A multi-parameter coupling fire detection method and system based on an adaptive algorithm

Abstract

This invention discloses a multi-parameter coupled fire detection method and system based on an adaptive algorithm, relating to the field of intelligent detection technology. The method includes: acquiring a preset fire monitoring coverage area and a multi-parameter sensing monitoring module; collecting gas data from the preset fire monitoring coverage area using an air intake module to obtain multi-dimensional sensing monitoring data; sending the multi-dimensional sensing monitoring data to a single-parameter discrimination module to perform single-parameter fire risk identification and generate multiple single-parameter risk indicators; performing fire alarm trigger judgment based on the multiple single-parameter risk indicators; if none of the multiple single-parameter risk indicators meet the preset alarm trigger threshold, sending the multiple single-parameter risk indicators to a multi-parameter coupling analysis module for coupling weighting to obtain coupled risk indicators; and performing fire early warning based on the coupled risk indicators. This solves the technical problems of low early warning accuracy and poor stability in existing cable tunnel fire early warning technologies.

Description

Technical Field

[0001] This application relates to the field of intelligent detection technology, specifically to a multi-parameter coupled fire detection method and system based on an adaptive algorithm. Background Technology

[0002] As a crucial component of urban power transmission lines, the safety of cable tunnels is vital for ensuring the stable operation of the urban power grid. The enclosed space within tunnels and the presence of potentially flammable materials, such as cable insulation, make fire risk management particularly critical. Traditional cable tunnel fire early warning technologies often rely on single-parameter detectors, such as smoke or heat detectors. However, under the complex operating conditions of cable tunnels, using single-parameter detectors results in low accuracy and poor stability in early warning systems.

[0003] Therefore, existing cable tunnel fire early warning technologies suffer from low accuracy and poor stability. Summary of the Invention

[0004] This application provides a multi-parameter coupled fire detection method and system based on an adaptive algorithm, which solves the technical problems of low accuracy and poor stability in existing cable tunnel fire early warning technologies. By collecting multiple fire early warning parameters and processing them using an adaptive algorithm, the accuracy and stability of cable tunnel fire detection and early warning are improved.

[0005] This application provides a multi-parameter coupled fire detection method based on an adaptive algorithm. The method includes: performing historical fire monitoring record index analysis based on the cable distribution characteristics of a target cable tunnel to obtain a preset fire monitoring coverage area; obtaining a multi-parameter sensing monitoring module, wherein the multi-parameter sensing monitoring module includes a temperature sensor, a smoke sensor, a hydrogen chloride sensor, and a pyrolysis particle sensor, and any one of the sensors is connected to an air intake module; collecting gas from the preset fire monitoring coverage area through the air intake module and sending the gas sample to the multi-parameter sensing monitoring module for multi-parameter detection to obtain multi-dimensional sensing monitoring data, wherein the multi-dimensional sensing monitoring data includes temperature, smoke occlusion, hydrogen chloride concentration, and pyrolysis particle concentration; sending the multi-dimensional sensing monitoring data to a single-parameter discrimination module to perform single-parameter fire risk identification and generate multiple single-parameter risk indicators; performing fire alarm trigger judgment based on the multiple single-parameter risk indicators; if none of the multiple single-parameter risk indicators meet a preset alarm trigger threshold, sending the multiple single-parameter risk indicators to a multi-parameter coupling analysis module to perform coupling weighting to obtain coupled risk indicators; and issuing a fire warning based on the coupled risk indicators.

[0006] In the implementation method, historical fire monitoring records are indexed and analyzed based on the cable distribution characteristics of the target cable tunnel to obtain a preset fire monitoring coverage area. This includes: the cable distribution characteristics include tunnel structural characteristics and cable topology characteristics; a set of fire monitoring record samples of similar cable tunnels with a feature similarity greater than a preset similarity is indexed based on the cable distribution characteristics; the fire source is located based on the fire monitoring record sample set, and the temperature distribution, smoke shading distribution, hydrogen chloride concentration distribution, and pyrolysis particle concentration distribution corresponding to the fire source are extracted; multi-sensor sensitive areas are identified based on the temperature distribution, smoke shading distribution, hydrogen chloride concentration distribution, and pyrolysis particle concentration distribution; and the preset fire monitoring coverage area is generated based on the multi-sensor sensitive areas.

[0007] In the implementation method, the multiple single-parameter risk indicators are sent to the multi-parameter coupling analysis module for coupling weighting to obtain the coupling risk indicator, including: monitoring the environmental conditions within a preset range of the target cable tunnel; traversing preset weight modes with the environmental conditions to obtain a matching weight mode, wherein the preset weight mode includes a low humidity and high dust mode, a low humidity and high dust mode, a high humidity and low dust mode, and a high humidity and high dust mode; assigning weights to the multiple single-parameter risk indicators based on the matching weight mode and performing weighted calculation to obtain the coupling risk indicator.

[0008] In the implementation method, the multiple single-parameter risk indicators are weighted based on the matching weight mode, including: using the analytic hierarchy process (AHP) to perform expert subjective weighting on the multiple single-parameter risk indicators to obtain a parameter subjective weight matrix; and using bipolar analysis to iteratively correct and optimize the parameter subjective weight matrix to generate the optimal weight.

[0009] In the implementation, the parameter subjective weight matrix is ​​iteratively corrected and optimized through dual-base-point analysis to generate optimal weights. This includes: averaging the parameter subjective weight matrix to obtain initial positive base points; standardizing the parameter subjective weight matrix based on the initial positive base points to obtain relative evaluation values; constructing ideal and anti-ideal sequences, combining the relative evaluation values ​​to construct an optimal weighted distance function, and obtaining the optimal weighting factor; calculating the ideal and anti-ideal point distances on the optimal weighting factor to obtain ideal and anti-ideal point weighted distances; calculating the preference index based on the ideal and anti-ideal point weighted distances, and performing credibility standardization calculations to obtain a credibility sequence; generating a first group of decision weights based on the credibility sequence and the parameter subjective weight matrix; using the first group of decision weights as positive base points, repeating the analysis to obtain a second group of decision weights, and calculating the average difference between the first group of decision weights and the second group of decision weights. If the average difference is less than a preset difference, the second group of decision weights is taken as the optimal weight. The expression for the optimal weighted distance function is as follows:

[0010]

[0011] Where, r ij This refers to the relative evaluation value of the j-th expert's subjective weight for the i-th parameter, obtained by standardizing the subjective weight matrix of the parameters; k i is the weighting factor; n is the number of parameter types; m is the number of weighted experts in the parameter subjective weight matrix.

[0012] In one implementation, the method further includes: acquiring the gas delivery channel between the temperature sensor, smoke sensor, hydrogen chloride sensor, and pyrolysis particle sensor and the inhalation module; performing a gas delivery test on the gas delivery channel to acquire a transmission sample, wherein the transmission sample includes pre-delivery sensing data from the inhalation module and post-delivery sensing data detected by the temperature sensor, smoke sensor, hydrogen chloride sensor, and pyrolysis particle sensor; performing multi-parameter sensing loss analysis based on the transmission sample to acquire a multi-parameter loss index; and using the multi-parameter loss index to perform confidence correction on the multi-dimensional sensing monitoring data.

[0013] This application also provides a multi-parameter coupled fire detection system based on an adaptive algorithm, including:

[0014] The index analysis module is used to perform index analysis of historical fire monitoring records based on the cable distribution characteristics of the target cable tunnel to obtain the preset fire monitoring coverage area;

[0015] The sensor acquisition module is used to acquire information from the multi-parameter sensing and monitoring module, wherein the multi-parameter sensing and monitoring module includes a temperature sensor, a smoke sensor, a hydrogen chloride sensor, and a pyrolysis particle sensor, and any one of the sensors is connected to the inhalation module.

[0016] The monitoring data acquisition module is used to collect gas from the preset fire monitoring coverage area through the air intake module, and send the gas sample to the multi-parameter sensing monitoring module for multi-parameter detection to obtain multi-dimensional sensing monitoring data, wherein the multi-dimensional sensing monitoring data includes temperature, smoke occlusion, hydrogen chloride concentration and pyrolysis particle concentration.

[0017] The risk identification module is used to send the multi-dimensional sensing and monitoring data to the single-parameter discrimination module to perform single-parameter fire risk identification and generate multiple single-parameter risk indicators.

[0018] The risk indicator acquisition module is used to perform fire alarm trigger judgment based on the multiple single-parameter risk indicators. If none of the multiple single-parameter risk indicators meet the preset alarm trigger threshold, the multiple single-parameter risk indicators are sent to the multi-parameter coupling analysis module to perform coupling weighting to obtain the coupled risk indicator.

[0019] A fire early warning module is used to provide fire early warning based on the coupled risk indicators.

[0020] This application proposes a multi-parameter coupled fire detection method and system based on an adaptive algorithm. The method analyzes historical fire monitoring records based on the cable distribution characteristics of a target cable tunnel to obtain a preset fire monitoring coverage area. A multi-parameter sensing monitoring module is acquired, comprising a temperature sensor, a smoke sensor, a hydrogen chloride sensor, and a pyrolysis particle sensor, with any one sensor connected to an intake module. The intake module collects gas samples from the preset fire monitoring coverage area and sends these samples to the multi-parameter sensing monitoring module for multi-parameter detection, obtaining multi-dimensional sensing monitoring data, including temperature, smoke opacity, hydrogen chloride concentration, and pyrolysis particle concentration. This multi-dimensional sensing monitoring data is then sent to a single-parameter discrimination module for single-parameter fire risk identification, generating multiple single-parameter risk indicators. Based on these single-parameter risk indicators, a fire alarm triggering judgment is performed. If none of the single-parameter risk indicators meet a preset alarm triggering threshold, the indicators are sent to a multi-parameter coupling analysis module for coupling weighting to obtain a coupled risk indicator. A fire early warning is then issued based on the coupled risk indicator. This invention addresses the technical problems of low accuracy and poor stability in existing cable tunnel fire early warning technologies. By collecting various fire early warning parameters and processing them using an adaptive algorithm, the accuracy and stability of cable tunnel fire detection and early warning are improved. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0022] Figure 1 A schematic flowchart of a multi-parameter coupled fire detection method based on an adaptive algorithm is provided in this application embodiment;

[0023] Figure 2 This is a schematic diagram of a multi-parameter coupled fire detection system based on an adaptive algorithm, provided in an embodiment of this application.

[0024] Explanation of reference numerals in the attached diagram: Index analysis module 11, sensor acquisition module 12, monitoring data acquisition module 13, risk identification module 14, risk indicator acquisition module 15, fire early warning module 16. Detailed Implementation

[0025] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application.

[0026] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0028] This application provides a multi-parameter coupled fire detection method and system based on an adaptive algorithm, such as... Figure 1 As shown, the method includes:

[0029] Based on the cable distribution characteristics of the target cable tunnel, historical fire monitoring records are indexed and analyzed to obtain the preset fire monitoring coverage area.

[0030] A multi-parameter sensing and monitoring module is acquired, wherein the multi-parameter sensing and monitoring module includes a temperature sensor, a smoke sensor, a hydrogen chloride sensor and a pyrolysis particle sensor, and any one of the sensors is connected to the inhalation module;

[0031] The gas is collected from the preset fire monitoring coverage area by the air intake module, and the gas sample is sent to the multi-parameter sensing and monitoring module for multi-parameter detection to obtain multi-dimensional sensing and monitoring data, including temperature, smoke opacity, hydrogen chloride concentration and pyrolysis particle concentration.

[0032] The process involves acquiring the cable distribution characteristics of the target cable tunnel, indexing and analyzing historical fire monitoring records to determine these characteristics (including tunnel structural features and cable topology), and identifying a pre-defined fire monitoring coverage area. Multi-parameter sensor monitoring modules are then deployed within this area. These modules include temperature sensors, smoke sensors, hydrogen chloride sensors, and pyrolysis particle sensors, with each sensor connected to an intake module. Subsequently, the intake module collects gas samples from the pre-defined fire monitoring coverage area and sends these samples to the multi-parameter sensor monitoring modules for multi-parameter detection. The modules analyze these samples to obtain multi-dimensional sensor monitoring data, including temperature, smoke opacity, hydrogen chloride concentration, and pyrolysis particle concentration.

[0033] The method provided in this application also includes:

[0034] The cable distribution characteristics include tunnel structural characteristics and cable topology characteristics;

[0035] A set of fire monitoring records for similar cable tunnels whose similarity to the cable distribution feature index is greater than a preset similarity.

[0036] Based on the fire monitoring record sample set, the fire source point is located, and the temperature distribution, smoke occlusion distribution, hydrogen chloride concentration distribution and pyrolysis particle concentration distribution corresponding to the fire source point are extracted.

[0037] The sensitive areas of the multi-sensor system are identified based on the temperature distribution, smoke occlusion distribution, hydrogen chloride concentration distribution, and pyrolysis particle concentration distribution.

[0038] The preset fire monitoring coverage area is generated using the sensitive areas of the multi-sensor system.

[0039] The cable distribution features include tunnel structural features and cable topology features. Tunnel structural features include structural parameters such as tunnel length. Cable topology features include the overall layout image of the cable system. Further, a set of fire monitoring record samples of cable tunnels with a similarity greater than a preset similarity threshold is obtained based on the cable distribution features. When the similarity is greater than or equal to this threshold, the cable features in the corresponding fire monitoring record sample have a high similarity to the target cable, and the sample has high reference value. Image similarity calculation is used to calculate the similarity of the cable distribution features. Subsequently, the fire source is located based on the set of fire monitoring record samples, and the temperature distribution, smoke occlusion distribution, hydrogen chloride concentration distribution, and pyrolysis particle concentration distribution corresponding to the fire source are extracted. Based on the temperature distribution, smoke occlusion distribution, hydrogen chloride concentration distribution, and pyrolysis particle concentration distribution, multi-sensor sensitive areas are identified. When identifying these areas, data on temperature distribution, smoke occlusion distribution, hydrogen chloride concentration distribution, and pyrolysis particle concentration distribution are acquired. Distribution data that meets warning values ​​and whose variance within a preset length interval meets or is less than a preset variance are identified as the corresponding distribution location interval, which is then designated as the sensor sensitive area. Fire monitoring sensors are deployed within these multi-sensor sensitive areas to ensure stable monitoring data after a fire, facilitating accurate subsequent analysis. The preset fire monitoring coverage area is then generated from these multi-sensor sensitive areas.

[0040] The multi-dimensional sensing and monitoring data is sent to the single-parameter discrimination module to perform single-parameter fire risk identification and generate multiple single-parameter risk indicators.

[0041] Fire alarm triggering judgment is performed based on the multiple single-parameter risk indicators. If none of the multiple single-parameter risk indicators meet the preset alarm triggering threshold, the multiple single-parameter risk indicators are sent to the multi-parameter coupling analysis module to perform coupling weighting to obtain the coupling risk indicators.

[0042] Fire early warning is based on the aforementioned coupled risk indicators.

[0043] Multi-dimensional sensor monitoring data from the multi-parameter sensor monitoring module is sent to the single-parameter discrimination module for single-parameter fire risk identification. The single-parameter discrimination module performs fire alarm monitoring threshold judgment on each of the multi-dimensional sensor monitoring data, determining whether each monitoring data point meets the corresponding fire alarm monitoring threshold. The fire alarm monitoring threshold is a pre-set minimum monitoring parameter required to meet ignition conditions. The multi-dimensional sensor monitoring data is normalized to generate multiple single-parameter risk indicators, which record whether each sensor monitoring data point meets the fire alarm monitoring threshold and the normalized parameter. Based on these multiple single-parameter risk indicators, a fire alarm trigger judgment is performed. If none of the multiple single-parameter risk indicators meet the preset alarm trigger threshold, the multiple single-parameter risk indicators are sent to the multi-parameter coupling analysis module for coupled weighted calculation to obtain a coupled risk indicator. If any of the multiple single-parameter risk indicators meets the preset alarm trigger threshold, a fire alarm is directly triggered. Finally, a fire warning is issued based on the coupled risk indicators. The weighted calculated coupled risk indicator is compared with a preset warning value, which is the pre-set minimum coupled risk indicator value for issuing a fire warning. If the fire risk level exceeds the preset warning value, the fire risk level is determined to be out of bounds and a fire warning is required. If the fire risk level does not exceed the preset warning value, the fire risk level is determined to be out of bounds and no fire warning is required. Optionally, different levels can be set, each with a different preset warning value. When the preset warning value of a different level is triggered, the corresponding level of warning is triggered. This solves the technical problems of low accuracy and poor stability in existing cable tunnel fire warning technologies. By collecting multiple fire warning parameters and using an adaptive algorithm to calculate and process these parameters, the accuracy and stability of cable tunnel fire detection and warning are improved.

[0044] The method provided in this application also includes:

[0045] Monitor environmental conditions within a pre-defined area of ​​the target cable tunnel;

[0046] The preset weighted modes are traversed under the environmental conditions to obtain the matching weighted modes, wherein the preset weighted modes include low humidity and high dust mode, low humidity and high dust mode, high humidity and low dust mode and high humidity and high dust mode.

[0047] The multiple single-parameter risk indicators are weighted based on the matching weight mode and then weighted to obtain the coupled risk indicator.

[0048] The environmental conditions within a preset range of the target cable tunnel are monitored. These environmental conditions include dry season and rainy season information, as well as construction information within a 50m range. A preset weighted pattern is iterated through based on these environmental conditions to obtain a matching weighted pattern. The preset weighted patterns include a low humidity / low dust pattern, a low humidity / high dust pattern, a high humidity / low dust pattern, and a high humidity / high dust pattern. If it is the dry season and there is no construction, the weight is adjusted to the low humidity / low dust pattern. If it is the dry season and there is construction, the weight is adjusted to the low humidity / high dust pattern. If it is the rainy season and there is no construction, the weight is adjusted to the high humidity / low dust pattern. If it is the rainy season and there is construction, the weight is adjusted to the high humidity / high dust pattern. Different patterns serve as reference standards for expert subjective weighting. The matching weighted pattern assigns weights to the multiple single-parameter risk indicators and performs a weighted calculation to obtain the coupled risk indicator.

[0049] The method provided in this application also includes:

[0050] The analytic hierarchy process (AHP) is used to subjectively assign expert weights to the multiple single-parameter risk indicators, resulting in a parameter subjective weight matrix.

[0051] The subjective weight matrix of the parameters is iteratively corrected and optimized through two-point analysis to generate the optimal weights.

[0052] The weighting of the multiple single-parameter risk indicators based on the matching weighting pattern includes: using the analytic hierarchy process (AHP) to subjectively weight the multiple single-parameter risk indicators by experts, that is, by having multiple experts subjectively weight the multiple single-parameter risk indicators to obtain a subjective weight matrix. For example, the subjective weight matrix is ​​constructed as follows: w ij Let be the subjective weight given by the j-th expert for the i-th indicator. The subjective weight matrix of these parameters is iteratively corrected and optimized using bipolar analysis to generate the optimal weights.

[0053] The method provided in this application also includes:

[0054] The subjective weight matrix of the parameters is averaged to obtain the initial positive base points;

[0055] Based on the initial positive base point, the subjective weight matrix of the parameters is standardized to obtain a relative evaluation value;

[0056] Construct ideal and anti-ideal sequences, combine the relative evaluation values ​​to construct the optimal weighted distance function, and obtain the optimal weighting factor;

[0057] The ideal point distance and anti-ideal point distance are calculated for the optimal weighting factor to obtain the ideal point weighted distance and anti-ideal point weighted distance;

[0058] The optimization index is calculated by combining the weighted distance of the ideal point and the weighted distance of the anti-ideal point, and the credibility is standardized to obtain the credibility sequence.

[0059] The first group of decision weights is generated by combining the credibility sequence and the subjective weight matrix of parameters;

[0060] Using the first group decision weight as a positive base point, repeat the analysis to obtain the second group decision weight, and calculate the average difference between the first group decision weight and the second group decision weight. If the average difference is less than a preset difference, the second group decision weight is taken as the optimal weight.

[0061] The expression for the optimal weighted distance function is as follows:

[0062]

[0063] Where, r ij This refers to the relative evaluation value of the j-th expert's subjective weight for the i-th parameter, obtained by standardizing the subjective weight matrix of the parameters; k i is the weighting factor; n is the number of parameter types; m is the number of weighted experts in the parameter subjective weight matrix.

[0064] The subjective weight matrix of the parameters is iteratively corrected and optimized through dual-base-point analysis to generate optimal weights. This includes: calculating the mean of the subjective weight matrix of the parameters, obtaining the mean calculation result, and obtaining initial positive base points. Based on the initial positive base points, the subjective weight matrix of the parameters is standardized to obtain relative evaluation values. The standardization formula is as follows:

[0065] Where: i = 1, 2, ..., n; j = 1, 2, ..., m; w ij w represents the subjective weight given by the j-th expert for the i-th indicator. * i Let r be the ideal value of the weight of the i-th indicator, and let r be the mean of the subjective weights of the experts. ij For w ij The relative evaluation value obtained after standardization. The ideal sequence is set as (1, ..., 1). T Let the antiideal sequence be (0, ..., 0). T To ensure that the sum of squares of the distances between the evaluation parameters and the ideal sequence is as small as possible, the decision scheme approaches the ideal state; conversely, to ensure that the sum of squares of the distances between the evaluation parameters and the anti-ideal sequence is as negative as possible, the decision scheme deviates from the worst state. A weighted distance optimal function is constructed based on the relative evaluation values, and its expression is as follows:

[0066]

[0067] Where, r ij This refers to the relative evaluation value of the j-th expert's subjective weight for the i-th parameter, obtained by standardizing the subjective weight matrix of the parameters; k i Let K be the weighting factor; n be the number of parameter types; and m be the number of weighting experts in the parameter subjective weight matrix. The optimal weighting factor, K, is then obtained using the above formula. i = (k1, ..., k) n ) T The ideal point distance and anti-ideal point distance are calculated for the optimal weighting factor to obtain the ideal point weighted distance D. j+ Weighted distance D from the antiideal point j- .

[0068] The specific formula for calculating the weighted distance of ideal points is as follows:

[0069] The specific formula for calculating the weighted distance of the anti-ideal point is as follows:

[0070] Furthermore, the optimization index U is calculated by combining the weighted distance of the ideal point and the weighted distance of the anti-ideal point. j The formula for calculating the preference index is:

[0071] The preference index U j Substituting the values ​​and performing credibility standardization calculations, we obtain the credibility sequence, i.e., the credibility w of each expert weight. j The formula for standardizing credibility is: For example, the subjective weight matrix is ​​as follows: The credibility sequence calculated using the credibility standardization formula is (w1′, …, w′). m If the first group decision weights are obtained, then the first group decision weights are W'×W. T Furthermore, taking the decision weights of the first group as a positive baseline, for w... * i The process is updated, the analysis steps are repeated, the second group decision weights are obtained, and the average difference between the first group decision weights and the second group decision weights is calculated. If the average difference is less than a preset difference, the second group decision weight is taken as the optimal weight; otherwise, the analysis steps are repeated until the average difference is less than the preset difference to obtain the optimal weight. The formula for calculating the average difference is:

[0072] The method provided in this application also includes:

[0073] Obtain the gas delivery channel between the temperature sensor, smoke sensor, hydrogen chloride sensor, and pyrolysis particle sensor and the intake module;

[0074] A gas delivery test is performed on the gas delivery channel to obtain a delivery sample, wherein the delivery sample includes pre-delivery sensing data of the intake module and post-delivery sensing data detected by the temperature sensor, smoke sensor, hydrogen chloride sensor and pyrolysis particle sensor.

[0075] Based on the transmitted samples, perform multi-parameter sensing loss analysis to obtain multi-parameter loss indices;

[0076] The confidence correction of the multidimensional sensing and monitoring data is performed using the multi-parameter loss index.

[0077] The gas delivery channel between the temperature sensor, smoke sensor, hydrogen chloride sensor, and pyrolysis particle sensor and the inhalation module is obtained. A gas delivery test is performed on the gas delivery channel to obtain a transmission sample. This transmission sample includes pre-transmission sensing data from the inhalation module and post-transmission sensing data detected by the temperature sensor, smoke sensor, hydrogen chloride sensor, and pyrolysis particle sensor. Further, a sensing loss analysis of each parameter is performed on the transmission sample to obtain the parameter loss ratio, i.e., the ratio of each sensing data point after transmission to each sensing data point before transmission, thus obtaining a multi-parameter loss index. Finally, the confidence correction of the multi-dimensional sensing monitoring data is performed using the multi-parameter loss index, i.e., by dividing each monitoring data point in the multi-dimensional sensing monitoring data by the corresponding parameter loss index, and obtaining the calculation result to complete the confidence correction.

[0078] In the above text, refer to Figure 1 A multi-parameter coupled fire detection method based on an adaptive algorithm according to an embodiment of the present invention is described in detail. Next, reference will be made to... Figure 2 This invention describes a multi-parameter coupled fire detection system based on an adaptive algorithm according to an embodiment of the present invention.

[0079] According to an embodiment of the present invention, a multi-parameter coupled fire detection system based on an adaptive algorithm solves the technical problems of low accuracy and poor stability in existing cable tunnel fire early warning technologies. By collecting multiple fire early warning parameters and processing them using an adaptive algorithm, the system improves the accuracy and stability of cable tunnel fire detection and early warning. The multi-parameter coupled fire detection system based on an adaptive algorithm includes: an index analysis module 11, a sensor acquisition module 12, a monitoring data acquisition module 13, a risk identification module 14, a risk indicator acquisition module 15, and a fire early warning module 16.

[0080] Index analysis module 11 is used to perform index analysis of historical fire monitoring records based on the cable distribution characteristics of the target cable tunnel to obtain the preset fire monitoring coverage area;

[0081] The sensor acquisition module 12 is used to acquire the multi-parameter sensing and monitoring module, wherein the multi-parameter sensing and monitoring module includes a temperature sensor, a smoke sensor, a hydrogen chloride sensor and a pyrolysis particle sensor, and any one of the sensors is connected to the inhalation module.

[0082] The monitoring data acquisition module 13 is used to collect gas from the preset fire monitoring coverage area through the air intake module, and send the gas sample to the multi-parameter sensing and monitoring module for multi-parameter detection to obtain multi-dimensional sensing and monitoring data, wherein the multi-dimensional sensing and monitoring data includes temperature, smoke occlusion, hydrogen chloride concentration and pyrolysis particle concentration.

[0083] Risk identification module 14 is used to send the multi-dimensional sensing and monitoring data to the single-parameter discrimination module to perform single-parameter fire risk identification and generate multiple single-parameter risk indicators;

[0084] The risk indicator acquisition module 15 is used to perform fire alarm trigger judgment based on the multiple single-parameter risk indicators. If none of the multiple single-parameter risk indicators meet the preset alarm trigger threshold, the multiple single-parameter risk indicators are sent to the multi-parameter coupling analysis module to perform coupling weighting to obtain the coupling risk indicator.

[0085] The fire early warning module 16 is used to provide fire early warning based on the coupled risk indicators.

[0086] The specific configuration of the index analysis module 11 will be described in detail below. The index analysis module 11 may further include: performing historical fire monitoring record index analysis based on the cable distribution characteristics of the target cable tunnel to obtain a preset fire monitoring coverage area, including: the cable distribution characteristics including tunnel structural characteristics and cable topology characteristics; indexing a set of fire monitoring record samples of similar cable tunnels with a similarity greater than a preset similarity based on the cable distribution characteristics; locating the fire source point based on the fire monitoring record sample set, and extracting the temperature distribution, smoke shading distribution, hydrogen chloride concentration distribution, and pyrolysis particle concentration distribution corresponding to the fire source point; identifying multi-sensor sensitive areas based on the temperature distribution, smoke shading distribution, hydrogen chloride concentration distribution, and pyrolysis particle concentration distribution; and generating the preset fire monitoring coverage area using the multi-sensor sensitive areas.

[0087] The specific configuration of the risk indicator acquisition module 15 will be described in detail below. The risk indicator acquisition module 15 further includes: sending the plurality of single-parameter risk indicators to a multi-parameter coupling analysis module for coupling weighting to obtain coupled risk indicators, including: monitoring environmental conditions within a preset range of the target cable tunnel; traversing preset weighting modes based on the environmental conditions to obtain matching weighting modes, wherein the preset weighting modes include a low-humidity dust mode, a low-humidity high-dust mode, a high-humidity low-dust mode, and a high-humidity high-dust mode; assigning weights to the plurality of single-parameter risk indicators based on the matching weighting modes and performing weighted calculations to obtain the coupled risk indicators.

[0088] The specific configuration of the risk indicator acquisition module 15 will be described in detail below. The risk indicator acquisition module 15 may further include: assigning weights to the plurality of single-parameter risk indicators based on the matching weight mode, including: using the analytic hierarchy process (AHP) to perform expert subjective weighting on the plurality of single-parameter risk indicators to obtain a parameter subjective weight matrix; and iteratively correcting and optimizing the parameter subjective weight matrix through bipolar analysis to generate optimal weights.

[0089] The specific configuration of the risk indicator acquisition module 15 will be described in detail below. The risk indicator acquisition module 15 further includes: iteratively correcting and optimizing the parameter subjective weight matrix through dual-base point analysis to generate optimal weights, including: averaging the parameter subjective weight matrix to obtain initial positive base points; standardizing the parameter subjective weight matrix based on the initial positive base points to obtain relative evaluation values; constructing ideal and anti-ideal sequences, combining the relative evaluation values ​​to construct an optimal weighted distance function, and obtaining the optimal weighting factor; calculating the ideal point distance and anti-ideal point distance on the optimal weighting factor to obtain the ideal point weighted distance and anti-ideal point weighted distance; calculating the preference index by combining the ideal point weighted distance and anti-ideal point weighted distance, and performing credibility standardization calculation to obtain a credibility sequence; generating a first group of decision weights by combining the credibility sequence and the parameter subjective weight matrix; using the first group of decision weights as positive base points, repeating the analysis to obtain a second group of decision weights, and calculating the average difference between the first group of decision weights and the second group of decision weights. If the average difference is less than a preset difference, the second group of decision weights is used as the optimal weight.

[0090] The expression for the optimal weighted distance function is as follows:

[0091]

[0092] Where, r ij This refers to the relative evaluation value of the j-th expert's subjective weight for the i-th parameter, obtained by standardizing the subjective weight matrix of the parameters; ki is the weighting factor; n is the number of parameter types; m is the number of weighted experts in the parameter subjective weight matrix.

[0093] The specific configuration of the fire early warning module 16 will be described in detail below. The fire early warning module 16 further includes: acquiring the gas delivery channel between the temperature sensor, smoke sensor, hydrogen chloride sensor, and pyrolysis particle sensor and the intake module; performing a gas delivery test on the gas delivery channel to acquire a transmission sample, wherein the transmission sample includes pre-delivery sensing data from the intake module and post-delivery sensing data detected by the temperature sensor, smoke sensor, hydrogen chloride sensor, and pyrolysis particle sensor; performing multi-parameter sensing loss analysis based on the transmission sample to acquire a multi-parameter loss index; and using the multi-parameter loss index to perform confidence correction on the multi-dimensional sensing monitoring data.

[0094] The multi-parameter coupled fire detection system based on an adaptive algorithm provided in this embodiment of the invention can execute the multi-parameter coupled fire detection method based on an adaptive algorithm provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0095] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0096] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A multi-parameter coupled fire detection method based on an adaptive algorithm, characterized in that, The method includes: Based on the cable distribution characteristics of the target cable tunnel, historical fire monitoring records are indexed and analyzed to obtain the preset fire monitoring coverage area. A multi-parameter sensing and monitoring module is acquired, wherein the multi-parameter sensing and monitoring module includes a temperature sensor, a smoke sensor, a hydrogen chloride sensor and a pyrolysis particle sensor, and any one of the sensors is connected to the inhalation module; The gas is collected from the preset fire monitoring coverage area by the air intake module, and the gas sample is sent to the multi-parameter sensing and monitoring module for multi-parameter detection to obtain multi-dimensional sensing and monitoring data, including temperature, smoke opacity, hydrogen chloride concentration and pyrolysis particle concentration. The multi-dimensional sensing and monitoring data is sent to the single-parameter discrimination module to perform single-parameter fire risk identification and generate multiple single-parameter risk indicators. Fire alarm triggering judgment is performed based on the multiple single-parameter risk indicators. If none of the multiple single-parameter risk indicators meet the preset alarm triggering threshold, the multiple single-parameter risk indicators are sent to the multi-parameter coupling analysis module to perform coupling weighting to obtain the coupling risk indicators. Fire early warning is based on the aforementioned coupled risk indicators.

2. The method as described in claim 1, characterized in that, Based on the cable distribution characteristics of the target cable tunnel, historical fire monitoring records are indexed and analyzed to obtain the preset fire monitoring coverage area, including: The cable distribution characteristics include tunnel structural characteristics and cable topology characteristics; A set of fire monitoring records for similar cable tunnels whose similarity to the cable distribution feature index is greater than a preset similarity. Based on the fire monitoring record sample set, the fire source point is located, and the temperature distribution, smoke occlusion distribution, hydrogen chloride concentration distribution and pyrolysis particle concentration distribution corresponding to the fire source point are extracted. The sensitive areas of the multi-sensor system are identified based on the temperature distribution, smoke occlusion distribution, hydrogen chloride concentration distribution, and pyrolysis particle concentration distribution. The preset fire monitoring coverage area is generated using the sensitive areas of the multi-sensor system.

3. The method as described in claim 1, characterized in that, The multiple single-parameter risk indicators are sent to the multi-parameter coupling analysis module for coupling weighting to obtain coupled risk indicators, including: Monitor environmental conditions within a pre-defined area of ​​the target cable tunnel; The preset weighted modes are traversed under the environmental conditions to obtain the matching weighted modes, wherein the preset weighted modes include low humidity and high dust mode, low humidity and high dust mode, high humidity and low dust mode and high humidity and high dust mode. The multiple single-parameter risk indicators are weighted based on the matching weight mode and then weighted to obtain the coupled risk indicator.

4. The method as described in claim 3, characterized in that, The multiple single-parameter risk indicators are weighted based on the matching weight pattern, including: The analytic hierarchy process (AHP) is used to subjectively assign expert weights to the multiple single-parameter risk indicators, resulting in a parameter subjective weight matrix. The subjective weight matrix of the parameters is iteratively corrected and optimized through two-point analysis to generate the optimal weights.

5. The method as described in claim 4, characterized in that, The subjective weight matrix of the parameters is iteratively corrected and optimized through two-point analysis to generate the optimal weights, including: The subjective weight matrix of the parameters is averaged to obtain the initial positive base point; Based on the initial positive base point, the subjective weight matrix of the parameters is standardized to obtain a relative evaluation value; Construct ideal and anti-ideal sequences, combine the relative evaluation values ​​to construct the optimal weighted distance function, and obtain the optimal weighting factor; The ideal point distance and anti-ideal point distance are calculated for the optimal weighting factor to obtain the ideal point weighted distance and anti-ideal point weighted distance; The optimization index is calculated by combining the weighted distance of the ideal point and the weighted distance of the anti-ideal point, and the credibility is standardized to obtain the credibility sequence. The first group of decision weights is generated by combining the credibility sequence and the subjective weight matrix of parameters; Using the first group decision weight as a positive base point, repeat the analysis to obtain the second group decision weight, and calculate the average difference between the first group decision weight and the second group decision weight. If the average difference is less than a preset difference, the second group decision weight is taken as the optimal weight.

6. The method as described in claim 5, characterized in that, The expression for the optimal weighted distance function is as follows: Where, r ij This refers to the relative evaluation value of the j-th expert's subjective weight for the i-th parameter, obtained by standardizing the subjective weight matrix of the parameters; k i is the weighting factor; n is the number of parameter types; m is the number of weighted experts in the parameter subjective weight matrix.

7. The method as described in claim 1, characterized in that, The method further includes: Obtain the gas delivery channel between the temperature sensor, smoke sensor, hydrogen chloride sensor, and pyrolysis particle sensor and the intake module; A gas delivery test is performed on the gas delivery channel to obtain a delivery sample, wherein the delivery sample includes pre-delivery sensing data of the intake module and post-delivery sensing data detected by the temperature sensor, smoke sensor, hydrogen chloride sensor and pyrolysis particle sensor. Multi-parameter sensing loss analysis is performed based on the transmitted samples to obtain multi-parameter loss indices. The confidence correction of the multidimensional sensing and monitoring data is performed using the multi-parameter loss index.

8. A multi-parameter coupled fire detection system based on an adaptive algorithm, characterized in that, The system includes: The index analysis module is used to perform index analysis of historical fire monitoring records based on the cable distribution characteristics of the target cable tunnel to obtain the preset fire monitoring coverage area; The sensor acquisition module is used to acquire information from the multi-parameter sensing and monitoring module, wherein the multi-parameter sensing and monitoring module includes a temperature sensor, a smoke sensor, a hydrogen chloride sensor, and a pyrolysis particle sensor, and any one of the sensors is connected to the inhalation module. The monitoring data acquisition module is used to collect gas from the preset fire monitoring coverage area through the air intake module, and send the gas sample to the multi-parameter sensing monitoring module for multi-parameter detection to obtain multi-dimensional sensing monitoring data, wherein the multi-dimensional sensing monitoring data includes temperature, smoke occlusion, hydrogen chloride concentration and pyrolysis particle concentration. The risk identification module is used to send the multi-dimensional sensing and monitoring data to the single-parameter discrimination module to perform single-parameter fire risk identification and generate multiple single-parameter risk indicators. The risk indicator acquisition module is used to perform fire alarm trigger judgment based on the multiple single-parameter risk indicators. If none of the multiple single-parameter risk indicators meet the preset alarm trigger threshold, the multiple single-parameter risk indicators are sent to the multi-parameter coupling analysis module to perform coupling weighting to obtain the coupled risk indicator. A fire early warning module is used to provide fire early warning based on the coupled risk indicators.

Images