Residential building fire area prediction method, system, storage medium and device

Abstract

The application discloses a kind of residential building fire area prediction method, system, storage medium and device, the method includes based on fire heat release rate and ventilation control type combustion mode, the fire area prediction model of residential building is established;According to the combustion parameter of fire area prediction model determination burnt area;Obtain the fire history data of combustion parameter, determine the probability distribution of combustion parameter according to fire history data;The probability distribution of combustion parameter is randomly sampled to obtain the combustion parameter corresponding to multiple groups of different probabilities;Based on the fire area prediction model and the combustion parameter corresponding to multiple groups of different probabilities, Monte Carlo simulation is carried out to obtain the cumulative probability distribution of residential building's burnt area;According to the cumulative probability distribution of burnt area, determine the fire risk level of residential building.The application can accurately predict the fire area of residential building, determine the fire risk level, so as to formulate effective fire prevention and control measures.

Description

Methods, systems, storage media and devices for predicting the fire area of ​​residential buildings in fires Technical Field

[0001] This invention relates to the field of residential building fire prediction technology, and more specifically, to a method, system, storage medium, and device for predicting the burned area of ​​a residential building fire. Background Technology

[0002] Residential building fires are a common type of fire in my country, accounting for about 30% of all fires, but causing over 70% of all fire-related deaths. They are characterized by rapid spread, strong localized damage, high threat level, and suddenness. Once a residential building fire occurs, it will cause serious casualties, enormous property losses, and severe social impact.

[0003] The burned area refers to the extent affected by the high temperature of a fire, and it is an important factor in assessing the fire risk of residential buildings. However, due to the uncertainty of residential building fires, it is impossible to accurately predict the burned area of ​​a residential building fire, thus making it impossible to determine the fire risk level of the residential building and to formulate effective fire prevention and control measures for residential buildings.

[0004] In view of this, it is necessary to design a method, system, storage medium and device for predicting the fire area of ​​residential buildings, to accurately predict the fire area of ​​residential buildings through Monte Carlo simulation, to determine the fire risk level of residential buildings based on the fire area, and thus to formulate effective fire prevention and control measures. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, storage medium, and device for predicting the fire-affected area of ​​residential buildings. By using Monte Carlo simulation, the fire-affected area of ​​residential buildings can be accurately predicted, and the fire risk level of residential buildings can be determined based on the fire-affected area, thereby enabling the formulation of effective fire prevention and control measures.

[0006] To achieve the above objectives, the present invention provides a method for predicting the fire-affected area of ​​residential buildings in fires, comprising the following steps:

[0007] A fire area prediction model for residential buildings is established based on fire heat release rate and ventilation-controlled combustion mode.

[0008] The combustion parameters of the burned area are determined based on the fire burned area prediction model;

[0009] Obtain historical fire data of combustion parameters and determine the probability distribution of combustion parameters based on the historical fire data;

[0010] Random sampling is performed on the probability distribution of combustion parameters to obtain multiple sets of combustion parameters corresponding to different probabilities;

[0011] Based on the fire area prediction model and multiple sets of combustion parameters corresponding to different probabilities, Monte Carlo simulation is performed to obtain the cumulative probability distribution of fire area of ​​residential buildings.

[0012] The fire risk level of a residential building is determined based on the cumulative probability distribution of the burned area. Specifically, if 0 ≤ cumulative burned area ≤ 5m², the fire risk level is determined accordingly. 2 If the fire risk level is 5m, then the fire risk level is determined to be low risk; 2 <Cumulative fire area ≤10m² 2 If the fire risk level is determined to be medium risk I; if 10m 2 <Cumulative fire area ≤20m² 2 If the fire risk level is determined to be medium risk II; if 20m 2 <Cumulative fire area ≤50m² 2 If the fire risk level is determined to be severe risk I; if 50m 2 <Cumulative fire area ≤100m² 2 If the fire risk level is determined to be severe risk II; if 100m 2 If the cumulative burned area is less than the fire risk level, the fire risk level is determined to be severe risk III.

[0013] By employing the technical solution disclosed in this invention, the fire-affected area of ​​a residential building can be accurately predicted, and the fire risk level of the residential building can be determined based on the fire-affected area, thereby enabling the formulation of effective fire prevention and control measures.

[0014] The aforementioned method for predicting the burned area of ​​residential building fires uses combustion parameters including fire response time, fire growth coefficient, and fire load density.

[0015] The aforementioned method for predicting the burned area of ​​residential building fires, which involves obtaining historical fire data of combustion parameters and determining the probability distribution of these parameters based on that data, further includes the following steps:

[0016] Historical fire data on combustion parameters were preprocessed using data cleaning methods.

[0017] Based on the fire history data of the pre-processed combustion parameters, the probability density function is obtained through statistical methods.

[0018] The aforementioned method for predicting the burned area of ​​residential buildings in a fire, further includes the step of obtaining the probability density function through statistical methods, namely:

[0019] Construct a probability density distribution map of combustion parameters based on historical fire data to determine the probability distribution that the combustion parameters follow;

[0020] The probability density function of combustion parameters is obtained by fitting historical fire data.

[0021] The aforementioned method for predicting the fire-affected area of ​​residential buildings, including the step of performing Monte Carlo simulation to obtain the cumulative probability distribution of the fire-affected area, further includes:

[0022] Based on the combustion parameters corresponding to multiple sets of different probabilities, multiple fire area calculation results are obtained through the fire area prediction model.

[0023] Based on the calculation results of multiple burned areas, the cumulative distribution function was obtained through statistical methods.

[0024] The aforementioned method for predicting the burned area of ​​residential buildings in a fire, further includes the step of obtaining the cumulative distribution function through statistical methods, namely:

[0025] Based on the calculation results of multiple burned areas, a probability density distribution map of the burned area is constructed to determine the probability distribution that the burned area follows.

[0026] Data fitting is performed based on multiple fire area calculation results to obtain the probability density function of the fire area;

[0027] Based on the probability density function of the burned area, the cumulative distribution function of the burned area is obtained, and the cumulative distribution map of the burned area is constructed.

[0028] The aforementioned method for predicting the burned area of ​​residential buildings also includes developing corresponding fire prevention and control measures based on the fire risk level of the residential building.

[0029] To better achieve the objectives of this invention, this invention also provides a residential building fire area prediction system, comprising:

[0030] The fire area modeling module is used to establish a fire area prediction model for residential buildings based on the fire heat release rate and ventilation-controlled combustion mode.

[0031] The combustion parameter confirmation module is used to determine the combustion parameters of the burned area based on the fire burned area prediction model.

[0032] The combustion parameter probability modeling module is used to acquire historical fire data of combustion parameters and determine the probability distribution of combustion parameters based on the historical fire data.

[0033] The random sampling module is used to randomly sample the probability distribution of combustion parameters to obtain multiple sets of combustion parameters corresponding to different probabilities.

[0034] The fire area probability modeling module is used to perform Monte Carlo simulations based on the fire area prediction model and multiple sets of combustion parameters corresponding to different probabilities to obtain the cumulative probability distribution of the fire area of ​​residential buildings.

[0035] The fire risk level confirmation module is used to determine the fire risk level of residential buildings based on the cumulative probability distribution of burned area. Specifically, if 0 ≤ cumulative burned area ≤ 5m², the fire risk level is determined accordingly. 2 If the fire risk level is 5m, then the fire risk level is determined to be low risk; 2 <Cumulative fire area ≤10m² 2 If the fire risk level is determined to be medium risk I; if 10m 2 <Cumulative fire area ≤20m² 2 If the fire risk level is determined to be medium risk II; if 20m 2 <Cumulative fire area ≤50m² 2 If the fire risk level is determined to be severe risk I; if 50m 2 <Cumulative fire area ≤100m² 2 If the fire risk level is determined to be severe risk II; if 100m 2 If the cumulative burned area is less than the fire risk level, the fire risk level is determined to be severe risk III.

[0036] The aforementioned residential building fire-affected area prediction system includes fire parameters such as fire dispatch time, fire growth coefficient, and fire load density in its combustion parameter confirmation module.

[0037] The aforementioned residential building fire-affected area prediction system, in its fire parameter probability modeling module, further includes the following steps: acquiring historical fire data of fire parameters and determining the probability distribution of fire parameters based on the historical fire data.

[0038] Historical fire data on combustion parameters were preprocessed using data cleaning methods.

[0039] Based on the fire history data of the pre-processed combustion parameters, the probability density function is obtained through statistical methods.

[0040] The aforementioned residential building fire-affected area prediction system, in the combustion parameter probability modeling module, further includes the step of obtaining the probability density function through statistical methods, specifically:

[0041] Construct a probability density distribution map of combustion parameters based on historical fire data to determine the probability distribution that the combustion parameters follow;

[0042] The probability density function of combustion parameters is obtained by fitting historical fire data.

[0043] The aforementioned residential building fire-affected area prediction system, in the fire-affected area probability modeling module, further includes the step of performing Monte Carlo simulation to obtain the cumulative probability distribution of the fire-affected area of ​​the residential building, which further includes:

[0044] Based on the combustion parameters corresponding to multiple sets of different probabilities, multiple fire area calculation results are obtained through the fire area prediction model.

[0045] Based on the calculation results of multiple burned areas, the cumulative distribution function was obtained through statistical methods.

[0046] The aforementioned residential building fire-affected area prediction system, in the fire-affected area probability modeling module, further includes the step of obtaining the cumulative distribution function through statistical methods, specifically:

[0047] Based on the calculation results of multiple burned areas, a probability density distribution map of the burned area is constructed to determine the probability distribution that the burned area follows.

[0048] Data fitting is performed based on multiple fire area calculation results to obtain the probability density function of the fire area;

[0049] Based on the probability density function of the burned area, the cumulative distribution function of the burned area is obtained, and the cumulative distribution map of the burned area is constructed.

[0050] The aforementioned residential building fire-affected area prediction system also includes a scheme formulation module, which is used to formulate corresponding fire prevention and control measures based on the fire risk level of the residential building.

[0051] To better achieve the objectives of this invention, this invention also provides a computer-readable storage medium having a computer program stored thereon, the computer program being configured to execute the above-described method at runtime.

[0052] To better achieve the objectives of this invention, this invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the methods described above.

[0053] The residential building fire area prediction system, storage medium, and device provided by this invention correspond to the above-mentioned method and have the same beneficial technical effects.

[0054] To provide a better understanding of the above and other aspects of the present invention, specific embodiments are described below in conjunction with the accompanying drawings, but these are not intended to limit the scope of protection of the present invention. Attached Figure Description

[0055] Figure 1 is a flowchart of the steps of a residential building fire area prediction method according to an embodiment of the present invention.

[0056] Figure 2 is a flowchart of the steps for determining the probability distribution of combustion parameters according to an embodiment of the present invention.

[0057] Figure 3 is a flowchart of the steps of performing Monte Carlo simulation to obtain the cumulative probability distribution of fire area of ​​a residential building according to an embodiment of the present invention.

[0058] Figure 4 is a cumulative distribution map of the burned area of ​​a residential building fire in a city according to an embodiment of the present invention.

[0059] Figure 5 is a structural module diagram of a residential building fire area prediction system according to an embodiment of the present invention.

[0060] Figure 6 is a schematic diagram of an electronic device according to an embodiment of the present invention.

[0061] In the attached figures, the following labels are used:

[0062] Steps in the Method for Predicting the Fire Area of ​​Residential Buildings in Fires (S1~S7)

[0063] S31~S32, S321~S322 - Steps for determining the probability distribution of combustion parameters

[0064] Steps S51-S52 and S521-S523 – Perform Monte Carlo simulations to obtain the cumulative probability distribution of fire-affected area in residential buildings.

[0065] L1-Probability Cumulative Curve

[0066] L2-Historical Probability Cumulative Curve

[0067] 1- Residential Building Fire Area Prediction System

[0068] 11-Fire Area Modeling Module

[0069] 12-Combustion Parameter Confirmation Module

[0070] 13-Combustion Parameter Probabilistic Modeling Module

[0071] 14-Random Sampling Module

[0072] 15-Probability Modeling Module for Fire Area

[0073] 16-Fire Risk Level Confirmation Module

[0074] 17-Solution Development Module

[0075] 2-Electronic Devices

[0076] 21-Memory

[0077] 22-processor Detailed Implementation

[0078] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments to further understand the purpose, solution and beneficial technical effects of the present invention, but this is not intended to limit the scope of protection of the appended claims. It should be noted that in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0079] The core of this invention lies in providing a method, system, storage medium, and device for predicting the fire-affected area of ​​residential buildings. Through Monte Carlo simulation, the fire-affected area of ​​residential buildings is accurately predicted, and the fire risk level of the residential buildings is determined based on the fire-affected area, thereby enabling the formulation of effective fire prevention and control measures.

[0080] Please refer to Figure 1, which is a flowchart of a method for predicting the fire-affected area of ​​a residential building according to an embodiment of the present invention. Specifically, it includes the following steps:

[0081] S1: Based on the fire heat release rate and ventilation-controlled combustion mode, a fire area prediction model for residential buildings is established.

[0082] In this embodiment of the invention, the fire heat release rate is the amount of heat released per unit time during fire combustion. The total fire heat release rate is obtained by integration. Based on the assumption that the fire spreads outwards after complete combustion of a unit of combustible material, a formula for calculating the burned area is derived. Specifically, the formula for calculating the total heat release within a residential building is:

[0083] ∑Q1=ΔMh c (1)

[0084] In the formula, ∑Q1 is the total heat release (kJ), ΔM is the total mass loss of combustible material (kg), and h c Heat of combustion (kJ / kg).

[0085] According to the formula for the rate of heat release in a fire, Q = αt 2 Integrating over time t yields the formula for calculating the total heat release during the fire's growth time t:

[0086]

[0087] In the formula, ∑Q2 represents the total heat release (kJ), t represents the fire growth time (s), and α represents the fire growth coefficient (kW / s). 2 ).

[0088] Furthermore, the formula for calculating fire load density is:

[0089]

[0090] In the formula, q is the fire load density (kJ / m³). 2 M is the total mass of the combustible material (kg), h c Heat of combustion (kJ / kg), A t Ground area (m²) 2 ).

[0091] If the total heat release ∑Q2 generated during the fire growth time t is equal to the heat release ∑Q1 generated by the effective combustion of combustibles within the fire-affected area of ​​the residential building during that time, the formulas for calculating the fire-affected area of ​​the residential building within the fire growth time t in fuel-controlled combustion locations can be obtained through formulas (1), (2), and (3):

[0092] ∑Q2=δ∑Q1(4)

[0093]

[0094] In the formula, A is the fire-affected area (m²). 2 ), t1 is the time from fire ignition to alarm (s), t2 is the fire dispatch time, i.e., the time from the first fire truck's dispatch to effective water output (s), the fire growth time t is the sum of the time from fire ignition to alarm t1 and the fire dispatch time t2, and α is the fire growth coefficient (kW / s). 2 ), where q is the fire load density (kJ / m³). 2 δ is the combustion efficiency factor. At the time of a fire alarm, most fires have not yet entered the decay stage, and the combustion efficiency factor ranges from 0.5 to 0.7 based on historical fire data analysis.

[0095] According to the combustion triangle, combustion occurs when fuel, oxidizer (oxygen), and ignition energy are all present; controlling any one of these factors stops combustion. During a fire, there are three combustion states: fuel-controlled (limited quantity of combustible material, unlimited oxygen); ventilation-controlled (unlimited combustible material, limited oxygen); and unrestricted combustible material and oxygen supply. For residential buildings, fires are typically small-scale indoor fires, usually controlled by ventilation (oxygen supply). Based on the above basic formula and analysis of fire heat release rates, and assuming ventilation-controlled combustion, the fire-affected area A of a residential building fire is established. f Prediction model:

[0096] A f =f(α,t,q)=0.02951αt 2 q -1 / 3 (6)

[0097] In the formula, the fire-affected area A f The burned area (m²) when the fire department arrives. 2 ), where α is the fire growth coefficient (kW / s) 2 ), t is the fire growth time (s), and q is the fire load density (kJ / m³). 2 The fire area prediction model in this embodiment of the invention is based on the fire heat release rate, the total calorific value of the fire, and the fire combustion characteristics of residential buildings, thus the predicted fire area is more realistic and accurate.

[0098] S2: Determine the combustion parameters of the burned area based on the fire burned area prediction model. In a preferred embodiment, the combustion parameters include fire response time, fire growth coefficient, and fire load density.

[0099] In this embodiment of the invention, the fire-affected area of ​​a residential building in formula (6) is calculated using the fire growth time, fire growth coefficient, and fire load density. Therefore, the combustion parameters for determining the fire-affected area include the fire growth time, fire growth coefficient, and fire load density. Due to the uncertainty of residential building fires, the time from ignition to alarm in the fire growth time is usually difficult to determine. Therefore, the fire dispatch time is used as the fire growth time to calculate the fire-affected area of ​​the residential building.

[0100] Residential building fires are typically fought by fire brigades. The shorter the fire brigade's arrival time, the smaller the fire-affected area; conversely, the longer the arrival time, the larger the fire-affected area and the higher the risk. Fire load density refers to the amount of combustible material per unit area within a residential building. Furthermore, a higher fire growth coefficient indicates a faster fire spread, resulting in a larger fire-affected area and higher risk for the same time and fire load density. Therefore, using fire brigade response time, fire load density, and fire growth coefficient as combustion parameters leads to more accurate predictions of the fire-affected area in residential building fires.

[0101] S3: Obtain historical fire data of combustion parameters, and determine the probability distribution of combustion parameters based on the historical fire data. Please refer to Figure 2, which is a flowchart of the steps for determining the probability distribution of combustion parameters according to an embodiment of the present invention.

[0102] By acquiring historical fire data of residential buildings in various cities or communities across the country, the probability distribution of combustion parameters in each city or community can be obtained. The historical fire data includes fire dispatch time, fire load density, average heat release rate per unit area of ​​fire, burned area when the fire brigade arrives, and historical data of the house and / or room area of ​​the residential building where the fire occurred, but the present invention is not limited thereto.

[0103] In a preferred embodiment, the step of acquiring historical fire data of combustion parameters and determining the probability distribution of combustion parameters based on the historical fire data further includes:

[0104] S31: Use data cleaning methods to preprocess historical fire data of combustion parameters;

[0105] S32: Based on the fire history data of the pre-processed combustion parameters, the probability density function is obtained through statistical methods.

[0106] In this embodiment of the invention, because the historical fire data includes data from multiple years and comes from multiple different departments, there is a lack of consistency among the data. Therefore, it is necessary to preprocess the data in the dataset to remove incomplete, inaccurate, or irrelevant data to ensure the accuracy of the probability distribution of combustion parameters. The above data cleaning process can be achieved through big data analytics, and will not be elaborated further here.

[0107] In a preferred embodiment, the step of obtaining the probability density function using statistical methods further includes:

[0108] S321: Construct a probability density distribution map of combustion parameters based on historical fire data to determine the probability distribution followed by the combustion parameters;

[0109] S322: Obtain the probability density function of combustion parameters by fitting data based on historical fire data.

[0110] In this embodiment of the invention, since residential buildings in my country typically do not have automatic sprinkler systems and fire alarm systems, one of the main factors affecting the fire area is determined by formula (6). Historical fire data of a city is obtained, and historical data on fire response times are preprocessed using data cleaning methods. Based on the preprocessed historical data on fire response times, a probability density distribution map of fire response times is constructed. The probability density distribution map is a histogram, but this invention is not limited to this. The probability distribution that the histogram of fire response times follows is determined based on its characteristics. The probability distribution can be a normal distribution, a log-normal distribution, or a Weibull distribution, but this invention is not limited to this. Then, data fitting is performed based on the probability distribution that the fire response times follow to obtain the mean and standard deviation of the probability distribution, and subsequently, the probability density function of the fire response time is obtained. In a specific embodiment, the fire response times of a city obtained according to the above method conform to a normal distribution function, and its probability density function is:

[0111]

[0112] In the formula, P(t) is the probability of fire dispatch time, and t is the fire dispatch time (s).

[0113] The average fire growth coefficient for each fire can be obtained by considering the burned area at the time of fire discovery, the burned area upon fire arrival, and the time interval between fire discovery and fire department arrival. Based on historical fire data, fires are typically very small at discovery; therefore, the burned area at discovery is ignored as zero. The formula for calculating the fire growth coefficient is:

[0114]

[0115] In the formula, q″ is the average heat release rate per unit area of ​​the fire (kW / m²). 2 A1 is the burned area (m²) when the fire department arrives. 2 ), where t2 is the fire dispatch time (s).

[0116] In this embodiment of the invention, historical fire data of a city is obtained, including the average heat release rate per unit area of ​​each fire, the burned area when the fire brigade arrives, and the historical data of the fire dispatch time. Data cleaning is then used for preprocessing. Based on the preprocessed data, the fire growth coefficient of each fire is calculated using formula (8), and a probability density distribution map of the fire growth coefficient is constructed. The probability density distribution map is a histogram, but this invention is not limited to this. The probability distribution that the fire growth coefficient follows is determined based on the characteristics of the histogram. The probability distribution can be a normal distribution, a log-normal distribution, or a Weibull distribution, but this invention is not limited to this. Then, data fitting is performed based on the probability distribution that the fire growth coefficient follows to obtain the mean and standard deviation of the probability distribution, and subsequently, the probability density function of the fire growth coefficient is obtained. In a specific embodiment, the fire growth coefficient of a city obtained by the above method conforms to a log-normal distribution function, and its probability density function is:

[0117]

[0118] In the formula, P(α) is the probability of the fire growth coefficient, and α is the fire growth coefficient.

[0119] Furthermore, by acquiring historical fire data of a city, historical fire load density data is obtained and preprocessed using data cleaning methods. Based on the preprocessed historical fire load density data, a probability density distribution map of the fire load density is constructed. The probability density distribution map is a histogram, but this invention is not limited to this. The probability distribution that the fire load density histogram follows is determined based on its characteristics. The probability distribution can be a normal distribution, a log-normal distribution, or a Weibull distribution, but this invention is not limited to this. Then, data fitting is performed based on the probability distribution followed by the fire load density to obtain the mean and standard deviation of the probability distribution, thereby obtaining the probability density function of the fire load density. In a specific embodiment, the fire load density of a city obtained according to the above method conforms to a normal distribution function, and its probability density function is:

[0120]

[0121] In the formula, P(q) is the probability of fire load density, and q is the fire load density (kJ / m³). 2 ).

[0122] S4: Randomly sample the probability distribution of combustion parameters to obtain multiple sets of combustion parameters corresponding to different probabilities.

[0123] S5: Based on the fire-affected area prediction model and multiple sets of combustion parameters corresponding to different probabilities, Monte Carlo simulation is performed to obtain the cumulative probability distribution of the fire-affected area of ​​the residential building. Please refer to Figure 3, which is a flowchart of the steps for performing Monte Carlo simulation to obtain the cumulative probability distribution of the fire-affected area of ​​a residential building according to an embodiment of the present invention.

[0124] As a preferred embodiment, the step of performing Monte Carlo simulation to obtain the cumulative probability distribution of fire-affected area of ​​a residential building further includes:

[0125] S51: Based on the combustion parameters corresponding to multiple sets of different probabilities, multiple fire area calculation results are obtained through the fire area prediction model;

[0126] S52: Based on the calculation results of multiple burned areas, the cumulative distribution function is obtained through statistical methods.

[0127] Monte Carlo simulation analysis uses random variables that follow a certain probability distribution as input parameters, inputs them into a specific mathematical model, and uses the output results to represent possible random phenomena. Repeated sampling simulations can obtain several output results and determine the distribution law of the results. In this embodiment of the invention, after obtaining the probability density functions of fire dispatch time, fire growth coefficient, and fire load density of a city through the above method, according to formulas (7), (9), and (10), a random sampling method is used to obtain multiple sets of fire dispatch time, fire growth coefficient, and fire load density data corresponding to different probabilities. The multiple sets of combustion parameter data corresponding to different probabilities are input into formula (6) to obtain the calculation results of the fire-affected area of ​​multiple sets of residential buildings. Based on the calculation results of the fire-affected area of ​​multiple sets of residential buildings, the cumulative distribution function of the fire-affected area can be obtained through statistical methods. Since the fire dispatch time, fire growth coefficient, and fire load density corresponding to different residential buildings are uncertain when a fire occurs, this invention uses Monte Carlo simulation to perform a large number of repeated calculations to make the calculation results of the fire-affected area more accurate and reasonable.

[0128] In a preferred embodiment, the step of obtaining the cumulative distribution function using statistical methods further includes:

[0129] S521: Construct a probability density distribution map of the burned area based on the calculation results of multiple burned areas, and determine the probability distribution that the burned area follows;

[0130] S522: Based on the calculation results of multiple burned areas, perform data fitting to obtain the probability density function of the burned area;

[0131] S523: Based on the probability density function of the burned area, obtain the cumulative distribution function of the burned area and construct the cumulative distribution map of the burned area.

[0132] In this embodiment of the invention, a probability density distribution map of the burned area is constructed based on the calculation results of the burned area of ​​multiple sets of residential building fires. The probability density distribution map is a histogram, but this invention is not limited to this. The probability distribution that the burned area follows is determined based on the characteristics of the histogram. The probability distribution can be a normal distribution, a log-normal distribution, or a Weibull distribution, but this invention is not limited to this. Then, data fitting is performed based on the probability distribution followed by the burned area to obtain the mean and standard deviation of the probability distribution, thus obtaining the probability density function of the burned area, and subsequently, the cumulative distribution function of the burned area and the cumulative distribution map of the burned area.

[0133] S6: Determine the fire risk level of residential buildings based on the cumulative probability distribution of burned area.

[0134] In this embodiment of the invention, based on historical fire data of residential buildings, the cumulative fire-affected area of ​​residential buildings is typically less than 5m². 2 The cumulative fire-affected area of ​​a normal-sized room is 10m². 2 The cumulative fire-affected area of ​​a large room was 20m². 2 Therefore, the risk level of residential buildings is divided based on the cumulative fire area. The risk level and the corresponding cumulative fire area are shown in Table 1.

[0135] Table 1 Fire Risk Levels of Residential Buildings

[0136]

[0137]

[0138] Please refer to Figure 4, which is a cumulative distribution map of the burned area of ​​a residential building fire in a city according to an embodiment of the present invention. In a specific embodiment, the cumulative distribution map of the burned area of ​​a residential building fire is obtained by the above method, where L1 is the probability cumulative curve of the burned area obtained by the method of the present invention. The data 0.7205, 0.8067, 0.8715, 0.9364, and 0.9642 in the figure represent cumulative burned areas of 5m². 2 10m 2 20m 2 50m 2 and 100m 2 The corresponding cumulative frequency. As shown in L1 of Figure 4, according to the method of the present invention, the cumulative burned area of ​​the city is ≥0 and ≤5m². 2 The cumulative frequency of fires was 0.7205, therefore the probability that residential buildings in the city are at low risk is 72.05%; the cumulative fire area in the city is >5m². 2 and ≤10m 2The cumulative frequency of fire was 0.0862, therefore the probability that the residential buildings in this city are at medium risk level I is 8.62%; the cumulative fire area in this city is >10m². 2 and ≤20m 2 The cumulative frequency of fires is 0.0648, therefore the probability that residential buildings in this city are classified as medium-risk II is 6.48%; the cumulative fire area in this city is >20m². 2 and ≤50m 2 The cumulative frequency of fire is 0.0649, therefore the probability that the residential buildings in this city are at severe risk level I is 6.49%; the cumulative fire area in this city is >50m². 2 and ≤100m 2 The cumulative frequency of fires is 0.0278, therefore the probability that residential buildings in this city are at severe risk level II is 2.78%; the cumulative fire area in this city is >100m². 2 The cumulative frequency is 0.0358, therefore the probability that the residential buildings in this city are at severe risk level III is 3.58%. Therefore, the fire risk level of the residential buildings in this city is determined to be low risk.

[0139] Please refer to Figure 4. In Figure 4, L2 represents the historical probability cumulative curve of the burned area of ​​residential buildings in the city described in the above embodiment. The historical probability cumulative curve was obtained from historical fire data. The data in the figure, 0.7515, 0.8669, 0.9334, 0.9748, and 0.9881, represent cumulative burned areas of 5m². 2 10m 2 20m 2 50m 2 and 100m 2 The cumulative frequency corresponding to the time. Table 2 shows a comparison between the historical cumulative probability data of the burned area of ​​residential buildings in this city and the cumulative probability of burned area obtained by the method of this invention.

[0140] Table 2 Comparison of historical cumulative probability data with the cumulative probability of the present invention

[0141]

[0142]

[0143] According to Figure 4 and Table 2, the probability of fire area obtained by the present invention has an error rate of less than 10% compared with the historical probability data of fire area in the city. Therefore, it can accurately predict the fire area of ​​residential buildings and thus determine the risk level of residential buildings.

[0144] S7: Develop corresponding fire prevention and control measures based on the fire risk level of residential buildings.

[0145] After determining the fire risk level of residential buildings in a city through the above steps, corresponding fire prevention and control measures are formulated to effectively prevent and respond to fires in residential buildings. In a specific embodiment shown in Figure 4, the probability that a city's residential buildings are at low risk is 72.05%. Therefore, the fire risk level of the city's residential buildings is determined to be low risk, and corresponding fire prevention and control measures are then formulated for low-risk fires. The above method is applicable to all residential buildings in a city, and also to a specific residential complex within a city. That is, it can predict the fire area, determine the fire risk level, and formulate fire prevention and control measures for a city or a residential complex respectively. This invention is not limited to these limitations.

[0146] The present invention provides a method for predicting the fire-affected area of ​​residential buildings in fires. This method constructs a correct prediction model for the fire-affected area of ​​residential buildings, determines the combustion parameters affecting the fire-affected area based on the prediction model, constructs the probability distribution of each combustion parameter using statistical methods combined with historical fire data, and uses Monte Carlo simulation to determine the fire-affected area. Based on this, the probability distribution of the fire-affected area is constructed, thereby determining the fire risk level and formulating corresponding fire prevention and control measures. In contrast, existing technologies cannot accurately predict the fire-affected area of ​​residential buildings, thus failing to determine the fire risk level and formulate effective fire prevention and control measures. Therefore, this invention accurately predicts the fire-affected area of ​​residential buildings through Monte Carlo simulation, determines the fire risk level of residential buildings based on the fire-affected area, and thus formulates effective fire prevention and control measures.

[0147] Please refer to Figure 5, which is a structural block diagram of a residential building fire fire area prediction system 1 according to an embodiment of the present invention. Specifically, it includes the following modules:

[0148] The fire area modeling module 11 is used to establish a fire area prediction model for residential buildings based on the fire heat release rate and ventilation-controlled combustion mode. The specific implementation method and beneficial technical effects of the fire area modeling module 11 are as described in step S1 above, and will not be repeated here.

[0149] The combustion parameter confirmation module 12 is used to determine the combustion parameters of the burned area based on the fire burned area prediction model.

[0150] In a preferred embodiment, the combustion parameters in the combustion parameter confirmation module 12 include fire dispatch time, fire growth coefficient, and fire load density.

[0151] The specific implementation method and beneficial technical effects of the combustion parameter confirmation module 12 are as described in step S2 above, and will not be repeated here.

[0152] The combustion parameter probability modeling module 13 is used to acquire historical fire data of combustion parameters and determine the probability distribution of combustion parameters based on the historical fire data. The historical fire data includes fire dispatch time, fire load density, average heat release rate per unit area of ​​fire, burned area when the fire brigade arrives, and historical data on the building and / or room area of ​​the residential building where the fire occurred, but this invention is not limited thereto.

[0153] In a preferred embodiment, the step of acquiring historical fire data of combustion parameters and determining the probability distribution of combustion parameters based on the historical fire data in the combustion parameter probability modeling module 13 further includes:

[0154] Historical fire data on combustion parameters were preprocessed using data cleaning methods.

[0155] Based on the fire history data of the pre-processed combustion parameters, the probability density function is obtained through statistical methods.

[0156] In a preferred embodiment, the step of obtaining the probability density function using statistical methods in the combustion parameter probability modeling module 13 further includes:

[0157] Construct a probability density distribution map of combustion parameters based on historical fire data to determine the probability distribution that the combustion parameters follow;

[0158] The probability density function of combustion parameters is obtained by fitting historical fire data.

[0159] The specific implementation method and beneficial technical effects of the combustion parameter probability modeling module 13 are as described in step S3 above, and will not be repeated here.

[0160] The random sampling module 14 is used to randomly sample the probability distribution of combustion parameters to obtain multiple sets of combustion parameters corresponding to different probabilities. The specific implementation and beneficial technical effects of the random sampling module 14 are as described in step S4 above, and will not be repeated here.

[0161] The fire area probability modeling module 15 is used to perform Monte Carlo simulation based on the fire area prediction model and multiple sets of combustion parameters corresponding to different probabilities to obtain the cumulative probability distribution of the fire area of ​​residential buildings.

[0162] In a preferred embodiment, the step of performing Monte Carlo simulation to obtain the cumulative probability distribution of the fire area of ​​a residential building in the fire area probability modeling module 15 further includes:

[0163] Based on the combustion parameters corresponding to multiple sets of different probabilities, multiple fire area calculation results are obtained through the fire area prediction model.

[0164] Based on the calculation results of multiple burned areas, the cumulative distribution function was obtained through statistical methods.

[0165] In a preferred embodiment, the step of obtaining the cumulative distribution function through statistical methods in the fire area probability modeling module 15 further includes:

[0166] Based on the calculation results of multiple burned areas, a probability density distribution map of the burned area is constructed to determine the probability distribution that the burned area follows.

[0167] Data fitting is performed based on multiple fire area calculation results to obtain the probability density function of the fire area;

[0168] Based on the probability density function of the burned area, the cumulative distribution function of the burned area is obtained, and the cumulative distribution map of the burned area is constructed.

[0169] The specific implementation method and beneficial technical effects of the fire area probability modeling module 15 are as described in step S5 above, and will not be repeated here.

[0170] The fire risk level confirmation module 16 is used to determine the fire risk level of a residential building based on the cumulative probability distribution of the burned area.

[0171] If 0 ≤ cumulative fire area ≤ 5m 2 If so, the fire risk level is determined to be low risk;

[0172] If 5m 2 <Cumulative fire area ≤10m² 2 If so, the fire risk level is determined to be medium risk I;

[0173] If 10m 2 <Cumulative fire area ≤20m² 2 If so, the fire risk level is determined to be medium risk II;

[0174] If 20m 2 <Cumulative fire area ≤50m² 2 If so, the fire risk level is determined to be severe risk I;

[0175] If 50m 2 <Cumulative fire area ≤100m² 2 If so, the fire risk level is determined to be severe risk II;

[0176] If 100m 2 If the cumulative burned area is less than the fire risk level, the fire risk level is determined to be severe risk III.

[0177] The specific implementation method and beneficial technical effects of the fire risk level confirmation module 16 are as described in step S6 above, and will not be repeated here.

[0178] The scheme formulation module 17 is used to formulate corresponding fire prevention and control measures based on the fire risk level of residential buildings. The specific implementation method and beneficial technical effects of the scheme formulation module 17 are as described in step S7 above, and will not be repeated here.

[0179] Since the embodiments of the system part correspond to the embodiments of the method part, please refer to the description of the embodiments of the method part for the embodiments of the system part, which will not be repeated here. It should be noted that the system proposed above can also be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of the modules described above is merely a logical functional division. In actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed.

[0180] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, the computer program being configured to execute the residential building fire area prediction method as described above when running.

[0181] Please refer to Figure 6, which is a schematic diagram of an electronic device according to an embodiment of the present invention. An embodiment of the present invention also provides an electronic device 2, including a memory 21 and a processor 22. The memory 21 stores a computer program, and the processor 22 is configured to run the computer program to execute the residential building fire area prediction method as described above.

[0182] The present invention has provided a detailed description of the method, system, storage medium, and device for predicting the fire area of ​​residential buildings in fires. The various embodiments in the specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems, storage media, and devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.

[0183] Of course, the present invention may have other various embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and modifications according to the present invention, but these corresponding changes and modifications should all fall within the protection scope of the appended claims.

Claims

1. A method for predicting the burned area of ​​a residential building in a fire, characterized in that, Includes the following steps: Based on the fire heat release rate and fuel-controlled combustion mode, a fire-affected area prediction model for residential buildings is established. Combustion parameters for the affected area are determined according to the fire-affected area prediction model. Historical fire data for these parameters are acquired, and their probability distribution is determined based on the historical fire data. Random sampling is performed on the probability distribution of the combustion parameters to obtain multiple sets of combustion parameters corresponding to different probabilities. Monte Carlo simulation is conducted based on the fire-affected area prediction model and the multiple sets of combustion parameters corresponding to different probabilities to obtain the cumulative probability distribution of the fire-affected area for the residential building. Based on the cumulative probability distribution of the fire-affected area, the fire risk level of the residential building is determined. If 0 ≤ cumulative fire-affected area ≤ 5m², the fire risk level is determined. 2 If the fire risk level is 5m, then the fire risk level is determined to be low risk; 2 <Cumulative fire area ≤10m² 2 If the fire risk level is determined to be medium risk I; if 10m 2 <Cumulative fire area ≤20m² 2 If the fire risk level is determined to be medium risk II; if 20m 2 <Cumulative fire area ≤50m² 2 If the fire risk level is determined to be severe risk I; if 50m 2 <Cumulative fire area ≤100m² 2 If the fire risk level is determined to be severe risk II; if 100m 2 If the cumulative burned area is determined, the fire risk level is classified as Severe Risk III; wherein the combustion parameters include fire growth time, fire growth coefficient, and fire load density; and the fire burned area prediction model is... A is the fire-affected area (m²) 2 ), t1 is the time from fire ignition to alarm (s), t2 is the fire dispatch time (s), and the fire growth time is the sum of the time from fire ignition to alarm t1 and the fire dispatch time t2. Fire growth coefficient (kW / s) 2 ), where q is the fire load density (kJ / m³). 2 ), This is the combustion efficiency factor.

2. The method for predicting the fire-affected area of ​​a residential building according to claim 1, characterized in that, The step of obtaining historical fire data of the combustion parameters and determining the probability distribution of the combustion parameters based on the historical fire data further includes: preprocessing the historical fire data of the combustion parameters using a data cleaning method; and obtaining a probability density function using statistical methods based on the preprocessed historical fire data of the combustion parameters.

3. The method for predicting the fire-affected area of ​​a residential building according to claim 2, characterized in that, The step of obtaining the probability density function through statistical methods further includes: constructing a probability density distribution map of the combustion parameters based on the historical fire data, and determining the probability distribution followed by the combustion parameters; and performing data fitting based on the historical fire data to obtain the probability density function of the combustion parameters.

4. The method for predicting the fire-affected area of ​​a residential building according to claim 1, characterized in that, The step of performing Monte Carlo simulation to obtain the cumulative probability distribution of the fire area of ​​a residential building further includes: calculating multiple fire area calculation results using the fire area prediction model based on the combustion parameters corresponding to the multiple sets of different probabilities; and obtaining the cumulative distribution function using statistical methods based on the multiple fire area calculation results.

5. The method for predicting the fire-affected area of ​​a residential building according to claim 4, characterized in that, The step of obtaining the cumulative distribution function through statistical methods further includes: constructing a probability density distribution map of the burned area based on the multiple burned area calculation results, and determining the probability distribution followed by the burned area; performing data fitting based on the multiple burned area calculation results to obtain the probability density function of the burned area; obtaining the cumulative distribution function of the burned area based on the probability density function of the burned area, and constructing the cumulative distribution map of the burned area.

6. The method for predicting the fire-affected area of ​​a residential building according to any one of claims 1-5, characterized in that, It also includes developing corresponding fire prevention and control measures based on the fire risk level of the residential building.

7. A fire-affected area prediction system for residential buildings, characterized in that, include: The fire area modeling module is used to establish a fire area prediction model for residential buildings based on the fire heat release rate and fuel-controlled combustion mode. The system includes a combustion parameter confirmation module for determining combustion parameters of the burned area based on the fire burned area prediction model; a combustion parameter probability modeling module for acquiring historical fire data of the combustion parameters and determining the probability distribution of the combustion parameters based on the historical fire data; a random sampling module for randomly sampling the probability distribution of the combustion parameters to obtain multiple sets of combustion parameters corresponding to different probabilities; a burned area probability modeling module for performing Monte Carlo simulation based on the fire burned area prediction model and the multiple sets of combustion parameters corresponding to different probabilities to obtain the cumulative probability distribution of the burned area of ​​the residential building; and a fire risk level confirmation module for determining the fire risk level of the residential building based on the cumulative probability distribution of the burned area, where 0 ≤ cumulative burned area ≤ 5m². 2 If the fire risk level is 5m, then the fire risk level is determined to be low risk; 2 <Cumulative fire area ≤10m² 2 If the fire risk level is determined to be medium risk I; if 10m 2 <Cumulative fire area ≤20m² 2 If the fire risk level is determined to be medium risk II; if 20m 2 <Cumulative fire area ≤50m² 2 If the fire risk level is determined to be severe risk I; if 50m 2 <Cumulative fire area ≤100m² 2 If the fire risk level is determined to be severe risk II; if 100m 2 If the cumulative burned area is determined, the fire risk level is classified as Severe Risk III; wherein, in the combustion parameter confirmation module, the combustion parameters include fire growth time, fire growth coefficient, and fire load density; and, the fire burned area prediction model is... A is the fire-affected area (m²) 2 ), t1 is the time from fire ignition to alarm (s), t2 is the fire dispatch time (s), and the fire growth time is the sum of the time from fire ignition to alarm t1 and the fire dispatch time t2. Fire growth coefficient (kW / s) 2 ), where q is the fire load density (kJ / m³). 2 ), This is the combustion efficiency factor.

8. The residential building fire-affected area prediction system according to claim 7, characterized in that, In the combustion parameter probability modeling module, the step of acquiring the historical fire data of the combustion parameters and determining the probability distribution of the combustion parameters based on the historical fire data further includes: preprocessing the historical fire data of the combustion parameters using data cleaning methods; and obtaining the probability density function using statistical methods based on the preprocessed historical fire data of the combustion parameters.

9. The residential building fire-affected area prediction system according to claim 8, characterized in that, In the combustion parameter probability modeling module, the step of obtaining the probability density function through statistical methods further includes: constructing a probability density distribution map of the combustion parameters based on the historical fire data to determine the probability distribution followed by the combustion parameters; and performing data fitting based on the historical fire data to obtain the probability density function of the combustion parameters.

10. The residential building fire-affected area prediction system according to claim 7, characterized in that, In the fire area probability modeling module, the step of performing Monte Carlo simulation to obtain the cumulative probability distribution of the fire area of ​​the residential building further includes: calculating multiple fire area calculation results through the fire area prediction model based on the combustion parameters corresponding to the multiple sets of different probabilities; and obtaining the cumulative distribution function through statistical methods based on the multiple fire area calculation results.

11. The residential building fire-affected area prediction system according to claim 10, characterized in that, In the fire area probability modeling module, the step of obtaining the cumulative distribution function through statistical methods further includes: constructing a probability density distribution map of the fire area based on the multiple fire area calculation results, and determining the probability distribution followed by the fire area; performing data fitting based on the multiple fire area calculation results to obtain the probability density function of the fire area; obtaining the cumulative distribution function of the fire area based on the probability density function of the fire area, and constructing the cumulative distribution map of the fire area.

12. The residential building fire-affected area prediction system according to any one of claims 7-11, characterized in that, It also includes a scheme development module, which is used to formulate corresponding fire prevention and control measures based on the fire risk level of the residential building.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program is configured to execute the method of any one of claims 1-6 when it is run.

14. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the method of any one of claims 1-6.

Images