CN114511243A

CN114511243A - Method and system for dynamically evaluating fire risk based on Internet of things monitoring

Info

Publication number: CN114511243A
Application number: CN202210163285.5A
Authority: CN
Inventors: 田春伟; 徐勇
Original assignee: Shenzhen Graduate School Harbin Institute of Technology
Current assignee: Shenzhen Graduate School Harbin Institute of Technology
Priority date: 2022-02-22
Filing date: 2022-02-22
Publication date: 2022-05-17

Abstract

The method comprises the steps of obtaining the current temperature and the current environment monitoring data of a target area in a target building, and generating whether information exists in a current fire source or not according to the current temperature and the current environment monitoring data; acquiring current environmental information of a target area in a target building, and determining a current environmental information processing result corresponding to the current environmental information through a first relation pre-established by the first neural network; and determining the current fire risk level of the target area in the target building corresponding to the current environmental information processing result through a second relation pre-established by a second neural network. The current environmental information of the building is acquired in real time through the Internet of things, a data basis is provided for follow-up dynamic fire risk assessment, the fire risk level of the building is dynamically assessed in real time through the neural network, and the building can be effectively supervised.

Description

Method and system for dynamically evaluating fire risk based on Internet of things monitoring

Technical Field

The application relates to the technical field of fire risk assessment, in particular to a method and a system for dynamically assessing fire risks based on Internet of things monitoring.

Background

With the increase of urban scale and the development of science and technology, the causes of building fires are diversified gradually, and the difficulty of fire control management of buildings is increased accordingly. The accuracy of fire risk assessment can effectively help fire managers to control the fire risk of buildings, and a targeted fire protection scheme is provided according to the existing situation.

In the era of undeveloped software, the domestic evaluation of the fire risk depends on a daily check list of the building to a great extent, and the fire risk evaluation lacks real-time data, cannot effectively supervise the building, cannot scientifically manage the building, and hinders the effective fire-fighting supervision of the building. With the progress of computer technology and software engineering, an entity relation graph of a building fire risk assessment system can be established by analyzing the relation among various entities, and a database can be effectively established for carrying out effective fire risk assessment on a building. The use of databases, however, only reduces the management and labor costs, and accurate assessment of fire risk remains a significant challenge.

Foreign scholars have earlier focused on the field of fire risk assessment. In the last 70 th century, european scholars began using statistical methods to study the relationship between the frequency of fires and building attributes.

In recent years, along with the improvement of the mathematical theory basis, the machine learning algorithm has a great breakthrough, and more machine learning algorithms are applied to the fire risk assessment of buildings. For example: accident tree and event tree analysis is applied to fire risk assessment, and fire risk is assessed by analyzing factors of fire occurrence, fire-fighting facilities and various fire scene characteristics; the Bayesian network is combined with dynamic fire assessment, a network model is combined with a directed acyclic graph, and prior knowledge is added into the network according to the Bayesian thought, so that the network is favorable for processing uncertainty problems. However, these methods have the following problems:

1. the parameters of the network are difficult to adjust, which causes the modellability to be difficult to reach the optimum;

2. and more historical data are adopted, so that the requirement on the performance of the model is high, and the dynamic evaluation is difficult.

Disclosure of Invention

In view of the above, the present application is directed to a method and system for dynamic fire risk assessment based on internet of things monitoring, which overcomes or at least partially solves the above problems, and comprises:

a method for dynamically evaluating fire risk based on Internet of things monitoring is applied to establishing a corresponding relation between relevant information of a building and a corresponding fire risk level through an artificial intelligence model; wherein the artificial intelligence model comprises a first model;

the first model comprises a first neural network and a second neural network, the corresponding relation comprises a first relation and a second relation, and the related information comprises the temperature, the smoke concentration, the hazardous gas concentration, the environment monitoring data and corresponding basic parameters of a target area in the target building;

the method comprises the following steps:

acquiring current temperature and current environment monitoring data of a target area in a target building, and generating information whether a current fire source exists or not according to the current temperature and the current environment monitoring data;

acquiring current environmental information of a target area in a target building, and determining a current environmental information processing result corresponding to the current environmental information through a first relation pre-established by the first neural network; the current environmental information comprises information of whether a current fire source exists, current smoke concentration, current dangerous gas concentration and current basic parameters of a target building; the current environment information processing result comprises an information processing result of whether the current fire source exists, a current smoke concentration processing result, a current dangerous gas concentration processing result and a current basic parameter processing result of the target building;

and determining the current fire risk level of the target area in the target building corresponding to the current environmental information processing result through a second relation pre-established by a second neural network.

Further, the first model further comprises a third neural network, and the corresponding relationship further comprises a third relationship; the step of obtaining the current temperature and the current environment monitoring data of a target area in a target building and generating whether the current fire source exists according to the current temperature and the current environment monitoring data comprises the following steps:

acquiring current temperature and current environment monitoring data of a target area in a target building, and determining a current thermodynamic diagram corresponding to the current temperature through a third relation pre-established by a third neural network;

and generating information whether the current fire source exists or not according to the current thermodynamic diagram and the current environment monitoring data.

Further, the method further comprises:

and establishing the first relation between the environmental information of the target area in the target building and the environmental information processing result through the first neural network.

Further, the first neural network comprises a first sub-neural network, a second sub-neural network, a third sub-neural network and a fourth sub-neural network; the first relationship comprises a first sub-relationship, a second sub-relationship, a third sub-relationship and a fourth sub-relationship; the step of establishing the first relationship between the environmental information of the target area in the target building and the environmental information processing result through the first neural network includes:

establishing a first sub-relationship between whether the fire source has information and whether the fire source has an information processing result through the first sub-neural network;

establishing a second sub-relationship between the smoke concentration and the smoke concentration processing result through the second sub-neural network;

establishing a third sub-relationship between the concentration of the dangerous gas and the concentration processing result of the dangerous gas through the third sub-neural network;

and establishing a fourth sub-relationship between the basic parameters of the target building and the basic parameter processing result of the target building through the fourth sub-neural network.

Further, the method further comprises:

and establishing whether the second relation between the information processing result, the smoke concentration processing result, the dangerous gas concentration processing result, the basic parameter processing result of the target building and the fire risk grade of the target area in the target building exists or not through the second neural network.

Further, the second neural network includes a first prediction subnetwork and a second prediction subnetwork, the second relationship includes a fifth sub-relationship and a sixth sub-relationship; the step of establishing, by the second neural network, whether the second relationship between the information processing result, the smoke concentration processing result, the hazardous gas concentration processing result, the basic parameter processing result of the target building and the fire risk level of the target area in the target building exists or not includes:

establishing whether a fire source has a fifth sub-relationship among an information processing result, a smoke concentration processing result, a dangerous gas concentration processing result, a basic parameter processing result of a target building and a comprehensive processing result through the first prediction sub-network;

establishing, by the second prediction subnetwork, a sixth sub-relationship between the combined processing result and the fire risk level.

Further, the second neural network further comprises a third prediction sub-network, and the second relationship further comprises a seventh sub-relationship; when the fire risk level output by the second neural network is a preset fire risk level, calling the third prediction sub-network; the method further comprises the following steps:

and when the fire risk level is a danger level or a damage level, determining the current dangerous area coordinate corresponding to the target area with the current fire risk level being the danger level or the damage level through a seventh sub-relationship pre-established by the third prediction sub-network.

The invention also provides a fire risk dynamic evaluation system based on the monitoring of the Internet of things, which is applied to establishing the corresponding relation between the relevant information of the building and the corresponding fire risk level through an artificial intelligence model; wherein the artificial intelligence model comprises a first model;

the first model comprises a first neural network and a second neural network, the corresponding relation comprises a first relation and a second relation, and the related information comprises the temperature, the smoke concentration, the hazardous gas concentration, the environment monitoring data and corresponding basic parameters of the target building in the target building;

the system comprises:

the current fire source existence information generating module is used for acquiring the current temperature and the current environment monitoring data of a target area in a target building and generating whether the current fire source exists or not according to the current temperature and the current environment monitoring data;

the current processing result determining module is used for acquiring current environment information of a target area in a target building and determining a current environment information processing result corresponding to the current environment information through a first relation pre-established by the first neural network; the current environmental information comprises information of whether a current fire source exists, current smoke concentration, current dangerous gas concentration and current basic parameters of a target building; the current environment information processing result comprises an information processing result of whether the current fire source exists, a current smoke concentration processing result, a current dangerous gas concentration processing result and a current basic parameter processing result of the target building;

and the current fire risk grade determining module is used for determining the current fire risk grade of a target area in the target building corresponding to the current environmental information processing result through a second relation pre-established by a second neural network.

The invention also provides a computer device comprising a processor, a memory and a computer program stored on the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the steps of the method for dynamic assessment of fire risk based on monitoring of the internet of things as described above.

The invention also provides a computer-readable storage medium on which a computer program is stored, which computer program, when being executed by a processor, realizes the steps of the method for dynamic assessment of fire risk based on monitoring of the internet of things as described above.

The application has the following advantages:

in the embodiment of the application, the corresponding relation between the relevant information of the building and the corresponding fire risk level is established through an artificial intelligence model; wherein the artificial intelligence model comprises a first model; the first model comprises a first neural network and a second neural network, the corresponding relation comprises a first relation and a second relation, and the related information comprises the temperature, the smoke concentration, the hazardous gas concentration, the environment monitoring data and corresponding basic parameters of the target building in the target building; acquiring current temperature and current environment monitoring data of a target area in a target building, and generating information whether a current fire source exists or not according to the current temperature and the current environment monitoring data; acquiring current environmental information of a target area in a target building, and determining a current environmental information processing result corresponding to the current environmental information through a first relation pre-established by the first neural network; the current environmental information comprises information of whether a current fire source exists, current smoke concentration, current dangerous gas concentration and current basic parameters of a target building; the current environment information processing result comprises a current fire source information processing result, a current smoke concentration processing result, a current dangerous gas concentration processing result and a current basic parameter processing result of a target building; and determining the current fire risk level of the target area in the target building corresponding to the current environmental information processing result through a second relation pre-established by a second neural network. The current environmental information of the building is acquired in real time through the Internet of things, a data basis is provided for follow-up dynamic fire risk assessment, the fire risk level of the building is dynamically assessed in real time through the neural network, and the building can be effectively supervised.

Drawings

In order to more clearly illustrate the technical solutions of the present application, the drawings needed to be used in the description of the present application will be briefly introduced below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive labor.

Fig. 1 is a flowchart illustrating steps of a method for dynamically evaluating a fire risk based on monitoring of an internet of things according to an embodiment of the present application;

fig. 2a is a schematic structural diagram of a lower computer of a method for dynamically evaluating fire risk based on internet of things monitoring according to an embodiment of the present application;

fig. 2b is a connection topology diagram of a lower computer and an upper computer of a method for dynamically evaluating fire risk based on internet of things monitoring according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of a multi-modal attention neural network of a method for dynamically evaluating fire risk based on internet of things monitoring according to an embodiment of the present application;

fig. 4 is a schematic structural diagram of a fire source existence information generating module in a method for dynamically evaluating fire risk based on internet of things monitoring according to an embodiment of the present application;

fig. 5 is a schematic structural diagram of an enhanced convolutional neural network of a method for dynamically evaluating fire risk based on internet of things monitoring according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of an integrated processing module of a method for dynamically evaluating fire risk based on internet of things monitoring according to an embodiment of the present application;

fig. 7a is a schematic visualization diagram of a thermodynamic diagram of a safety area of a method for dynamic fire risk assessment based on internet of things monitoring, which is provided by an embodiment of the present application, in combination with environmental monitoring data;

fig. 7b is a schematic diagram illustrating a prediction of fire risk level of a safety area according to a method for dynamic fire risk assessment based on internet of things monitoring provided by an embodiment of the present application;

fig. 8a is a schematic visualization diagram of a thermodynamic diagram of a dangerous area and environment monitoring data in a method for dynamically evaluating a fire risk based on internet of things monitoring provided by an embodiment of the present application;

fig. 8b is a schematic diagram illustrating prediction of fire risk levels in a dangerous area according to a method for dynamically evaluating fire risk based on internet of things monitoring provided by an embodiment of the present application;

fig. 9 is a block diagram illustrating a system for dynamically evaluating a fire risk based on monitoring of an internet of things according to an embodiment of the present application;

fig. 10 is a schematic structural diagram of a computer device according to an embodiment of the present invention.

Detailed Description

In order to make the aforementioned objects, features and advantages of the present application more comprehensible, the present application is described in further detail with reference to the accompanying drawings and the detailed description. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

It should be noted that, referring to fig. 2, in any embodiment of the present invention, the present invention includes an upper computer and a lower computer, where the lower computer functions to collect data related to the building environment in real time, and the upper computer functions to analyze the data related to the building environment in real time, evaluate fire risk levels of different areas of the building, and determine a dangerous area of the building. An industrial personal computer with an operating system of Ubuntu can be selected as an upper computer hardware part, an upper computer software part is composed of an artificial intelligence model, a Multi-modal Attention and Attention Neural Network (MMANN) is selected, and a Network structure of the MMANN refers to FIG. 3 and is composed of a temperature processing module, a picture processing module, a smoke concentration processing module, a gas concentration processing module, a building attribute analysis module and a comprehensive processing module and used for jointly analyzing data; the lower computer comprises singlechip, WIFI module and 4 kinds of sensors, and wherein, 4 kinds of sensors are respectively: temperature and humidity sensor, smoke transducer, gaseous sensor and the camera module of detecting.

Fig. 2a is a schematic diagram of a hardware connection structure of a lower computer, wherein 4 types of sensors are respectively connected to one end of a single chip microcomputer, and the other end of the single chip microcomputer is connected with a WIFI transmission module; fig. 2b is a connection topological diagram of the lower computer and the upper computer, wherein firstly, the 4 sensors collect the environmental information in real time and send the data to the single chip microcomputer, and the single chip microcomputer arranges various data and then sends the data to the upper computer at regular time through the WIFI module. Because the collected environment information has different coding modes, the upper computer needs to carry out different preprocessing on various information and then inputs the information into the multi-mode attention neural network to predict the fire risk level and the coordinates of the dangerous area.

Referring to fig. 1, a method for dynamically evaluating a fire risk based on monitoring of an internet of things according to an embodiment of the present application is shown;

the method comprises the following steps:

s110, acquiring current temperature and current environment monitoring data of a target area in a target building, and generating information whether a current fire source exists or not according to the current temperature and the current environment monitoring data;

s120, obtaining current environment information of a target area in a target building, and determining a current environment information processing result corresponding to the current environment information through a first relation pre-established by the first neural network; the current environmental information comprises information of whether a current fire source exists, current smoke concentration, current dangerous gas concentration and current basic parameters of a target building; the current environment information processing result comprises an information processing result of whether the current fire source exists, a current smoke concentration processing result, a current dangerous gas concentration processing result and a current basic parameter processing result of the target building;

s130, determining the current fire risk level of the target area in the target building corresponding to the current environmental information processing result through a second relation pre-established by a second neural network.

In the embodiment of the application, the corresponding relation between the relevant information of the building and the corresponding fire risk level is established through an artificial intelligence model; wherein the artificial intelligence model comprises a first model; the first model comprises a first neural network and a second neural network, the corresponding relation comprises a first relation and a second relation, and the related information comprises the temperature, the smoke concentration, the hazardous gas concentration, the environment monitoring data and corresponding basic parameters of the target building in the target building; the method comprises the following steps: acquiring current temperature and current environment monitoring data of a target area in a target building, and generating information whether a current fire source exists or not according to the current temperature and the current environment monitoring data; acquiring current environmental information of a target area in a target building, and determining a current environmental information processing result corresponding to the current environmental information through a first relation pre-established by the first neural network; the current environmental information comprises information of whether a current fire source exists, current smoke concentration, current dangerous gas concentration and current basic parameters of a target building; the current environment information processing result comprises an information processing result of whether the current fire source exists, a current smoke concentration processing result, a current dangerous gas concentration processing result and a current basic parameter processing result of the target building; and determining the current fire risk level of the target area in the target building corresponding to the current environmental information processing result through a second relation pre-established by a second neural network. The current environmental information of the building is acquired in real time through the Internet of things, a data basis is provided for follow-up dynamic fire risk assessment, the fire risk level of the building is dynamically assessed in real time through the neural network, and the building can be effectively supervised.

In the following, a method for dynamically evaluating a fire risk based on monitoring of the internet of things in the exemplary embodiment will be further described.

And step S110, obtaining the current temperature and the current environment monitoring data of the target area in the target building, and generating information on whether the current fire source exists according to the current temperature and the current environment monitoring data.

In an embodiment of the present invention, a specific process of "obtaining the current temperature and the current environment monitoring data of the target area in the target building, and generating the information whether the current fire source exists according to the current temperature and the current environment monitoring data" in step S110 may be further described with reference to the following description.

In an embodiment of the present invention, the first model further includes a third neural network, and the corresponding relationship further includes a third relationship.

The method comprises the following steps of obtaining current temperature and current environment monitoring data of a target area in a target building, and determining a current thermodynamic diagram corresponding to the current temperature through a third relation pre-established by a third neural network.

As an example, the temperature in a building is an important index for judging the fire risk level, thermodynamic diagrams corresponding to the temperatures of different areas in the building are analyzed by using an artificial neural network algorithm, and a mapping rule between the temperature of a target area and the thermodynamic diagrams is found through the self-learning and self-adaptive characteristics of the artificial neural network.

The invention provides a temperature processing module which mainly comprises a third neural network, and a third relation between temperature and thermodynamic diagrams is established through the third neural network, so that the effect of corresponding thermodynamic diagrams can be obtained after temperature data are input to the module. The third Neural Network may be a Fully Connected Neural Network (FCNN) with an input size of 32 × 1, and 32 represents the values of 32 temperature sensors. The fully-connected layer is composed of 5 hidden layers, the number of neurons of each hidden layer is 2048, the activation function of each hidden layer adopts ReLU, and finally the neurons of the output layer are converted into a two-dimensional matrix of 224 x 224 to serve as a thermodynamic diagram, and the formula is as follows:

Map＝Reshape(f(a(f(a(f(a(f(a(f(a(x))))))))))) (1)

a(x)＝Wx+b (2)

wherein, formula (1) represents that the temperature is processed by a fully-connected neural network to generate a thermodynamic diagram; formula (2) is a (x) represents the linear calculation of the hidden layer, W is a weight matrix, linear transformation is carried out, and b is offset; f () is the activation function ReLU; the Reshape (×) function is a matrix that converts one-dimensional data into sizes; map represents the generated thermodynamic diagram.

In an embodiment of the present application, the input data of the third neural network is the temperature of the target area in the target building, and the temperature is periodically input to the fully-connected neural network model in a time sequence, so that the third neural network can theoretically acquire information from the past temperature data, and can also adjust thermodynamic diagram output according to the latest recorded temperature data. The output of the third neural network is a thermodynamic diagram corresponding to the current temperature.

When the third neural network is trained, the temperature diagrams corresponding to different area temperature conditions (including but not limited to one or more of the following temperature diagrams of minus 10-10 ℃, 10-30 ℃, 30-40 ℃, 40-200 ℃ and the like) of a large number of buildings are collected in a gathering way, the temperature diagrams and the thermodynamic diagrams corresponding to a plurality of different areas are selected as sample data, and the neural network is trained by combining with prestored expert experience information. For example: collecting temperature data and corresponding thermodynamic diagrams of buildings at different places; collecting temperature data and corresponding thermodynamic diagrams of the same building in different time periods; and collecting temperature data and corresponding thermodynamic diagrams of the same building under different use conditions. Therefore, data are collected through multiple ways, the data volume is increased, the learning capacity of the artificial neural network is improved, and the accuracy and the reliability of the determined corresponding relation are improved.

And fitting the relationship between the temperatures and the thermodynamic diagrams corresponding to different areas by the neural network through adjusting the weight between the network structure and the network nodes by using an artificial neural network algorithm, and finally enabling the neural network to accurately fit the corresponding relationship between the temperatures and the thermodynamic diagrams corresponding to different areas of different buildings.

The temperature processing module uses the third neural network to identify current temperature data for the target area of the building and to output the task of the thermodynamic diagram. It should be noted that temperature data are collected by a temperature and humidity sensor in a target area in a building, and then are uploaded to an upper computer through a WIFI module to be used as an input of a first sub-neural network, wherein the WIFI module is preferably an ESP8266 module.

In a particular implementation, after the trained third neural network is used to generate a corresponding current thermodynamic diagram for current temperature data for different areas in the building. Every 50m in a building²And 32 temperature and humidity sensors are respectively installed and used for collecting the current temperature in the region, so that the current temperature distribution of the target region closer to the real temperature is conveniently obtained, wherein the hardware type of the temperature and humidity sensors is selected to be a DHT11 module. After each temperature and humidity sensor uploads the collected current temperature data to the single chip microcomputer, the single chip microcomputer sends the data to an upper computer at intervals of 5 minutes through the WIFI module, wherein the time interval difference of sending the data between the single chip microcomputers is

And n is the number of lower computers in the building, and a data basis is provided for dynamically evaluating the fire risk of the subsequent upper computers by sending dynamic data of the building environment at regular time.

And generating information whether the current fire source exists or not according to the current thermodynamic diagram and the current environment monitoring data as follows.

It should be noted that the current environment monitoring data is collected in the building by the camera module, and then the environment monitoring data is uploaded to the upper computer through the WIFI module.

As an example, referring to fig. 4, the invention provides an information generation module for determining whether a fire source exists, which cuts current environment monitoring data acquired by a camera module, and then splices the cut current environment monitoring data with the thermodynamic diagram in a channel dimension to obtain information about whether the current fire source exists, and obtains detailed current temperature information and current open fire service conditions in the picture area according to the information about whether the current fire source exists, so that the data fusion promotes the accuracy of a network model.

In one implementation, every 50m in a building²And a camera is installed, the current environment monitoring data in the area is collected, the use condition of open fire in the building can be better detected, and the hardware model of the camera module is an OV7670 module. After each camera uploads the collected current environment monitoring data to the single chip microcomputer, the single chip microcomputer sends the data to the upper computer at intervals of 5 minutes through the WIFI module, wherein the time interval difference of sending the data between the single chip microcomputers is

As described in step S120, obtaining current environmental information of a target area in a target building, and determining a current environmental information processing result corresponding to the current environmental information through a first relationship pre-established by the first neural network; the current environmental information comprises information of whether a current fire source exists, current smoke concentration, current dangerous gas concentration and current basic parameters of a target building; the current environment information processing result comprises an information processing result of whether the current fire source exists, a current smoke concentration processing result, a current dangerous gas concentration processing result and a current basic parameter processing result of the target building.

In an embodiment of the present invention, the step S120 of "obtaining current environment information of a target area in a target building" may be further described with reference to the following description, and determining a current environment information processing result corresponding to the current environment information through a first relationship pre-established by the first neural network; the current environmental information comprises information of whether a current fire source exists, current smoke concentration, current dangerous gas concentration and current basic parameters of a target building; the current environment information processing result comprises the specific processes of whether the current fire source has an information processing result, a current smoke concentration processing result, a current dangerous gas concentration processing result and a current basic parameter processing result of the target building.

In an embodiment of the present invention, the first model includes a first neural network and a second neural network, the corresponding relationship includes a first relationship and a second relationship, and the related information includes a temperature, a smoke concentration, a hazardous gas concentration, environmental monitoring data of a target area in the target building and a corresponding basic parameter of the target building.

Establishing, by the first neural network, the first relationship between the environmental information of the target area in the target building and the environmental information processing result, as described in the following steps.

In an embodiment of the present invention, a specific process of the step "establishing the first relationship between the environmental information of the target area in the target building and the environmental information processing result through the first neural network" may be further described in conjunction with the following description.

In an embodiment of the present invention, the first neural network includes a first sub-neural network, a second sub-neural network, a third sub-neural network, and a fourth sub-neural network; the first relationship includes a first sub-relationship, a second sub-relationship, a third sub-relationship, and a fourth sub-relationship.

And establishing a first sub-relation between the information whether the fire source exists and the information processing result whether the fire source exists through the first sub-neural network as described in the following steps.

The invention provides an image processing module which is mainly composed of a first sub-neural network and used for preprocessing information whether a fire source exists or not, and an artificial neural network algorithm is used for analyzing whether information processing result rules exist in the fire source corresponding to the information whether the fire source exists or not in different areas in a building or not.

In a specific implementation, referring to fig. 5, the first sub-Neural Network may be an enhanced Convolutional Neural Network (enhanced Convolutional Neural Network), and a Network architecture design of the enhanced Convolutional Neural Network takes advantage of an Encoder-Decoder concept, which may effectively reduce a computation amount without losing performance. The structure of the enhanced convolutional neural network is schematically shown in fig. 5, and the enhanced convolutional neural network is divided into a main network and a branch network, wherein the main network comprises 10 convolutional layers, 3 pooling layers and 3 upsampling layers; in the tributary network, each tributary has only one convolutional layer. The input size of the enhanced convolutional neural network is 4 × 224 × 224, 4 represents the sum of the input picture data and the number of channels of the thermodynamic diagram, and 224 × 224 represents the resolution of the picture. Convolution kernels of all convolution layers are 3 multiplied by 3, the pooling layers adopt average pooling, the resolution is reduced to half of the original resolution, the upper sampling layer adopts transposition convolution, and the resolution is doubled. The output results of 2 nd, 3 rd, 4 th and 5 th convolutional layers on the trunk are respectively input into the branch convolutional layers for feature enhancement, and then are fused with the features of the 6 th, 7 th, 8 th and 9 th convolutional layers. And finally, performing pooling and flattening treatment on the extracted features to obtain a final result, wherein the final output size is 588 multiplied by 1. The formula is as follows:

F₁＝g(I_input) (3)

F₂＝z(g(F₁)) (4)

F′₂＝z(g(F₂)) (5)

F₃＝z(g(P(F₂))) (6)

F′₃＝z(g(F₃)) (7)

F₄＝z(g(P(F₃))) (8)

F′₄＝z(g(F₄)) (9)

F₅＝z(g(P(F₄))) (10)

F′₅＝z(g(F₅)) (11)

F₆＝z(g(F₅)) (12)

V_output＝R(g(z(g(Up(z(g(Up(z(g(Up(F₆+F′₅)+F′₄)))+F′₃)))+F′₂)))) (13)

wherein formula (3) represents input I_inputCalculating to obtain characteristic F through the 1 st layer of convolution layer₁The formula (4) represents the feature F₁Calculating to obtain characteristic F through 2 nd layer convolution layer₂The formula (5) represents the feature F₂F 'is obtained by calculating a branch winding layer'₂The formula (6) represents the feature F₂Calculating to obtain characteristic F through the pooling layer and the 3 rd convolution layer₃The formula (7) represents the feature F₃Calculating to obtain characteristic F 'through the branch winding layer'₃The formula (8) represents the feature F₃Calculating to obtain characteristic F through the pooling layer and the 4 th convolution layer₄The formula (9) represents the feature F₄Calculating to obtain characteristic F 'through the branch winding layer'₄The formula (10) represents the feature F₄Through a pooling layer andcalculating the 5 th convolution layer to obtain the characteristic F₅The formula (11) represents the feature F₅Calculating to obtain characteristic F 'through the branch winding layer'₅The formula (12) represents the feature F₅Calculating to obtain characteristic F through the 6 th layer of convolution layer₆(ii) a Equation (13) represents feature F₆And characteristic F'₂And characteristic F'₃And characteristic F'₄And characteristic F'₅Calculating an up-sampling layer, convolution layers of 6 th, 7 th, 8 th and 9 th layers and a down-sampling layer, flattening two-dimensional features into a one-dimensional vector to obtain an output vector V_output(ii) a g represents convolution operation of the convolutional layer, z represents an activation function ReLU, P represents average pooling operation of the pooling layer, Up represents upper sampling layer transposition convolution operation, and R represents average pooling operation and flattening of the two-dimensional matrix into one-dimensional vector operation.

As an example, when training the first sub-neural network, using information on whether a large number of fire sources exist in different areas of the target building as input of a training set, and using a result of processing whether an information on whether a fire source exists corresponding to each data of the information on whether a fire source exists in the training set as a training output corresponding to each corresponding data point, thereby constructing a structure of the first sub-neural network and performing supervised training thereon. After the first sub-neural network is trained, the first sub-neural network can be used for generating corresponding information processing results of whether the fire source exists in different areas in the building or not.

Establishing, by the second sub-neural network, a second sub-relationship between smoke concentration and a smoke concentration processing result, as described in the following steps.

As an example, when training the second sub-neural network, the smoke concentration and the smoke concentration processing result of different areas in the building are collected as sample data, and the second sub-neural network is trained. For example: sample collection: the smoke concentration of different buildings at the same time and the corresponding smoke concentration processing results; and smoke concentration processing results corresponding to the smoke concentrations of the same building at different times, and the like. After the second sub-neural network is trained, it can be used to generate corresponding smoke concentration processing results for the smoke concentrations in different areas of the building.

In one implementation, every 50m in a building²And 32 smoke concentration sensors are arranged, the smoke concentration in the area is collected, the distribution of the smoke concentration which is closer to the real smoke concentration is convenient to obtain, and the hardware type of the smoke concentration sensor is an MQ-2 sensor module. After each smoke concentration sensor uploads the collected smoke concentration to the single chip microcomputer, the single chip microcomputer sends data to an upper computer at intervals of 5 minutes through the WIFI module, wherein the time interval difference of sending the data between the single chip microcomputers is

Establishing a third sub-relationship between the hazardous gas concentration and the hazardous gas concentration processing result through the third sub-neural network as described in the following steps;

as an example, when training the third sub-neural network, the dangerous gas concentrations and the dangerous gas concentration processing results of different areas in the building are collected as sample data, and the third sub-neural network is trained. For example: sample collection: the concentration of flammable and explosive gases (such as ethanol, methane, liquefied gas and the like) is collected before a building fire occurs; the concentration of specific gases (such as carbon monoxide and carbon dioxide, etc.) is collected after a building fire occurs. After the third sub-neural network is trained, it can be used to generate corresponding hazardous gas concentration processing results for the hazardous gas concentrations in different areas of the building.

In one implementation, due to the variety of flammable and explosive gases and hazardous gases in the natural world, every 50m in a building²The area is respectively provided with 32 gas detection sensors of 3 different types, the concentration of methane, ethanol, liquefied gas, carbon monoxide and carbon dioxide gas at different positions in the building is respectively collected, the hidden danger brought to the building by different dangerous gases can be more accurately analyzed, wherein the hardware types of the 3 gas sensors are respectively MQ-3 module and MQ-5 moduleAnd an MQ-7 module. After each gas sensor uploads the collected temperature data to the single chip microcomputer, the single chip microcomputer sends the data to an upper computer at intervals of 5 minutes through the WIFI module, wherein the time interval difference of sending the data between the single chip microcomputers is

And establishing a fourth sub-relation between the basic parameters of the target building and the basic parameter processing result of the target building through the fourth sub-neural network as described in the following steps.

As an example, when the fourth sub-neural network is trained, the basic parameters of the target building and the basic parameter processing result of the target building are collected as sample data, and the fourth sub-neural network is trained. Basic parameters of the target building include, but are not limited to, the type of building, fire protection facilities, building area, and building service life. After the fourth sub-neural network is trained, it can be used to generate corresponding hazardous gas concentration processing results for the hazardous gas concentrations in different areas of the building.

In a specific implementation, the invention provides a smoke concentration processing module, a hazardous gas concentration processing module and a building attribute analysis module, where the modules are respectively composed of a second sub-Neural Network, a third sub-Neural Network and a fourth sub-Neural Network, the second sub-Neural Network, the third sub-Neural Network and the fourth sub-Neural Network may all use Fully Connected Neural networks (Fully Connected Neural networks) with the same Network structure, and are respectively used to analyze smoke concentration data, hazardous gas concentration data and basic parameters of a target building, the input size of the Fully Connected Neural Network is 32 × 1, there are 5 hidden layers in total, the number of neurons in each hidden layer is 6144, and an activation function of neurons in the hidden layers is a Sigmoid function; the number of neurons in the output layer is 588, and the formula is as follows:

Out＝f(a(f(a(f(a(f(a(f(a(x)))))))))) (14)

a(x)＝Wx+b (15)

wherein, formula (14) represents that smoke concentration data, dangerous gas concentration data or basic parameters of a target building are analyzed through a fully-connected neural network; equation (15), a (x), represents the linear calculation of the hidden layer; w is a weight matrix; b is an offset; f (, is the activation function Sigmoid function of hidden layer neurons, Out is the output of the fully-connected neural network.

As described in step S130, the current fire risk level of the target area in the target building corresponding to the current environmental information processing result is determined through the second relationship pre-established by the second neural network.

In an embodiment of the present invention, a specific process of "determining the current fire risk level of the target area in the target building corresponding to the current environmental information processing result through the second relationship pre-established by the second neural network" in step S130 may be further described with reference to the following description.

In an embodiment of the invention, the first model comprises a first neural network and a second neural network, the corresponding relationship comprises a first relationship and a second relationship, and the related information comprises the temperature, the smoke concentration, the hazardous gas concentration, the environment monitoring data and corresponding basic parameters of the target building of the target area.

Establishing, by the second neural network, the second relationship between the presence or absence of the information processing result, the smoke concentration processing result, the hazardous gas concentration processing result, and the basic parameter processing result of the target building and the fire risk level of the target area in the target building, as described in the following steps.

In an embodiment of the present invention, a specific process of the step of establishing, by the second neural network, the second relationship between the information processing result, the smoke concentration processing result, the hazardous gas concentration processing result, and the basic parameter processing result of the target building and the fire risk level of the target area in the target building may be further described in conjunction with the following description.

In an embodiment of the invention, the second neural network comprises a first prediction subnetwork and a second prediction subnetwork, and the second relationship comprises a fifth sub-relationship and a sixth sub-relationship.

Establishing whether the fire source has a fifth sub-relationship among the information processing result, the smoke concentration processing result, the dangerous gas concentration processing result, the basic parameter processing result of the target building and the comprehensive processing result through the first prediction sub-network as described in the following steps.

In a specific implementation, referring to fig. 6, the present invention provides a comprehensive processing module, which mainly comprises a second neural network, wherein the second neural network includes a first prediction subnetwork, the first prediction subnetwork may select an attention mechanism network, and the attention mechanism network includes a self-attention function, so as to achieve an effect of inputting to the neural network whether an information processing result, a smoke concentration processing result, a hazardous gas concentration processing result, and a basic parameter processing result of a target building exist, and then obtaining a comprehensive processing result, wherein a formula of the comprehensive processing module is as follows:

Out_attention＝f_{self-attention}(W_qx,W_kx,W_vx) (16)

wherein, formula (16) indicates that the input x is self-attentive, f_{self-attention}(. X) denotes the self-attention function, W_q、W_k、W_vRespectively representing three different matrixes, and inputting the three different matrixes into the self-attention function after three different linear transformations are performed; formula (17) represents f_{self-attention}The calculation process of (x), softmax (x) represents the softmax function, and d represents the normalization factor.

Whether the fire source has the sum of the information processing result, the smoke concentration processing result, the dangerous gas concentration processing result and the basic parameter processing result of the target building is used as the input of the first prediction sub-network to generate a comprehensive processing result, the network is controlled by the attention force to find more important neurons, further more weight is given to input data, and the performance of the network model can be greatly improved.

As an example, when the first prediction sub-network is trained, the information processing result, the smoke concentration processing result, the dangerous gas concentration processing result and the basic parameter processing result of the target building are collected as sample data, and the first prediction sub-network is trained. After the first prediction subnetwork is trained, the first prediction subnetwork can be used for generating corresponding comprehensive processing results according to the information processing results, smoke concentration processing results, dangerous gas concentration processing results and basic parameter processing results of a target building, wherein the information processing results, the smoke concentration processing results, the dangerous gas concentration processing results and the basic parameter processing results exist in fire sources of different areas in the building.

Establishing, by the second prediction sub-network, a sixth sub-relationship between the integrated processing result and the fire risk level, as described in the following steps.

In a specific implementation, the second Neural Network further includes a second prediction sub-Network, the second prediction sub-Network is a fire risk level prediction Network, which may be a Fully Connected Neural Network (full Connected Neural Network), the input size of the Network is 588 × 1, there are 5 hidden layers in total, the number of neurons in the hidden layers is 6,144, and the activation function of neurons in the hidden layers is a Sigmoid function; the number of neurons in the output layer is 4. The formula is as follows:

Out_grade＝f(a(f(a(f(a(f(a(f(a(Out_attention)))))))))) (18)

a(x)＝Wx+b (19)

Wherein, Out_attentionRepresenting the output of the attention mechanism network; formula (18) represents the prediction of the fire risk level by the output of the attention mechanism network through the fully-connected neural network; out_GradeIndicating a predicted fire risk level. Formula (19) is a (x) represents the linear calculation of a certain hidden layer, W is the weight matrix, b is the bias; f (—) is the Sigmoid function of the activation function of hidden layer neurons.

As an example, when training the second prediction subnetwork, the second prediction subnetwork is trained by collecting the integrated processing results of different areas in the building as sample data. After the second prediction subnetwork is trained, it can be used to synthesize the results of processing different areas in the building to generate corresponding fire risk levels.

In one implementation, the predicted fire risk levels for different areas in a building may be classified into 4 categories, respectively: level 1, safe; grade 2, critical; grade 3, hazardous; grade 4, destruction; the corresponding preventive measures are as follows: grade 1, normal; level 2, implementation of required assurance measures; grade 3, taking emergency measures; 4, level and rectification.

In an embodiment of the present invention, the specific process of "establishing, by the second neural network, whether the information processing result, the smoke concentration processing result, the hazardous gas concentration processing result, and the second relationship between the basic parameter processing result of the target building and the fire risk level of the target area in the target building exist or not" may be further described with reference to the following description.

In an embodiment of the invention, the second neural network comprises a third prediction subnetwork, and the second relationship comprises a seventh sub-relationship; and when the second neural network outputs that the fire risk level is a preset fire risk level, calling the third prediction sub-network.

As an example, the second neural network further comprises a third prediction sub-network, the role of the third prediction sub-network is to predict the current dangerous area coordinates when the second prediction sub-network predicts an area where the target building is at a dangerous level or a damage level. The third prediction sub-network is a dangerous area coordinate prediction network, and a fully-connected neural network with the same network structure as the second prediction sub-network can be selected, and the formula is as follows:

Out_{coordinates of the object}＝f(a(f(a(f(a(f(a(f(a(Out_attention)))))))))) (20)

a(x)＝Wx+b (21)

Wherein, Out_attentionRepresenting the output of the attention mechanism network; equation (20) represents the prediction of the output of the attention mechanism network on the danger zone, Out, via the fully-connected neural network_{Coordinates of the object}Four coordinates representing the predicted hazard zone; formula (21) is a (x) represents the linear calculation of a certain hidden layer, W is the weight matrix, b is the bias; (x) is the hidden layer neuron's activation function Sigmoid function.

As an example, in training the third prediction subnetwork, the fire risk levels of different areas in the target building are collected as sample data and the third prediction subnetwork is trained. After the third prediction subnetwork is trained, it can be used to generate corresponding danger area coordinates for the fire risk levels of different areas in the target building, and the danger areas are marked out with rectangles.

In a specific embodiment, taking two areas of a supervised building as an example, fig. 7a and 7b are the same area of the supervised building, where fig. 7a is a visual schematic diagram of a thermodynamic diagram output by a temperature processing module in an upper computer combined with environment monitoring data, and fig. 7b represents an output of a multimodal neural network when the current area is safe and the fire risk level is 1.

Fig. 8a and 8b are another area of the supervised building, wherein fig. 8a is a visual schematic diagram of the thermodynamic diagram output by the temperature processing module in the lower computer combined with the environment monitoring data, fig. 8b shows the output of the multi-modal attention neural network, when the current area is dangerous and the fire risk level is 4, the dangerous area in the graph is identified by the method provided by the embodiment, and the coordinates predicted by the multi-modal attention neural network are marked by rectangles.

For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.

Referring to fig. 9, a system for dynamically evaluating fire risk based on monitoring of the internet of things according to an embodiment of the present application is shown; the system is applied to establishing a corresponding relation between the relevant information of the building and the corresponding fire risk level through an artificial intelligence model; wherein the artificial intelligence model comprises a first model;

the method specifically comprises the following steps:

a current fire source existence information generating module 910, configured to obtain a current temperature and current environment monitoring data of a target area in a target building, and generate information whether a current fire source exists according to the current temperature and the current environment monitoring data;

a current processing result determining module 920, configured to obtain current environment information of a target area in a target building, and determine, through a first relationship pre-established by the first neural network, a current environment information processing result corresponding to the current environment information; the current environmental information comprises information of whether a current fire source exists, current smoke concentration, current dangerous gas concentration and current basic parameters of a target building; the current environment information processing result comprises an information processing result of whether the current fire source exists, a current smoke concentration processing result, a current dangerous gas concentration processing result and a current basic parameter processing result of the target building;

a current fire risk level determining module 930, configured to determine, through a second relationship pre-established by a second neural network, a current fire risk level of a target area in the target building corresponding to the current environmental information processing result.

In an embodiment of the present invention, the module 910 for generating information whether there is a current fire source includes:

the current temperature processing module is used for acquiring the current temperature of a target area in a target building and determining a current thermodynamic diagram corresponding to the current temperature through a third relation pre-established by the third neural network;

and the current fire source existence information generating module is used for generating current fire source existence information according to the current thermodynamic diagram and the current environment monitoring data.

In an embodiment of the present invention, the current processing result determining module 920 includes:

and the first relation determining module is used for establishing the first relation between the environmental information of the target area in the target building and the environmental information processing result through the first neural network.

In an embodiment of the present invention, the first relationship determining module includes:

the first subrelationship determining submodule is used for establishing a first subrelationship between information whether the fire source exists and information processing results whether the fire source exists through the first subneural network;

the second sub-relation determining sub-module is used for establishing a second sub-relation between the smoke concentration and the smoke concentration processing result through the second sub-neural network;

the third sub-relation determining sub-module is used for establishing a third sub-relation between the concentration of the dangerous gas and the concentration processing result of the dangerous gas through the third sub-neural network;

and the fourth sub-relationship determining sub-module is used for establishing a fourth sub-relationship between the basic parameters of the target building and the basic parameter processing result of the target building through the fourth sub-neural network.

In an embodiment of the present invention, the current fire risk level determining module 930 includes:

and the second relation determining module is used for establishing a second relation between the information processing result, the smoke concentration processing result, the hazardous gas concentration processing result, the basic parameter processing result of the target building and the fire risk level of the target area in the target building through the second neural network.

In an embodiment of the present invention, the second relationship determining module includes:

a fifth sub-relationship determining sub-module, configured to establish, through the first prediction sub-network, a fifth sub-relationship between the information processing result, the smoke concentration processing result, the hazardous gas concentration processing result, and the basic parameter processing result and the comprehensive processing result of the target building;

a sixth sub-relationship determining sub-module for establishing a sixth sub-relationship between the integrated processing result and the fire risk level through the second prediction sub-network.

In an embodiment of the present invention, the module for determining the current fire risk level 930 further includes:

and the seventh sub-relationship determining module is used for determining the current dangerous area coordinate corresponding to the target area with the current fire risk level as the dangerous level or the damage level through a seventh sub-relationship pre-established by the third prediction sub-network when the fire risk level is the dangerous level or the damage level.

Referring to fig. 10, a computer device of the method for dynamically evaluating fire risk based on internet of things monitoring according to the present application is shown, which may specifically include the following:

the computer device 12 described above is embodied in the form of a general purpose computing device, and the components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a memory 28, and a bus 18 that couples various system components including the memory 28 and the processing unit 16.

Bus 18 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, such architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, micro-channel architecture (MAC) bus, enhanced ISA bus, audio Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

Computer device 12 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by computer device 12 and includes both volatile and nonvolatile media, removable and non-removable media.

The memory 28 may include computer system readable media in the form of volatile memory, such as random access memory 30 and/or cache memory 32. Computer device 12 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 34 may be used to read from and write to non-removable, nonvolatile magnetic media (commonly referred to as "hard drives"). Although not shown in FIG. 10, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 by one or more data media interfaces. The memory may include at least one program product having a set (e.g., at least one) of program modules 42, with the program modules 42 configured to carry out the functions of embodiments of the application.

A program/utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules 42, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. Program modules 42 generally perform the functions and/or methodologies of the embodiments described herein.

Computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, camera, etc.), with one or more devices that enable an operator to interact with computer device 12, and/or with any devices (e.g., network card, modem, etc.) that enable computer device 12 to communicate with one or more other computing devices. Such communication may be through the I/O interface 22. Also, computer device 12 may communicate with one or more networks (e.g., a Local Area Network (LAN)), a Wide Area Network (WAN), and/or a public network (e.g., the Internet) via network adapter 20. As shown in FIG. 10, the network adapter 20 communicates with the other modules of the computer device 12 via the bus 18. It should be appreciated that although not shown in FIG. 10, other hardware and/or software modules may be used in conjunction with computer device 12, including but not limited to: microcode, device drivers, redundant processing units 16, external disk drive arrays, RAID systems, tape drives, and data backup storage systems 34, etc.

The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, for example, implementing a method for dynamic fire risk assessment based on internet of things monitoring provided by the embodiments of the present invention.

In an embodiment of the present invention, the present invention further provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor, implements a method for dynamic fire risk assessment based on monitoring of the internet of things as provided in all embodiments of the present application.

Any combination of one or more computer-readable media may be employed. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Computer program code for carrying out operations for aspects of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the operator's computer, partly on the operator's computer, as a stand-alone software package, partly on the operator's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the operator's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.

While preferred embodiments of the present application have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the true scope of the embodiments of the application.

Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.

The method and the system for dynamically evaluating the fire risk based on monitoring of the internet of things are introduced in detail, specific examples are applied in the method for explaining the principle and the implementation mode of the method, and the description of the embodiments is only used for helping to understand the method and the core idea of the method; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Claims

1. A method for dynamically evaluating fire risk based on Internet of things monitoring is characterized in that the method is applied to establishing a corresponding relation between relevant information of a building and a corresponding fire risk level through an artificial intelligence model; wherein the artificial intelligence model comprises a first model; the first model comprises a first neural network and a second neural network, the corresponding relation comprises a first relation and a second relation, and the related information comprises the temperature, the smoke concentration, the hazardous gas concentration, the environment monitoring data and corresponding basic parameters of the target building in the target building;

the method comprises the following steps:

2. The method of claim 1, wherein the first model further comprises a third neural network, and wherein the correspondence further comprises a third relationship; the step of obtaining the current temperature and the current environment monitoring data of a target area in a target building and generating whether the current fire source exists according to the current temperature and the current environment monitoring data comprises the following steps:

3. The method of claim 1, further comprising:

4. The method of claim 3, wherein the first neural network comprises a first sub-neural network, a second sub-neural network, a third sub-neural network, and a fourth sub-neural network; the first relationship comprises a first sub-relationship, a second sub-relationship, a third sub-relationship and a fourth sub-relationship; the step of establishing the first relationship between the environmental information of the target area in the target building and the environmental information processing result through the first neural network includes:

establishing a third sub-relation between the concentration of the dangerous gas and a dangerous gas concentration processing result through the third sub-neural network;

5. The method of claim 1, further comprising:

6. The method of claim 5, wherein the second neural network comprises a first prediction subnetwork and a second prediction subnetwork, and wherein the second relationship comprises a fifth sub-relationship and a sixth sub-relationship; the step of establishing, by the second neural network, whether the second relationship between the information processing result, the smoke concentration processing result, the hazardous gas concentration processing result, the basic parameter processing result of the target building and the fire risk level of the target area in the target building exists or not includes:

7. The method of claim 5, wherein the second neural network further comprises a third prediction sub-network, and wherein the second relationship further comprises a seventh sub-relationship; when the fire risk level output by the second neural network is a preset fire risk level, calling the third prediction sub-network; the method further comprises the following steps:

8. A fire risk dynamic evaluation system based on Internet of things monitoring is characterized in that the system is applied to establishing a corresponding relation between relevant information of a building and a corresponding fire risk level through an artificial intelligence model; wherein the artificial intelligence model comprises a first model;

the system comprises:

and the current fire risk level determining module is used for determining the current fire risk level of the target area in the target building corresponding to the current environmental information processing result through a second relation pre-established by a second neural network.

9. An apparatus comprising a processor, a memory, and a computer program stored on the memory and capable of running on the processor, the computer program when executed by the processor implementing the method for dynamic assessment of fire risk based on internet of things monitoring of any one of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the method for dynamic assessment of fire risk based on internet of things monitoring according to any one of claims 1 to 7.