CN117495110A - Fire rescue risk assessment method, device, equipment and readable storage medium - Google Patents
Fire rescue risk assessment method, device, equipment and readable storage medium Download PDFInfo
- Publication number
- CN117495110A CN117495110A CN202311845234.1A CN202311845234A CN117495110A CN 117495110 A CN117495110 A CN 117495110A CN 202311845234 A CN202311845234 A CN 202311845234A CN 117495110 A CN117495110 A CN 117495110A
- Authority
- CN
- China
- Prior art keywords
- fire
- real
- time
- data
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000012502 risk assessment Methods 0.000 title claims abstract description 35
- 238000012544 monitoring process Methods 0.000 claims description 60
- 238000012545 processing Methods 0.000 claims description 28
- 238000004458 analytical method Methods 0.000 claims description 27
- 238000005516 engineering process Methods 0.000 claims description 25
- 239000000779 smoke Substances 0.000 claims description 24
- 238000010801 machine learning Methods 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 18
- 230000007480 spreading Effects 0.000 claims description 18
- 230000007246 mechanism Effects 0.000 claims description 17
- 238000004891 communication Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000013527 convolutional neural network Methods 0.000 claims description 10
- 230000008447 perception Effects 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 125000004122 cyclic group Chemical group 0.000 claims description 8
- 238000007405 data analysis Methods 0.000 claims description 8
- 238000013499 data model Methods 0.000 claims description 8
- 238000013210 evaluation model Methods 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 6
- 238000010223 real-time analysis Methods 0.000 claims description 6
- 230000010354 integration Effects 0.000 claims description 5
- 238000013480 data collection Methods 0.000 claims description 3
- 230000008713 feedback mechanism Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000012876 topography Methods 0.000 claims description 3
- 238000010200 validation analysis Methods 0.000 claims 1
- 230000009471 action Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000003111 delayed effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The invention discloses a fire rescue risk assessment method, a device, equipment and a readable storage medium, which relate to the technical field of fire safety.
Description
Technical Field
The invention relates to the technical field of fire safety, in particular to a fire rescue risk assessment method, a device, equipment and a readable storage medium.
Background
With frequent occurrence of various social rescue accidents in recent years, the fire rescue task is increasingly heavy. In particular to rescue actions in water areas, earthquakes and mountains, which have the defects of more potential safety hazards and large risk coefficients, accidents can be possibly caused by slightly paying attention to the actions, the completion of rescue tasks is seriously influenced, and the risk assessment work of executing the fire-fighting rescue tasks can effectively reduce the rescue risk and improve the rescue efficiency.
For example, chinese patent discloses a fire rescue risk assessment method, apparatus, device and readable storage medium, CN116993144a, comprising: the vehicle-road cooperative platform detects the parking time of the vehicle in the fire-fighting channel; the vehicle-road cooperation platform obtains estimated time, wherein the estimated time is the estimated time for a fire-fighting vehicle to reach a position corresponding to the fire-fighting channel under the condition that an accident occurs at the position; the vehicle-road cooperative platform determines the total arrival time of the fire rescue based on the parking time and the estimated time; and the vehicle-road cooperation platform evaluates the fire rescue risk of the position based on the total arrival time of the fire rescue.
Although the above scheme has the advantages, the traditional fire rescue risk assessment method relies on offline or periodically collected data, so that the latest information of a fire scene cannot be obtained in real time, and due to the fact that the development speed of the fire is high, the delayed data update causes the assessment result to not reflect the actual situation in time, the problem of poor real-time performance is faced, the assessment result is delayed, the dynamic change of the fire scene is difficult to respond in time, emergency decision and rescue actions are affected, in addition, the analysis on the fire spreading trend and smoke density is simpler, the dynamic change of the fire is difficult to accurately simulate, and a large deviation between the assessment result and the actual situation can be caused, so that a fire rescue risk assessment method, device, equipment and readable storage medium for monitoring high flexibility in real time are needed to solve the problem.
Disclosure of Invention
(one) solving the technical problems
Aiming at the defects of the prior art, the invention provides a fire rescue risk assessment method, a fire rescue risk assessment device, fire rescue risk assessment equipment and a readable storage medium, and solves the problem that offline or periodically acquired data cannot respond to dynamic changes of fire scenes in real time in the prior art.
(II) technical scheme
In order to achieve the above object, the present invention provides a fire rescue risk assessment method, including:
step 1, rescue scene information is obtained in real time, data are obtained through real-time monitoring, data integration is performed, a real-time data stream processing system is established, the data stream processing system comprises a sensor, monitoring equipment and a satellite data real-time obtaining unit, comprehensive perception of a fire scene is provided, the real-time data are processed through edge calculation, and processing analysis of large-scale data is performed based on cloud calculation;
step 2, real-time data analysis and model updating, wherein real-time intelligent analysis is adopted to process real-time data in real time, identify the fire spreading trend and smoke density, and adaptively adjust and evaluate parameters of the model according to new real-time data feedback;
step 3, calibrating and predicting a dynamic risk area, updating the boundary of the fire risk area through real-time data by utilizing a GIS technology, evaluating a model to reflect the dynamic change of the fire, carrying out dynamic prediction by adopting a machine learning algorithm, and predicting the path and the speed of fire spreading based on the real-time data;
step 4, monitoring the positions of the rescue workers and the communication system in real time, adopting a GPS positioning technology to monitor the positions of the rescue workers in real time, providing the latest position and fire information for rescue teams by combining the real-time dynamics of fire sources, and providing a real-time communication and command system for information transmission among the rescue organizations;
and 5, establishing an iterative model updating mechanism by using a model updating and flexibility mechanism, and allowing the evaluation model to be updated rapidly according to new data and scene experience.
The invention is further arranged to: in the step 1, the rescue scene information real-time acquisition method comprises the following steps:
sensor deployment and data collection, wherein the sensors comprise a temperature sensor, a smoke sensor and a gas sensor;
deploying a sensor in a fire risk area, and collecting data generated by the sensor in real time;
setting a monitoring camera and monitoring equipment to monitor a fire scene, and configuring the monitoring equipment to transmit image and video stream data in real time;
simultaneously acquiring satellite data including geographic and meteorological related information, and integrating the satellite data to provide comprehensive fire scene perception;
the invention is further arranged to: in the step 1, the method for acquiring the rescue scene information in real time further comprises the following steps:
a real-time monitoring system is deployed, and a fire scene is monitored in real time through a sensor and monitoring equipment;
establishing a real-time data stream processing system;
performing preliminary data processing and analysis by using edge calculation near the data acquisition point;
transmitting the data monitored in real time to a cloud end, and carrying out deep analysis on the sensor, the monitoring equipment and the satellite data by utilizing a large-scale data processing and analyzing technology at the cloud end;
the invention is further arranged to: in the step 2, the real-time intelligent analysis method comprises the following steps:
adopting a convolutional neural network CNN as a basic model of real-time analysis, deploying a real-time analysis module, and inputting image data to perform fire scene analysis;
by using the position of the fire sourceWind direction->Temperature->Parameters, establishing a fire spread model->:
Wherein f represents a fire spread function;
smoke density measured in combination with sensorImage data of a monitoring device->Establishing a smoke density model->;
Wherein g represents a smoke density recognition function;
a dynamic adjustment mechanism is established, model parameters are adjusted based on real-time data feedback, and the adjustment mode is as follows:
,
wherein the method comprises the steps ofRepresenting new model parameters->Then it is the old model parameter,/->Expressed as learning rate->Representing the gradient of the loss function versus the parameter;
adopting random gradient descent SGD, and improving real-time performance by utilizing a hardware accelerator;
periodically updating the model, establishing a feedback mechanism, and collecting an actual scene data adjustment model;
the invention is further arranged to: in the step 3, the dynamic risk area calibration and prediction method specifically comprises the following steps:
based on the real-time monitoring of the fire scene in the step 1, the acquired related data including wind direction, temperature and topography are utilized to draw a map of the fire area, and the boundary of the fire area is updated through the real-time data;
a machine learning model is built by adopting a cyclic neural network (RNN), historical data is input for training, and the possible path and speed of fire spread are predicted;
updating the boundary of the fire danger area according to the prediction result of the machine learning model;
real-time fire hazard early warning is carried out, and an alarm is sent out when a prediction result shows that the fire hazard possibly spreads to a new area;
the invention is further arranged to: in the step 3, the step of establishing a machine learning model includes:
collecting fire scene data monitored in real time, and setting the wind direction as W, the temperature as T and the terrain as Ter;
drawing a map of the fire disaster area by using a GIS technology, and updating the boundary of the fire disaster area through real-time data;
establishing a cyclic neural network (RNN) model to learn a possible path and a possible speed of fire spread;
the inputs of the RNN model include historically monitored fire scene data sequences, specifically:
wherein t represents current time data, t-1 and t-n represent the last time step observation data and the observation data of the first n time steps respectively;
dividing the data into a training set and a verification set;
the invention is further arranged to: in the step 3, the step of establishing a machine learning model further includes:
establishing a loss function L, performing model training by using historical data, optimizing model parameters, and minimizing the loss function, and specifically:
wherein->And->Respectively indicating the possible fire spread path and speed at time i, N indicating the number of samples, ++>And->Respectively representing the actual fire spreading path and the actual fire spreading speed in the historical data;
predicting fire scene data at the current moment by using the trained RNN model to obtain a possible path for fire spreadAnd speed->;
Updating the boundary of the fire risk area according to the path and the speed predicted by the RNN model;
establishing an update function prediction path and a speed map as a new boundary;
when the model prediction result shows that the fire possibly spreads to a new area, an alarm is sent out;
the invention also provides a fire rescue risk assessment device, which comprises:
the rescue scene information real-time acquisition module is used for monitoring and acquiring data in real time, integrating sensors, monitoring equipment and satellite data and providing comprehensive perception of fire scenes;
the monitoring deployment module is used for deploying temperature, smoke and gas sensors in a fire risk area, and the monitoring camera transmits image and video stream data in real time;
the real-time data stream processing system is used for carrying out data processing and analysis by adopting edge computing and cloud computing technologies;
the real-time data analysis and model updating module adopts real-time intelligent analysis to identify the fire spreading trend and the smoke density, and dynamically adjusts and evaluates model parameters;
the dynamic risk area calibration and prediction module is used for calibrating a dynamic fire risk area by using a GIS technology, and performing dynamic prediction by using a machine learning algorithm to provide real-time fire risk early warning;
the real-time monitoring personnel position and communication system module is used for monitoring the position of the rescue personnel in real time by utilizing a GPS positioning technology and providing real-time position and fire information by combining the dynamic information of the fire source;
the invention also provides a fire rescue risk assessment device comprising: a memory, a processor, and a program stored on the memory and executable on the processor, characterized in that:
the processor is used for reading the program in the memory to realize the steps in the method;
the present invention also provides a readable storage medium storing a program, wherein the program when executed by a processor implements the steps of the above method.
(III) beneficial effects
The invention provides a fire rescue risk assessment method, a fire rescue risk assessment device, fire rescue equipment and a readable storage medium. The beneficial effects are as follows:
according to the fire rescue risk assessment method, in the aspect of acquiring rescue scene information in real time, the comprehensive sensing is realized by adopting various sensors including the temperature sensor, the smoke sensor and the gas sensor, the fire scene is sensed more comprehensively through integration of monitoring equipment and satellite data, the boundary of a fire hazard area is updated in real time through a GIS technology, the data of the sensor and the monitoring equipment are integrated, preliminary processing is performed through edge calculation, large-scale data processing and analysis are performed through cloud computing, and a richer real-time data basis is provided for subsequent steps.
The method is characterized in that a convolutional neural network CNN is adopted as a basic model in the aspects of real-time data analysis and model updating, a model parameter dynamic adjustment mechanism is adopted, parameters of an evaluation model are adaptively adjusted through real-time data feedback, different fire scenes are adapted, a deep learning model and real-time data dynamic adjustment are combined, the real-time performance and accuracy of the fire scenes are improved, meanwhile, a cyclic neural network RNN is adopted as a machine learning model for predicting the possible path and speed of fire spreading, the boundary of a fire risk area is updated through the real-time data, and a real-time alarm is sent when a prediction result shows that the fire possibly spreads to a new area, so that the risk area is calibrated more accurately and the prediction is more dynamic.
In addition, the GPS positioning technology and the real-time communication system are adopted to realize real-time monitoring of the positions of rescue workers, the real-time dynamics of a fire source is combined to provide the latest position and fire information for rescue teams, the position dynamic model and the command system are integrated, the rapid and coordinated tightness of information transmission among rescue organizations is ensured, and the overall emergency response efficiency is improved.
Finally, through the established flexible iterative model updating mechanism, the assessment model is allowed to be updated rapidly according to new data and scene experience so as to adapt to complex and rapidly changing fire scenes, and online learning and incremental learning strategies are adopted, so that the model is updated in real time under the condition of continuously receiving the new data, and timeliness and flexibility of the assessment model are ensured.
The method solves the problem that the offline or periodically acquired data in the prior art cannot respond to the dynamic change of the fire scene in real time.
Drawings
FIG. 1 is a flow chart of a fire rescue risk assessment method of the present invention;
fig. 2 is a frame diagram of the fire rescue risk assessment device according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Examples
Referring to fig. 1-2, the present invention provides a fire rescue risk assessment method, which includes:
step 1, rescue scene information is obtained in real time, data are obtained through real-time monitoring, data integration is performed, a real-time data stream processing system is established, the data stream processing system comprises a sensor, monitoring equipment and a satellite data real-time obtaining unit, comprehensive perception of a fire scene is provided, the real-time data are processed through edge calculation, and processing analysis of large-scale data is performed based on cloud calculation;
in step 1, the rescue scene information real-time acquisition method comprises the following steps:
sensor deployment and data collection, wherein the sensors comprise a temperature sensor, a smoke sensor and a gas sensor;
deploying a sensor in a fire risk area, and collecting data generated by the sensor in real time;
setting a monitoring camera and monitoring equipment to monitor a fire scene, and configuring the monitoring equipment to transmit image and video stream data in real time;
simultaneously acquiring satellite data including geographic and meteorological related information, and integrating the satellite data to provide comprehensive fire scene perception;
in step 1, the rescue scene information real-time acquisition method further includes:
a real-time monitoring system is deployed, and a fire scene is monitored in real time through a sensor and monitoring equipment;
establishing a real-time data stream processing system;
performing preliminary data processing and analysis by using edge calculation near the data acquisition point;
transmitting the data monitored in real time to a cloud end, and carrying out deep analysis on the sensor, the monitoring equipment and the satellite data by utilizing a large-scale data processing and analyzing technology at the cloud end;
integrating data from the sensor, the monitoring equipment and the satellite to form comprehensive perceived fire scene information;
step 2, real-time data analysis and model updating, wherein real-time intelligent analysis is adopted to process real-time data in real time, identify fire spreading trend and smoke density, improve evaluation instantaneity, design model parameter dynamic adjustment mechanism, and adaptively adjust parameters of an evaluation model according to new real-time data feedback so as to adapt to different fire scenes;
in step 2, the real-time intelligent analysis method comprises the following steps:
adopting a convolutional neural network CNN as a basic model of real-time analysis, deploying a real-time analysis module, and inputting image data to perform fire scene analysis;
by using the position of the fire sourceWind direction->Temperature->Parameters, establishing a fire spread model->:
Wherein f represents a fire spread function;
smoke density measured in combination with sensorImage data of a monitoring device->Establishing a smoke density model->;
Wherein g represents a smoke density recognition function;
a dynamic adjustment mechanism is established, model parameters are adjusted based on real-time data feedback, and the adjustment mode is as follows:
,
wherein the method comprises the steps ofRepresenting new model parameters->Then it is the old model parameter,/->Expressed as learning rate->Representing the gradient of the loss function versus the parameter;
adopting random gradient descent SGD, and improving real-time performance by utilizing a hardware accelerator;
periodically updating the model, establishing a feedback mechanism, and collecting an actual scene data adjustment model;
step 3, calibrating and predicting a dynamic risk area, updating the boundary of a fire risk area through real-time data by utilizing a GIS technology, evaluating a model to reflect dynamic change of fire, carrying out dynamic prediction by adopting a machine learning algorithm, predicting the path and speed of fire spreading based on the real-time data, and providing real-time fire risk early warning;
in the step 3, the dynamic risk area calibration and prediction method specifically comprises the following steps:
based on the real-time monitoring of the fire scene in the step 1, the acquired related data including wind direction, temperature and topography are utilized to draw a map of the fire area, and the boundary of the fire area is updated through the real-time data;
a machine learning model is built by adopting a cyclic neural network (RNN), historical data is input for training, and the possible path and speed of fire spread are predicted;
updating the boundary of the fire danger area according to the prediction result of the machine learning model;
real-time fire hazard early warning is carried out, and an alarm is sent out when a prediction result shows that the fire hazard possibly spreads to a new area;
in step 3, the step of establishing a machine learning model includes:
collecting fire scene data monitored in real time, and setting the wind direction as W, the temperature as T and the terrain as Ter;
drawing a map of the fire disaster area by using a GIS technology, and updating the boundary of the fire disaster area through real-time data;
establishing a cyclic neural network (RNN) model to learn a possible path and a possible speed of fire spread;
the inputs of the RNN model include historically monitored fire scene data sequences, specifically:
wherein t represents current time data, t-1 and t-n represent the last time step observation data and the observation data of the first n time steps respectively;
dividing the data into a training set and a verification set;
in step 3, the step of building a machine learning model further includes:
establishing a loss function L, performing model training by using historical data, optimizing model parameters, and minimizing the loss function, and specifically:
wherein->And->Respectively indicating the possible fire spread path and speed at time i, N indicating the number of samples, ++>And->Respectively representing the actual fire spreading path and the actual fire spreading speed in the historical data;
predicting fire scene data at the current moment by using the trained RNN model to obtain a possible path for fire spreadAnd speed->;
Updating the boundary of the fire risk area according to the path and the speed predicted by the RNN model;
establishing an update function prediction path and a speed map as a new boundary;
when the model prediction result shows that the fire possibly spreads to a new area, an alarm is sent out;
step 4, monitoring the positions of the rescue workers and the communication system in real time, adopting a GPS positioning technology to monitor the positions of the rescue workers in real time, providing the latest position and fire information for rescue teams by combining the real-time dynamics of fire sources, and providing a real-time communication and command system for information transmission among various rescue organizations, wherein the coordination is tighter, and the overall emergency response efficiency is improved;
in step 4, the method for positioning and communicating the position of the detecting personnel comprises the following steps:
a global positioning system GPS is deployed on a monitor, and the position of a rescue worker is monitored in real time by adopting Bluetooth and Wi-Fi;
position data are collected by using sensors and equipment and transmitted to a central system in real time;
real-time monitoring the position of a rescue person and the real-time dynamics of a fire source, and establishing a position dynamic model by combining the real-time data of the position of the rescue person and the fire source:
establishing a command system, integrating position information of rescue workers and dynamic information of fire sources, and providing a real-time command view;
transmitting real-time position and fire source dynamic information to related rescue organizations;
a close coordination mechanism is established by utilizing a communication system, so that rapid and close coordination of information transmission among rescue organizations is ensured;
step 5, a model updating and flexibility mechanism is established, and an iterative model updating mechanism is established, so that an evaluation model is allowed to be updated rapidly according to new data and scene experience; to accommodate complex, rapidly changing fire scenarios;
in step 5, the model update mechanism is established in the following manner:
the model updating mechanism allows the evaluation model to be flexibly and iteratively updated according to new data and scene experience;
online learning and incremental learning are adopted to enable the model to be updated in real time under the condition of continuously receiving new data;
making an incremental learning strategy model to efficiently adjust when new data is processed;
and establishing an iterative updating framework to adapt the model to new data and scene changes.
The invention also provides a fire rescue risk assessment device, which comprises:
the rescue scene information real-time acquisition module is used for monitoring and acquiring data in real time, integrating sensors, monitoring equipment and satellite data and providing comprehensive perception of fire scenes;
the monitoring deployment module is used for deploying temperature, smoke and gas sensors in a fire risk area, and the monitoring camera transmits image and video stream data in real time;
the real-time data stream processing system is used for carrying out data processing and analysis by adopting edge computing and cloud computing technologies;
the real-time data analysis and model updating module adopts real-time intelligent analysis to identify the fire spreading trend and the smoke density, and dynamically adjusts and evaluates model parameters;
the dynamic risk area calibration and prediction module is used for calibrating a dynamic fire risk area by using a GIS technology, and performing dynamic prediction by using a machine learning algorithm to provide real-time fire risk early warning;
the real-time monitoring personnel position and communication system module is used for monitoring the position of the rescue personnel in real time by utilizing a GPS positioning technology and providing real-time position and fire information by combining the dynamic information of the fire source;
the invention also provides fire-fighting rescue risk assessment equipment, which comprises: a memory, a processor, and a program stored on the memory and executable on the processor, characterized in that:
a processor for reading the program in the memory to implement the steps in the above method;
the invention also provides a readable storage medium for storing a program, characterized in that the program, when executed by a processor, implements the steps of the above method.
In combination with the above, in the present application:
according to the fire rescue risk assessment method, in the aspect of acquiring rescue scene information in real time, the comprehensive sensing is realized by adopting various sensors including the temperature sensor, the smoke sensor and the gas sensor, the fire scene is sensed more comprehensively through integration of monitoring equipment and satellite data, the boundary of a fire hazard area is updated in real time through a GIS technology, the data of the sensor and the monitoring equipment are integrated, preliminary processing is performed through edge calculation, large-scale data processing and analysis are performed through cloud computing, and a richer real-time data basis is provided for subsequent steps.
The method is characterized in that a convolutional neural network CNN is adopted as a basic model in the aspects of real-time data analysis and model updating, a model parameter dynamic adjustment mechanism is adopted, parameters of an evaluation model are adaptively adjusted through real-time data feedback, different fire scenes are adapted, a deep learning model and real-time data dynamic adjustment are combined, the real-time performance and accuracy of the fire scenes are improved, meanwhile, a cyclic neural network RNN is adopted as a machine learning model for predicting the possible path and speed of fire spreading, the boundary of a fire risk area is updated through the real-time data, and a real-time alarm is sent when a prediction result shows that the fire possibly spreads to a new area, so that the risk area is calibrated more accurately and the prediction is more dynamic.
In addition, the GPS positioning technology and the real-time communication system are adopted to realize real-time monitoring of the positions of rescue workers, the real-time dynamics of a fire source is combined to provide the latest position and fire information for rescue teams, the position dynamic model and the command system are integrated, the rapid and coordinated tightness of information transmission among rescue organizations is ensured, and the overall emergency response efficiency is improved.
Finally, through the established flexible iterative model updating mechanism, the assessment model is allowed to be updated rapidly according to new data and scene experience so as to adapt to complex and rapidly changing fire scenes, and online learning and incremental learning strategies are adopted, so that the model is updated in real time under the condition of continuously receiving the new data, and timeliness and flexibility of the assessment model are ensured.
It is to be understood that the above examples of the present invention are provided by way of illustration only and not by way of limitation of the embodiments of the present invention. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are desired to be protected by the following claims.
Claims (10)
1. The fire rescue risk assessment method is characterized by comprising the following steps of:
step 1, rescue scene information is obtained in real time, data are obtained through real-time monitoring, data integration is performed, a real-time data stream processing system is established, the data stream processing system comprises a sensor, monitoring equipment and a satellite data real-time obtaining unit, comprehensive perception of a fire scene is provided, the real-time data are processed through edge calculation, and processing analysis of large-scale data is performed based on cloud calculation;
step 2, real-time data analysis and model updating, wherein real-time intelligent analysis is adopted to process real-time data in real time, identify the fire spreading trend and smoke density, and adaptively adjust and evaluate parameters of the model according to new real-time data feedback;
step 3, calibrating and predicting a dynamic risk area, updating the boundary of the fire risk area through real-time data by utilizing a GIS technology, evaluating a model to reflect the dynamic change of the fire, carrying out dynamic prediction by adopting a machine learning algorithm, and predicting the path and the speed of fire spreading based on the real-time data;
step 4, monitoring the positions of the rescue workers and the communication system in real time, adopting a GPS positioning technology to monitor the positions of the rescue workers in real time, providing the latest position and fire information for rescue teams by combining the real-time dynamics of fire sources, and providing a real-time communication and command system for information transmission among the rescue organizations;
and 5, establishing an iterative model updating mechanism by using a model updating and flexibility mechanism, and allowing the evaluation model to be updated rapidly according to new data and scene experience.
2. The fire rescue risk assessment method according to claim 1, wherein in step 1, the rescue scene information real-time acquisition method comprises:
sensor deployment and data collection, wherein the sensors comprise a temperature sensor, a smoke sensor and a gas sensor;
deploying a sensor in a fire risk area, and collecting data generated by the sensor in real time;
setting a monitoring camera and monitoring equipment to monitor a fire scene, and configuring the monitoring equipment to transmit image and video stream data in real time;
and meanwhile, satellite data comprising geographic and meteorological related information is acquired, and comprehensive fire scene perception is provided by integrating the satellite data.
3. The fire rescue risk assessment method according to claim 2, wherein in step 1, the rescue scene information real-time acquisition method further comprises:
a real-time monitoring system is deployed, and a fire scene is monitored in real time through a sensor and monitoring equipment;
establishing a real-time data stream processing system;
performing preliminary data processing and analysis by using edge calculation near the data acquisition point;
and transmitting the data monitored in real time to a cloud, and performing deep analysis on the sensor, the monitoring equipment and the satellite data by using a large-scale data processing and analyzing technology at the cloud.
4. The fire rescue risk assessment method according to claim 3, wherein in the step 2, the real-time intelligent analysis method comprises:
adopting a convolutional neural network CNN as a basic model of real-time analysis, deploying a real-time analysis module, and inputting image data to perform fire scene analysis;
by using the position of the fire sourceWind direction->Temperature->Parameters, establishing a fire spread model->:
Wherein f represents a fire spread function;
smoke density measured in combination with sensorImage data of a monitoring device->Establishing a smoke density model->;
Wherein g represents a smoke density recognition function;
a dynamic adjustment mechanism is established, model parameters are adjusted based on real-time data feedback, and the adjustment mode is as follows:
,
wherein the method comprises the steps ofRepresenting new model parameters->Then it is the old model parameter,/->Expressed as learning rate->Representing the gradient of the loss function versus the parameter;
adopting random gradient descent SGD, and improving real-time performance by utilizing a hardware accelerator;
and periodically updating the model, establishing a feedback mechanism, and collecting the actual scene data adjustment model.
5. The fire rescue risk assessment method according to claim 4, wherein in the step 3, the dynamic risk area calibration and prediction method specifically comprises:
based on the real-time monitoring of the fire scene in the step 1, the acquired related data including wind direction, temperature and topography are utilized to draw a map of the fire area, and the boundary of the fire area is updated through the real-time data;
a machine learning model is built by adopting a cyclic neural network (RNN), historical data is input for training, and the possible path and speed of fire spread are predicted;
updating the boundary of the fire danger area according to the prediction result of the machine learning model;
and carrying out real-time fire hazard early warning, and sending out an alarm when the prediction result shows that the fire hazard possibly spreads to a new area.
6. The fire rescue risk assessment method according to claim 5, wherein in the step 3, the step of establishing a machine learning model includes:
collecting fire scene data monitored in real time, and setting the wind direction as W, the temperature as T and the terrain as Ter;
drawing a map of the fire disaster area by using a GIS technology, and updating the boundary of the fire disaster area through real-time data;
establishing a cyclic neural network (RNN) model to learn a possible path and a possible speed of fire spread;
the inputs of the RNN model include historically monitored fire scene data sequences, specifically:
wherein t represents current time data, and t-1 and t-n represent the last time step observation data and the observation data of the first n time steps respectively;
and divide the data into a training set and a validation set.
7. The fire rescue risk assessment method according to claim 6, wherein in the step 3, the step of establishing a machine learning model further comprises:
establishing a loss function L, performing model training by using historical data, optimizing model parameters, and minimizing the loss function, and specifically:
wherein->And->Respectively indicating the possible fire spread path and speed at time i, N indicating the number of samples, ++>And->Respectively representing the actual fire spreading path and the actual fire spreading speed in the historical data;
predicting fire scene data at the current moment by using the trained RNN model to obtain a possible path for fire spreadAnd speed->;
Updating the boundary of the fire risk area according to the path and the speed predicted by the RNN model;
establishing an update function prediction path and a speed map as a new boundary;
an alarm is raised when the model prediction indicates that the fire may spread to a new area.
8. Fire rescue risk assessment device, its characterized in that includes:
the rescue scene information real-time acquisition module is used for monitoring and acquiring data in real time, integrating sensors, monitoring equipment and satellite data and providing comprehensive perception of fire scenes;
the monitoring deployment module is used for deploying temperature, smoke and gas sensors in a fire risk area, and the monitoring camera transmits image and video stream data in real time;
the real-time data stream processing system is used for carrying out data processing and analysis by adopting edge computing and cloud computing technologies;
the real-time data analysis and model updating module adopts real-time intelligent analysis to identify the fire spreading trend and the smoke density, and dynamically adjusts and evaluates model parameters;
the dynamic risk area calibration and prediction module is used for calibrating a dynamic fire risk area by using a GIS technology, and performing dynamic prediction by using a machine learning algorithm to provide real-time fire risk early warning;
and the real-time monitoring personnel position and communication system module is used for monitoring the position of the rescue personnel in real time by utilizing a GPS positioning technology and providing real-time position and fire information by combining the dynamic information of the fire source.
9. Fire rescue risk assessment device, the fire rescue risk assessment device comprising: a memory, a processor, and a program stored on the memory and executable on the processor, characterized in that:
the processor for reading a program in a memory to implement the steps in the fire rescue risk assessment method according to any one of claims 1 to 7.
10. A readable storage medium storing a program, wherein the program when executed by a processor implements the steps in the fire rescue risk assessment method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311845234.1A CN117495110A (en) | 2023-12-29 | 2023-12-29 | Fire rescue risk assessment method, device, equipment and readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311845234.1A CN117495110A (en) | 2023-12-29 | 2023-12-29 | Fire rescue risk assessment method, device, equipment and readable storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117495110A true CN117495110A (en) | 2024-02-02 |
Family
ID=89673021
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311845234.1A Pending CN117495110A (en) | 2023-12-29 | 2023-12-29 | Fire rescue risk assessment method, device, equipment and readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117495110A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117745083A (en) * | 2024-02-20 | 2024-03-22 | 山东居安特消防科技有限公司 | Fire control management system and method based on big data |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20230042856A (en) * | 2021-09-23 | 2023-03-30 | 노아에스앤씨 주식회사 | Method, apparatus and program for multi-dense facility large-scale fire situation management and decision support services |
CN116527846A (en) * | 2023-04-28 | 2023-08-01 | 王乐廷 | Fire scene real-time monitoring command system based on digital cloud technology |
CN116664359A (en) * | 2023-03-31 | 2023-08-29 | 山东浪潮数字服务有限公司 | Intelligent fire early warning decision system and method based on multi-sensor fusion |
-
2023
- 2023-12-29 CN CN202311845234.1A patent/CN117495110A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20230042856A (en) * | 2021-09-23 | 2023-03-30 | 노아에스앤씨 주식회사 | Method, apparatus and program for multi-dense facility large-scale fire situation management and decision support services |
CN116664359A (en) * | 2023-03-31 | 2023-08-29 | 山东浪潮数字服务有限公司 | Intelligent fire early warning decision system and method based on multi-sensor fusion |
CN116527846A (en) * | 2023-04-28 | 2023-08-01 | 王乐廷 | Fire scene real-time monitoring command system based on digital cloud technology |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117745083A (en) * | 2024-02-20 | 2024-03-22 | 山东居安特消防科技有限公司 | Fire control management system and method based on big data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN117495110A (en) | Fire rescue risk assessment method, device, equipment and readable storage medium | |
KR102203135B1 (en) | Method and system for detecting disaster damage information based on artificial intelligence using drone | |
CN113283324B (en) | Forest fire prevention early warning method and system based on dynamic image | |
CN109035665A (en) | A kind of novel forest fire early-warning system and fire alarm method | |
WO2023082550A1 (en) | Wind speed early-warning method and apparatus and operation machinery | |
CN114665608B (en) | Intelligent sensing inspection system and method for transformer substation | |
CN116822969A (en) | Water conservancy model cloud computing method and system based on model combination | |
CN113778137A (en) | Unmanned aerial vehicle autonomous inspection method for power transmission line | |
CN111343287B (en) | Helicopter laser radar remote monitoring system and method for power transmission line inspection | |
CN106546703A (en) | Air quality surveillance system, method and device | |
CN113468724B (en) | Digital twin system simulation method and device for airport aircraft landing guidance | |
CN116485066B (en) | GIS-based intelligent gas safety line inspection management method and Internet of things system | |
CN111243215B (en) | Low-altitude unmanned monitoring and early warning system and method for forest fire scene | |
Thakkar et al. | Environmental fire hazard detection and prediction using random forest algorithm | |
CN116931596A (en) | Unmanned aerial vehicle flight system with flight program automatically arranged | |
US20240103537A1 (en) | Methods, systems, and devices for inspecting structures and objects | |
CN111866464B (en) | Agricultural tractor remote control system based on virtual reality technology | |
CN117576920B (en) | Traffic control system based on unmanned aerial vehicle | |
CN114162134B (en) | Method, device and storage medium for predicting vehicle track on sea-crossing bridge | |
CN116647651B (en) | Unmanned aerial vehicle construction monitoring method and system based on Beidou satellite | |
CN116485160B (en) | Power transmission line inspection processing system and method | |
CN116700070B (en) | Safety supervision method and system for flight state of unmanned aerial vehicle | |
KR102541215B1 (en) | Underwater living thing monitoring device and method using Virtual Reality | |
CN117077238B (en) | Method and system for accurately tracking fire-fighting points | |
CN116880573B (en) | Collaborative control method and system for unmanned missile-borne unmanned aerial vehicle and unmanned detection aerial vehicle |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |