Building fire hazard risk prediction system and method based on artificial intelligence

Abstract

The invention discloses a building fire hazard risk prediction system and method based on artificial intelligence, and relates to the technical field of building fire hazard risk prediction, and the system comprises the following modules: an internal hidden danger prediction module which obtains the building structure coefficient, the fire source type number and the inflammable material number in a building; obtaining an internal hidden danger coefficient through a building structure coefficient, a fire source type number and an inflammable material number; according to the scheme, the parameters of external factors of the building are comprehensively summarized through the external hidden danger module, so that the conventional and unchanged building parameters and the external parameters which can be changed every day but affect the occurrence of the fire are comprehensively summarized, and the possibility of the occurrence of the fire can be predicted according to the real situation in the building every day; therefore, the accuracy of the fire prediction result can be improved, and misjudgment or missed judgment of the prediction result can be avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of building fire hazard prediction, and in particular to a building fire hazard prediction system and method based on artificial intelligence. Background Art

[0002] The building fire hazard prediction system can use advanced data processing and analysis technology to make efficient and accurate predictions of building fire hazards, so that managers can understand the predicted probability of building fires, and then manage and prevent fires before they occur, which is beneficial to avoid the occurrence of fire hazards.

[0003] Most of the existing building fire prediction systems can only retrieve prediction parameters through conventional facilities in the building, such as smoke sensors, temperature sensors, and the location of fire passages in the building layout, etc. However, in real life, the conditions in buildings are diverse and changeable. Predicting the occurrence of fires solely through conventional facilities and layout conditions in buildings is seriously lacking in accuracy, resulting in frequent misjudgments and missed judgments. Moreover, it is difficult for managers to formulate evacuation plans within a limited time after receiving the fire prediction results. Therefore, the use effect of this fire prediction system in actual use is not ideal. In order to facilitate more accurate and intelligent prediction of the occurrence of building fires, and to facilitate the timely formulation of evacuation plans when a fire occurs, so as to assist managers in timely evacuating people in the building and avoid economic losses and casualties caused by fires, we propose a building fire hazard prediction system and method based on artificial intelligence. Summary of the invention

[0004] 1. Technical issues to be resolved In view of the deficiencies in the prior art, the present invention provides a building fire hazard prediction system and method based on artificial intelligence to solve the above-mentioned problems in the prior art.

[0005] (II) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: a building fire hazard prediction system based on artificial intelligence, comprising the following modules: The internal hidden danger prediction module obtains the building structure coefficient, the number of fire source types and the number of flammable materials in the building, and obtains the internal hidden danger coefficient through the building structure coefficient, the number of fire source types and the number of flammable materials; The external hidden danger prediction module obtains the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, obtains the external hidden danger coefficient through the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, and obtains the building hidden danger value by weighted summation of the external hidden danger coefficient and the internal hidden danger coefficient; The fire score prediction module establishes a historical building fire database, captures the historical building hidden danger value closest to the building hidden danger value on the day in the historical building fire database, and obtains the fire prediction score on the day based on the historical building hidden danger value; The fire simulation evacuation module defines a preset threshold for the fire prediction score, compares the preset threshold for the fire prediction score with the fire prediction score of the day, and if the fire prediction score of the day is greater than the preset threshold for the fire prediction score, a simulated evacuation plan is formulated and fed back to the management personnel.

[0006] Preferably, the building structure coefficient in the building is obtained in the internal hidden danger prediction module, specifically: Get the number of building floors, the building height, the building area, the size of the fire passage, and the size of the vent, and get the building structure coefficient through the number of building floors, the building height, the building area, the size of the fire passage, and the size of the vent; The calculation method of the building structure coefficient is as follows:

[0007] Among them, Jz represents the building structure coefficient, Lc represents the number of building floors, Jd represents the building height, Jm represents the building area, Xf represents the size of the fire escape, and Tf represents the size of the ventilation opening. , , , and are weights, , , , , .

[0008] Preferably, the calculation method of the internal hidden danger coefficient in the internal hidden danger prediction module is specifically as follows:

[0009] Among them, Nh represents the internal coefficient, Jz represents the building structure coefficient, Hy represents the number of fire source types, and Yr represents the number of flammable materials. , and are weights, , , .

[0010] Preferably, the building personnel coefficient is obtained in the external hidden danger prediction module, specifically: Get the total number of people in the building, get the number of permanent people in the building, get the number of visiting people by subtracting the total number of people in the building from the number of permanent people, and get the building personnel coefficient by the total number of people in the building and the number of visiting people; The calculation method of the building personnel coefficient is as follows:

[0011] Among them, Rx represents the building personnel coefficient, Zs represents the total number of people in the building, and Cz represents the number of visiting personnel. and are weights, , .

[0012] Preferably, the security inspection score is obtained in the external hidden danger prediction module, specifically: Step 1: formulate N different security inspection routes, select N security inspection personnel, have the N security inspection personnel conduct security inspections along the N security inspection routes, and score the inspection results after the inspections; Step 2: Sum and average multiple scoring results to obtain a reference security inspection score, obtain the security inspection scores of the previous two days, and sum and average the security inspection scores of the previous two days and the reference security inspection scores to obtain the security inspection score of the day.

[0013] Preferably, the calculation method of the external hidden danger coefficient in the external hidden danger prediction module is specifically as follows:

[0014] Among them, Wy represents the external hidden danger coefficient, Wd represents the weather temperature on the day, Sd represents the air humidity on the day, Rx represents the building personnel coefficient on the day, and Ax represents the security inspection score on the day. , and are weights, , , .

[0015] Preferably, the fire score prediction module is specifically: Step 1: Establish a historical building fire database, obtain the building hidden danger value, the external hidden danger coefficient and the internal hidden danger coefficient of the day, arrange the historical building hidden danger values ​​in the historical building fire database in ascending order, and capture the historical building hidden danger value closest to the building hidden danger value of the day in the historical building fire database, obtain the external hidden danger coefficient of the historical building hidden danger value on the day, mark it as the historical external hidden danger coefficient, obtain the internal hidden danger coefficient of the historical building hidden danger value on the day, mark it as the historical internal hidden danger coefficient; Step 2: Compare the size relationship between the historical external hidden danger coefficient and the external hidden danger coefficient of the day to obtain the external hidden danger comparison result. If the external hidden danger coefficient of the day is greater than the historical external hidden danger coefficient, the external hidden danger comparison result is marked as 2. If the external hidden danger coefficient of the day is less than the historical external hidden danger coefficient, the external hidden danger comparison result is marked as 0.5. Compare the size relationship between the historical internal hidden danger coefficient and the internal hidden danger coefficient of the day to obtain the internal hidden danger comparison result. If the internal hidden danger coefficient of the day is greater than the historical internal hidden danger coefficient, the internal hidden danger comparison result is marked as 1.5. If the internal hidden danger coefficient of the day is less than the historical internal hidden danger coefficient, the internal hidden danger comparison result is marked as 0.7. Step 3: Define the preset threshold of the building hidden danger value for the day, and obtain the fire prediction score for the day through the preset threshold of the building hidden danger value for the day, the building hidden danger value for the day, the external hidden danger comparison result and the internal hidden danger comparison result.

[0016] Preferably, the calculation method of the fire prediction score for the day is as follows:

[0017] Among them, Pf represents the fire prediction score for the day, Ys represents the preset threshold of the building hidden danger value for the day, Jh represents the preset threshold of the building hidden danger value for the day, Wj represents the external hidden danger comparison result, and Nj represents the internal hidden danger comparison result. .

[0018] Preferably, a simulated evacuation plan is formulated in the fire simulation evacuation module and fed back to the management personnel, specifically: Step 1: Obtain the building layout, the location of the building's emergency exits, and the distribution of people in the building, and use the genetic algorithm to develop three shortest evacuation routes based on the building layout, the location of the building's emergency exits, and the distribution of people in the building; Step 2: Obtain the age of people in the building, the safety knowledge of people in the building, and the building personnel coefficient. Use the network model to predict the evacuation time required for the three shortest routes based on the age of people in the building, the safety knowledge of people in the building, the building personnel coefficient, and the three shortest evacuation routes; Step 3: Compare the size relationships of the three evacuation times, select the evacuation route with the shortest evacuation time and mark it as the optimal evacuation route, and feed back the optimal evacuation route to the management personnel.

[0019] A method for predicting building fire hazards based on artificial intelligence comprises the following steps: Step 1: Obtain the building structure coefficient, the number of fire source types and the number of flammable materials in the building, and obtain the internal hidden danger coefficient through the building structure coefficient, the number of fire source types and the number of flammable materials; Step 2: Obtain the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, obtain the external hidden danger coefficient through the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, and obtain the building hidden danger value by weighted summation of the external hidden danger coefficient and the internal hidden danger coefficient; Step 3: Establish a historical building fire database, capture the historical building hidden danger value closest to the building hidden danger value on the day in the historical building fire database, and obtain the fire prediction score on the day through the historical building hidden danger value; Step 4: Define the preset threshold of the fire prediction score, compare the preset threshold of the fire prediction score with the fire prediction score of the day, and if the fire prediction score of the day is greater than the preset threshold of the fire prediction score, formulate a simulated evacuation plan and feedback it to the management personnel.

[0020] (III) Beneficial effects The present invention provides a building fire risk prediction system and method based on artificial intelligence, which has the following beneficial effects: (1) In this scheme, the parameters of the external factors of the building are comprehensively summarized through the external hidden danger module. The external hidden danger coefficient is obtained through the weather temperature, air humidity, building personnel coefficient and security inspection score of the day. The internal hidden danger coefficient and the external hidden danger coefficient are comprehensively summarized to obtain the building hidden danger value. This makes it easier to comprehensively summarize the conventional unchanging building parameters and the external parameters that change every day but affect the occurrence of fire. It is then easier to predict the possibility of fire based on the actual situation in the building every day, which is beneficial to improve the accuracy of fire prediction results and avoid misjudgment or omission of prediction results.

[0021] (2) In this scheme, by combining the historical fire data of the building with the building hidden danger value of the day for comparison, the historical building hidden danger value closest to the building hidden danger value of the day is captured in the building historical fire database, and the historical external hidden danger coefficient and the historical internal hidden danger coefficient are obtained respectively. Then, the internal hidden danger comparison result and the external hidden danger comparison result are determined by comparing the historical external hidden danger coefficient, the historical internal hidden danger coefficient, the external hidden danger coefficient of the day and the internal hidden danger coefficient of the day. Finally, the fire prediction score of the day is calculated, so that the fire prediction score can be fed back to the management personnel, so that the management personnel can understand the fire prediction situation in the building on the day, which is further beneficial to combine historical data to more accurately predict the possibility of fire, and further help to avoid misjudgment or omission of prediction results.

[0022] (3) In this scheme, a simulated evacuation drill is conducted through the fire simulation evacuation module. When the fire prediction score for the day is greater than the preset threshold of the fire prediction score, an evacuation plan is formulated and the judgment result and the evacuation plan are fed back to the management personnel, so that the management personnel can check the building before the fire. At the same time, it is also convenient to evacuate the crowd according to the evacuation plan as soon as possible after the fire occurs, thereby saving precious evacuation time for the management personnel to formulate an evacuation plan after the fire occurs, which is more beneficial to assist the management personnel in evacuating the people in the building, thereby helping to avoid economic losses and casualties caused by the fire. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a schematic diagram of the structure of a building fire hazard prediction system based on artificial intelligence according to the present invention; Figure 2 The present invention is a flowchart of a method for predicting building fire hazards based on artificial intelligence. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] See also Figure 1-Figure 2 The present invention provides a building fire hazard prediction system based on artificial intelligence, comprising the following modules: The internal hidden danger prediction module obtains the building structure coefficient, the number of fire source types and the number of flammable materials in the building, and obtains the internal hidden danger coefficient through the building structure coefficient, the number of fire source types and the number of flammable materials; The external hidden danger prediction module obtains the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, obtains the external hidden danger coefficient through the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, and obtains the building hidden danger value by weighted summation of the external hidden danger coefficient and the internal hidden danger coefficient; The fire score prediction module establishes a historical building fire database, captures the historical building hidden danger value closest to the building hidden danger value on the day in the historical building fire database, and obtains the fire prediction score on the day based on the historical building hidden danger value; The fire simulation evacuation module defines a preset threshold for the fire prediction score, compares the preset threshold for the fire prediction score with the fire prediction score of the day, and if the fire prediction score of the day is greater than the preset threshold for the fire prediction score, formulates a simulated evacuation plan and feeds it back to the management personnel; The calculation method of the internal hidden danger coefficient in the internal hidden danger prediction module is as follows:

[0026] Among them, Nh represents the internal coefficient, Jz represents the building structure coefficient, Hy represents the number of fire source types, and Yr represents the number of flammable materials. , and are weights, , , ; The calculation method of the external hidden danger coefficient in the external hidden danger prediction module is as follows:

[0027] Among them, Wy represents the external hidden danger coefficient, Wd represents the weather temperature on the day, Sd represents the air humidity on the day, Rx represents the building personnel coefficient on the day, and Ax represents the security inspection score on the day. , and are weights, , , .

[0028] In this embodiment, the conventional parameters in the building are comprehensively summarized by the internal hidden danger prediction module, and the internal hidden danger coefficient is calculated by the building structure, the number of fire source types and the number of flammable materials. The possibility of fire occurrence is predicted by the internal hidden danger coefficient. In this scheme, the parameters of the external factors of the building are comprehensively summarized through the external hidden danger module, and the external hidden danger coefficient is obtained through the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, and then the internal hidden danger coefficient and the external hidden danger coefficient are comprehensively summarized to obtain the building hidden danger value, so as to facilitate the comprehensive summary of the conventional unchanged building parameters and the external parameters that change every day but affect the occurrence of fire, and then facilitate the prediction of the possibility of fire according to the actual situation in the building every day, which is beneficial to improve the accuracy of the fire prediction results, and then help to avoid the situation of misjudgment or omission of the prediction results; In this scheme, the possibility of fire is predicted through the fire scoring prediction module, and the historical fire data of the building and the building hidden danger value of the day are combined for comparison, which is further beneficial to more accurately predict the possibility of fire by combining historical data, and further helps to avoid misjudgment or omission of prediction results; In this scheme, a simulated evacuation drill is conducted through the fire simulation evacuation module. When the fire prediction score of the day is greater than the preset threshold of the fire prediction score, an evacuation plan is formulated and the judgment result and the evacuation plan are fed back to the management personnel, so that the management personnel can check the building before the fire. At the same time, it is also convenient to evacuate the crowd according to the evacuation plan as soon as possible after the fire occurs, thereby saving precious evacuation time for the management personnel to formulate an evacuation plan after the fire occurs, which is more beneficial to assist the management personnel to evacuate the people in the building, thereby helping to avoid economic losses and casualties caused by the fire; It is worth mentioning that the weight values ​​in this scheme can be obtained through the hierarchical analysis method, and the values ​​of the preset thresholds in this scheme can be obtained through the weight analysis method, which will not be elaborated here.

[0029] In the internal hidden danger prediction module, the building structure coefficient in the building is obtained, specifically: Get the number of building floors, the building height, the building area, the size of the fire passage, and the size of the vent, and get the building structure coefficient through the number of building floors, the building height, the building area, the size of the fire passage, and the size of the vent; The calculation method of the building structure coefficient is as follows:

[0030] Among them, Jz represents the building structure coefficient, Lc represents the number of building floors, Jd represents the building height, Jm represents the building area, Xf represents the size of the fire escape, and Tf represents the size of the ventilation opening. , , , and are weights, , , , , .

[0031] In this embodiment, the building structure coefficient is calculated by the number of building floors, building height, building area, size of fire escape passages and size of ventilation openings, so that the building structure coefficient can be obtained by comprehensively summarizing the actual situation of the building and the conditions that affect fire evacuation, thereby further facilitating the improvement of the accuracy of fire prediction results, and thus helping to avoid misjudgment or omission of prediction results.

[0032] The building personnel coefficient is obtained in the external hidden danger prediction module, specifically: Get the total number of people in the building, get the number of permanent people in the building, get the number of visiting people by subtracting the total number of people in the building from the number of permanent people, and get the building personnel coefficient by the total number of people in the building and the number of visiting people; The calculation method of the building personnel coefficient is as follows:

[0033] Among them, Rx represents the building personnel coefficient, Zs represents the total number of people in the building, and Cz represents the number of visiting personnel. and are weights, , .

[0034] In this embodiment, since the number of people in the building affects the evacuation of the fire and the possibility of fire, and the permanent staff are more familiar with the building layout, it is more convenient to evacuate after the fire occurs. Therefore, the number of permanent staff, the number of visiting staff and the total number of people in the building are combined to make it easier to improve the accuracy of the fire prediction results, which is beneficial to avoid misjudgment or omission of the prediction results.

[0035] Obtain security inspection scores in the external hidden danger prediction module, specifically: Step 1: formulate N different security inspection routes, select N security inspection personnel, have the N security inspection personnel conduct security inspections along the N security inspection routes, and score the inspection results after the inspections; Step 2: Sum and average multiple scoring results to obtain a reference security inspection score, obtain the security inspection scores of the previous two days, and sum and average the security inspection scores of the previous two days and the reference security inspection scores to obtain the security inspection score of the day.

[0036] In this embodiment, N security patrol personnel conduct security inspections along N security patrol routes respectively, and score the inspection results after the inspection, so as to facilitate the recording and scoring of the actual situation in the building on that day, which is beneficial to avoid the influence of changing external unstable factors on the accuracy of fire prediction results. Finally, the security patrol score of the day is obtained by summing and averaging the security patrol scores of the previous two days and the reference security patrol scores, which is beneficial to data cleaning of the scoring results, and further beneficial to improving the accuracy of the security patrol scores.

[0037] Fire score prediction module, specifically: Step 1: Establish a historical building fire database, obtain the building hidden danger value, the external hidden danger coefficient and the internal hidden danger coefficient of the day, arrange the historical building hidden danger values ​​in the historical building fire database in ascending order, and capture the historical building hidden danger value closest to the building hidden danger value of the day in the historical building fire database, obtain the external hidden danger coefficient of the historical building hidden danger value on the day, mark it as the historical external hidden danger coefficient, obtain the internal hidden danger coefficient of the historical building hidden danger value on the day, mark it as the historical internal hidden danger coefficient; Step 2: Compare the size relationship between the historical external hidden danger coefficient and the external hidden danger coefficient of the day to obtain the external hidden danger comparison result. If the external hidden danger coefficient of the day is greater than the historical external hidden danger coefficient, the external hidden danger comparison result is marked as 2. If the external hidden danger coefficient of the day is less than the historical external hidden danger coefficient, the external hidden danger comparison result is marked as 0.5. Compare the size relationship between the historical internal hidden danger coefficient and the internal hidden danger coefficient of the day to obtain the internal hidden danger comparison result. If the internal hidden danger coefficient of the day is greater than the historical internal hidden danger coefficient, the internal hidden danger comparison result is marked as 1.5. If the internal hidden danger coefficient of the day is less than the historical internal hidden danger coefficient, the internal hidden danger comparison result is marked as 0.7. Step 3: define the preset threshold of the building hidden danger value on the day, and obtain the fire prediction score on the day through the preset threshold of the building hidden danger value on the day, the building hidden danger value on the day, the external hidden danger comparison result and the internal hidden danger comparison result; The calculation method of the fire prediction score for the day is as follows:

[0038] Among them, Pf represents the fire prediction score for the day, Ys represents the preset threshold of the building hidden danger value for the day, Jh represents the preset threshold of the building hidden danger value for the day, Wj represents the external hidden danger comparison result, and Nj represents the internal hidden danger comparison result. .

[0039] In this embodiment, by combining and comparing the historical fire data of the building with the building hidden danger value of the day, the historical building hidden danger value closest to the building hidden danger value of the day is captured in the historical building fire database, and the historical external hidden danger coefficient and the historical internal hidden danger coefficient are obtained respectively. Then, the internal hidden danger comparison result and the external hidden danger comparison result are determined by comparing the historical external hidden danger coefficient, the historical internal hidden danger coefficient, the external hidden danger coefficient of the day and the internal hidden danger coefficient of the day, and finally the fire prediction score of the day is calculated, so as to facilitate the feedback of the fire prediction score to the management personnel, so that the management personnel can understand the fire prediction situation in the building on the day, which is further beneficial to combine historical data to more accurately predict the possibility of fire, and further help to avoid misjudgment or omission of prediction results.

[0040] In the fire simulation evacuation module, a simulated evacuation plan is developed and fed back to the management personnel, specifically: Step 1: Obtain the building layout, the location of the building's emergency exits, and the distribution of people in the building, and use the genetic algorithm to develop three shortest evacuation routes based on the building layout, the location of the building's emergency exits, and the distribution of people in the building; Step 2: Obtain the age of people in the building, the safety knowledge of people in the building, and the building personnel coefficient. Use the network model to predict the evacuation time required for the three shortest routes based on the age of people in the building, the safety knowledge of people in the building, the building personnel coefficient, and the three shortest evacuation routes; Step 3: Compare the size relationships of the three evacuation times, select the evacuation route with the shortest evacuation time and mark it as the optimal evacuation route, and feed back the optimal evacuation route to the management personnel.

[0041] In this embodiment, three shortest evacuation routes are formulated by using a genetic algorithm according to the building layout, the location of the building's emergency exits, and the distribution of people in the building, so that it is convenient to select the most suitable evacuation route according to the building conditions and the personnel conditions. The network model is used to predict the evacuation time required for the three shortest routes according to the age of the people in the building, the safety knowledge mastery of the people in the building, the building's personnel coefficient, and the three shortest evacuation routes, so that it is convenient to select a route with the shortest evacuation time as a backup evacuation route among the three shortest evacuation routes in combination with the personnel conditions. When a fire occurs, it is convenient for management personnel to evacuate people in the building through this route, and it is more convenient to formulate an evacuation plan for management personnel after a fire occurs, saving precious evacuation time, and it is more beneficial to assist management personnel in evacuating people in the building, and it is beneficial to avoid economic losses and casualties caused by fire. It is worth mentioning that the network model and genetic algorithm in this scheme can be generated through training, which will not be elaborated in detail here.

[0042] See also Figure 1-Figure 2 The present invention provides a method for predicting building fire hazards based on artificial intelligence, comprising the following steps: Step 1: Obtain the building structure coefficient, the number of fire source types and the number of flammable materials in the building, and obtain the internal hidden danger coefficient through the building structure coefficient, the number of fire source types and the number of flammable materials; Step 2: Obtain the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, obtain the external hidden danger coefficient through the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, and obtain the building hidden danger value by weighted summation of the external hidden danger coefficient and the internal hidden danger coefficient; Step 3: Establish a historical building fire database, capture the historical building hidden danger value closest to the building hidden danger value on the day in the historical building fire database, and obtain the fire prediction score on the day through the historical building hidden danger value; Step 4: Define the preset threshold of the fire prediction score, compare the preset threshold of the fire prediction score with the fire prediction score of the day, and if the fire prediction score of the day is greater than the preset threshold of the fire prediction score, formulate a simulated evacuation plan and feedback it to the management personnel.

[0043] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0044] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0045] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A building fire hazard prediction system based on artificial intelligence, characterized in that: Includes the following modules: The internal hidden danger prediction module obtains the building structure coefficient, the number of fire source types and the number of flammable materials in the building, and obtains the internal hidden danger coefficient through the building structure coefficient, the number of fire source types and the number of flammable materials; The external hidden danger prediction module obtains the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, obtains the external hidden danger coefficient through the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, and obtains the building hidden danger value by weighted summation of the external hidden danger coefficient and the internal hidden danger coefficient; The fire score prediction module establishes a historical building fire database, captures the historical building hidden danger value closest to the building hidden danger value on the day in the historical building fire database, and obtains the fire prediction score on the day based on the historical building hidden danger value; The fire simulation evacuation module defines a preset threshold for the fire prediction score, compares the preset threshold for the fire prediction score with the fire prediction score of the day, and if the fire prediction score of the day is greater than the preset threshold for the fire prediction score, a simulated evacuation plan is formulated and fed back to the management personnel.

2. The artificial intelligence-based building fire hazard prediction system according to claim 1 is characterized in that: In the internal hidden danger prediction module, the building structure coefficient in the building is obtained, specifically: Get the number of building floors, the building height, the building area, the size of the fire passage, and the size of the vent, and get the building structure coefficient through the number of building floors, the building height, the building area, the size of the fire passage, and the size of the vent; The calculation method of the building structure coefficient is as follows: ； Among them, Jz represents the building structure coefficient, Lc represents the number of building floors, Jd represents the building height, Jm represents the building area, Xf represents the size of the fire escape, and Tf represents the size of the ventilation opening. , , , and are weights, , , , , .

3. The artificial intelligence-based building fire hazard prediction system according to claim 1 is characterized in that: The calculation method of the internal hidden danger coefficient in the internal hidden danger prediction module is as follows: ； Among them, Nh represents the internal coefficient, Jz represents the building structure coefficient, Hy represents the number of fire source types, and Yr represents the number of flammable materials. , and are weights, , , .

4. The artificial intelligence-based building fire hazard prediction system according to claim 1 is characterized in that: The building personnel coefficient is obtained in the external hidden danger prediction module, specifically: Get the total number of people in the building, get the number of permanent people in the building, get the number of visiting people by subtracting the total number of people in the building from the number of permanent people, and get the building personnel coefficient by the total number of people in the building and the number of visiting people; The calculation method of the building personnel coefficient is as follows: ； Among them, Rx represents the building personnel coefficient, Zs represents the total number of people in the building, and Cz represents the number of visiting personnel. and are weights, , .

5. The artificial intelligence-based building fire hazard prediction system according to claim 1 is characterized in that: Obtain security inspection scores in the external hidden danger prediction module, specifically: Step 1: formulate N different security inspection routes, select N security inspection personnel, have the N security inspection personnel conduct security inspections along the N security inspection routes, and score the inspection results after the inspections; Step 2: Sum and average multiple scoring results to obtain a reference security inspection score, obtain the security inspection scores of the previous two days, and sum and average the security inspection scores of the previous two days and the reference security inspection scores to obtain the security inspection score of the day.

6. The artificial intelligence-based building fire hazard prediction system according to claim 1 is characterized in that: The calculation method of the external hidden danger coefficient in the external hidden danger prediction module is as follows: ； Among them, Wy represents the external hidden danger coefficient, Wd represents the weather temperature on the day, Sd represents the air humidity on the day, Rx represents the building personnel coefficient on the day, and Ax represents the security inspection score on the day. , and are weights, , , .

7. The artificial intelligence-based building fire hazard prediction system according to claim 1 is characterized in that: Fire score prediction module, specifically: Step 1: Establish a historical building fire database, obtain the building hidden danger value, the external hidden danger coefficient and the internal hidden danger coefficient of the day, arrange the historical building hidden danger values ​​in the historical building fire database in ascending order, and capture the historical building hidden danger value closest to the building hidden danger value of the day in the historical building fire database, obtain the external hidden danger coefficient of the historical building hidden danger value on the day, mark it as the historical external hidden danger coefficient, obtain the internal hidden danger coefficient of the historical building hidden danger value on the day, mark it as the historical internal hidden danger coefficient; Step 2: Compare the size relationship between the historical external hidden danger coefficient and the external hidden danger coefficient of the day to obtain the external hidden danger comparison result. If the external hidden danger coefficient of the day is greater than the historical external hidden danger coefficient, the external hidden danger comparison result is marked as 2. If the external hidden danger coefficient of the day is less than the historical external hidden danger coefficient, the external hidden danger comparison result is marked as 0.

5. Compare the size relationship between the historical internal hidden danger coefficient and the internal hidden danger coefficient of the day to obtain the internal hidden danger comparison result. If the internal hidden danger coefficient of the day is greater than the historical internal hidden danger coefficient, the internal hidden danger comparison result is marked as 1.

5. If the internal hidden danger coefficient of the day is less than the historical internal hidden danger coefficient, the internal hidden danger comparison result is marked as 0.

7. Step 3: Define the preset threshold of the building hidden danger value for the day, and obtain the fire prediction score for the day through the preset threshold of the building hidden danger value for the day, the building hidden danger value for the day, the external hidden danger comparison result and the internal hidden danger comparison result.

8. The artificial intelligence-based building fire hazard prediction system according to claim 7 is characterized in that: The calculation method of the fire prediction score for the day is as follows: ； Among them, Pf represents the fire prediction score for the day, Ys represents the preset threshold of the building hidden danger value for the day, Jh represents the preset threshold of the building hidden danger value for the day, Wj represents the external hidden danger comparison result, and Nj represents the internal hidden danger comparison result. .

9. The artificial intelligence-based building fire hazard prediction system according to claim 1, characterized in that: In the fire simulation evacuation module, a simulated evacuation plan is developed and fed back to the management personnel, specifically: Step 1: Obtain the building layout, the location of the building's emergency exits, and the distribution of people in the building, and use the genetic algorithm to develop three shortest evacuation routes based on the building layout, the location of the building's emergency exits, and the distribution of people in the building; Step 2: Obtain the age of people in the building, the safety knowledge of people in the building, and the building personnel coefficient. Use the network model to predict the evacuation time required for the three shortest routes based on the age of people in the building, the safety knowledge of people in the building, the building personnel coefficient, and the three shortest evacuation routes; Step 3: Compare the size relationships of the three evacuation times, select the evacuation route with the shortest evacuation time and mark it as the optimal evacuation route, and feed back the optimal evacuation route to the management personnel.

10. A method for predicting building fire hazards based on artificial intelligence, applied to a system for predicting building fire hazards based on artificial intelligence as claimed in any one of claims 1 to 9, characterized in that: The steps include: Step 1: Obtain the building structure coefficient, the number of fire source types and the number of flammable materials in the building, and obtain the internal hidden danger coefficient through the building structure coefficient, the number of fire source types and the number of flammable materials; Step 2: Obtain the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, obtain the external hidden danger coefficient through the weather temperature, air humidity, building personnel coefficient and security inspection score of the day, and obtain the building hidden danger value by weighted summation of the external hidden danger coefficient and the internal hidden danger coefficient; Step 3: Establish a historical building fire database, capture the historical building hidden danger value closest to the building hidden danger value on the day in the historical building fire database, and obtain the fire prediction score on the day through the historical building hidden danger value; Step 4: Define the preset threshold of the fire prediction score, compare the preset threshold of the fire prediction score with the fire prediction score of the day, and if the fire prediction score of the day is greater than the preset threshold of the fire prediction score, formulate a simulated evacuation plan and feedback it to the management personnel.

Images