Space-time smoke box system and method for testing space-time response characteristics of fire detector

By designing the space-time smoke box system and data fusion technology, the problem of false alarms of fire detectors in narrow and long spaces is solved, efficient and accurate detection of fire detectors is achieved, and the impact of environmental interference on the detectors is reduced.

CN119068647APending Publication Date: 2024-12-03PEOPLES POLICE UNIV OF CHINA (INT LAW ENFORCEMENT COOP INST OF THE MINISTRY OF PUBLIC SECURITY CHINA PEACEKEEPING POLICE TRAINING CENT)

Patent Information

Application Number
CN202411204677.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The false alarm conditions of existing fire detectors in narrow and long spaces are difficult to accurately judge, and the traditional smoke box detection efficiency is low, so it is impossible to effectively simulate the environmental interference conditions in narrow and long spaces such as tunnels, resulting in high difficulty in identifying fires and low accuracy.

Method used

Design a space-time smoke box system, including a unit smoke box made of transparent acrylic plates, equipped with temperature and humidity sensors, dust generators, smoke generators and axial flow fans, and simulate different fires through data fusion technology and adaptive weighting algorithms. Environmental parameters, evaluate the spatiotemporal response characteristics of the fire detector.

Benefits of technology

It effectively reduces the false alarm rate of fire detectors due to environmental interference, improves the fire detection accuracy and detection efficiency in narrow and long spaces, and can accurately judge the alarm accuracy and response time of the detectors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119068647A_ABST
    Figure CN119068647A_ABST
Patent Text Reader

Abstract

The invention discloses a space-time smoke box system and method for testing space-time response characteristics of a fire detector, a space-time smoke box comprises a support frame and a long and narrow box body, the long and narrow box body comprises a plurality of unit smoke boxes, the unit smoke boxes are tightly connected and spliced through buckles, the side wall of each unit smoke box is provided with a temperature and humidity sensor, and the temperature and humidity sensor is connected with the support frame. A sampling hole is formed in the top of each unit smoke box, a to-be-tested fire detector is arranged at each sampling hole, and the device further comprises a dust generator, a temperature and humidity generator and a smoke generator. According to the invention, the problem of building the environment interference condition of the detector in a long and narrow space is solved, and the anti-environment interference experiment research of the fire detector under the environment condition is realized. According to the method for testing the space-time response characteristics of the fire detectors, the data fusion value of the multiple fire detectors is obtained through the self-adaptive weighted data fusion alarm algorithm, the value is close to the measurement average value of the non-interference detectors, and the probability of false alarm of the single detector due to an interference source can be effectively reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of fire detection, and in particular to a space-time smoke box system and method for testing the space-time response characteristics of a fire detector. Background Art

[0002] Most of the existing fire detection smoke boxes are annular smoke boxes, whose main function is to calibrate the sensitivity of fire detectors. They are used to calibrate or detect a certain detector. There is a lack of straight smoke boxes that can create environmental interference conditions and simulate the design of narrow and long spaces such as tunnels. At the same time, the fire protection system used in narrow and long spaces requires many types and a large number of detectors, and the collected data is easily affected by environmental interference, which makes it difficult to identify fires and has low accuracy. Traditional smoke boxes can only detect a single fire detector at a time, and the detection efficiency is low. Therefore, it is urgent to design an accurate test method to test the spatiotemporal response characteristics of multiple fire detectors used in the space in a targeted manner, so as to accurately determine the false alarm situation of the fire detectors used in the environment. Summary of the invention

[0003] In order to solve the technical problems mentioned in the background technology, the present invention provides a spatiotemporal smoke box system and method for testing the spatiotemporal response characteristics of fire detectors.

[0004] The present invention adopts the following technical solution: a spatiotemporal smoke box system for testing the spatiotemporal response characteristics of fire detectors, comprising:

[0005] Support frame;

[0006] A narrow and long box body arranged on a supporting frame;

[0007] The narrow and long box body includes a plurality of unit cigarette boxes made of transparent acrylic plates, each unit cigarette box is tightly connected and spliced ​​by buckles; the unit cigarette boxes are interconnected;

[0008] Each unit smoke box side wall is equipped with a temperature and humidity sensor to monitor the internal environment temperature and humidity in real time;

[0009] A sampling hole is set on the top of each unit smoke box;

[0010] A fire detector to be tested is arranged at each sampling hole;

[0011] At least one dust generator, one temperature and humidity generator, and one smoke generator are used to create different fire simulation environment parameters.

[0012] Furthermore, a smoke barrier is provided at the joint of each unit smoke box. The height of the smoke barrier is between 1 / 3 and 1 / 2 of the height of the side panel, and the smoke barrier is electrically raised and lowered by a transmission mechanism to simulate smoke flow obstruction in a fire environment.

[0013] Furthermore, it also includes at least one axial flow fan for exhausting smoke and ventilating.

[0014] Furthermore, the present invention also provides a method for testing the spatiotemporal response characteristics of a fire detector using the spatiotemporal smoke box system, comprising the following steps:

[0015] Assemble the unit smoke box according to the experimental requirements to form a narrow and long box with a predetermined length;

[0016] Install the fire detectors at the sampling holes and number them from 1 to N in sequence, and check whether the working status of each fire detector is normal;

[0017] Adjust the dust generator, smoke generator and temperature and humidity generator to achieve the environmental parameters required for the test in the narrow and long box;

[0018] Release the candle smoke interference source to the fire detector to sound an alarm;

[0019] Collect real-time temperature and humidity data in Yanxiangnei through temperature and humidity sensors;

[0020] Data fusion technology is used to process the data collected by several fire detectors and evaluate the alarm accuracy and response time of each detector.

[0021] Furthermore, the data fusion technology is used to process the data collected by several fire detectors and evaluate the alarm accuracy and response time of each detector, including:

[0022] Collect the detection concentration data of N fire detectors within 1 hour near the alarm time. The sampling interval of the fire detectors is 2s, and draw the concentration curve through the origin software;

[0023] Calculate the mean and variance of the concentration data of N fire detectors;

[0024] By applying the statistical process control (SPC) method, abnormal data corresponding to detectors with large errors in the mean and variance of concentration data are identified and eliminated;

[0025] Draw the control line of the group of data according to the N-1 group of data after the above elimination;

[0026] The confidence matrix and support matrix formed by the above N-1 groups of data are calculated according to the confidence distance measurement formula, the support of the above N-1 fire detectors is calculated by the support calculation formula, the number of the fire detector with the highest support is obtained, and the fire detector with the highest support is determined as the fire detector participating in data fusion;

[0027] The weighting factor corresponding to each detector is obtained according to the adaptive weight calculation formula;

[0028] After the weighting factor is obtained, it is substituted into the adaptive weighted data fusion algorithm to obtain the data fusion value of all N fire detectors;

[0029] Substitute the data fusion value and compare it with the data fusion value of a fire detector to evaluate the alarm instruction of the fire detector.

[0030] Furthermore, the data fusion value is substituted and compared with the data fusion value of a certain fire detector to evaluate the alarm instruction of the fire detector, including when the difference between the data fusion value of a certain fire detector and the data fusion values ​​of all N fire detectors increases three times in a row, it is determined that the fire detector has not falsely alarmed.

[0031] Compared with the prior art, the advantages of the present invention are: the space-time smoke box system designed by the present invention for testing the space-time response characteristics of fire detectors solves the problem of establishing environmental interference conditions for detectors in narrow and long spaces, and realizes experimental research on the anti-environmental interference of fire detectors under such environmental conditions.

[0032] The method for testing the spatiotemporal response characteristics of fire detectors designed in the present invention, combined with the above-mentioned spatiotemporal smoke box system, adopts an adaptive weighted data fusion alarm algorithm to obtain the data fusion value of multiple fire detectors. This value is close to the average value of the measurement of the undisturbed detectors, thereby effectively reducing the probability of false alarm of a single detector due to interference sources. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the overall structure of the spatiotemporal smoke box system used for testing the spatiotemporal response characteristics of fire detectors according to the present invention;

[0034] Figure 2 The detection concentration variation curves of the four groups of fire detectors of the present invention are as follows;

[0035] Figure 3 It is the detection concentration mean-variance control chart of four groups of fire detectors of the present invention;

[0036] Figure 4 This is the variation curve of the detection concentration value of the fourth fire detector.

[0037] in:

[0038] 1-support frame, 2-unit smoke box, 3-clip, 4-temperature and humidity sensor, 5-smoke barrier wall, 6-axial flow fan, 7-dust generator, 8-smoke generator, 9-sampling hole. DETAILED DESCRIPTION

[0039] In order to facilitate those skilled in the art to understand the technical solution of the present invention, further description will be given below with reference to the accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention.

[0040] In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may be implemented without these specific details. In addition, in the following description, descriptions of known structures and techniques are omitted to avoid unnecessary confusion of concepts of the present invention.

[0041] like Figure 1 As shown, it is a schematic diagram of the overall structure of the spatiotemporal smoke box system for testing the spatiotemporal response characteristics of fire detectors of the present invention, including a support frame 1, a narrow and long box body arranged on the support frame 1, and the narrow and long box body includes a plurality of unit smoke boxes 2 made of transparent acrylic plates, and the transparent acrylic plates are convenient for observing the flow process of smoke, water vapor, etc. in the box body. Each unit smoke box 2 is tightly connected and spliced ​​by a buckle 3, and the unit smoke boxes are interconnected. A temperature and humidity sensor 4 is installed on the side wall of each unit smoke box 2 for real-time monitoring of the internal environment temperature and humidity. A sampling hole is arranged on the top of each unit smoke box 2, and a fire detector to be tested is arranged at each sampling hole 9. The fire detector can be a point-type smoke detector or an active suction detector. When testing the active suction detector, a straight pipeline and a U-shaped pipeline of the suction detector are arranged in the smoke box body. The fire detector is connected with a data analysis and calculation processing module for collecting concentration data and obtaining the spatiotemporal response characteristics of the fire detector. The system also includes at least one dust generator 7, a temperature and humidity generator 4, and a smoke generator 8, which are used to create different fire simulation environment parameters.

[0042] In the present system, a smoke barrier wall 5 is provided at the joint of each unit smoke box 2. The height of the smoke barrier wall 5 is between 1 / 3 and 1 / 2 of the height of the side panel, and the smoke barrier wall 5 is electrically lifted and lowered by a transmission mechanism to simulate the smoke flow obstruction state in a fire environment.

[0043] In addition, the system also includes at least one axial flow fan 6 for exhausting smoke and ventilating.

[0044] The spatiotemporal smoke box system designed by the present invention for testing the spatiotemporal response characteristics of fire detectors solves the problem of setting up environmental interference conditions for detectors in narrow and long spaces, and realizes experimental research on the anti-environmental interference of fire detectors under such environmental conditions, including obtaining the data regularity of the fire source location and the alarm response time, and performing data processing and curve fitting on the change regularity of the alarm time of different detection points over time, to obtain the regularity of the detector attenuation with distance for different fire source powers and different pyrolytic materials. The algorithm compensation coefficient of the detection distance on the attenuation of the alarm time is studied.

[0045] In addition, the present invention also designs a method for testing the spatiotemporal response characteristics of a fire detector based on the spatiotemporal smoke box system, the method comprising the following steps:

[0046] Step 1: Assemble the unit smoke box according to the experimental requirements to form a narrow and long box with a predetermined length;

[0047] Step 2: Install the fire detectors at the sampling holes and number them from 1 to N in sequence, and check whether the working status of each fire detector is normal;

[0048] Step 3: Adjust the dust generator, smoke generator and temperature and humidity generator to achieve the environmental parameters required for the test in the narrow and long box;

[0049] Step 4: Release the candle smoke interference source to the fire detector to sound an alarm;

[0050] Step 5: Collect real-time temperature and humidity data in the smoke box through the temperature and humidity sensor;

[0051] Step 6: Use data fusion technology to process the data collected by several fire detectors and evaluate the alarm accuracy and response time of each detector. The specific implementation includes:

[0052] Collect the detection concentration data of N fire detectors within 1 hour near the alarm time. The sampling interval of the fire detectors is 2s, and draw the concentration curve through the origin software;

[0053] Calculate the mean and variance of the concentration data of N fire detectors;

[0054] By applying the statistical process control (SPC) method, abnormal data corresponding to detectors with large errors in the mean and variance of concentration data are identified and eliminated;

[0055] Draw the control line of the group of data according to the N-1 group of data after the above elimination;

[0056] The confidence matrix and support matrix formed by the above N-1 groups of data are calculated according to the confidence distance measurement formula, the support of the above N-1 fire detectors is calculated by the support calculation formula, the number of the fire detector with the highest support is obtained, and the fire detector with the highest support is determined as the fire detector participating in data fusion;

[0057] The weighting factor corresponding to each detector is obtained according to the adaptive weight calculation formula;

[0058] After the weighting factor is obtained, it is substituted into the adaptive weighted data fusion algorithm to obtain the data fusion value of all N fire detectors;

[0059] Substitute the data fusion value and compare it with the data fusion value of a fire detector to evaluate the alarm instruction of the fire detector. The specific implementation includes: when the difference between the data fusion value of a fire detector and the data fusion values ​​of all N fire detectors increases three times in a row, it is determined that the fire detector has not falsely alarmed.

[0060] The following describes the method for testing the spatiotemporal response characteristics of fire detectors by taking the setting of 4 groups of fire detectors as an example.

[0061] First, light a candle under fire detector No. 3 to release the candle smoke interference source to detector No. 3 to sound an alarm. Fire detectors No. 1, 2, and 3 maintain normal working conditions. Collect the detection concentration data of detectors No. 1-4 for a total of 1 hour near the alarm time and draw the concentration curve through origin software. The sampling interval of the detector is 2s, so there are about 1500 points in the horizontal axis. The detection curves of the four fire detectors are as follows: Figure 2 As shown, Figure 2 a is the concentration variation curve of fire detector No. 1, Figure 2 b is the concentration variation curve of fire detector No. 2, Figure 2 c is the concentration variation curve of fire detector No. 3, Figure 2 d is the concentration variation curve of fire detector No. 4.

[0062] At the same time, based on the above data, the detection average and variance of the four detectors are calculated as shown in Table 1:

[0063]

[0064] Table 1 Mean and variance of four fire detectors

[0065] After the data is calculated, according to the algorithm flow chart, the data with large errors are first eliminated through the SPC method. After calculation according to the SPC method, the number of subgroups of the algorithm, that is, the number of detectors n=4, refer to GB / T 17989.2-2020 to know the control coefficients B3=0, B4=2.266 in the formula, and the mean-variance control line is calculated as: center line CLs=19.74, upper control line UCLs=43.94, lower control line LCLs=0. After sorting out the data of each fire detector in Table 1, Figure 3 As shown, fire detector No. 3 is abnormal data with large error. Its data in Table 1 is removed, and the remaining data after removal is shown in Table 2.

[0066]

[0067] Table 2. Mean and variance of detectors after removing abnormal data

[0068] According to the algorithm flow, the next step is to calculate the confidence matrix and support matrix composed of the remaining detector numbers according to the confidence distance measurement formula. Through the support calculation formula, it can be calculated that the support of fire detector No. 2 is the highest, followed by fire detector No. 4. After the algorithm calculation and processing, the detector participating in data fusion is No. 2, as shown in Table 3.

[0069]

[0070] Table 3 Detector data involved in data fusion At the same time, the corresponding weights of each detector are obtained according to the adaptive weight calculation formula as shown in Table 4.

[0071]

[0072] Table 4 Corresponding weights of each detector

[0073] After the weight is obtained, it is substituted into the adaptive weighted data fusion alarm algorithm to obtain the data fusion value X = 23.45, which is close to the average value of the uninterfered detector measurement, effectively reducing the probability of a single detector making a false alarm due to an interference source.

[0074] In order to verify whether the algorithm can accurately determine the occurrence of fire and alarm in the case of a real fire, the environmental conditions are maintained unchanged, and wood is ignited under the No. 4 fire detector. At the same time, the detection value of the No. 4 fire detector and the overall result of the algorithm are continuously monitored. The change curve of the detection value of the No. 4 fire detector is shown in the figure below. Figure 4 shown.

[0075] The fusion values ​​of four sets of data obtained at four consecutive time points starting from 750s are shown in Table 5:

[0076]

[0077] Table 5 Fusion values ​​of four groups of data

[0078] According to the difference results, the difference rises three times in a row, and the detector sends an alarm command to the system at this time, which proves that the algorithm is running effectively.

[0079] The method for testing the spatiotemporal response characteristics of fire detectors designed in the present invention, combined with the above-mentioned spatiotemporal smoke box system, adopts an adaptive weighted data fusion alarm algorithm to obtain the data fusion value of multiple fire detectors. This value is close to the average value of the measurement of the undisturbed detectors, thereby effectively reducing the probability of false alarm of a single detector due to interference sources.

[0080] The above embodiments are merely descriptions of preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A spatiotemporal smoke chamber system for testing the spatiotemporal response characteristics of fire detectors, characterized in that: include: Support frame (1); A narrow and long box body arranged on a support frame (1); The narrow and long box body comprises a plurality of unit cigarette boxes (2) made of transparent acrylic plates, each unit cigarette box (2) being tightly connected and spliced ​​by means of buckles (3); the unit cigarette boxes are interconnected; A temperature and humidity sensor (4) is installed on the side wall of each unit smoke box (2) for real-time monitoring of the internal environment temperature and humidity; A sampling hole is arranged on the top of each unit smoke box (2); A fire detector to be tested is arranged at each sampling hole; It includes at least one dust generator, a temperature and humidity generator, and a smoke generator, which are used to create different fire simulation environment parameters.

2. The spatiotemporal smoke chamber system for testing the spatiotemporal response characteristics of fire detectors according to claim 1, characterized in that: A smoke barrier vertical wall (5) is arranged at the joint of each unit smoke box (2), the height of the smoke barrier vertical wall (5) is between 1 / 3 and 1 / 2 of the height of the side plate, and the smoke barrier vertical wall (5) is electrically lifted and lowered by a transmission mechanism to simulate the obstruction of smoke flow in a fire environment.

3. The spatiotemporal smoke chamber system for testing the spatiotemporal response characteristics of fire detectors according to claim 2, characterized in that: It also includes at least one axial flow fan (6) for exhausting smoke and exchanging air.

4. A method for testing the spatiotemporal response characteristics of a fire detector using the spatiotemporal response smoke box system according to any one of claims 1 to 3, characterized in that: The following steps are involved: Assemble the unit smoke box according to the experimental requirements to form a narrow and long box with a predetermined length; Install the fire detectors at the sampling holes and number them from 1 to N in sequence, and check whether the working status of each fire detector is normal; Adjust the dust generator, smoke generator and temperature and humidity generator to achieve the environmental parameters required for the test in the narrow and long box; Release the candle smoke interference source to the fire detector to sound an alarm; Collect real-time temperature and humidity data in the smoke box through temperature and humidity sensors; Data fusion technology is used to process the data collected by several fire detectors and evaluate the alarm accuracy and response time of each detector.

5. The method for testing the spatiotemporal response characteristics of a fire detector according to claim 4, characterized in that: The data fusion technology is used to process the data collected by several fire detectors and evaluate the alarm accuracy and response time of each detector, including: Collect the detection concentration data of N fire detectors within 1 hour near the alarm time. The sampling interval of the fire detectors is 2s, and draw the concentration curve through the origin software; Calculate the mean and variance of the concentration data of N fire detectors; By applying the statistical process control (SPC) method, abnormal data corresponding to detectors with large errors in the mean and variance of concentration data are identified and eliminated; Draw the control line of the group of data according to the N-1 group of data after the above elimination; The confidence matrix and support matrix formed by the above N-1 groups of data are calculated according to the confidence distance measurement formula, the support of the above N-1 fire detectors is calculated by the support calculation formula, the number of the fire detector with the highest support is obtained, and the fire detector with the highest support is determined as the fire detector participating in data fusion; The weighting factor corresponding to each detector is obtained according to the adaptive weight calculation formula; After the weighting factor is obtained, it is substituted into the adaptive weighted data fusion algorithm to obtain the data fusion value of all N fire detectors; Substitute the data fusion value and compare it with the data fusion value of a fire detector to evaluate the alarm instruction of the fire detector.

6. The method for testing the spatiotemporal response characteristics of a fire detector according to claim 5, characterized in that: The substituting data fusion value and comparing it with the data fusion value of a certain fire detector to evaluate the alarm instruction of the fire detector includes: when the difference between the data fusion value of a certain fire detector and the data fusion values ​​of all N fire detectors increases three times in a row, it is determined that the fire detector has not falsely alarmed.

Citation Information

Patent Citations

  • Combined underground pipe gallery space fire safety research model device

    CN109345938A

  • Small-size tunnel fire simulation test system

    CN110223590A

  • Small-size tunnel fire simulation experiment table

    CN110838259A

  • Cable fire prediction monitoring system and method in transformer substation

    CN112116776A

  • General Internet of Things data processing system

    CN112565457A

Cited By

  • Self-adaptive smoke-sensing fire detection method suitable for ancient building environment

    CN122266126A