CN115909641B - Fire detection method, system, device and medium with balanced black and white smoke response - Google Patents

Abstract

The invention discloses a fire detection method, a system, equipment and a medium for balancing black-and-white smoke response, belonging to the field of fire smoke detection, wherein the method comprises the following steps: designing particles with various shapes to respectively represent black smoke and white smoke, and calculating and combining different scattered light receiving angles theta _i and scattered light intensities I _ij of the particles with different particle diameters d _j for each particle to obtain an original scattered light intensity matrix I _M×N; calculating the ratio of each row to the ith row in I _M×N to obtain an ith intermediate matrix, and combining to obtain an extended light intensity ratio matrixDetecting target particles by using different theta _i, and combining the detected scattered light intensities after the ratio of the scattered light intensities is calculated to obtain a light intensity ratio vectorCalculation ofAnd each ofThe distance between each column in (b) willAnd taking the ratio between the columns corresponding to the minimum distance as a scattering light intensity balance value of the target particles, and judging that fire disaster occurs when the scattering light intensity balance value exceeds an alarm threshold value. The method has higher detection precision for black smoke and white smoke.

Description

Fire detection method, system, device and medium with balanced black and white smoke response

Technical Field

The present invention belongs to the field of fire smoke detection, and more specifically, relates to a fire detection method, system, equipment and medium for balanced black and white smoke response.

Background technique

In the early stage of a fire, objects undergo pyrolysis under the action of external heat sources, generating combustible volatile substances and solid carbon, namely fire smoke. Smoke detectors can respond quickly and accurately to the trace smoke particles emitted by the fire source and issue alarms in a timely manner, which is of great significance in disaster suppression and fire control. Among them, photoelectric smoke detectors based on the scattering principle are more suitable for use as point-type smoke detectors and are widely used in large-area detection.

The smoldering process of smoldering objects can quickly produce a large number of spherical white smoke particles, and the light scattering effect of white smoke particles is very regular. Other combustibles such as polymer compounds will produce solid carbon during pyrolysis, and solid carbon will be pyrolyzed before combustion to generate a large number of free carbon black particles. In addition, when the amount of oxygen is low, the combustible gas will not burn completely, and a large number of free carbon black particles will also be produced. Therefore, the smoke in the early stage of the fire caused by these substances is often black or grayish black, usually called black smoke. Carbon black particles have a strong ability to absorb light, and are prone to aggregation, resulting in more irregular shapes of black smoke particles and more complex light scattering patterns. This leads to differences in the sensitivity of smoke detectors to white smoke and black smoke during fire detection, that is, when the detector's response threshold is just right for white smoke detection, black smoke needs to reach a higher concentration to trigger an alarm, making the detector susceptible to the type of combustion material and causing different degrees of false alarms or missed alarms.

Summary of the invention

In view of the defects of the prior art and the need for improvement, the present invention provides a fire detection method, system, equipment and medium with balanced black and white smoke responses, with the aim of solving the problem in the existing scattered smoke photoelectric detection technology that the concentration detector has a low alarm accuracy rate due to the irregular shape of black smoke and its strong light absorption.

To achieve the above-mentioned purpose, according to one aspect of the present invention, a fire detection method for balancing black and white smoke responses is provided, comprising a design stage and a detection stage, wherein the design stage comprises S1-S2, and the detection stage comprises S3-S4; S1, pre-designing particles of various shapes to respectively characterize black smoke and white smoke, for each particle, calculating and combining the scattered light intensity I _ij at different scattered light receiving angles θ _i and different particle sizes d _j to obtain its original scattered light intensity matrix I _M×N , i=1, 2, ..., M, j=1, 2, ..., N, M and N are set parameters; S2, calculating the ratio of the scattered light intensity of each row in the original scattered light intensity matrix I _M×N to the scattered light intensity of its i-th row to obtain the i-th intermediate matrix, combining the i-th intermediate matrices to obtain the corresponding extended light intensity ratio matrix S3, detect the target particles at different scattered light receiving angles _θi , calculate the ratio of the detected scattered light intensities in pairs and combine them to obtain the light intensity ratio vector S4, calculate the light intensity ratio vector With each extended light intensity ratio matrix The distance between each column in the image is the intensity ratio vector The ratio between the columns corresponding to the minimum distance is used as the scattered light intensity balance value of the target particles. When the scattered light intensity balance value exceeds the alarm threshold, it is determined that a fire has occurred.

Furthermore, the design stage also includes: combining each extended light intensity ratio matrix Calculate the average standard deviation of each row, for each extended light intensity ratio matrix L rows with the smallest average standard deviation are selected to form a corresponding optimal light intensity ratio matrix _IL×N , where L is a given parameter between 1 and M.

Furthermore, in S3, the target particles are detected at a number of scattered light receiving angles θ _i corresponding to the L rows with the smallest average standard deviation, and the detected scattered light intensities are ratioed in pairs and combined to obtain an optimal light intensity ratio vector IL _×1 ; in S4, the distance between the optimal light intensity ratio vector _IL×1 and each column in each optimal light intensity ratio matrix _IL×N is calculated, and the ratio between the optimal light intensity ratio vector IL _×1 and the column corresponding to the minimum distance is used as the scattered light intensity balance value of the target particles.

Furthermore, in S1, spherical particles are designed to represent white smoke, and k ₁ types of rotating ellipsoidal particles with set major-minor axis ratios and/or k ₂ types of cylindrical particles with set bottom radii and set heights are designed to represent black smoke, and k ₁ and k ₂ are both natural numbers and cannot be 0 at the same time.

Furthermore, the design stage also includes: using the extended light intensity ratio matrix of the first particle As a benchmark, the expanded light intensity ratio matrix of the remaining particles is Normalization processing is performed to obtain relative response coefficients of the remaining particles, wherein the first particles are any one of particles characterizing black smoke and white smoke.

Furthermore, the S4 includes: calculating the light intensity ratio vector With each extended light intensity ratio matrix The distance between each column in the target particle type is determined based on the column with the minimum distance; if the determination result is consistent with the first particle type, the light intensity ratio vector When the alarm threshold is exceeded, it is determined that a fire has occurred. If it is inconsistent, the light intensity ratio vector and The ratio between them is used as the scattered light intensity balance value of the target particles. When the scattered light intensity balance value exceeds the alarm threshold, it is determined that a fire has occurred. is the column corresponding to the minimum distance corresponding column selected from the relative response coefficients.

Furthermore, the distance calculated in S4 is any one of Euclidean distance, Mahalanobis distance, Manhattan distance, Chebyshev distance, higher-order norm distance, cosine distance, and information entropy.

According to another aspect of the present invention, a fire detection system for balancing black and white smoke responses is provided, comprising: a design module for pre-designing particles of various shapes to respectively characterize black smoke and white smoke, for each particle, calculating and combining the scattered light intensity I _ij at different scattered light receiving angles θ _i and different particle sizes d _j to obtain its original scattered light intensity matrix I _M×N , i=1, 2, ..., M, j=1, 2, ..., N, M and N are set parameters; a calculation module for calculating the ratio of the scattered light intensity of each row in the original scattered light intensity matrix I _M×N to the scattered light intensity of its i-th row to obtain the i-th intermediate matrix, combining the i-th intermediate matrices to obtain the corresponding extended light intensity ratio matrix The detection module is used to detect the target particles at different scattered light receiving angles θ _i , and the detected scattered light intensities are combined after calculating the ratios of each two to obtain the light intensity ratio vector Determination module, used to calculate the light intensity ratio vector With each extended light intensity ratio matrix The distance between each column in the image is the intensity ratio vector The ratio between the columns corresponding to the minimum distance is used as the scattered light intensity balance value of the target particles. When the scattered light intensity balance value exceeds the alarm threshold, it is determined that a fire has occurred.

According to another aspect of the present invention, an electronic device is provided, characterized in that it includes: a processor; and a memory storing a computer executable program, wherein when the program is executed by the processor, the processor executes the fire detection method of balanced black and white smoke response as described above.

According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the program is executed by a processor, the fire detection method with balanced black and white smoke response as described above is implemented.

In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects:

(1) A fire detection method for balancing the response of black and white smoke is provided. In the design stage, the scattered light intensity distribution characteristics of typical particles similar to black and white smoke are preprocessed to extract the main scattering characteristics between different smokes. In the detection stage, the particle characteristics are first obtained by a simple feature distance discrimination method, and then the scattered light intensity ratio of black smoke and white smoke is balanced accordingly. Then, the scattered photoelectric smoke detector generates an alarm when the smoke generated by the smoldering of different substances reaches the same concentration;

(2) It can solve the problem that the existing detector uses the white smoke scattered light intensity corresponding to the threshold concentration as the alarm threshold, and the actual concentration of black smoke when the alarm is triggered is much greater than the threshold concentration, resulting in missed alarms. When the black smoke scattered light intensity corresponding to the threshold concentration is used as the alarm threshold, the white smoke concentration is much lower than the threshold, resulting in false alarms. This effectively avoids the problem that the accuracy of the concentration detector alarm is low due to the irregular shape of black smoke and its strong absorption of light.

(3) The main scattering features between the extracted different smokes are further extracted, and only the main scattering features with the smallest average standard deviation, that is, the smallest fluctuation, are extracted for subsequent equalization processing. On the basis of ensuring the equalization effect and alarm accuracy, the calculation complexity and the hardware resources required for calculation are greatly reduced.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a flow chart of a fire detection method with balanced black and white smoke response provided by an embodiment of the present invention;

FIG2 is a set of light intensity ratios with the smallest standard deviation of typical particle side angle scattered light intensity ratios provided by an embodiment of the present invention;

FIG3 is a normalized relative response coefficient of ellipsoidal particles provided in an embodiment of the present invention;

4 is a schematic diagram of determining the type of particles by distance comparison during measurement of a scattering type photoelectric smoke detector provided by an embodiment of the present invention;

FIG5 is a block diagram of a fire detection system with balanced black and white smoke response provided by an embodiment of the present invention;

FIG6 is a block diagram of an electronic device provided by an embodiment of the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

In the present invention, the terms "first", "second", etc. (if any) in the present invention and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

Fig. 1 is a flow chart of a fire detection method with balanced black and white smoke response provided by an embodiment of the present invention. Referring to Fig. 1 and in combination with Fig. 2 to Fig. 4, a fire detection method with balanced black and white smoke response in this embodiment is described in detail. The method includes a design phase and a detection phase. The design phase includes operations S1 to S2, and the detection phase includes operations S3 to S4.

Operation S1, pre-design various shapes of particles to represent black smoke and white smoke respectively. For each particle, calculate and combine the scattered light intensity I _ij at different scattered light receiving angles θ _i and different particle sizes d _j to obtain its original scattered light intensity matrix I _M×N , i=1, 2, ..., M, j=1, 2, ..., N, M and N are set parameters.

According to an embodiment of the present invention, in operation S1, spherical particles are designed to represent white smoke, k ₁ types of rotating ellipsoidal particles with set major-minor axis ratios and/or k ₂ types of cylindrical particles with set bottom radii and set heights are designed to represent black smoke, and k ₁ and k ₂ are both natural numbers and are not 0 at the same time.

In this embodiment, K typical particles are used to characterize white smoke and black smoke as an example, wherein white smoke is replaced by only one type of spherical particles with a complex refractive index of 1.55+0.02i, and black smoke is replaced by K-1 types of regular non-spherical particles such as rotating ellipsoidal particles and/or cylindrical particles with a complex refractive index of 1.55+0.55i, and K≥2. For each type of particle, the following calculation process and the calculation process in operation S2 are repeated.

According to the theoretical model of optical scattering of submicron/micron particles, the scattered light intensity distribution values of particles of N particle sizes are calculated under the wavelength of the light source to be used by the detector and M scattered light receiving angles, and recorded in the original scattered light intensity matrix I _M×N with a dimension of _M×N .

Specifically, for calculating the optical scattering intensity distribution characteristics of submicron/micron particles, theoretical models such as the Lorentz-Mie scattering theory model, the discrete dipole theory model, and the T-matrix calculation model can be used. Preferably, for spherical particles, the Lorentz-Mie scattering theory is the simplest and most efficient; for non-spherical particles, the discrete dipole theory model is more efficient.

Operation S2, calculate the ratio of the scattered light intensity of each row in the original scattered light intensity matrix I _M×N to the scattered light intensity of the i-th row, obtain the i-th intermediate matrix, combine the i-th intermediate matrices, and obtain the corresponding extended light intensity ratio matrix

Calculate the ratio of the scattered light intensity of each row in the original scattered light intensity matrix I _M×N to the scattered light intensity of the i-th row, and get the i-th intermediate matrix Thus, M new M×N dimensional light intensity ratio matrices are obtained. Further, these M matrices are concatenated into an extended light intensity ratio matrix with M ² rows and N columns. Each row represents the ratio of scattered light at two scattering angles. K typical particles get K M ² ×N dimensional expanded light intensity ratio matrices

Operation S3, detecting the target particles at different scattered light receiving angles θ _i , calculating the ratio of the detected scattered light intensities in pairs and combining them to obtain the light intensity ratio vector

The detector is designed to detect the target particles at M different scattered light receiving angles θ _i , i = 1, 2, ..., M, and M scattered light intensities are obtained. These M scattered light intensities are compared and combined to obtain the light intensity ratio vector

Operation S4, calculate the light intensity ratio vector With each extended light intensity ratio matrix The distance between each column in the image is the intensity ratio vector The ratio between the columns corresponding to the minimum distance is used as the scattered light intensity balance value of the target particles. When the scattered light intensity balance value exceeds the alarm threshold, it is determined that a fire has occurred.

Preferably, the calculated distance is any one of Euclidean distance, Mahalanobis distance, Manhattan distance, Chebyshev distance, high-order norm distance, cosine distance, and information entropy. The ratio between the columns corresponding to the minimum distance is used as the balanced value of the scattered light intensity of the target particles to balance the scattered light intensity of black smoke and white smoke, thereby improving the alarm accuracy.

According to an embodiment of the present invention, the above operations S1 to S4 are optimized to obtain a better solution, which is as follows:

The design phase also includes the following operations: combining the extended light intensity ratio matrix Calculate the average standard deviation of each row, for each extended light intensity ratio matrix L rows with the smallest average standard deviation are selected to form a corresponding optimal light intensity ratio matrix _IL×N , where L is a given parameter between 1 and M.

At this time, the operation performed in operation S3 is: detecting the target particles at a plurality of scattered light receiving angles θ _i corresponding to L rows with the smallest average standard deviation, calculating the ratios of the detected scattered light intensities in pairs and combining them to obtain the optimal light intensity ratio vector _IL×1 .

At this time, the operation performed in operation S4 is: calculating the distance between the optimal light intensity ratio vector _IL×1 and each column in each optimal light intensity ratio matrix _IL×N , and taking the ratio between the optimal light intensity ratio vector _IL×1 and the column corresponding to the minimum distance as the scattered light intensity balance value of the target particles. When the scattered light intensity balance value exceeds the alarm threshold, it is determined that a fire has occurred.

According to an embodiment of the present invention, the above operations S1 to S4 are optimized to obtain another better solution, which is as follows:

The design phase also includes the following operations: using the expanded light intensity ratio matrix of the first particle As a benchmark, the expanded light intensity ratio matrix of the remaining particles is Normalization is performed to obtain the relative response coefficients of the remaining particles. The first particle is any one of the particles that characterize black smoke and white smoke. Since the particle size, shape and refractive index of white smoke are relatively stable, in this embodiment, the first particle is preferably a particle of white smoke, and the corresponding alarm threshold is directly determined by the measurement results of the white smoke detection experiment.

At this time, the operation performed in operation S4 is: calculating the light intensity ratio vector With each extended light intensity ratio matrix The distance between each column in the target particle type is determined based on the column with the minimum distance; if the determination result is consistent with the first particle type, the light intensity ratio vector When the alarm threshold is exceeded, it is determined that a fire has occurred. If it is inconsistent, the light intensity ratio vector and The ratio between them is used as the scattered light intensity balance value of the target particles. When the scattered light intensity balance value exceeds the alarm threshold, it is determined that a fire has occurred. is the column corresponding to the minimum distance corresponding column selected from the relative response coefficients.

According to an embodiment of the present invention, the above operations S1 to S4 are optimized to obtain another better solution, which is as follows:

The design phase also includes the following operations: combining the extended light intensity ratio matrix Calculate the average standard deviation of each row, for each extended light intensity ratio matrix L rows with the smallest average standard deviation are selected to form the corresponding optimal light intensity ratio matrix _IL×N , where L is a given parameter between 1 and M. Taking the optimal light intensity ratio matrix _IL×N of the first particle as a reference, the optimal light intensity ratio matrix _IL×N of the remaining particles is normalized to obtain the relative response coefficients of the remaining particles.

At this time, the operation performed in operation S3 is: detecting the target particles at a plurality of scattered light receiving angles θ _i corresponding to L rows with the smallest average standard deviation, calculating the ratios of the detected scattered light intensities in pairs and combining them to obtain the optimal light intensity ratio vector _IL×1 .

At this time, the operation performed in operation S4 is: calculate the distance between the optimal light intensity ratio vector _IL×1 and each column in each optimal light intensity ratio matrix _IL×N , and judge the type of target particles based on the column corresponding to the minimum distance; if the judgment result is consistent with the first particle type, when the optimal light intensity ratio vector _IL×1 exceeds the alarm threshold, it is judged that a fire has occurred; if not, the ratio between the optimal light intensity ratio vector _IL×1 and _LL×1 is used as the scattered light intensity balance value of the target particles, and when the scattered light intensity balance value exceeds the alarm threshold, it is judged that a fire has occurred, and _LL×1 is the column corresponding to the column corresponding to the minimum distance selected from the relative response coefficient.

In the following, spherical particles with a refractive index of 1.55+0.02i are used to replace white smoke, and rotating ellipsoidal particles with a refractive index of 1.55+0.55i and a major-minor axis ratio of 2:1 are used to replace black smoke, and the wavelength of the light source used is 650nm, M=15, N=10. The 15 scattered light receiving angles are 30°, 40°, 50°, ..., 170°, respectively. This example is used to illustrate the specific implementation process of the embodiment of the present invention.

1) Calculate 10 particle sizes d _j = λ*α _j /π, where λ is the wavelength of the light source, α _j is the dimensionless particle size, and α _j takes values of 0.1, 0.2, 0.3, ..., 1.0; establish the original scattered light intensity matrix of spherical particles and the original scattered light intensity matrix of ellipsoidal particles

2) Calculation The corresponding extended light intensity ratio matrix And calculate The corresponding extended light intensity ratio matrix

3) Calculate the average standard deviation σ _i of each row:

4) Select L rows with the smallest average standard deviation to form the corresponding optimal light intensity ratio matrix _IL×N . Take L=2 and the 23rd and 173rd rows with the smallest average standard deviation as an example, that is, the ratio of the scattered light intensity in the direction of the 100° scattered light receiving angle to the scattered light intensity in the direction of the 40° scattered light receiving angle to the scattered light intensity in the direction of the 100° scattered light receiving angle to, and the ratio of the scattered light intensity in the direction of the 40° scattered light receiving angle to the scattered light intensity in the direction of the 100° scattered light receiving angle to. These two ratios are reciprocals of each other, and only one of them needs to be recorded, such as recording the ratio of the scattered light intensity in the direction of the 100° scattered light receiving angle to the scattered light intensity in the direction of the 40° scattered light receiving angle to, as shown in Figure 2. Further, from Select the 23rd and 173rd rows to form the optimal light intensity ratio matrix Select the 23rd and 173rd rows to form the optimal light intensity ratio matrix Correspondingly, the mean values of spherical particles and ellipsoidal particles were calculated to be 0.522 and 0.225, respectively.

5) Optimal light intensity ratio matrix for spherical particles Based on the optimal light intensity ratio matrix for ellipsoidal particles Normalization processing is performed, that is, the relative response coefficients of spherical particles of various particle sizes at each angle ratio are all set to 1. The relative response coefficients of ellipsoidal particles of different particle sizes are shown in FIG3 , and their average value is 1.10.

6) Design detectors for fire detection with the selected optimal scattered light receiving angles of 40° and 100°, calculate the ratios of the scattered light intensities at the two receiving angles and combine them to obtain the optimal light intensity ratio vector I _2×1 .

7) Calculate the optimal light intensity ratio vector I _2×1 and The distance between each column in the , based on the minimum distance corresponding column to determine the type of target particles, when the minimum distance corresponding column is When the target particle is the white smoke corresponding to the spherical particle, when the minimum distance corresponds to When the target particles are black smoke corresponding to ellipsoidal particles, as shown in FIG4 , particles 1 and 4 are closer to the ellipsoidal particles, and are determined to be ellipsoidal particles, while particles 2 and 3 are closer to the spherical particles, and are determined to be spherical particles.

8) Taking the four particles shown in Figure 4 as an example, since particles 2 and 3 are determined to be spherical particles, no balancing is required, and their measured scattered light intensities are directly used to determine whether the alarm threshold has been reached; since particles 1 and 4 are determined to be ellipsoidal particles, balancing is required, and their measured scattered light intensities are divided by the relative response coefficients obtained after the corresponding normalization processing to serve as the scattered light intensity balance value, and whether the alarm threshold has been reached is determined based on the scattered light intensity balance value.

FIG5 is a block diagram of a fire detection system with balanced black and white smoke response according to an embodiment of the present invention. Referring to FIG5 , the fire detection system with balanced black and white smoke response 500 includes a design module 510 , a calculation module 520 , a detection module 530 and a determination module 540 .

The design module 510, for example, performs operation S1 to pre-design particles of various shapes to respectively characterize black smoke and white smoke. For each particle, the scattered light intensity I _ij at different scattered light receiving angles θ _i and different particle sizes d _j is calculated and combined to obtain its original scattered light intensity matrix I _M×N , i=1, 2, …, M, j=1, 2, …, N, where M and N are set parameters.

The calculation module 520, for example, performs operation S2 to calculate the ratio of the scattered light intensity of each row in the original scattered light intensity matrix I _M×N to the scattered light intensity of the i-th row, obtains the i-th intermediate matrix, combines the i-th intermediate matrices, and obtains the corresponding extended light intensity ratio matrix

The detection module 530, for example, performs operation S3 to detect the target particles at different scattered light receiving angles _θi , and calculates the ratios of the detected scattered light intensities in pairs and then combines them to obtain a light intensity ratio vector

The determination module 540, for example, performs operation S4 to calculate the light intensity ratio vector With each extended light intensity ratio matrix The distance between each column in the image is the intensity ratio vector The ratio between the columns corresponding to the minimum distance is used as the scattered light intensity balance value of the target particles. When the scattered light intensity balance value exceeds the alarm threshold, it is determined that a fire has occurred.

The fire detection system 500 with balanced black and white smoke response is used to execute the fire detection method with balanced black and white smoke response in the embodiments shown in Figures 1 to 4. For details not covered in this embodiment, please refer to the fire detection method with balanced black and white smoke response in the embodiments shown in Figures 1 to 4, which will not be described here.

The embodiment of the present disclosure also shows an electronic device, as shown in Fig. 6, the electronic device 600 includes a processor 610 and a readable storage medium 620. The electronic device 600 can execute the fire detection method of balanced black and white smoke response described in Figs. 1 to 4 above.

Specifically, the processor 610 may include, for example, a general-purpose microprocessor, an instruction set processor and/or a related chipset and/or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 610 may also include an onboard memory for cache purposes. The processor 610 may be a single processing unit or multiple processing units for executing different actions of the method flow according to the embodiment of the present disclosure described with reference to FIG. 1-FIG 4.

The readable storage medium 620 may be, for example, any medium capable of containing, storing, conveying, propagating, or transmitting instructions. For example, the readable storage medium may include, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, component, or propagation medium. Specific examples of readable storage media include: a magnetic storage device, such as a magnetic tape or a hard disk (HDD); an optical storage device, such as a compact disk (CD-ROM); a memory, such as a random access memory (RAM) or flash memory; and/or a wired/wireless communication link.

The readable storage medium 620 may include a computer program 621 , which may include code/computer executable instructions, which when executed by the processor 610 causes the processor 610 to perform, for example, the method flow described above in conjunction with Figures 1-4 and any variations thereof.

The computer program 621 may be configured to have, for example, a computer program code including a computer program module. For example, in an exemplary embodiment, the code in the computer program 621 may include one or more program modules, for example, including 621A, module 621B, ... It should be noted that the division method and number of modules are not fixed, and those skilled in the art may use appropriate program modules or program module combinations according to actual conditions. When these program module combinations are executed by the processor 610, the processor 610 may execute, for example, the method flow described above in conjunction with FIG. 1 to FIG. 4 and any variation thereof.

It will be easily understood by those skilled in the art that 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 and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. A fire detection method for balancing black-and-white smoke response is characterized by comprising a design stage and a detection stage, wherein the design stage comprises S1-S2, and the detection stage comprises S3-S4;

S1, designing particles with various shapes in advance to respectively represent black smoke and white smoke, and calculating and combining different scattered light receiving angles theta _i and scattered light intensities I _ij under different particle diameters d _j for each particle to obtain an original scattered light intensity matrix I _M×N, i=1, 2, …, M, j=1, 2, …, N, M and N as set parameters;

s2, calculating the ratio of the scattered light intensity of each row in the original scattered light intensity matrix I _M×N to the scattered light intensity of the ith row to obtain an ith intermediate matrix, and combining the ith intermediate matrices to obtain a corresponding expanded light intensity ratio matrix

S3, detecting target particles by using different scattered light receiving angles theta _i, and combining the detected scattered light intensities after the ratio of every two to obtain a light intensity ratio vector

S4, calculating a light intensity ratio vectorAnd each extended light intensity ratio matrixThe distance between each row of the light intensity ratio vectorAnd taking the ratio between the columns corresponding to the minimum distance as a scattering light intensity balance value of the target particles, and judging that fire disaster occurs when the scattering light intensity balance value exceeds an alarm threshold value.

2. The fire detection method of equalizing a black and white smoke response according to claim 1, wherein said design phase further comprises: combining the ratio matrix of each expansion light intensityCalculating average standard deviation of each row, and matrix of ratio of light intensity for each expansionAnd selecting L rows with the smallest average standard deviation from the two rows to form a corresponding optimal light intensity ratio matrix I _L×N, wherein L is a given parameter ranging from 1 to M.

3. The fire detection method for equalizing black-and-white smoke response according to claim 2, wherein in S3, the target particles are detected by a plurality of scattered light receiving angles θ _i corresponding to L rows with the smallest average standard deviation, and the detected scattered light intensities are combined after being subjected to ratio of two by two to obtain an optimal light intensity ratio vector I _L×1;

In the step S4, the distance between the optimal light intensity ratio vector I _L×1 and each column in each optimal light intensity ratio matrix I _L×N is calculated, and the ratio between the optimal light intensity ratio vector I _L×1 and the column corresponding to the minimum distance is used as the scattered light intensity balance value of the target particle.

4. A fire detection method for equalizing black and white smoke response according to any one of claims 1 to 3, wherein in S1, spherical particles are designed to represent white smoke, k ₁ kinds of ellipsoidal particles of revolution with a set length-axis ratio and/or k ₂ kinds of cylindrical particles with a set bottom radius and a set height are designed to represent black smoke, and k ₁、k₂ is a natural number and is not 0 at the same time.

5. The method of fire detection for equalizing a black and white smoke response according to claim 4, wherein said design phase further comprises: matrix of expanded light intensity ratios with first particlesAs a benchmark, the extended light intensity ratio matrix for the remaining particlesAnd carrying out normalization treatment to obtain the relative response coefficient of each other particle, wherein the first particle is any one of particles representing black smoke and white smoke.

6. The fire detection method of equalizing a black and white smoke response according to claim 5, wherein said S4 comprises:

Calculating the light intensity ratio vector And each extended light intensity ratio matrixJudging the type of the target particles based on the distance between each column and the corresponding column of the minimum distance;

if the judging result is consistent with the first particle type, the light intensity ratio vector When the alarm threshold value is exceeded, judging that fire disaster occurs, if the fire disaster is inconsistent, and carrying out vector on the light intensity ratioAnd (3) withThe ratio of the two is used as a scattered light intensity balance value of the target particles, when the scattered light intensity balance value exceeds an alarm threshold value, the fire disaster is judged to occur,Is a column corresponding to the minimum distance corresponding column selected from the relative response coefficients.

7. The fire detection method of equalizing a black and white smoke response according to claim 1, wherein the distance calculated in S4 is any one of a euclidean distance, a mahalanobis distance, a manhattan distance, a chebyshev distance, a higher order norm distance, a cosine distance, and an information entropy.

8. A fire detection system for equalizing black and white smoke response, comprising:

The design module is used for designing particles with various shapes in advance to respectively represent black smoke and white smoke, calculating different scattered light receiving angles theta _i and the scattered light intensities I _ij of the particles with different particle diameters d _j for each particle, and combining to obtain an original scattered light intensity matrix I _M×N, i=1, 2, …, M, j=1, 2, …, N, M and N of the particles;

A calculation module for calculating the ratio of the scattered light intensity of each row in the original scattered light intensity matrix I _M×N to the scattered light intensity of the ith row to obtain the ith intermediate matrix, and combining the ith intermediate matrices to obtain the corresponding expanded light intensity ratio matrix

The detection module is used for detecting the target particles at different scattered light receiving angles theta _i, and combining the detected scattered light intensities after the ratio of the two pairs to obtain a light intensity ratio vector

A judging module for calculating the light intensity ratio vectorAnd each extended light intensity ratio matrixThe distance between each row of the light intensity ratio vectorAnd taking the ratio between the columns corresponding to the minimum distance as a scattering light intensity balance value of the target particles, and judging that fire disaster occurs when the scattering light intensity balance value exceeds an alarm threshold value.

9. An electronic device, comprising:

A processor;

A memory storing a computer executable program that when executed by the processor causes the processor to perform the method of equalizing black and white smoke response fire detection of any one of claims 1-7.

10. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when executed by a processor, implements a fire detection method of equalizing black-and-white smoke responses according to any one of claims 1-7.