding safety measures. For example, the temperature change and smoke concentration can be recorded every 10 seconds until the simulation ends.
[0160] Step S406, performing abnormal threshold analysis based on the environmental status data to identify potential risk points and corresponding risk assessment levels within the safe operation area;
[0161] Specifically, by analyzing the recorded environmental status data, for example, identifying which areas have abnormal conditions based on temperature and smoke concentration, these abnormal conditions may represent potential risk points and grading them according to their severity, it helps to quickly locate safety hazards and take timely measures to prevent accidents.
[0162] For example, if the temperature at a certain location within the safe working area rises rapidly to exceed a preset threshold in a short period of time, or the smoke concentration exceeds the normal range, then the location is considered to be a potential risk point, and the corresponding risk assessment level can be determined based on the degree to which the temperature exceeds the normal threshold or the smoke concentration exceeds the normal range.
[0163] Step S407: Generate a risk warning signal according to the risk assessment level corresponding to each potential risk point and send it to the management terminal.
[0164] Among them, according to the different risk assessment levels, corresponding risk warning signals are generated and sent to managers through the network or other communication means so that they can respond in time.
[0165] For example, if the risk assessment level is "high", a red warning signal is generated and immediately sent to the on-site manager's mobile phone or monitoring center to notify relevant personnel in real time to take measures in advance to reduce risks, such as adjusting operating hours, increasing protective measures, etc.
[0166] In the above implementation, the comprehensive use of environmental risk index analysis, virtual environment modeling and simulation, abnormal threshold analysis and other technical means can effectively identify and evaluate the potential risks in the process of hot work, and generate corresponding early warning signals according to the risk level, which not only improves the safety of hot work, but also reduces the possibility of safety accidents caused by human errors. Through real-time monitoring and early warning mechanisms, the safety management level of the work site can be significantly improved to ensure the safety of life and property of personnel.
[0167] It is understandable that when it is identified that the environmental risk index exceeds the risk index threshold, it is very necessary to build a virtual environment model to simulate the hot work process in advance. By building an accurate virtual environment model, computer simulation technology can be used to reproduce the actual conditions of hot work, including but not limited to factors such as temperature, humidity, wind speed, and chemical reactions and thermodynamic processes involved in the operation. This simulation can predict the possible consequences of hot work under specific conditions, such as the speed of fire spread, the direction of smoke diffusion, etc., so as to determine the possible risk points in the safe operation area in advance.
[0168] In actual application, preventive measures can be taken in this way to reduce the risk of accidents and protect the lives and property of personnel. For example, the simulation results can guide on-site staff to take more effective preventive measures, such as adjusting the operation position, adding fire isolation belts, and preparing fire-fighting equipment. In addition, the risk assessment level obtained by simulation can also be used to optimize emergency plans to ensure that they can respond quickly and effectively in the event of an accident, thereby improving the safety and reliability of hot work.
[0169] Reference Figure 5 As a further implementation method of the safety monitoring method, after the step of granting the hot work person the right to use the hot work equipment, the method further includes:
[0170] Step S501, obtaining monitoring image data and real-time environmental data of a safe operation area;
[0171] Specifically, multiple high-definition cameras installed in the hot work area can be used to capture monitoring image data of the on-site conditions, and various sensors (such as temperature sensors, humidity sensors, wind speed sensors, etc.) can be used to collect real-time environmental data to fully understand the conditions at the hot work site and provide a basis for subsequent data analysis.
[0172] Step S502, performing data preprocessing on the monitoring image data and the real-time environment data;
[0173] In some embodiments, the preprocessing step may include: video stream decoding, decoding the original video stream into frame image data; image enhancement, adjusting the brightness and contrast of each frame image to improve image quality; noise filtering, using filtering technology to remove noise interference in environmental data, such as using a low-pass filter to smooth temperature data to reduce fluctuations.
[0174] Step S503, performing target detection on the pre-processed monitoring image data, identifying the target object in the safe operation area and marking the target frame;
[0175] Among them, the target objects include personnel and equipment;
[0176] In one embodiment of the present application, the YOLOv5 model can be used to classify and locate people and equipment in the image, draw a rectangular box around the detected target, and mark the category, and accurately identify important target objects in the work area to provide a basis for subsequent feature extraction and risk assessment. For example, the identified person is marked as "person", the equipment is marked as "equipment", and a bounding box is drawn around them.
[0177] Step S504, performing feature extraction based on the surveillance image data after the target frame annotation to obtain the visual features of the target object;
[0178] In some embodiments, ResNet-50 can be used to extract feature vectors of people and equipment in the image, and the most representative features can be selected for retention. For example, the first 100 principal component features can be retained through the PCA dimensionality reduction method. The extracted feature vectors can effectively express the visual information of the target object, which is convenient for subsequent fusion analysis.
[0179] Step S505, extracting environmental parameter features according to the preprocessed real-time environmental data;
[0180] The extracted environmental parameter features are used to reflect the changes in the working environment, so as to judge potential risks. Specifically, time series analysis can be used to analyze the time series pattern of environmental data and extract trend features, such as the rate of change of temperature and gas concentration; at the same time, statistical feature extraction can be used to calculate statistical features such as the mean and variance of environmental data.
[0181] For example, time series analysis includes, for example, performing sliding window averaging on temperature data to extract temperature variation trends; statistical feature extraction includes, for example, calculating the mean and variance of temperature data over the past 10 minutes.
[0182] Step S506, fusing the visual features with the environmental parameter features to obtain a multimodal feature vector;
[0183] Among them, the multimodal feature vector contains both visual and environmental information, which more comprehensively represents the status of the hot work site;
[0184] Step S507, based on a machine learning algorithm, input the multimodal feature vector into a pre-trained anomaly recognition model to obtain an anomaly recognition result;
[0185] Specifically, supervised learning methods can be used to pre-train anomaly recognition models. The model can use historical data to learn normal and abnormal behavior patterns, thereby automatically identifying potential abnormal behaviors or environmental changes.
[0186] Abnormal behaviors include workers not wearing protective equipment correctly, workers violating regulations, missing protective / fire-fighting equipment, abnormal environmental changes, etc.
[0187] Step S508: Send an abnormal warning signal to the management terminal according to the abnormal identification result.
[0188] Among them, when the model predicts an abnormality, the abnormal warning mechanism is immediately triggered. After receiving the abnormal warning signal, the management terminal takes corresponding measures and responds quickly to reduce safety accidents caused by human negligence.
[0189] In the above implementation, after the hot work begins, a comprehensive safety monitoring system is constructed using advanced image processing technology and environmental monitoring methods, which can not only accurately identify the status of personnel and equipment in the hot work area, but also promptly discover potential risks in environmental changes. Through the application of multimodal feature fusion and anomaly recognition models, effective early warning of safety accidents is achieved, which greatly improves the safety management level of the work site, reduces the probability of accidents, and protects the life and property safety of workers.
[0190] Reference Figure 6As an implementation of the anomaly recognition model, the training steps of the anomaly recognition model include:
[0191] Step S601, obtaining a historical sample data set and performing preprocessing;
[0192] The historical sample data set includes historical sample image data of normal or abnormal behavior, historical environmental data, and pre-labeled sample labels;
[0193] Specifically, each sample data in the data set has a pre-labeled label indicating whether it is normal or abnormal; for example, images can be captured from past hot work monitoring videos, which may include images of personnel wearing protective equipment correctly, not wearing protective equipment, and images of workers violating regulations, lacking protective / fire-fighting equipment, and fully equipped protective / fire-fighting equipment. In addition, historical environmental data such as temperature, humidity, gas concentration, etc. can also be collected and labeled as normal or abnormal.
[0194] Step S602, performing feature extraction and feature fusion on the preprocessed historical sample data set to obtain a multimodal sample feature vector set;
[0195] In one of the embodiments of the present application, ResNet-50 can be used to extract visual sample features from historical sample image data, while time series analysis methods and statistical methods can be used to extract environmental parameter sample features from historical environmental data; by fusing the visual sample features and the environmental parameter sample features, a multimodal sample feature vector is formed.
[0196] Step S603, dividing the multimodal sample feature vector set into a training set, a validation set and a test set;
[0197] Exemplarily, the data set can be divided into a training set, a validation set, and a test set in a ratio of 70%, 15%, and 15% to ensure the generalization ability of the model and prevent overfitting.
[0198] Step S604, inputting the training set into a pre-built machine learning model for initial training;
[0199] In one of the embodiments of the present application, the pre-built machine learning model may adopt a random forest algorithm model, which can effectively process the multimodal feature data involved in the present application, and can achieve accurate identification and early warning of abnormal behaviors at the hot work site through multi-step operations in the model training and tuning process.
[0200] When building a random forest model, you need to pre-set model parameters, such as the number of trees, the number of feature selections, etc. For example, you can set the number of trees to 100 to ensure that the model has sufficient complexity to capture the patterns in the data, but also be careful to avoid overfitting; for the number of feature selections, you can set the number of features considered at each split to the square root of the number of all features, which helps reduce the variance of the model and improve the stability and accuracy of the prediction.
[0201] Step S605, verifying and optimizing the model parameters of the initially trained machine learning model based on the validation set and the test set to obtain a trained anomaly recognition model.
[0202] Specifically, the constructed anomaly recognition model includes the following parts: an input layer, which is used to receive multimodal feature vectors, which contain visual features extracted from image data and environmental parameter features extracted from environmental data; a random forest model composed of multiple decision trees, each of which independently predicts the input data. For classification tasks, the final prediction result is determined by majority voting, and for regression tasks, it is determined by the average of the prediction results of all decision trees; an output layer: based on the prediction results of the random forest model, outputs the anomaly recognition result, that is, the identified potential abnormal behavior or environmental changes.
[0203] Among them, the validation set can be used to evaluate the model performance and perform hyperparameter tuning. For example, a grid search or random search strategy can be used to adjust the hyperparameters in the random forest, such as the number of trees, the number of feature selections, etc., to improve the accuracy and stability of the model; and then the test set can be used to evaluate the generalization ability of the model, such as evaluating the model's accuracy, recall rate and other indicators to ensure the performance of the model on unknown data.
[0204] In the above implementation, an abnormality recognition model is constructed to realize the automatic detection of abnormal behaviors at the hot work site. By combining multimodal sample features for training, it is ensured that the model can distinguish normal and abnormal patterns from the multimodal sample feature vectors, and then identify violations and abnormal environmental changes in actual monitoring, so as to quickly identify potential safety hazards and improve the safety management level of hot work.
[0205] Reference Figure 7 As an implementation method of step S604 and step S605, the training set is input into a pre-built machine learning model for initial training, and the steps of verifying the machine learning model after initial training and optimizing model parameters based on the verification set and the test set include:
[0206] Step S701, input the training set into a pre-built machine learning model for iterative training, optimize the model parameters and calculate the loss function of the model until the loss function meets the preset conditions, thereby obtaining a trained anomaly recognition model;
[0207] Specifically, the random forest model can be iteratively trained using the training set. In each iteration, the model parameters are adjusted to optimize the model performance, and the model loss function is calculated to evaluate the model performance, such as the classification error rate or the cross entropy loss function. The training is iterative until the loss function reaches the preset condition, so that the model gradually learns the patterns in the data.
[0208] Among them, the preset condition is, for example, when the loss function changes very little (less than a certain threshold, such as 0.05) in several consecutive iterations, it is considered that the model has achieved the expected performance and the training is stopped.
[0209] It is understandable that by setting preset conditions, it is possible to ensure that the random forest model can learn the patterns in the data and avoid overfitting during the training process, which helps to improve the generalization ability of the model and ensure that the model can also perform well on unseen data. In practical applications, these preset conditions can be adjusted according to specific needs to balance the training time and performance of the model.
[0210] Step S702, verifying the anomaly recognition model based on the verification set, evaluating the performance of the anomaly recognition model and adjusting the hyperparameters of the model according to the verification results;
[0211] The validation set is used to evaluate the performance of the model, such as accuracy and recall, and the hyperparameters of the model are adjusted according to the performance on the validation set, such as the number of trees and the number of feature selections. The training and validation process is repeated until the best hyperparameter combination is found.
[0212] Step S703: testing the capability of the adjusted anomaly recognition model based on the test set.
[0213] Among them, an independent test set is used to perform a final evaluation on the adjusted model to measure the performance of the model on unseen data. The evaluation indicators may include accuracy, recall, F1 score, etc.
[0214] In the above implementation, the training set is used for iterative training to optimize the model parameters, ensuring that the model can learn the patterns in the data. The validation set is used to adjust the model hyperparameters to improve the generalization ability and prediction accuracy of the model. Finally, the model is comprehensively evaluated through the test set to ensure the effectiveness of the model in practical applications. The overall technical solution of this application can not only accurately identify potential safety hazards, but also send abnormal warning signals in a timely manner, thereby improving the safety management level of hot work operations and reducing the occurrence of accidents.
[0215] In the actual application process, the intelligent construction fire safety monitoring method of the embodiment of the present application has important practical significance. It can not only effectively prevent the occurrence of safety accidents such as fires, protect the life safety and property safety of personnel, but also help enterprises improve construction efficiency and reduce costs; through real-time monitoring and intelligent early warning, it can timely discover potential safety risks and take corresponding measures to avoid unnecessary shutdowns and delays. In addition, intelligent management can also reduce the workload of on-site supervisors and improve supervision efficiency. In summary, this application not only helps to improve the safety conditions of the construction environment, but also promotes the sustainable development of the construction industry, and has broad application prospects and promotion value.
[0216] The embodiment of the present application also discloses an intelligent construction fire safety monitoring system.
[0217] An intelligent construction fire safety monitoring system, the monitoring system comprising:
[0218] The first acquisition module is used to acquire corresponding hot work element information according to the current hot work task; wherein the hot work element information includes hot work location information, hot work level, hot work person information, fire watcher information and fire extinguishing equipment information;
[0219] An initial safety radius determination module, used to determine the corresponding initial safety radius according to the hot work level based on the first preset mapping table;
[0220] The second acquisition module is used to obtain the fire hazard level and real-time environmental parameters of the hot work location;
[0221] The environmental risk index calculation module is used to calculate the environmental risk index corresponding to the hot work location according to the fire hazard level and real-time environmental parameters;
[0222] A dynamic adjustment module is used to dynamically adjust the initial safety radius according to the environmental risk index to obtain the hot work safety radius;
[0223] A safe operation area determination module is used to determine a safe operation area centered on the hot work location according to the hot work location information and the hot work safety radius;
[0224] A safety isolation module is used to send safety isolation instructions according to the safe operation area;
[0225] Position monitoring module, used to monitor the real-time position information of the person starting the fire, the person watching the fire, and the fire-fighting equipment;
[0226] A judgment module is used to judge in real time whether the person starting the fire, the person watching the fire and the fire extinguishing equipment are all in a safe operation area according to the real-time location information; if so, a first judgment result is output; if not, a second judgment result is output;
[0227] A hot work equipment authority management module, used for granting the hot work person the authority to use the hot work equipment in response to the first judgment result;
[0228] The power-off warning module is used to control the power off of the hot equipment and send a warning signal to the management terminal in response to the second judgment result.
[0229] In the above implementation, by integrating multiple functional modules, safety management and risk control of the entire process of hot work are achieved; specifically, the initial safety radius is determined according to the level of the hot work, and the safety radius is dynamically adjusted in combination with the fire hazard level and real-time environmental parameters, ensuring that the safety distance of the hot work is reasonable and effective; by real-time monitoring of the location information of the person who starts the fire, the person who watches the fire, and the fire-fighting equipment, the system can ensure that all relevant personnel and equipment are in a safe area, thereby avoiding safety accidents caused by improper location. In addition, the system can also cut off the power supply to the hot work equipment that violates safety regulations and send early warning signals, further improving the safety management level of the construction site. Overall, the system can significantly improve the safety and efficiency of hot work and reduce safety hazards caused by human factors.
[0230] An intelligent construction fire safety monitoring system in an embodiment of the present application can implement any of the above-mentioned safety monitoring methods, and the specific working process of each module in the intelligent construction fire safety monitoring system can refer to the corresponding process in the above-mentioned method embodiment.
[0231] In the several embodiments provided in this application, it should be understood that the provided methods and systems can be implemented in other ways. For example, the system embodiments described above are only illustrative; for example, the division of a certain module is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.
[0232] The embodiment of the present application also discloses a computer device.
[0233] The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the intelligent construction fire safety monitoring method as described above is implemented.
[0234] The embodiment of the present application also discloses a computer-readable storage medium.
[0235] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any of the above-mentioned intelligent construction fire safety monitoring methods.
[0236] Among them, computer-readable storage media can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus or device; the program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.
[0237] It should be noted that in the above embodiments, the description of each embodiment has different emphases, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0238] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.
Claims
1. An intelligent construction fire safety monitoring method, characterized in that: The monitoring method comprises: According to the current hot work task, the corresponding hot work element information is obtained; wherein the hot work element information includes the hot work location information, the hot work level, the hot work person information, the fire watcher information and the fire extinguishing equipment information; Based on the first preset mapping table, determining the corresponding initial safety radius according to the hot work level; Obtaining the fire hazard level and real-time environmental parameters of the hot work location; Calculate the environmental risk index corresponding to the hot work location according to the fire hazard level and real-time environmental parameters; Dynamically adjusting the initial safety radius according to the environmental risk index to obtain a hot work safety radius; According to the hot work location information and the hot work safety radius, determine a safe working area centered on the hot work location and send a safety isolation instruction; Monitor the real-time location information of the fire starter, fire watcher and fire extinguishing equipment; According to the real-time location information, it is determined in real time whether the person starting the fire, the person watching the fire, and the fire-fighting equipment are all in the safe working area; If so, the user of the hot work is allowed to use the hot work equipment; If not, the power supply of the hot-fire equipment is cut off and an early warning signal is sent to the management terminal.
2. According to claim 1, a smart construction fire safety monitoring method is characterized in that: The step of calculating the environmental risk index corresponding to the hot work location according to the fire hazard level and the real-time environmental parameters comprises: Based on the second preset mapping table, determining the corresponding regional hazard index according to the fire hazard level; Standardizing real-time environmental parameters, wherein the real-time environmental parameters include temperature, humidity and wind speed; Assign preset weights according to environmental parameter categories, multiply and sum the standardized environmental parameter values with the corresponding preset weights to obtain an environmental parameter index; The environmental risk index is obtained according to the sum of the regional hazard index and the environmental parameter index.
3. The intelligent construction fire safety monitoring method according to claim 1 is characterized in that: The step of dynamically adjusting the initial safety radius according to the environmental risk index to obtain the hot work safety radius includes: Based on a third preset mapping table, determining an index interval according to the environmental risk index, and determining an environmental risk level corresponding to the environmental risk index according to the index interval; Based on a fourth preset mapping table, determining a corresponding adjustment factor according to the environmental risk level; The hot work safety radius is calculated based on the initial safety radius and the adjustment factor.
4. The intelligent construction fire safety monitoring method according to claim 1 is characterized in that: After dynamically adjusting the initial safety radius according to the environmental risk index to obtain the hot work safety radius, the method further includes: Determining whether the environmental risk index exceeds a risk index threshold; If yes, then obtain hot work operation related parameters according to the current hot work operation task; According to the hot work related parameters, a virtual environment model is constructed, and initial conditions and boundary conditions of the virtual environment model are configured; the initial conditions include the environmental state at the beginning of the hot work, and the boundary conditions include the safe operation area determined according to the hot work safety radius; Based on the configured virtual environment model, simulate the hot work process; the simulated hot work process includes the stages of ignition and combustion spread; Recording environmental status data at various time points during the simulated hot work process; wherein the environmental status data includes temperature and smoke concentration; Performing abnormal threshold analysis based on the environmental status data to identify potential risk points and corresponding risk assessment levels within the safe operation area; According to the risk assessment level corresponding to each potential risk point, a risk warning signal is generated and sent to the management terminal.
5. An intelligent construction fire safety monitoring method according to any one of claims 1 to 4, characterized in that: After the step of granting the hot work person the right to use the hot work equipment, the following steps are also included: Acquire monitoring image data and real-time environmental data of the safe operation area; Performing data preprocessing on the monitoring image data and real-time environmental data; Performing target detection on the preprocessed monitoring image data, identifying the target objects in the safe operation area and marking the target frames; wherein the target objects include personnel and equipment; Perform feature extraction based on the surveillance image data after the target frame annotation to obtain the visual features of the target object; Extracting environmental parameter features based on preprocessed real-time environmental data; Fusing the visual feature with the environmental parameter feature to obtain a multimodal feature vector; Based on a machine learning algorithm, the multimodal feature vector is input into a pre-trained anomaly recognition model to obtain an anomaly recognition result; An abnormal warning signal is sent to the management terminal according to the abnormal identification result.
6. The intelligent construction fire safety monitoring method according to claim 5 is characterized in that: The safety monitoring method further comprises a training step of the abnormality recognition model, wherein the training step comprises: Acquire a historical sample data set and perform preprocessing; wherein the historical sample data set includes historical sample image data of normal or abnormal behavior, historical environmental data, and pre-annotated sample labels; Perform feature extraction and feature fusion on the preprocessed historical sample data set to obtain a multimodal sample feature vector set; Dividing the multimodal sample feature vector set into a training set, a validation set, and a test set; Inputting the training set into a pre-built machine learning model for initial training; The machine learning model after initial training is verified and model parameters are optimized based on the verification set and the test set to obtain the trained anomaly recognition model.
7. The intelligent construction fire safety monitoring method according to claim 6 is characterized in that: The steps of inputting the training set into a pre-built machine learning model for initial training, and verifying and optimizing model parameters of the machine learning model after initial training based on the verification set and the test set include: Inputting the training set into a pre-built machine learning model for iterative training, optimizing the model parameters and calculating the loss function of the model until the loss function meets the preset conditions, thereby obtaining a trained anomaly recognition model; Validating the anomaly recognition model based on the validation set, evaluating the performance of the anomaly recognition model and adjusting the hyperparameters of the model according to the validation results; The ability of the adjusted anomaly recognition model is tested based on the test set.
8. An intelligent construction fire safety monitoring system, characterized in that: The monitoring system comprises: The first acquisition module is used to acquire corresponding hot work element information according to the current hot work task; wherein the hot work element information includes hot work location information, hot work level, hot work person information, fire watcher information and fire extinguishing equipment information; An initial safety radius determination module, configured to determine a corresponding initial safety radius according to the hot work level based on a first preset mapping table; A second acquisition module is used to obtain the fire hazard level and real-time environmental parameters of the hot work location; An environmental risk index calculation module, used to calculate the environmental risk index corresponding to the hot work location according to the fire hazard level and real-time environmental parameters; A dynamic adjustment module, used to dynamically adjust the initial safety radius according to the environmental risk index to obtain a hot work safety radius; A safe operation area determination module, used to determine a safe operation area centered on the hot work position according to the hot work position information and the hot work safety radius; A safety isolation module, used for sending a safety isolation instruction according to the safe operation area; A location monitoring module, used to monitor the real-time location information of the fire starter, fire watcher and fire extinguishing equipment; A judgment module, used to judge in real time whether the person starting the fire, the person watching the fire and the fire extinguishing equipment are all in the safe operation area according to the real-time position information; if so, output a first judgment result; if not, output a second judgment result; A hot work equipment authority management module, used for granting the hot work person the authority to use the hot work equipment in response to the first judgment result; A power-off warning module is used to, in response to the second judgment result, perform power-off control on the hot work equipment and send a warning signal to the management terminal.
9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the method according to any one of claims 1 to 7 is implemented when the processor executes the program.
10. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Construction site fire operation safety supervision method using tower crane video image detection
CN115861875A
Power transmission line forest fire early warning method and system
CN116543518A
Electric power intelligent anti-electric shock safety method and system based on deep learning
CN117611410A
Anti-collision method based on tower crane anti-collision early warning system
CN118419787A
Deep learning system for disaster detection and automatic fire suppression in the underground power facility
KR102553690B1
Cited By
Building construction safety monitoring system
CN120317605A
Safety protection method and device in operation scene
CN121143235A
Fire operation whole-process acousto-optic identification monitoring and early warning system
CN121236858A
Dynamic Bayesian network construction risk prediction method based on sliding window adaptive learning and particle filtering
CN121481228A
Method and system for preventing combustion caused by fire operation in gas tunnel
CN121556934A
