A performance-based optimization method for building fire detector layout

Abstract

The present invention discloses a performance-based building fire detector layout optimization method, which relates to the technical field of fire safety and layout optimization. A fire scene data set is obtained, and the fire scene includes two fire parameters: the location of the fire source and the fire growth rate; a batch calculation method of the fire scene data set is designed using fire simulation software, which is used to batch calculate the detection time corresponding to each fire scene; a detector layout scheme evaluation function is constructed, and a fitness function is defined to evaluate the advantages and disadvantages of the detector layout scheme, including two indicators: the shortest detection time and full spatial coverage; a detector layout scheme optimization model based on a genetic algorithm is established, and a genetic algorithm is used to solve the detector layout scheme corresponding to the maximum fitness function value to obtain the optimal detector layout scheme. On the premise of meeting the specifications, the present invention considers the fire hazard level of a specific place and provides an optimal detector layout scheme with the shortest detection time.

Description

Technical Field

[0001] The invention relates to the technical field of fire safety and layout optimization, in particular to a performance-based building fire detector layout optimization method. Background Art

[0002] Frequent building fires pose a huge threat to people's lives and property safety. Fire detection systems play a vital role in building fire safety. Through early detection of fires, potential fire threats can be quickly identified and alarm responses can be made in a timely manner. With the increasing emphasis on sustainability in building design, optimizing existing fire protection systems has become an increasingly important issue in fire protection engineering. Regarding fire detection systems, developing more efficient and cost-effective fire detector deployment strategies has attracted widespread attention.

[0003] At present, the installation and layout of fire detectors in buildings are mainly based on the specifications of various countries. For example, the US specification NFPA72-2022: National Fire Alarm and Signaling Code; the British standard BS 5839-1:2017: Fire Detection and Alarm Systems for Buildings; the Singapore standard SS 645:2019: Code of Practice for the Installation and Servicing of Electrical Fire Alarm Systems; and the Chinese standard GB 50116-2013: Design Specification for Automatic Fire Alarm Systems. These specifications and standards give the installation spacing requirements for fire detectors. For example, for flat roofs, if point-type smoke detectors are installed, NFPA 72-2022 requires the maximum spacing between fire detectors to be 9.1 m; BS 5839-1:2017 requires the maximum spacing between fire detectors to be 10.6 m; SS645:2019 requires the maximum spacing between fire detectors to be 10.0 m; GB 50116-2013 requires the maximum spacing between fire detectors to be 10.4 m (roof height less than 6 m). It can be seen from the above specifications and standards that there are certain differences in the installation spacing of the same type of fire detectors in various countries, and the maximum installation spacing is between 9.1 m and 10.6 m. It can be seen that even if the specifications are met, the maximum installation spacing of fire detectors still has room for optimization. However, in the design stage of the actual fire detector installation layout, designers mainly consider whether the local specifications are met, such as whether the maximum installation spacing requirements are met, and will not perform performance-based installation layout design for places with unique fire risks. For example, as long as the maximum installation spacing of fire detectors does not exceed 10.6m, the installation spacing parameters of the detectors are considered to meet the specifications, and the rationality and economy of the installation layout will not be considered.

[0004] In summary, the existing technology does not consider the impact of the fire hazard level of a specific place on the layout of detectors. It only judges whether the maximum installation spacing meets the specifications, and the installation layout plan is rigid and not economical and optimal. Summary of the invention

[0005] In order to overcome the defects in the above-mentioned prior art, the present invention provides a performance-based building fire detector layout optimization method, aiming to solve the problems in the prior art that the influence of the fire hazard level of a specific place on the detector layout is not considered, the detector layout scheme is rigid, and is not economical and optimal.

[0006] To achieve the above object, the present invention adopts the following technical solutions, including:

[0007] A performance-based building fire detector layout optimization method comprises the following steps:

[0008] S1, obtain a fire scene dataset; the fire scene includes two fire parameters: fire source location and fire growth rate;

[0009] S2, using fire simulation software to design a batch calculation method for fire scene data sets, which is used to batch calculate the detection time corresponding to each fire scene, and obtain the detection time data set corresponding to the fire scene data set;

[0010] S3, construct a detector layout scheme evaluation function, define a fitness function to evaluate the detector layout scheme, including two indicators: the shortest detection time and full spatial coverage;

[0011] S4, establish a detector layout optimization model based on a genetic algorithm, use the genetic algorithm to solve the detector layout corresponding to the maximum fitness function value, and obtain the optimal detector layout.

[0012] Preferably, in step S3, the fitness function is specifically as follows:

[0013] ;

[0014] ;

[0015] ;

[0016] ;

[0017] ;

[0018] In formula (1), Arrangement plan for detector i Ind In the fire scene S m The shortest detection time under i Ind Indicated by i The detector arrangement scheme consists of detectors. i Ind By detector i 1 to detector i i composition; f ( i 1, S m ) is the detector i 1In the fire scene S m The detection time required to detect a fire; S m Indicates m A fire scene;

[0019] In formula (2), F 1 is the detector layout plan i Ind exist M The average shortest detection time under each fire scenario; M is the number of fire scenes;

[0020] In formula (3), the function I d ( X,Y ) is used to determine any point in space ( X,Y ) falls within the protection radius of a certain detector; X , Y are the horizontal and vertical coordinates of any point in space respectively; , The detector i c The horizontal and vertical coordinates of c =1,2,..., i ; R is the protection radius of the detector; X ∈[0, a ], Y ∈[0, b ], a , b are the length and width of the space respectively; if any point in the space ( X,Y ) falls within the protection radius of a detector, then I d ( X,Y ) value is 1, otherwise

[0021] I d ( X,Y ) value is 0;

[0022] In formula (4), the function F 2 is the detector layout plan i Ind The spatial coverage factor is used to measure the detector layout scheme. i Ind Whether the entire space is fully covered; N is the number of times the experiment is repeated, and each time any point in the space is selected for judgment; Indicates n The results of the experiment, ifn If any point in the space extracted in this experiment falls within the protection radius of a detector, then The value is 1, otherwise The value is 0;

[0023] In formula (5), F 1' and F 2' are F 1 and F 2 Normalized value in the range [0,1]; w 1 and w 2 are F 1' and F 2' weight value; fitness ( i Ind ) is the fitness function. The larger the fitness function value, the better the corresponding detector layout scheme. i Ind The better.

[0024] Preferably, the specific method of step S1 is:

[0025] S11, define the value range of each fire parameter;

[0026] S12, for each fire parameter value range, divide the value range into The interval of

[0027] S13, for each fire parameter, a sampling point is randomly selected in each interval;

[0028] S14, randomly combining the sampling points of various fire parameters to form a number of fire scenes, forming a fire scene data set.

[0029] Preferably, in step S2, a batch calculation method for a fire scene data set is designed using fire simulation software Fire Dynamic Simulator.

[0030] Preferably, in step S2, the batch calculation method is specifically as follows:

[0031] The FDS sample file generator is used to realize the conversion of fire scene data set to FDS code file, the fire parameters in the fire scene are replaced with the corresponding items in the FDS code template file, and the FDS code executable file of the fire scene is constructed; the batch processing program is used to realize the batch running of the FDS code executable file under the FDS calculation program.

[0032] Preferably, in step S4, a batch calculation method based on the fire scene data set constructed in step S1 and the fire scene data set designed in step S2 is used to calculate the fitness function value of the detector layout plan, and a genetic algorithm is used to optimize the detector layout plan to find the maximum fitness function value and obtain the optimal detector layout plan.

[0033] Preferably, the specific optimization process of step S4 is as follows:

[0034] S41, initializing the population: randomly initialize a certain number of detector layout schemes to construct an initial population; each detector layout scheme is each individual in the population;

[0035] S42, evaluate the fitness of individuals in the population: calculate the fitness function value of each individual in the population;

[0036] S43, selection: according to the fitness function value of each individual in the population, select a number of individuals with larger fitness function values ​​from the population as parents;

[0037] S44, crossover and mutation: From the selected parent individuals, crossover and mutation operations are performed according to a certain probability to generate new offspring individuals;

[0038] S45, generate a new population: the new population is composed of offspring individuals;

[0039] S46, determine whether the termination condition is met. If so, end the genetic algorithm and output a new population. The individual with the largest fitness function value in the new population is the optimal detector layout plan. Otherwise, jump to step S42, continue to iterate, and continue to generate new populations until the termination condition is met to obtain the optimal detector layout plan.

[0040] The present invention also provides a computer program product, which includes a computer program / instruction. When the computer program / instruction is executed by a processor, the above-mentioned performance-based building fire detector layout optimization method is implemented.

[0041] The present invention also provides a readable storage medium on which a computer program is stored. When the computer program is executed, the above-mentioned performance-based building fire detector layout optimization method is implemented.

[0042] The present invention also provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned performance-based building fire detector layout optimization method is implemented.

[0043] The advantages of the present invention are:

[0044] (1) On the premise of meeting the specifications, the present invention takes into account the fire hazard level of a specific place and provides an optimal detector layout plan with the shortest detection time. This can solve the problem that the prior art does not consider the impact of the fire hazard level of a specific place on the detector layout, and the detector layout plan is rigid and not economical and optimal.

[0045] (2) The detector layout evaluation function (i.e., fitness function) of the present invention takes into account two indicators: the shortest detection time and full spatial coverage. The evaluation is more comprehensive. The optimal detector layout scheme selected based on the fitness function value has the shortest detection time while meeting the specifications.

[0046] (3) The present invention establishes a detector layout optimization model based on a genetic algorithm. The genetic algorithm is used to greatly reduce the calculation time, quickly find the maximum fitness function value, and quickly determine the optimal detector layout plan.

[0047] (4) The present invention uses the Latin hypercube sampling method to perform representative and effective sampling of fire parameters, randomly reorganizes the sampled values ​​of fire parameters (i.e., fire source location and fire growth rate), and constructs several fire scenarios for analysis. These fire scenarios represent various potential fire events, taking into account the fire hazard level of specific places, and can provide a more objective assessment of fire detection in the target environment.

[0048] (5) The present invention designs a batch calculation method for fire scene data sets based on fire simulation software, realizes batch calculation of fire scene data sets in fire simulation software, and generates corresponding fire detection time data sets, which has higher calculation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 The present invention is a flow chart of a performance-based building fire detector layout optimization method.

[0050] Figure 2 Fire scenes with different fire source locations.

[0051] Figure 3 This is a statistical distribution diagram of the fire growth rate.

[0052] Figure 4 Fire heat release rate curve.

[0053] Figure 5 This is the optimal detector arrangement scheme obtained according to the method of the present invention.

[0054] Figure 6 This is the detector arrangement scheme obtained according to traditional specifications.

[0055] Figure 7This is the fitness function value variation curve based on the genetic algorithm of the present invention. DETAILED DESCRIPTION

[0056] 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.

[0057] Example 1

[0058] In order to achieve the optimal layout of fire detectors in a building, this embodiment provides a performance-based building fire detector layout optimization method. Figure 1 As shown, the following steps are included:

[0059] S1: Obtain a fire scene dataset. In order to cover various potential fire events in the target environment, this method generates a comprehensive set of fire scenes. These fire scenes are designed considering two fire parameters: the location of the fire source (i.e., the horizontal coordinate of the fire source) and the position of the fire source. X and the vertical coordinate Y ) and the fire growth rate (i.e., the fire source growth coefficient The fire source location is selected to take into account the distribution of combustibles in the target environment and the uncertainty of the fire source location in a single fire event. X and the vertical coordinate Y The fire growth rate is generated by using a normal distribution function. The fire growth rate is selected to take into account the differences in the flammability of the combustibles in the target environment, that is, the fire growth rate after the combustibles are ignited is different for places with different hazard levels. In this embodiment, the fire growth rate is generated by using a uniform distribution function. The specific fire parameters used for fire scene design are shown in Table 1 below.

[0060] Table 1 Fire parameters

[0061] ;

[0062] Subsequently, in order to ensure representative and effective sampling of the input space of fire parameters, the present invention adopts a probability-based sampling method, specifically the Latin hypercube sampling method. This method can select finite samples and effectively capture the distribution of the data range through the stratified Monte Carlo sampling method. In this embodiment, the Latin hypercube sampling method is used to sample the fire source location and the fire growth rate respectively, by dividing the given parameter probability distribution into 1000 groups of parameter intervals, and then extracting random values ​​from each parameter interval respectively, and finally combining a total of 1000 random fire scenes, such as Figure 2 , Figure 3 shown.

[0063] The Latin hypercube sampling method is an existing technology, and the processing method is as follows:

[0064] S11, define the value range of each variable (fire parameter).

[0065] S12, interval division: For the value range of each variable (fire parameter), divide it into an equal interval.

[0066] S13, randomly select sampling points: randomly select a sampling point in each interval, and each interval can only be sampled once, ensuring that each interval of each variable has a sampling point.

[0067] S14, permutation and combination: combine the sampling points of each variable to form several samples (fire scenarios) to ensure that different interval combinations of each variable are considered.

[0068] Figure 2 The fire source location sampling points shown in the figure conform to the normal distribution centered on the hazardous material storage area. These fire source location sampling points can accurately reflect the potential fire source locations in the target environment, that is, they take into account the key factor of the fire hazard level of the specific place. Figure 3 The statistical distribution of the sampled fire growth rates is depicted. The selected sample points are evenly distributed within the specified data range. The sampled fire growth rates are converted to HRR (fire heat release rate) using the time squared fire growth model. Figure 4 The fire heat release rate curve (i.e., HRR curve) was plotted. It can be seen that the HRR curve covers the entire range of fire growth rate, i.e., [0.01172, 0.1876] in Table 1.

[0069] By randomly reorganizing the sampled values ​​of fire parameters (i.e., fire source location and fire growth rate), 1,000 fire scenes were constructed for analysis. These scenes represent various potential fire events and are used to make a more objective assessment of fire detection in the target environment. The 1,000 fire scenes constitute the fire scene dataset in this embodiment.

[0070] S2, based on the fire simulation software Fire Dynamic Simulator (FDS), a batch calculation method for fire scene data sets is designed to batch calculate the detection time corresponding to each fire scene and obtain the detection time data set corresponding to the fire scene data set.

[0071] The specific batch calculation method is as follows: After the fire scene is sampled, the FDS sample file generator is used to realize the conversion of the fire scene data set to the FDS code file. By replacing the sampled fire parameters with the corresponding items in the FDS code template file prepared in advance, the FDS code executable files of a large number of fire scenes can be quickly constructed. The FDS code file will be operated with the automatically generated batch file running under the DOS or Windows operating system to realize the batch operation of the FDS code input executable file under the FDS calculation program.

[0072] The present invention implements batch calculation of fire scene data sets (a total of 1000 fire scenes in this embodiment) in fire simulation software (FDS) by writing program codes, and generates corresponding fire detection time data sets.

[0073] S3, construct the detector layout scheme evaluation function, by defining the fitness function fitness ( i Ind ) is used to evaluate the advantages and disadvantages of the detector layout scheme, as shown in the following formulas (1) to (5):

[0074] ;

[0075] ;

[0076] ;

[0077] ;

[0078] ;

[0079] In formula (1), For fire scene S m Detector layout plan i Ind The shortest detection time under i Ind Indicated by i The detector arrangement scheme consists of detectors, namely i Ind By detector i 1 to detector i i composition; f ( i 1, S m ) is the detector i 1In the fire scene S m The detection time required to detect a fire is as follows.

[0080] In formula (2), F 1 is the detector layout plan i Ind exist M The average shortest detection time under each fire scenario; M is the number of fire scenes.

[0081] In formula (3), the function I d ( X,Y ) is used to determine any point in space ( X,Y ) falls within the protection radius of a certain detector; X , Y are the horizontal and vertical coordinates of any point in space respectively; , The detector i c The horizontal and vertical coordinates of c =1,2,..., i ; R is the protection radius of the detector; X ∈[0, a ], Y ∈[0, b ], a , b are the length and width of the space respectively; if any point in the space ( X,Y ) falls within the protection radius of a detector, then I d ( X,Y ) value is 1, otherwise I d ( X,Y ) value is 0.

[0082] In formula (4), the function F 2 Used to measure the detector layout plan i Ind Is the entire space fully covered? Provide a plan for detector layout i Ind The spatial coverage factor of N It is the number of repeated experiments in the Monte Carlo method. Each experiment selects any point in the space for judgment. Indicates n The results of the experiment, if n The spatial point extracted in this experiment falls within the protection radius of a certain detector, then The value is 1, otherwise The value is 0.

[0083] In formula (5), F 1' and F 2' areF 1 and F 2 Normalized value in the range [0,1]; w 1 and w 2 are F 1' and F 2' weight value; fitness ( i Ind ) is the fitness function. The larger the fitness function value, the better the corresponding detector layout scheme. i Ind The better.

[0084] The present invention obtains the detector arrangement scheme through formula (1): i Ind In the fire scene S m The shortest detection time under the condition of F 1 is the detector layout plan i Ind exist M The average shortest detection time under a fire scene is M is 1000, M The fire scenes are the 1000 fire scenes generated by Latin hypercube sampling in step S1. F 1 can be used to measure the detector layout plan i Ind The advantages and disadvantages of the scheme in terms of detection time. On the other hand, the definition function F 2 Used to measure the detector layout plan i Ind Whether the entire space is fully covered, specifically, the Monte Carlo method, namely formula (3) and formula (4), is used to determine F 2 calculation formula; the function in formula (3) I d ( X,Y ) is used to determine any point in space ( X,Y ) falls within the protection radius of a certain detector. If any point in space falls within the protection radius of the detector, then I d ( X,Y ) value is 1, otherwise it is 0; Formula (4) gives F 2, can be used to quantify any point in space ( X,Y ) falls within the detector protection radius; N Repeat the experiment for times and you can get F The value of 2, F The value of 2 can be used to measure the detector layout plan i IndIn order to improve the accuracy of the model, the present invention uses the maximum and minimum normalization method to F 1 and F The value of 2 is normalized to [0,1], that is, F 1' and F 2'. Finally, formula (5) takes into account F 1' and F Under the influence of 2', the fitness function is constructed fitness ( i Ind ), which is done by giving F 1' and F 2' assign weight values ​​and calculate the fitness function values ​​under the influence of the two. The larger the fitness function value, the stronger the fitness function value. i Detector arrangement consisting of detectors i Ind The better.

[0085] S4, establish a detector layout optimization model based on genetic algorithm, and use genetic algorithm to solve the detector layout corresponding to the maximum fitness function value, that is, the goal of genetic algorithm is , argmax Function is used to select the fitness function fitness ( i Ind ) when the maximum value is obtained (detector layout scheme) i Ind ). Specifically, based on the fire scene dataset constructed in step S1 and the batch calculation method of the fire scene dataset designed in step S2, the fitness function is calculated. fitness ( i Ind ), the genetic algorithm is used to optimize the detector layout scheme, quickly find the maximum fitness function value, and obtain the optimal detector layout scheme. The specific optimization process is as follows:

[0086] S41, initialize the population: randomly initialize a certain number of detector layout schemes to construct an initial population; each detector layout scheme is each individual in the population.

[0087] S42, evaluate the fitness of the population, i.e., the quality of the individuals: evaluate according to the fitness function, and calculate the fitness function value of each individual in the population. The fitness function value represents the quality of each solution (individual), and the individual with a larger fitness function is better.

[0088] S43, selection: According to the fitness function value of each individual in the population, several individuals with larger fitness function values ​​are selected from the population as parents to generate the next generation. Individuals with larger fitness function values ​​are more likely to be selected, thereby ensuring that excellent solutions are transmitted in the population.

[0089] S44, crossover and mutation: From the selected parent individuals, a crossover operation (also called recombination operation) is performed with a certain probability to generate new offspring individuals. The crossover operation generates new offspring individuals by exchanging part of the chromosomes of the parent individuals. At the same time, according to a certain mutation probability, the gene values ​​at certain positions on the chromosomes of the parent individuals are randomly changed to generate new offspring individuals. The mutation operation helps to maintain the diversity of the population and avoid premature convergence to the local optimal solution.

[0090] S45, generate a new population: the new population is composed of offspring individuals.

[0091] S46, determine whether the termination condition is met. If so, end the genetic algorithm and output a new population. The individual with the largest fitness function value in the new population is the optimal detector layout plan; otherwise, jump to step S42 and continue to iterate to generate a new population until the termination condition is met and the optimal detector layout plan is obtained.

[0092] Among them, the termination condition is that the number of iterations reaches the set termination number of iterations, or the change of the fitness function value reaches stability.

[0093] In this embodiment, 6 detectors are taken as an example, namely i 1 to i 6, and the protection radius of each detector is 7m. Figure 5 , 6 As shown in the figure, the shortest detection time, detector layout scheme and coverage are all presented in the figure. Figure 5 The figure shows the optimal detector arrangement scheme obtained according to the method of the present invention, and the shortest detection time is 21.8s. Figure 6The figure shows the detector arrangement scheme obtained according to the requirements of the traditional specifications, and the shortest detection time is 24.81s. Based on this, it can be seen that the method provided in this embodiment can shorten the detection time by 2.14%, greatly improving the warning efficiency of the detector. In the actual application of the method of the present invention, the required parameters, such as the number of detectors, the minimum protection radius of the detector, etc., can be determined according to the actual situation of the target environment and the corresponding detector arrangement specifications, and then the detector arrangement scheme that is superior to the standard arrangement requirements can be obtained through the method of the present invention. For example, the number of detectors can be set to 5, 6 or 7, respectively, and the optimal detector arrangement scheme under various situations (various numbers of detectors) can be obtained by using the method of the present invention, and the shortest detection time of the optimal detector arrangement scheme under various situations can be obtained. In combination with the requirements of economic benefits and warning efficiency, the appropriate number of detectors can be selected.

[0094] It should be noted that, in this embodiment, if the optimal detector arrangement scheme is obtained according to the traditional method, it is necessary to perform approximately Tens of thousands of calculations are required to achieve this. Figure 7 is the fitness function value variation curve based on the genetic algorithm of the present invention, Figure 7 It can be seen that the optimal detector arrangement scheme can be determined by using the genetic algorithm provided by the present invention with only 30 iterations. It should be noted that in this embodiment, the iteration time only takes a few minutes, so the optimization method provided by the present invention can quickly determine the optimal detector arrangement scheme, while the time required for the traditional method to complete 20 million calculations cannot meet the actual needs.

[0095] Example 2

[0096] A computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, enables the processor to execute the steps of the decision-making behavior decision method described in the above-mentioned embodiment 1 according to various embodiments of the present application.

[0097] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0098] Example 3

[0099] An electronic device comprises a processor, a memory and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a performance-based building fire detector layout optimization method of the above-mentioned embodiment 1 is implemented.

[0100] The electronic device of the embodiment of the present application can be the mobile device itself, or a stand-alone device independent of the mobile device, which can communicate with the mobile device to receive the collected input signals from them and send the selected target decision behavior to them.

[0101] The electronic device includes one or more processors and a memory. The processor may be a central processing unit (CPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. The memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. One or more computer programs / instructions may be stored on the computer-readable storage medium, and the processor may run the program / instruction to implement the decision behavior decision method and / or other desired functions of the various embodiments of the present application described above.

[0102] Electronic devices may also include input devices and output devices.

[0103] Example 4

[0104] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes a computer program / instruction, which, when executed by a processor, enables the processor to execute the steps in the decision-making behavior decision method according to various embodiments of the present application described in the above-mentioned embodiment 1 of this specification.

[0105] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0106] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A performance-based building fire detector layout optimization method, characterized in that: The following steps are involved: S1, obtain a fire scene dataset; the fire scene includes two fire parameters: fire source location and fire growth rate; S2, using fire simulation software to design a batch calculation method for fire scene data sets, which is used to batch calculate the detection time corresponding to each fire scene, and obtain the detection time data set corresponding to the fire scene data set; S3, construct a detector layout scheme evaluation function, define a fitness function to evaluate the detector layout scheme, including two indicators: the shortest detection time and full spatial coverage; S4, establishing a detector arrangement optimization model based on a genetic algorithm, using the genetic algorithm to solve the detector arrangement corresponding to the maximum fitness function value, and obtaining the optimal detector arrangement; In step S3, the fitness function is specifically as follows: ； ； ； ； ； In formula (1), Arrangement plan for detector θ Ind In the fire scene S m The shortest detection time under θ Ind Indicated by i The detector arrangement scheme consists of detectors. θ Ind By detector θ 1 to detector θ i composition; f ( θ 1, S m ) is the detector θ 1In the fire scene S m The detection time required to detect a fire; S m Indicates m A fire scene; In formula (2), F 1 is the detector layout plan θ Ind exist M The average shortest detection time under each fire scenario; M is the number of fire scenes; In formula (3), the function I d ( X,Y ) is used to determine any point in space ( X,Y ) falls within the protection radius of a certain detector; X , Y are the horizontal and vertical coordinates of any point in space respectively; , The detector θ c The horizontal and vertical coordinates of c =1,2,..., i ; R is the protection radius of the detector; X ∈[0, a ], Y ∈[0, b ], a , b are the length and width of the space respectively; If any point in space ( X,Y ) falls within the protection radius of a detector, then I d ( X,Y ) value is 1, otherwise I d ( X,Y ) value is 0; In formula (4), the function F 2 is the detector layout plan θ Ind The spatial coverage factor is used to measure the detector layout plan θ Ind Whether the entire space is fully covered; N is the number of times the experiment is repeated, and each time any point in the space is selected for judgment; Indicates n The results of the experiment, if n If any point in the space extracted in this experiment falls within the protection radius of a detector, then The value is 1, otherwise The value is 0; In formula (5), F 1' and F 2' are F 1 and F 2 Normalized value in the range [0,1]; w 1 and w 2 are F 1' and F 2' weight value; fitness ( θ Ind ) is the fitness function. The larger the fitness function value, the better the corresponding detector layout scheme. θ Ind The better.

2. A performance-based building fire detector layout optimization method according to claim 1, characterized in that: The specific method of step S1 is: S11, define the value range of each fire parameter; S12, for each fire parameter value range, divide the value range into 𝑛 intervals; S13, for each fire parameter, a sampling point is randomly selected in each interval; S14, randomly combining the sampling points of various fire parameters to form a number of fire scenes, forming a fire scene data set.

3. The performance-based building fire detector layout optimization method according to claim 1 is characterized in that: Step S2, using the fire simulation software Fire Dynamic Simulator to design a batch calculation method for the fire scene data set.

4. A performance-based building fire detector layout optimization method according to claim 3, characterized in that: In step S2, the batch calculation method is specifically as follows: The FDS sample file generator is used to realize the conversion of fire scene data set to FDS code file, the fire parameters in the fire scene are replaced with the corresponding items in the FDS code template file, and the FDS code executable file of the fire scene is constructed; the batch processing program is used to realize the batch running of the FDS code executable file under the FDS calculation program.

5. The performance-based building fire detector layout optimization method according to claim 1 is characterized in that: In step S4, based on the batch calculation method of the fire scene data set constructed in step S1 and the fire scene data set designed in step S2, the fitness function value of the detector layout plan is calculated, and the genetic algorithm is used to optimize the detector layout plan to find the maximum fitness function value and obtain the optimal detector layout plan.

6. A performance-based building fire detector layout optimization method according to claim 5, characterized in that: The specific optimization process of step S4 is as follows: S41, initializing the population: randomly initialize a certain number of detector layout schemes to construct an initial population; each detector layout scheme is each individual in the population; S42, evaluate the fitness of individuals in the population: calculate the fitness function value of each individual in the population; S43, selection: according to the fitness function value of each individual in the population, select a number of individuals with larger fitness function values ​​from the population as parents; S44, crossover and mutation: From the selected parent individuals, crossover and mutation operations are performed according to a certain probability to generate new offspring individuals; S45, generate a new population: the new population is composed of offspring individuals; S46, determine whether the termination condition is met. If so, end the genetic algorithm and output a new population. The individual with the largest fitness function value in the new population is the optimal detector layout plan. Otherwise, jump to step S42, continue to iterate, and continue to generate new populations until the termination condition is met to obtain the optimal detector layout plan.

7. A computer program product, characterized in that It includes a computer program / instruction, which, when executed by a processor, implements a performance-based building fire detector layout optimization method as described in any one of claims 1 to 6.

8. A readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed, the method for optimizing the layout of building fire detectors based on performance is implemented as described in any one of claims 1 to 6.

9. An electronic device, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a performance-based building fire detector layout optimization method as described in any one of claims 1 to 6 is implemented.

Images