Evaluation method for measuring alarm performance of pyrolysis particle type electrical fire monitoring detector

Abstract

The invention discloses an evaluation method for measuring the alarm performance of a pyrolysis particle type electrical fire monitoring detector. The method comprises the following steps: acquiring pyrolysis particle data, temperature monitoring data, interference data and alarm data; deviation data of the alarm data are obtained based on the pyrolysis particle data, and the pyrolysis particle data, the temperature monitoring data and the interference data are extracted and selected to obtain influence characteristics; linear dependence among the influence characteristics is removed to obtain associated data, and an alarm performance evaluation index is established based on the associated data; constructing an alarm performance evaluation model according to the alarm performance evaluation index, and optimizing the alarm performance evaluation model by adopting a differential evolution algorithm; and inputting to-be-evaluated data into the alarm performance evaluation model, and outputting an evaluation result. The method not only can improve the accuracy of electrical fire monitoring detector alarm performance evaluation, but also has good interpretability, and can be directly applied to an electrical fire monitoring detector alarm performance evaluation system.

Description

Technical Field

[0001] The invention relates to the technical field of electrical fire monitoring, and in particular to a method for evaluating the alarm performance of a pyrolytic particle type electrical fire monitoring detector. Background Art

[0002] Electrical fire is a common disaster in modern society, and its occurrence is often accompanied by serious economic losses and casualties. In order to effectively prevent electrical fires, pyrolytic particle electrical fire monitoring detectors are widely used in various buildings and industrial facilities. By monitoring the changes in the concentration of pyrolytic particles in the environment, such detectors can send out alarm signals in time at the early stage of the fire, thereby providing valuable time for the early control and extinguishing of the fire. However, the existing pyrolytic particle electrical fire monitoring detectors have problems such as unstable alarm performance and high false alarm rate in practical applications, which seriously affect their reliability and practicality. Especially in complex environments, detectors are easily affected by factors such as dust, humidity, temperature fluctuations and electromagnetic interference, causing the alarm time to deviate from the preset threshold, reducing the accuracy and timeliness of the alarm. Therefore, how to accurately evaluate and optimize the alarm performance of pyrolytic particle electrical fire monitoring detectors has become a key issue that needs to be urgently solved in the current field of electrical fire monitoring. Existing evaluation methods mostly rely on a single data indicator, lack a comprehensive analysis of the influence of multiple factors, and it is difficult to fully reflect the actual performance of the detector. Therefore, there is an urgent need for an alarm performance evaluation method that can comprehensively consider multi-dimensional information such as pyrolysis particle data, temperature monitoring data, interference data, etc., in order to improve the accuracy and interpretability of the evaluation. Summary of the invention

[0003] The purpose of the present invention is to provide a method for measuring the alarm performance of a pyrolytic particle type electrical fire monitoring detector.

[0004] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0005] The present invention comprises the following steps:

[0006] Acquire pyrolysis particle data, temperature monitoring data, interference data and alarm data, wherein the interference data includes dust, humidity, temperature and electromagnetic interference in the test environment;

[0007] Obtain deviation data of the alarm data based on the pyrolysis particle data, and extract and select the pyrolysis particle data, the temperature monitoring data and the interference data based on the deviation data to obtain an influencing feature;

[0008] Removing the linear dependencies between the influencing features to obtain correlation data, and establishing an alarm performance evaluation index based on the correlation data;

[0009] Constructing an alarm performance evaluation model according to the alarm performance evaluation index, and optimizing the alarm performance evaluation model using a differential evolution algorithm;

[0010] The data to be evaluated is input into the alarm performance evaluation model, and the evaluation results are output.

[0011] Furthermore, the method for obtaining deviation data of the alarm data based on the pyrolysis particle data comprises:

[0012] The standard alarm time is obtained based on the preset pyrolytic particle alarm threshold and pyrolytic particle data of the pyrolytic particle type electrical fire monitoring detector, and the deviation data is obtained according to the offset of the alarm data relative to the standard alarm time.

[0013] Furthermore, the method of extracting and selecting the pyrolysis particle data, the temperature monitoring data and the interference data based on the deviation data to obtain the influencing features includes:

[0014] Extract the statistical features of pyrolysis particle data, temperature monitoring data and interference data, perform standardization preprocessing on the statistical features, and normalize the deviation data.

[0015] The decision tree algorithm is used to establish a regression model of statistical characteristics and deviation data, and the goodness of fit is:

[0016]

[0017] Where L is the goodness of fit, n is the number of target values, and the target value is the deviation from the data. It is target value, It is before The pair y obtained by the iteration i The predicted value of , l(·) is the loss function, It is The decision tree added in the iteration The value of It is Statistical features, is the initial weight, ξ is the decay coefficient, K is the number of decision trees, It is decision tree, Ω(·) is the regularization term used to control the complexity of the prediction model, β is the depth weight, D is the maximum depth of the leaf node in the decision tree, and e is the base of the natural logarithm.

[0018] The calculation formula for split gain is:

[0019]

[0020] Among them, I is the sample set of the current node, I Land I R are the sample sets of the left child node and the right child node after splitting, ε is the regularization coefficient, U n For the uncertainty of division,

[0021] Based on the recursive feature elimination method, the influence coefficient of the statistical feature on the Pearson correlation coefficient between the predicted value and the target value is obtained, and the statistical feature is selected based on the influence coefficient to obtain the influencing feature.

[0022] Furthermore, the method of removing the linear dependency between the influencing features to obtain the associated data includes:

[0023] A data matrix is ​​established based on the influencing features and the corresponding deviation data, and the elastic network model is trained using the data matrix. The loss function of the elastic network model is:

[0024]

[0025] Where v is the parameter vector of the elastic network model, v0 is the intercept term, and s t,j is the j-th feature data in the t-th column of the data matrix affecting the feature, v j For t,j The feature weight, T is the number of columns in the data matrix, is the number of data sets obtained, J is the number of features in the influencing feature, z t represents the deviation data of the tth column of the data matrix, λ1 and λ2 are regularization parameters, ρ is the smoothing coefficient, ρ∈[0,1], ‖·‖ represents the Euclidean distance, and the optimal parameter vector is:

[0026]

[0027] The elastic network model is trained with the optimal parameter vector, and the influencing features whose absolute values ​​of feature weights are less than the preset weight threshold are removed. The screened influencing features and their feature weights are used as influencing data.

[0028] Furthermore, a method for establishing an alarm performance evaluation index based on the associated data includes:

[0029]

[0030] Where Y is the alarm performance evaluation index, is the impact weight of the deviation data, is the influence weight of the i-th influencing feature in the influencing data, v i is the feature weight of the i-th influencing feature in the influencing data, s t,i is the eigenvalue of the i-th influencing feature in the influencing data in the t-th column of the data matrix, and I is the number of influencing features in the influencing data.

[0031] Furthermore, a method for constructing an alarm performance evaluation model according to the alarm performance evaluation index includes:

[0032] A data set is established using the alarm performance evaluation index and impact characteristics. The data set is divided into a training set and a test set by random sampling. The test set is used to train the alarm performance evaluation model using a machine learning algorithm to learn the mapping relationship between the impact characteristics and the alarm performance evaluation index. The test set is used to optimize the hyperparameters of the alarm performance evaluation model using a differential evolution algorithm.

[0033] Furthermore, the method of optimizing the alarm performance evaluation model using a differential evolution algorithm includes:

[0034] Use the hyperparameters of the alarm performance evaluation model as individuals to establish the initial population, perform crossover, mutation, and selection on the individuals, and optimize the population. The fitness function of the population is the mean square error, and the energy function is:

[0035]

[0036] in Indicates The individual in The energy function value of the generation, Φ min represents the minimum value of the energy function during the optimization process, Φ max represents the maximum value of the energy function during the optimization process, Indicates the current iteration number, represents the total number of iterations, For individuals The fitness function, ψ worst is the worst fitness value in the population, ψ best is the optimal fitness value in the population, N is the total number of individuals in the population,

[0037] Update formula for mutation rate and crossover rate:

[0038]

[0039] in For the The individual in The mutation rate of each generation, I base is the basic mutation rate, I base ∈[0.4,1], and Respectively The maximum and minimum values ​​of the energy function of the generation population, and Respectively The maximum and minimum values ​​of the energy function of the generation population, For the The individual in The crossover rate of each generation, Γ base is the basic crossover rate, Γ base ∈[0,1], and are the smoothing factors for mutation rate and crossover rate respectively, and the population is optimized iteratively until the maximum number of iterations is reached.

[0040] In a second aspect, an embodiment of the present application further provides an electronic device, including:

[0041] A processor; and a memory arranged to store computer executable instructions, which when executed cause the processor to perform the method steps described in the first aspect.

[0042] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple applications, the electronic device executes the method steps described in the first aspect.

[0043] The beneficial effects of the present invention are:

[0044] The present invention is a method for measuring the alarm performance of a pyrolytic particle type electrical fire monitoring detector. Compared with the prior art, the present invention has the following technical effects:

[0045] The present invention significantly improves the accuracy and reliability of the alarm performance evaluation of electrical fire monitoring detectors by integrating pyrolysis particle data, temperature monitoring data and interference data, combining with the differential evolution algorithm optimization model, effectively reduces the false alarm rate, and enhances the practicality and interpretability of the detectors. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flowchart of the steps of the method for evaluating the alarm performance of a pyrolytic particle type electrical fire monitoring detector according to the present invention;

[0047] Figure 2 The figure is the appearance of the test cabin and the back panel layout;

[0048] Figure 3 It is a schematic diagram of the structure of an electronic device in an embodiment of this specification. DETAILED DESCRIPTION

[0049] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0050] The method for measuring the alarm performance of a pyrolytic particle type electrical fire monitoring detector of the present invention comprises the following steps:

[0051] like Figure 1 As shown, in this embodiment, the following steps are included:

[0052] Acquire pyrolysis particle data, temperature monitoring data, interference data and alarm data, wherein the interference data includes dust, humidity, temperature and electromagnetic interference in the test environment;

[0053] In the actual evaluation, the test is carried out in a test chamber. The appearance and back panel layout of the test chamber are as follows: Figure 2 As shown, the left side is the appearance diagram, and the right side is the back panel layout diagram, wherein the index numbers are: 1 - test chamber, 2 - test chamber cover, 3 - test chamber back panel, 4 - fan, 5 - detector, 6 - heating furnace, 7 - pyrolysis particle sampling port;

[0054] Three pyrolytic materials were used in the evaluation:

[0055] Polyvinyl chloride (PVC) pyrolysis sheet, mass 10g ± 1g, length 90mm ± 5mm, width 90mm ± 5mm, thickness 1mm ± 0.3mm, formula:

[0056]

[0057]

[0058] Acrylonitrile-butadiene-styrene copolymer (ABS) pyrolysis sheet, mass 10g±1g, length 53mm±2mm, width 53mm±2mm, thickness 3mm±0.5mm, formula:

[0059]

[0060] FR-4 epoxy resin glass fiber cloth laminate pyrolysis sheet, mass 10g ± 1g, length 60mm ± 2mm, width 60mm ± 2mm, thickness 2mm ± 0.5mm, the organic solvent of the curing reaction is acetone, the chemical composition after curing and lamination accounts for 43% brominated epoxy resin and 57% alkali-free glass fiber cloth, the formula of FR-4 epoxy resin glass fiber cloth laminate pyrolysis sheet is:

[0061] Material Category Main resin Curing agent Accelerator Element Brominated Epoxy Resin Dicyandiamide 2-Methylimidazole Number of copies 100 3 0.3

[0062] Test conditions:

[0063] Heat the polyvinyl chloride (PVC) pyrolysis sheet at a heating rate of 3°C / min to 80°C, hold for 5 minutes, then heat to 180°C at a heating rate of 10°C / min, hold for 3 minutes. Start timing after the heating furnace is heated to 80°C and kept at a constant temperature for 5 minutes, calculate the alarm time of the detector, and record the concentration of particles with a diameter less than 1μm (PM1.0) in the test chamber;

[0064] The acrylonitrile-butadiene-styrene copolymer (ABS) pyrolysis sheet was heated to 90°C at a heating rate of 3°C / min, maintained for 5 minutes, and then heated to 190°C at a heating rate of 10°C / min, maintained for 3 minutes; the timing was started after the heating furnace was heated to 90°C and kept at a constant temperature for 5 minutes, the alarm time of the detector was calculated, and the concentration of particles with a diameter less than 1μm (PM1.0) in the test chamber was recorded;

[0065] The FR-4 epoxy glass fiber cloth laminate pyrolysis sheet was heated at a heating rate of 3°C / min to 90°C, maintained for 5 minutes, and then heated to 190°C at a heating rate of 10°C / min, maintained for 3 minutes; the timing was started after the heating furnace was heated to 90°C and kept at a constant temperature for 5 minutes, the alarm time of the detector was calculated, and the concentration of particles with a diameter of less than 1μm (PM1.0) in the test chamber was recorded;

[0066] The test ends when the detector sends out an alarm signal or the following conditions are met:

[0067] Polyvinyl chloride (PVC) pyrolysis sheet: heat the furnace to 180℃ and keep constant temperature for 3 minutes;

[0068] Acrylonitrile-butadiene-styrene copolymer (ABS) pyrolysis sheet: heat the furnace to 190°C and keep the temperature constant for 3 minutes;

[0069] FR-4 epoxy glass fiber cloth laminate: heat the furnace to 190℃ and keep it constant for 3 minutes;

[0070] The temperature monitoring data includes the temperature of the detector surface, the temperature of the pyrolysis sheet and the heating furnace. The interference data is the normal data in the test environment. The alarm data is the alarm time of the detector after the pyrolysis sheet is heated to the preset condition.

[0071] Obtain deviation data of the alarm data based on the pyrolysis particle data, and extract and select the pyrolysis particle data, the temperature monitoring data and the interference data based on the deviation data to obtain an influencing feature;

[0072] Removing the linear dependencies between the influencing features to obtain correlation data, and establishing an alarm performance evaluation index based on the correlation data;

[0073] Constructing an alarm performance evaluation model according to the alarm performance evaluation index, and optimizing the alarm performance evaluation model using a differential evolution algorithm;

[0074] The data to be evaluated is input into the alarm performance evaluation model, and the evaluation results are output.

[0075] In this embodiment, the method for obtaining the deviation data of the alarm data based on the pyrolysis particle data includes:

[0076] The standard alarm time is obtained based on the preset pyrolytic particle alarm threshold and pyrolytic particle data of the pyrolytic particle type electrical fire monitoring detector, and the deviation data is obtained according to the offset of the alarm data relative to the standard alarm time.

[0077] In this embodiment, the method of extracting and selecting the pyrolysis particle data, the temperature monitoring data and the interference data based on the deviation data to obtain the influencing features includes:

[0078] Extract the statistical features of pyrolysis particle data, temperature monitoring data and interference data, perform standardization preprocessing on the statistical features, and normalize the deviation data.

[0079] The decision tree algorithm is used to establish a regression model of statistical characteristics and deviation data, and the goodness of fit is:

[0080]

[0081] Where L is the goodness of fit, n is the number of target values, and the target value is the deviation from the data. It is target value, It is before The pair y obtained by the iteration i The predicted value of , l(·) is the loss function, It is The decision tree added in the iteration The value of It is Statistical features, is the initial weight, ξ is the decay coefficient, K is the number of decision trees, It is decision tree, Ω(·) is the regularization term used to control the complexity of the prediction model, β is the depth weight, D is the maximum depth of the leaf node in the decision tree, and e is the base of the natural logarithm.

[0082] The calculation formula for split gain is:

[0083]

[0084] Among them, I is the sample set of the current node, I L and I R are the sample sets of the left child node and the right child node after splitting, ε is the regularization coefficient, U n For the uncertainty of division,

[0085] Based on the recursive feature elimination method, the influence coefficient of the statistical feature on the Pearson correlation coefficient between the predicted value and the target value is obtained, and the statistical feature is selected based on the influence coefficient to obtain the influencing feature.

[0086] In this embodiment, the method of removing the linear dependency between the influencing features to obtain the associated data includes:

[0087] A data matrix is ​​established based on the influencing features and the corresponding deviation data, and the elastic network model is trained using the data matrix. The loss function of the elastic network model is:

[0088]

[0089] Where v is the parameter vector of the elastic network model, v0 is the intercept term, and s t,j is the j-th feature data in the t-th column of the data matrix affecting the feature, v j For t,j The feature weight, T is the number of columns in the data matrix, 3<T<50, is the number of acquired data sets, J is the number of features in the influencing feature, z t represents the deviation data of the tth column of the data matrix, λ1 and λ2 are regularization parameters, ρ is the smoothing coefficient, ρ∈[0,1], ‖·‖ represents the Euclidean distance, and the optimal parameter vector is:

[0090]

[0091] The elastic network model is trained with the optimal parameter vector, and the influencing features whose absolute values ​​of feature weights are less than the preset weight threshold are removed. The screened influencing features and their feature weights are used as influencing data.

[0092] In this embodiment, the method for establishing an alarm performance evaluation index based on the associated data includes:

[0093]

[0094] Where Y is the alarm performance evaluation index, is the impact weight of the deviation data, is the influence weight of the i-th influencing feature in the influencing data, v i is the feature weight of the i-th influencing feature in the influencing data, s t,i is the eigenvalue of the i-th influencing feature in the influencing data in the t-th column of the data matrix, and I is the number of influencing features in the influencing data.

[0095] In this embodiment, the method for constructing an alarm performance evaluation model according to the alarm performance evaluation index includes:

[0096] A data set is established using the alarm performance evaluation index and impact characteristics. The data set is divided into a training set and a test set by random sampling. The test set is used to train the alarm performance evaluation model using a machine learning algorithm to learn the mapping relationship between the impact characteristics and the alarm performance evaluation index. The test set is used to optimize the hyperparameters of the alarm performance evaluation model using a differential evolution algorithm.

[0097] In this embodiment, the method of optimizing the alarm performance evaluation model using a differential evolution algorithm includes:

[0098] Use the hyperparameters of the alarm performance evaluation model as individuals to establish the initial population, perform crossover, mutation, and selection on the individuals, and optimize the population. The fitness function of the population is the mean square error, and the energy function is:

[0099]

[0100] in Indicates The individual in The energy function value of the generation, Φ min represents the minimum value of the energy function during the optimization process, Φ max represents the maximum value of the energy function during the optimization process, Indicates the current iteration number, represents the total number of iterations, For individuals The fitness function, ψ worst is the worst fitness value in the population, ψ best is the optimal fitness value in the population, N is the total number of individuals in the population,

[0101] Update formula for mutation rate and crossover rate:

[0102]

[0103] in For the The individual in The mutation rate of each generation, I base is the basic mutation rate, I base ∈[0.4,1], and Respectively The maximum and minimum values ​​of the energy function of the generation population, and Respectively The maximum and minimum values ​​of the energy function of the generation population, For the The individual in The crossover rate of generations, Γ base is the basic crossover rate, Γ base∈[0,1], and are the smoothing factors for mutation rate and crossover rate respectively, and the population is optimized iteratively until the maximum number of iterations is reached.

[0104] Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 2 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0105] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0106] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0107] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a device for evaluating the alarm performance of a pyrolytic particle-type electrical fire monitoring detector at the logical level. The processor executes the program stored in the memory and is specifically used to execute any of the aforementioned methods for evaluating the alarm performance of a pyrolytic particle-type electrical fire monitoring detector.

[0108] The above application Figure 1The method for evaluating the alarm performance of a pyrolytic particle type electrical fire monitoring detector disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or by instructions in the form of software. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in a decoding processor. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0109] The electronic device may also perform Figure 1 A method for evaluating the alarm performance of pyrolytic particle electrical fire monitoring detectors was developed and implemented Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0110] An embodiment of the present application also proposes a computer-readable storage medium, which stores one or more programs, and the one or more programs include instructions. When the instructions are executed by an electronic device including multiple application programs, they execute any of the aforementioned methods for evaluating the alarm performance of pyrolytic particle electrical fire monitoring detectors.

[0111] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0112] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0113] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0114] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0115] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0116] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0117] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0118] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0119] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0120] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for evaluating the alarm performance of a pyrolytic particle type electrical fire monitoring detector, characterized in that: The following steps are involved: Acquire pyrolysis particle data, temperature monitoring data, interference data and alarm data, wherein the interference data includes dust, humidity, temperature and electromagnetic interference in the test environment; Obtain deviation data of the alarm data based on the pyrolysis particle data, and extract and select the pyrolysis particle data, the temperature monitoring data and the interference data based on the deviation data to obtain an influencing feature; Removing the linear dependencies between the influencing features to obtain correlation data, and establishing an alarm performance evaluation index based on the correlation data; Constructing an alarm performance evaluation model according to the alarm performance evaluation index, and optimizing the alarm performance evaluation model using a differential evolution algorithm; The data to be evaluated is input into the alarm performance evaluation model, and the evaluation results are output.

2. The method for evaluating the alarm performance of a pyrolytic particle type electrical fire monitoring detector according to claim 1, characterized in that: The method for obtaining deviation data of the alarm data based on the pyrolysis particle data comprises: The standard alarm time is obtained based on the preset pyrolytic particle alarm threshold and pyrolytic particle data of the pyrolytic particle type electrical fire monitoring detector, and the deviation data is obtained according to the offset of the alarm data relative to the standard alarm time.

3. The method for evaluating the alarm performance of a pyrolytic particle type electrical fire monitoring detector according to claim 1, characterized in that: The method for extracting and selecting the pyrolysis particle data, the temperature monitoring data and the interference data based on the deviation data to obtain the influencing features comprises: Extract the statistical features of pyrolysis particle data, temperature monitoring data and interference data, perform standardization preprocessing on the statistical features, and normalize the deviation data. The decision tree algorithm is used to establish a regression model of statistical characteristics and deviation data, and the goodness of fit is: Where L is the goodness of fit, n is the number of target values, and the target value is the deviation from the data. It is target value, It is before The pair y obtained by the iteration i The predicted value of , l(·) is the loss function, It is The decision tree added in the iteration The value of It is Statistical features, is the initial weight, ξ is the decay coefficient, K is the number of decision trees, It is decision tree, Ω(·) is the regularization term used to control the complexity of the prediction model, β is the depth weight, D is the maximum depth of the leaf node in the decision tree, and e is the base of the natural logarithm. The calculation formula for split gain is: Among them, I is the sample set of the current node, I L and I R are the sample sets of the left child node and the right child node after splitting, ε is the regularization coefficient, U n For the uncertainty of division, Based on the recursive feature elimination method, the influence coefficient of the statistical feature on the Pearson correlation coefficient between the predicted value and the target value is obtained, and the statistical feature is selected based on the influence coefficient to obtain the influencing feature.

4. The method for evaluating the alarm performance of a pyrolytic particle type electrical fire monitoring detector according to claim 1 is characterized in that: The method for removing the linear dependency between the influencing features to obtain the associated data includes: A data matrix is ​​established based on the influencing features and the corresponding deviation data, and the elastic network model is trained using the data matrix. The loss function of the elastic network model is: Where v is the parameter vector of the elastic network model, v0 is the intercept term, and s t,j is the j-th feature data in the t-th column of the data matrix affecting the feature, v j For t,j The feature weight, T is the number of columns in the data matrix, is the number of data sets obtained, J is the number of features in the influencing feature, z t represents the deviation data of the tth column of the data matrix, λ1 and λ2 are regularization parameters, ρ is the smoothing coefficient, ρ∈[0,1], ‖·‖ represents the Euclidean distance, and the optimal parameter vector is: The elastic network model is trained with the optimal parameter vector, and the influencing features whose absolute values ​​of feature weights are less than the preset weight threshold are removed. The screened influencing features and their feature weights are used as influencing data.

5. The method for evaluating the alarm performance of a pyrolytic particle type electrical fire monitoring detector according to claim 1 is characterized in that: The method for establishing an alarm performance evaluation index based on the associated data includes: Where Y is the alarm performance evaluation index, is the impact weight of the deviation data, is the influence weight of the i-th influencing feature in the influencing data, v i is the feature weight of the i-th influencing feature in the influencing data, s t,i is the eigenvalue of the i-th influencing feature in the influencing data in the t-th column of the data matrix, and I is the number of influencing features in the influencing data.

6. The method for evaluating the alarm performance of a pyrolytic particle type electrical fire monitoring detector according to claim 1, characterized in that: The method for constructing an alarm performance evaluation model according to the alarm performance evaluation index comprises: A data set is established using the alarm performance evaluation index and impact characteristics. The data set is divided into a training set and a test set by random sampling. The test set is used to train the alarm performance evaluation model using a machine learning algorithm to learn the mapping relationship between the impact characteristics and the alarm performance evaluation index. The test set is used to optimize the hyperparameters of the alarm performance evaluation model using a differential evolution algorithm.

7. The method for evaluating the alarm performance of a pyrolytic particle type electrical fire monitoring detector according to claim 1, characterized in that: The method of optimizing the alarm performance evaluation model using a differential evolution algorithm comprises: Use the hyperparameters of the alarm performance evaluation model as individuals to establish the initial population, perform crossover, mutation, and selection on the individuals, and optimize the population. The fitness function of the population is the mean square error, and the energy function is: in Indicates The individual in The energy function value of the generation, Φ min represents the minimum value of the energy function during the optimization process, Φ max represents the maximum value of the energy function during the optimization process, Indicates the current iteration number, represents the total number of iterations, For individuals The fitness function, ψ worst is the worst fitness value in the population, ψ best is the optimal fitness value in the population, N is the total number of individuals in the population, Update formula for mutation rate and crossover rate: in For the The individual in The mutation rate of each generation, I base is the basic mutation rate, I base ∈[0.4,1], and Respectively The maximum and minimum values ​​of the energy function of the generation population, and Respectively The maximum and minimum values ​​of the energy function of the generation population, For the The individual in The crossover rate of generations, Γ base is the basic crossover rate, Γ base ∈[0,1], and are the smoothing factors for mutation rate and crossover rate respectively, and the population is optimized iteratively until the maximum number of iterations is reached.

8. An electronic device comprising: processor; as well as A memory arranged to store computer executable instructions, which when executed cause the processor to perform the method according to any one of claims 1 to 7.

9. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, enables the electronic device to execute the method according to any one of claims 1 to 7.

Images