A hierarchical detection method and system for aspirating smoke fires in complex environments

Abstract

The present invention discloses a hierarchical detection method and system for aspirated smoke fires in complex environments, and relates to the field of fire early warning technology. The method comprises: identifying vents in a monitoring area under complex environments, and placing aspirated smoke detectors on the vents; monitoring the status of the vents to obtain the working mode parameters of the vents, including ventilation start / stop status, wind speed, and ventilation direction; analyzing and obtaining an adaptive strategy that matches the working mode parameters to control an exhaust pump to drive the aspirated smoke detector to collect smoke gas samples, perform multi-time series hierarchical detection, and output a fire early warning signal. The present invention solves the technical problem that existing fire detection technologies are difficult to accurately collect smoke samples and issue early warnings in places with complex ventilation structures and significant air disturbances, and achieves the technical effect of improving the accuracy of smoke collection and the real-time performance of fire early warnings by placing aspirated smoke detectors on vents and dynamically adjusting the sampling strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of fire warning, and in particular to a hierarchical detection method and system for aspiration-type smoke fires in complex environments. Background Art

[0002] In modern buildings and industrial environments, fire detection systems face challenges posed by complex ventilation conditions. This is especially true in locations with complex ventilation structures and significant air disturbance, such as large shopping malls, underground parking lots, data centers, and industrial plants. Traditional fire detection equipment often struggles to effectively capture smoke signals in the early stages of a fire. Ventilation openings in these locations are key nodes for air flow and are most sensitive to fire smoke under load. However, existing technologies for aspirating smoke sensors create airflow conflicts with the air velocity at the vents under different ventilation modes, making it difficult to collect accurate gas samples. This not only affects the timeliness and accuracy of fire detection, but can also lead to false or missed alarms, increasing the risk of fire. Summary of the Invention

[0003] The present application provides a hierarchical detection method and system for aspirated smoke fires in complex environments, which is used to solve the technical problem that existing fire detection technologies are difficult to accurately collect smoke samples and issue timely warnings in places with complex ventilation structures and significant air disturbances.

[0004] The first aspect of the present application provides a hierarchical detection method for suction-type smoke fires in a complex environment, the method comprising: identifying ventilation openings in a monitoring area under a complex environment, placing suction-type smoke sensing devices on the ventilation openings, the suction-type smoke sensing devices comprising a micro vacuum pump, and the vacuum pump being controlled by communication with a control unit; performing status monitoring on the ventilation openings to obtain working mode parameters of the ventilation openings, the working mode parameters comprising ventilation start and stop status, wind speed, and ventilation direction; analyzing the working mode parameters to obtain an adaptive strategy matching the working mode parameters, the control unit controlling the vacuum pump according to the matching adaptive strategy, and driving the suction-type smoke sensing device to collect smoke gas samples; performing multi-time series hierarchical detection based on the smoke gas samples, and outputting a fire warning signal.

[0005] The second aspect of the present application provides a graded detection system for suction-type smoke fires in complex environments, the system comprising: a smoke sensor deployment module, the smoke sensor deployment module being used to identify vents in a monitoring area under a complex environment, and deploying suction-type smoke sensors on the vents, the suction-type smoke sensors comprising a micro vacuum pump, and the vacuum pump being controlled by communication with a control unit; a status monitoring module, the status monitoring module being used to perform status monitoring on the vents to obtain working mode parameters of the vents, the working mode parameters comprising ventilation start and stop status, wind speed and ventilation direction; a smoke gas sample collection module, the smoke gas sample collection module being used to analyze the working mode parameters to obtain an adaptive strategy matching the working mode parameters, the control unit controlling the vacuum pump according to the matched adaptive strategy, and driving the suction-type smoke sensor to collect smoke gas samples; a multi-time series graded detection module, the multi-time series graded detection module being used to perform multi-time series graded detection based on the smoke gas samples, and output a fire warning signal.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The present application provides a hierarchical detection method and system for aspiration smoke fires in a complex environment, which relates to the field of fire warning technology. By identifying ventilation openings in a complex environment and deploying aspiration smoke detection equipment, the device sampling is adjusted by analyzing and matching adaptive strategies according to the working mode parameters of the ventilation openings. Multi-time series hierarchical detection is performed based on the collected smoke samples, and a fire warning signal is output. This solves the technical problem that the existing fire detection technology is difficult to accurately collect smoke samples and issue warnings in a timely manner in places with complex ventilation structures and significant air disturbances. It achieves the technical effect of improving the accuracy of smoke collection and the real-time nature of fire warnings by deploying aspiration smoke detection equipment at ventilation openings and dynamically adjusting the sampling strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A schematic flow chart of a hierarchical detection method for aspirating smoke fires in a complex environment provided by an embodiment of the present application;

[0010] Figure 2 This is a schematic diagram of the structure of a hierarchical detection system for aspiration-type smoke fires in complex environments provided in an embodiment of the present application.

[0011] Description of the accompanying drawings: smoke detection equipment layout module 11, status monitoring module 12, smoke gas sample collection module 13, multi-time series hierarchical detection module 14. DETAILED DESCRIPTION

[0012] The present application provides a hierarchical detection method and system for aspirated smoke fires in complex environments, which is used to solve the technical problem that existing fire detection technologies are difficult to accurately collect smoke samples and issue timely warnings in places with complex ventilation structures and significant air disturbances.

[0013] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0014] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, 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 clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices.

[0015] Example 1, as Figure 1 As shown, the present application provides a hierarchical detection method for aspirating smoke fires in complex environments, the method comprising:

[0016] P10: Identify the ventilation holes in the monitoring area under complex environment, and place an air suction type smoke sensor on the ventilation holes. The air suction type smoke sensor includes a micro air pump, and the air pump is controlled by the control unit through communication.

[0017] Specifically, in complex environments with intricate ventilation structures and significant air disturbance, the propagation path and diffusion rate of fire smoke are often difficult to predict. Vents, as key nodes of air flow, are crucial for the early detection of fire smoke. Therefore, to effectively monitor fires in these locations, it is first necessary to identify the ventilation openings within the monitoring area, including natural and mechanical vents, as well as the inlets and outlets of various types of air handling equipment. Installing aspirating smoke sensors at these vents can leverage the air flow characteristics of these vents to quickly capture smoke particles generated by the fire.

[0018] Among them, the suction-type smoke detector is an advanced fire detection device whose core component is a micro-vacuum pump. The function of the micro-vacuum pump is to generate negative pressure, drawing smoke particles from the surrounding air into the device for detection. This type of vacuum pump is generally characterized by small size, low energy consumption, and high extraction efficiency, and can adapt to various installation and operating conditions in complex environments. The micro-vacuum pump is controlled by a control unit through communication. The control unit can precisely control the operating status of the vacuum pump based on preset programs or real-time monitoring data. For example, the control unit can adjust the vacuum pump's extraction speed and frequency based on parameters such as wind speed and wind direction at the vent to ensure that smoke samples can be effectively collected under different ventilation conditions.

[0019] The control unit's communication control function is key to implementing the adaptive strategy. Through the communication module, the control unit receives real-time data from sensors, such as the vent's on / off status and wind speed. After processing and analyzing this data, it generates an adaptive strategy that matches the vent's operating mode and controls the micro-extraction pump accordingly. This adaptive control mechanism enables aspirating smoke sensors to consistently maintain optimal smoke collection performance in complex and changing ventilation environments, thereby improving the accuracy and reliability of fire detection.

[0020] In practical applications, the identification of ventilation openings and the placement of aspiration smoke detectors require comprehensive consideration of factors such as the building structure, ventilation system design, and fire risk distribution. By accurately identifying ventilation openings and rationally placing aspiration smoke detectors, combined with the efficient extraction capabilities of micro-vacuum pumps and the intelligent communication and control capabilities of control units, rapid and accurate fire smoke detection can be achieved in complex environments, providing reliable technical support for fire early warning.

[0021] Furthermore, in an embodiment of the present application, a method for identifying vents in a monitoring area under a complex environment and detecting whether the current environment is a complex environment includes:

[0022] P11a: Analyze the spatial structure information of the current environment, which includes the building structure distribution and the ventilation system distribution; P12a: Collect aerodynamic characteristics based on the building structure distribution and the ventilation system distribution, simulate based on the aerodynamic characteristics and the preset smoke diffusion path, and output the smoke diffusion disturbance index; P13a: When the smoke diffusion disturbance index is greater than the preset disturbance index, identify the current environment as a complex environment, and identify the ventilation outlets in the monitoring area under the complex environment.

[0023] Optionally, before identifying the ventilation holes in the monitoring area under a complex environment, it is first necessary to determine whether the current environment is a complex environment.

[0024] Specifically, the complexity of the environment is assessed by analyzing its spatial structure. Spatial structure information includes building structure distribution and ventilation system distribution. Building structure distribution refers to the spatial layout of various areas within a building, such as room partitions, floor heights, and wall thicknesses. These factors affect the path and intensity of air flow. Ventilation system distribution refers to the distribution of various ventilation equipment, air ducts, exhaust vents, and other elements within the building. These elements determine the primary routes of air flow. By comprehensively analyzing this information, we can gain a preliminary understanding of the basic conditions of air flow within the building, providing foundational data for further assessment.

[0025] Next, based on the aforementioned building structure and ventilation system distribution information, aerodynamic characteristics are collected. Aerodynamic characteristics include parameters such as wind speed, wind direction, and air pressure differential, which reflect the actual flow state of air in the current environment. The collected aerodynamic characteristic data will be used in the subsequent smoke diffusion path simulation. By establishing a mathematical model that combines the collected aerodynamic characteristics with the pre-set smoke diffusion path, smoke diffusion simulation is performed. The smoke diffusion disturbance index output by the simulation can quantitatively reflect the smoke diffusion in the current environment, including the smoke propagation speed, diffusion range, and the degree of air flow interference.

[0026] When the smoke diffusion disturbance index exceeds the preset disturbance index, it indicates that the air flow in the current environment is complex, causing significant disturbances to smoke diffusion, which may cause large changes in the smoke diffusion path and speed, thereby affecting the effectiveness of the smoke detection equipment. At this time, the system identifies the current environment as a complex environment and identifies the vents within this complex environment. The identification of vents can be based on a comprehensive analysis of spatial structure information and aerodynamic characteristics. By locating the inlet and outlet of the ventilation duct, the position of the fan, and the convergence point of air flow, the location of the vents can be accurately determined. These vents are key locations for the subsequent deployment of aspirated smoke detection equipment because they are key nodes for smoke propagation in complex environments.

[0027] Through the above steps, the present embodiment can accurately identify ventilation openings in complex environments, providing a scientific basis for the deployment of aspirating smoke detectors. This method not only helps identify the characteristics of complex environments but also optimizes the placement of devices within the monitoring area, improving the accuracy and response speed of smoke detection.

[0028] P20: Monitor the status of the vent to obtain the working mode parameters of the vent, where the working mode parameters include ventilation start / stop status, wind speed, and ventilation direction.

[0029] Specifically, in the embodiments of this application, monitoring the status of vents is a key step in implementing the aspirating smoke fire detection method. Vents are key nodes in air flow, and in complex environments, their operating conditions directly affect the flow path of smoke and the effectiveness of smoke detectors. Therefore, accurately monitoring the status of vents is crucial to ensuring that smoke detectors can accurately capture smoke samples.

[0030] First, appropriate sensors and monitoring equipment are deployed to monitor the vent's operating status in real time. These sensors can capture various vent operating parameters in real time, including but not limited to key indicators such as ventilation start / stop status, wind speed, and ventilation direction. Specifically, ventilation start / stop status refers to whether the ventilation system is on or temporarily shut down for some reason. Changes in ventilation start / stop status can interrupt or alter air flow, affecting smoke dispersion and collection. This parameter can be monitored using a switch detection sensor.

[0031] Secondly, monitor the wind speed at the vents. Wind speed refers to the rate of air flow through the vents and is a key factor influencing the speed and direction of smoke transmission. Installing wind speed sensors at the vents allows for real-time wind speed measurement. These sensors provide precise wind speed data for subsequent analysis and control. Accurate wind speed data is crucial for determining the extraction intensity and frequency of the vacuum pump, which needs to adjust its operating state based on wind speed to ensure effective smoke sampling.

[0032] Finally, monitor the ventilation direction of the vents—that is, the direction of air flow. This can be achieved using directional sensors installed at the vents. Changes in ventilation direction can affect the path of smoke propagation. Therefore, accurately monitoring ventilation direction can help the system better predict the direction of smoke diffusion, thereby optimizing the sampling strategy of the aspirating smoke detector.

[0033] By comprehensively monitoring the vent's on / off status, wind speed, and ventilation direction, the system can fully understand the vent's operating mode. These operating mode parameters serve as the basis for subsequent adaptive strategy analysis, enabling the control unit to dynamically adjust the micro-vacuum pump's operating parameters based on the vent's actual operating status. This ensures that the aspirating smoke sensor can effectively collect smoke samples under different ventilation conditions, improving the accuracy and reliability of fire detection.

[0034] P30: Analyze the working mode parameters to obtain an adaptive strategy that matches the working mode parameters, and the control unit controls the air pump according to the matched adaptive strategy to drive the suction-type smoke sensor to collect smoke gas samples.

[0035] Furthermore, the operating mode parameters are analyzed to obtain an adaptive strategy that matches the operating mode parameters. In this embodiment of the present application, step P30 further includes:

[0036] P31: Set up a strategy mapping library containing multiple ventilation scenarios, and the strategy mapping library includes a mapping relationship between each working mode parameter and the adaptive strategy for controlling the vacuum pump; P32: Based on the KNN algorithm, the working mode parameters are input into the strategy mapping library for strategy matching to obtain an adaptive strategy that matches the working mode parameters, wherein the adaptive strategy includes the vacuum pump speed, sampling period, filtering strength and pressure stabilization time for controlling the vacuum pump.

[0037] It should be understood that by analyzing the operating mode parameters obtained in the previous step, an adaptive strategy that matches these parameters is derived. Through such analysis, the control unit can formulate a reasonable strategy to control the operation of the air extraction pump according to different environmental conditions and the operating mode of the vent, thereby ensuring that the aspiration smoke sensor can efficiently collect smoke gas samples.

[0038] After completing vent status monitoring and obtaining operating mode parameters (ventilation start / stop status, wind speed, and ventilation direction), the first step is to set up a strategy mapping library containing multiple ventilation scenarios. This strategy mapping library records in detail the mapping relationship between each operating mode parameter and the adaptive strategy used to control the vacuum pump. Specifically, the strategy mapping library contains the optimal operating parameters that the vacuum pump should adopt under different combinations of ventilation start / stop status, wind speed, and ventilation direction, such as vacuum pump speed, sampling period, filter intensity, and voltage stabilization time. These mapping relationships are based on a large amount of experimental data and simulation analysis, and can ensure that the vacuum pump can operate in an optimal manner under various ventilation conditions, thereby improving the efficiency and accuracy of smoke collection.

[0039] Next, based on the KNN (K-Nearest Neighbors) algorithm, the monitored operating mode parameters are input into the strategy mapping library for strategy matching. The KNN algorithm is an example-based learning method that calculates the similarity between the input parameters and the known parameters in the mapping library to find the closest K neighbors. The final adaptive strategy is generated based on the strategy information of these neighbors. This method can effectively handle complex ventilation scenarios and select the most appropriate control strategy for the current environmental conditions.

[0040] The adaptive strategy thus matched includes the following key parameters: pump speed, sampling period, filter strength, and stabilization time. Pump speed determines the air sampling rate, the sampling period controls the time interval between each sampling, the filter strength adjusts the sensitivity to smoke signals during air sampling, and the stabilization time ensures stable pump operation, preventing pressure fluctuations from affecting sample quality.

[0041] According to the adaptive strategy obtained through analysis, the control unit will adjust the operating parameters of the vacuum pump in real time to ensure that the vacuum pump can effectively collect smoke samples. For example, when the wind speed is high, the rotation speed can be increased to overcome the resistance of the airflow; when the wind speed is low, the rotation speed can be appropriately reduced to save energy. In scenarios where ventilation conditions change rapidly, the sampling period needs to be shortened so that samples can be collected more frequently and signs of fire can be detected in a timely manner. In scenarios where ventilation conditions are complex and there is a lot of interference, the filtering intensity needs to be increased. In scenarios where the ventilation system starts and stops frequently or the wind speed changes greatly, the voltage stabilization time needs to be appropriately extended to ensure that the vacuum pump can collect smoke samples in a stable state.

[0042] In this way, this embodiment can adjust the working mode of the smoke detection device in real time according to the working status of the ventilation port, so that the device can maintain efficient smoke collection capabilities under different environmental conditions, significantly improving the reliability and response speed of fire warning.

[0043] Furthermore, step P32 of the embodiment of the present application further includes:

[0044] P32-1: Input the working mode parameters into the strategy mapping library, obtain the weighted distances between the working mode parameters and the working mode parameters corresponding to each adaptive strategy under the working mode parameters, and output a weighted distance set; P32-2: Select the first K nearest neighbor adaptive strategies in the weighted distance set based on the KNN algorithm; P32-3: Output the adaptive strategy that matches the working mode parameters based on the K historical call probabilities corresponding to the K nearest neighbor adaptive strategies.

[0045] Optionally, the matching process of the adaptive strategy using the KNN algorithm is further refined.

[0046] Specifically, the current working mode parameters are input into the strategy mapping library. In this library, each adaptive strategy has a clear mapping relationship with multiple working mode parameters. By inputting the working mode parameters, the system can calculate the weighted distance between the parameter and all the existing adaptive strategies in the library. The weighted distance refers to the distance value calculated based on the degree of difference between a specific working mode parameter and the strategy through a preset weight function. The weight function can be adjusted according to the importance of different parameters. The output of this step is a weighted distance set, which contains the weighted distance values ​​between the current working mode parameters and each adaptive strategy, reflecting the degree of match between these strategies and the current environment.

[0047] Next, the KNN algorithm selects the top K nearest neighbor adaptive strategies from the weighted distance set. The KNN algorithm analyzes each weighted distance in the weighted distance set and selects the top K adaptive strategies closest to the current operating mode parameters. The core of this process is to compare the similarity between historical environmental data and current parameters, identify the closest strategy, and select it as a potential candidate strategy.

[0048] Finally, based on the historical call probabilities of the K nearest neighbor adaptive strategies, the adaptive strategy that best matches the current operating mode parameters is output. The historical call probability refers to the frequency with which a particular adaptive strategy was selected in past operations. By incorporating this probability, the selected adaptive strategy is ensured to not only match the current operating mode parameters but also possess high applicability and reliability. In other words, based on the historical call data of the K nearest neighbor adaptive strategies, a comprehensive evaluation is performed to select the strategy that best suits the current environment.

[0049] Through the above steps, the embodiment of the present application can dynamically generate a matching adaptive strategy based on the operating mode parameters of the vents using the KNN algorithm and strategy mapping library. This method not only improves the efficiency and accuracy of smoke collection, but also enhances the system's adaptability to complex ventilation environments.

[0050] Furthermore, the aspirating smoke sensor is driven to collect smoke gas samples. Step P30 of the embodiment of the present application further includes:

[0051] P33: Output the smoke diffusion disturbance sub-index of the ventilation port according to the working mode parameters of the ventilation port; P34: Obtain the gas sampling compensation coefficient according to the smoke diffusion disturbance sub-index of the ventilation port, and use the gas sampling compensation coefficient to compensate for the frequency of collecting the smoke gas sample.

[0052] In a possible embodiment of the present application, in the process of further improving smoke collection and accurate judgment, a ventilation state-smoke distribution interference model can be introduced to perform compensation correction on smoke samples under specific ventilation conditions, thereby improving the accuracy of smoke concentration judgment.

[0053] Specifically, after completing vent status monitoring and obtaining operating mode parameters (ventilation on / off status, wind speed, and ventilation direction), the system analyzes the degree of interference caused by the vent's operating mode on the smoke diffusion path, quantifies the stability and consistency of smoke distribution, and outputs the vent's smoke diffusion disturbance sub-index. This sub-index is calculated based on the vent's specific operating mode parameters and quantifies the degree of disturbance caused by the vent on smoke diffusion. Factors such as the vent's on / off status, wind speed, and ventilation direction all affect the smoke propagation path and speed. Therefore, by weighting the impact of these factors and calculating the smoke diffusion disturbance sub-index, a more accurate assessment of the vent's impact on smoke collection can be achieved.

[0054] Next, the gas sampling compensation coefficient is obtained according to the smoke diffusion disturbance sub-index of the ventilation port. This compensation coefficient is derived based on the smoke diffusion disturbance sub-index and can reflect the sampling deviation caused by changes in ventilation conditions. For example, the gas sampling compensation coefficient is set to be proportional to the smoke diffusion disturbance sub-index to calculate the gas sampling compensation coefficient to compensate for the impact of the ventilation port on the smoke collection frequency. For example, when the degree of disturbance at the ventilation port is high, it may be necessary to increase the sampling frequency to ensure that the smoke signal can be captured in time; when the degree of disturbance is low, the sampling frequency can be appropriately reduced to save resources. By obtaining the gas sampling compensation coefficient, the sampling frequency of the aspirating smoke sensor can be dynamically adjusted, thereby improving the accuracy of smoke concentration determination.

[0055] Finally, the frequency of smoke gas sampling is compensated using a gas sampling compensation coefficient. Specifically, the control unit adjusts the micro-pump's operating parameters, such as pump speed and sampling period, based on the gas sampling compensation coefficient to ensure optimal smoke sampling frequency under varying ventilation conditions. This approach not only accounts for the impact of the vent's operating mode parameters on smoke collection but also further improves the accuracy of smoke concentration determination through a dynamic compensation mechanism.

[0056] P40: Perform multi-time series hierarchical detection based on the smoke gas sample and output a fire warning signal.

[0057] Specifically, after completing vent identification, status monitoring, and adjusting the sampling frequency based on operating mode parameters, the aspirating smoke detector collects a series of smoke gas samples. These samples contain key information about the initial stages of a fire, but a single sample at a single time point may not accurately reflect the fire's development trend. Therefore, the present embodiment employs a multi-time series hierarchical detection method to comprehensively analyze the collected smoke gas samples.

[0058] Multi-time series hierarchical detection involves analyzing smoke gas samples at different time scales to assess the likelihood and severity of a fire. Specifically, short-term time series analysis begins with continuous monitoring of smoke gas samples over short time intervals (e.g., minutes). This phase aims to capture rapid changes in the early stages of a fire, such as a sharp rise in smoke concentration. Short-term time series analysis enables rapid response to early fire signals and timely early warnings. Next, medium-term time series analysis involves analyzing smoke gas samples over longer time intervals (e.g., 30 minutes to several hours). This phase aims to assess the fire's development trend, determining whether smoke concentration continues to rise and whether the rate of increase conforms to typical fire development patterns. Medium-term time series analysis helps distinguish between brief smoke disturbances and true fire signals. Finally, long-term time series analysis involves analyzing smoke gas samples over longer time intervals (e.g., hours to days). This phase aims to assess the long-term impact of the fire and the long-term effects of the ventilation system on smoke dispersion. Long-term time series analysis helps identify potential fire hazards and assess the impact of a fire on the overall building environment.

[0059] Furthermore, graded detection means that fire warnings are divided into different levels according to different thresholds of smoke concentration. For example, when the smoke concentration exceeds the preset low threshold, the system issues a low-level warning signal, which indicates that there may be a fire risk, but it has not yet reached an emergency state. The low-level warning can remind relevant personnel to conduct a preliminary inspection to confirm whether there are fire hazards; when the smoke concentration exceeds the preset high threshold, the system issues a high-level warning signal, which indicates that the possibility of fire is high and immediate action is required. The high-level warning can trigger the automatic fire extinguishing system or notify the fire department to intervene; when the smoke concentration continues to rise and exceeds the preset emergency threshold, the system issues an emergency warning signal, which indicates that a fire has occurred and may have posed a serious threat to the safety of people and property, and can then trigger the emergency evacuation procedure to ensure the safe evacuation of personnel.

[0060] Based on the results of multi-time series hierarchical detection, a fire warning signal is ultimately output. The warning signal can be in the form of an audible alarm, flashing lights, SMS notification, email notification, etc., to ensure that relevant personnel receive the warning in time and take appropriate measures.

[0061] Through the above steps, the embodiment of the present application can perform multi-time series hierarchical detection based on collected smoke and gas samples, thereby more accurately assessing the likelihood and severity of a fire and promptly outputting corresponding fire warning signals. This approach not only improves the accuracy and reliability of fire detection, but also enhances the system's adaptability to complex environments, providing strong support for fire prevention and response.

[0062] Furthermore, when the monitoring area includes multiple ventilation openings, the embodiment of the present application further includes step P50, which further includes:

[0063] P51: When there are multiple ventilation holes in the monitoring area, air-suction smoke detectors are respectively arranged on the multiple ventilation holes, the multiple air-suction smoke detectors are connected for NB-IOT communication, and the multiple air-suction smoke detectors are distributedly controlled based on the integrated processor; P52: Multiple matching adaptive strategies corresponding to the multiple air-suction smoke detectors are obtained, and multiple smoke gas samples are obtained according to the multiple matching adaptive strategies; P53: Multi-time series graded detection is performed based on the multiple smoke gas samples, and the fire warning signal is updated.

[0064] It should be understood that in the embodiments of the present application, the layout and control of the smoke detection equipment can be further expanded and optimized to ensure that smoke collection and fire warning can be effectively carried out within the monitoring areas of multiple ventilation openings.

[0065] First, if there are more than one ventilation opening within the monitoring area, an aspirating smoke detector can be deployed at each vent. These devices are connected via NB-IoT communications, ensuring real-time synchronization and interoperability among all smoke detectors. NB-IoT (Narrowband Internet of Things) is a low-power wide-area network technology that ensures stable communication over a wide area, making it suitable for remote data transmission and control. All aspirating smoke detectors are controlled in a distributed manner via an integrated processor. The integrated processor coordinates the operation of multiple devices, ensuring they work together according to environmental conditions and operating modes. This distributed control allows each smoke detector to adapt to the specific environmental conditions of the vent, achieving comprehensive and efficient fire detection.

[0066] Next, for each aspirating smoke detector at each vent, an adaptive strategy is generated based on its operating mode parameters (ventilation on / off status, wind speed, and ventilation direction). These adaptive strategies include operating parameters such as the pump speed and sampling period, ensuring that each device can accurately collect smoke gas samples under different ventilation conditions. In this way, each smoke detector can independently and efficiently perform its smoke collection task based on its specific environment.

[0067] Next, multi-time series detection is performed based on smoke gas samples collected by multiple aspirating smoke detectors. This process is similar to the single vent scenario described above, but it incorporates data from multiple vents. By performing time series analysis on smoke concentration data collected from different vents, a more comprehensive understanding of smoke diffusion within the entire monitoring area can be obtained.

[0068] Furthermore, fire warning signals are updated based on the results of multi-time series hierarchical detection. For example, if smoke data from a majority of ventilation openings indicates an increased fire risk, the system will promptly adjust the warning level and issue a higher-level fire warning signal. This distributed monitoring and control approach ensures that even in complex environments where monitoring data from some ventilation openings is disrupted, the entire system can still accurately assess fire risk and issue timely warnings.

[0069] This multi-vent layout and coordinated control not only improves the sensitivity and accuracy of fire detection in the monitoring area, but also provides efficient early warning response in more complex environments, ensuring that fire hazards are discovered in time and appropriate safety measures are taken.

[0070] Furthermore, step P53 of the embodiment of the present application further includes:

[0071] P53-1: Obtain the spatial position distribution of the multiple ventilation openings; P53-2: Perform multi-time series hierarchical detection on the multiple smoke gas samples, obtain the time series smoke concentration changes at each sampling point, and construct multiple smoke concentration distribution maps; P53-3: Use an interpolation algorithm to analyze the multiple smoke concentration distribution maps, and establish a corresponding fitted smoke concentration distribution map based on the spatial position distribution; P53-4: Identify the fitted smoke concentration distribution map to obtain fire location information, and update the fire warning signal according to the fire location information.

[0072] Specifically, when the monitoring area contains multiple ventilation openings, it is possible to further refine how to perform multi-time series graded detection based on smoke gas samples from multiple ventilation openings, and ultimately update the fire warning signal.

[0073] First, determine the spatial distribution of multiple ventilation openings. This information can be obtained from building structural drawings or ventilation system design diagrams, or it can be determined through on-site measurements or sensor positioning. The spatial distribution of ventilation openings directly affects the smoke diffusion path. Therefore, accurately understanding the specific location of each ventilation opening is fundamental to smoke concentration distribution modeling, reflecting the smoke propagation path and speed within the building.

[0074] Next, multiple smoke gas samples collected from each ventilation outlet are subjected to multi-time series hierarchical detection to obtain the time-series smoke concentration changes at each sampling point. The smoke concentration changes at each sampling point reflect the smoke diffusion trend and changes at that location. By analyzing this time-series data, multiple smoke concentration distribution maps are constructed, each representing the smoke concentration changes at different time points or different ventilation outlet locations. These maps can reveal the spread of smoke within the monitoring area during a fire, helping to identify peak smoke concentration areas and diffusion paths.

[0075] Then, an interpolation algorithm is used to analyze multiple smoke concentration distribution maps and create a fitted smoke concentration distribution map based on the spatial location distribution. Interpolation algorithms (such as linear interpolation and spline interpolation) can be used to estimate smoke concentration at unknown locations between known sampling points. Through interpolation, a continuous smoke concentration distribution map covering the entire monitoring area can be obtained. This map can more intuitively visualize smoke distribution trends and help identify areas with significant concentration fluctuations.

[0076] Finally, the system analyzes the resulting fitted smoke concentration distribution map to identify fire location information. Based on the spatial distribution of smoke concentration, the system can accurately determine the fire's origin or concentrated area of ​​smoke concentration, thereby narrowing the fire's scope and providing guidance on firefighting and evacuation measures. Based on this fire location information, the system promptly updates fire warning signals, providing relevant personnel with accurate fire location data to ensure a fast and efficient response.

[0077] These steps not only enable comprehensive collection and analysis of smoke samples from multiple ventilation openings, but also enable precise fire location using spatial distribution factors. This approach improves the spatial resolution of fire detection, enabling effective tracking of fire sources within a large monitoring area, enhancing the accuracy of early warnings and the timeliness of responses.

[0078] Furthermore, step P53-4 of the embodiment of the present application also includes:

[0079] P53-41: Identify the peak concentration area of ​​the fitted smoke concentration distribution map, where the peak concentration area is a concentration area where the smoke concentration is greater than a preset concentration threshold; P53-42: Calculate the fitted smoke concentration distribution map to perform gradient calculation to obtain a concentration gradient vector field; P53-43: Analyze whether there is a positive gradient vector field in the peak concentration area in combination with the vector direction of the concentration gradient vector field. If there is a positive gradient vector field in the peak concentration area, mark it as a fire candidate area; P53-44: Output the fire location information of the first fire area in the fire candidate area.

[0080] Optionally, the process of identifying candidate fire areas by analyzing and fitting the smoke concentration distribution map can be further refined to accurately identify the fire source and locate the fire area.

[0081] After establishing a fitted smoke concentration distribution map, the first step is to identify peak concentration areas within the map. Peak concentration areas are regions where smoke concentration exceeds a preset threshold. This threshold, set based on historical data and experimental results, distinguishes between normal smoke concentration fluctuations and abnormally high concentrations caused by fires. By identifying peak concentration areas, we can initially determine locations where smoke concentration is abnormally concentrated, which may be the source of a fire.

[0082] Next, a gradient calculation is performed on the fitted smoke concentration distribution map to obtain a concentration gradient vector field. The concentration gradient vector field reflects the spatial trend of smoke concentration. The direction of each vector points to the direction of the fastest concentration increase, and the magnitude indicates the rate of concentration change. Specifically, the gradient calculation estimates the direction and rate of concentration change by analyzing the concentration changes near each point. In this way, the system can obtain a more intuitive smoke concentration change map, helping to identify areas with more drastic concentration changes.

[0083] Then, based on the vector direction of the concentration gradient vector field, we analyze whether the peak concentration region has a positive gradient vector field. A positive gradient vector field refers to an area where the overall gradient direction converges and increases toward a specific point. If a positive gradient vector field exists in a peak concentration region, it indicates that the smoke concentration in that area is increasing from the surrounding area toward that point, indicating that this area may be the source of the fire. Therefore, peak concentration regions with positive gradient vector fields are marked as candidate fire regions.

[0084] Finally, based on the above analysis, the fire location information for the first fire zone within the candidate fire zones is output. This first fire zone is the area within the candidate fire zones with the most pronounced concentration gradient and the highest smoke density. By accurately locating this area, including the specific location coordinates of the fire and its severity, the fire warning signal is updated, allowing personnel and the automatic fire extinguishing system to take prompt action.

[0085] Through the above steps, not only can the fire source be accurately identified based on the spatial distribution and changing trend of smoke concentration, but the accuracy of fire location can also be further improved through gradient analysis, and the system's adaptability to complex environments can be enhanced.

[0086] Furthermore, steps P53-44 of the embodiment of the present application also include:

[0087] P53-441: Obtain the concentration mean, gradient centrality and diffusion rate of the fire candidate area, perform fusion calculation on the concentration mean, gradient centrality and diffusion rate, and output the location reliability of each fire candidate area; P53-442: Output the fire location information of the first fire area in the fire candidate area with the location reliability.

[0088] In a possible embodiment of the present application, the evaluation of the candidate fire area can be further optimized, and the positioning confidence can be calculated to improve the positioning accuracy and reliability of the fire source.

[0089] After identifying the candidate fire areas, we first obtain the mean concentration, gradient centrality, and diffusion rate for each candidate fire area. The mean concentration represents the average level of smoke concentration within the candidate fire area, reflecting the intensity of smoke concentration in that area. The gradient centrality analyzes the concentration gradient vector field to determine the center point where the gradient vectors converge, reflecting the geometric center of the candidate fire area. The diffusion rate indicates how quickly smoke spreads within the candidate fire area, reflecting the activity of the fire.

[0090] Next, these characteristic parameters are fused and calculated to output the location confidence of each candidate fire area. For example, this fusion calculation can be implemented through weighted averaging, machine learning algorithms, or other comprehensive assessment methods. Location confidence is a comprehensive indicator that reflects the likelihood that each candidate fire area is the source of the fire. This fusion calculation allows for a more comprehensive assessment of the fire risk of each candidate fire area.

[0091] Finally, based on the calculated location confidence, the system outputs the fire location information for the area with the highest location confidence among the candidate fire areas, namely the primary fire area. This information guides the emergency response team in accurately locating the fire source and implementing appropriate firefighting or evacuation measures. By integrating multi-dimensional data such as smoke concentration, gradient analysis, and diffusion rate, the system provides highly accurate fire location results, enabling rapid response and effectively minimizing fire losses.

[0092] In summary, the embodiments of the present application have at least the following technical effects:

[0093] This application arranges suction-type smoke sensing equipment at the ventilation openings and dynamically adjusts the equipment operating parameters according to the working mode parameters of the ventilation openings to ensure that smoke samples can be accurately collected under different ventilation conditions, thereby improving the accuracy of fire detection; adopts multi-time series hierarchical detection technology to conduct short-term, medium-term and long-term analysis of the collected smoke gas samples, which can timely capture the signals at the early stage of fire, realize early fire warning, and buy precious time for personnel evacuation and fire fighting; by analyzing and fitting the smoke concentration distribution map and combining it with the concentration gradient vector field, the specific location of the fire can be accurately located, providing more accurate information for fire response; by dynamically adjusting the sampling frequency and equipment operating parameters, it can optimize resource utilization, reduce energy consumption and operating costs while ensuring detection effect; through distributed control and multi-time series hierarchical detection, not only the accuracy and reliability of fire detection are improved, but also the system's adaptability to complex environments is enhanced, providing strong technical support for fire prevention and response.

[0094] The technical effect of improving the accuracy of smoke collection and the real-time nature of fire warning is achieved by deploying suction-type smoke sensing equipment at ventilation openings and dynamically adjusting the sampling strategy.

[0095] The second embodiment is based on the same inventive concept as the hierarchical detection method of aspirating smoke fire in a complex environment in the above embodiment. Figure 2 As shown, the present application provides a hierarchical detection system for aspirating smoke fires in complex environments. The system and method embodiments in the present application are based on the same inventive concept. The system includes:

[0096] The smoke sensor device deployment module 11 is used to identify the ventilation holes in the monitoring area under complex environments, and to deploy suction-type smoke sensors on the ventilation holes. The suction-type smoke sensors include a micro vacuum pump, and the vacuum pump is controlled by the control unit through communication.

[0097] The state monitoring module 12 is used to monitor the state of the vents and obtain the working mode parameters of the vents. The working mode parameters include ventilation start and stop states, wind speed and ventilation direction.

[0098] The smoke gas sample collection module 13 is used to analyze the working mode parameters to obtain an adaptive strategy that matches the working mode parameters. The control unit controls the vacuum pump according to the matched adaptive strategy to drive the suction type smoke sensor to collect smoke gas samples.

[0099] The multi-time-sequence hierarchical detection module 14 is configured to perform multi-time-sequence hierarchical detection based on the smoke gas sample and output a fire warning signal.

[0100] Furthermore, the smoke detection equipment deployment module 11 is further configured to perform the following steps:

[0101] Analyze the spatial structure information of the current environment, wherein the spatial structure information includes the building structure distribution and the ventilation system distribution; collect aerodynamic characteristics based on the building structure distribution and the ventilation system distribution; simulate based on the aerodynamic characteristics and the preset smoke diffusion path, and output the smoke diffusion disturbance index; when the smoke diffusion disturbance index is greater than the preset disturbance index, identify the current environment as a complex environment, and identify the ventilation outlets in the monitoring area under the complex environment.

[0102] Furthermore, the smoke gas sample collection module 13 is further configured to perform the following steps:

[0103] A strategy mapping library containing multiple ventilation scenarios is set up, and the strategy mapping library includes a mapping relationship between each working mode parameter and an adaptive strategy for controlling the air pump; based on the KNN algorithm, the working mode parameters are input into the strategy mapping library for strategy matching to obtain an adaptive strategy matching the working mode parameters, wherein the adaptive strategy includes an air pump speed, sampling period, filtering strength and voltage stabilization time for controlling the air pump.

[0104] Furthermore, the smoke gas sample collection module 13 is further configured to perform the following steps:

[0105] The working mode parameters are input into the strategy mapping library, the weighted distances between the working mode parameters and the working mode parameters corresponding to each adaptive strategy under the working mode parameters are obtained, and a weighted distance set is output; the first K nearest neighbor adaptive strategies in the weighted distance set are selected based on the KNN algorithm; and the adaptive strategy matching the working mode parameters is output based on the K historical call probabilities corresponding to the K nearest neighbor adaptive strategies.

[0106] Furthermore, the smoke gas sample collection module 13 is further configured to perform the following steps:

[0107] According to the working mode parameters of the ventilation port, the smoke diffusion disturbance sub-index of the ventilation port is output; according to the smoke diffusion disturbance sub-index of the ventilation port, a gas sampling compensation coefficient is obtained, and the gas sampling compensation coefficient is used to compensate for the frequency of collecting the smoke gas sample.

[0108] Furthermore, the system further includes a distributed control module for performing the following steps:

[0109] When the monitoring area includes multiple ventilation openings, air-suction smoke detection devices are respectively arranged on the multiple ventilation openings, the multiple air-suction smoke detection devices are connected for NB-IOT communication, and the multiple air-suction smoke detection devices are distributedly controlled based on the integrated processor; multiple matching adaptive strategies corresponding to the multiple air-suction smoke detection devices are obtained, and multiple smoke gas samples are obtained according to the multiple matching adaptive strategies; multi-time series graded detection is performed based on the multiple smoke gas samples, and the fire warning signal is updated.

[0110] Furthermore, the distributed control module is further configured to perform the following steps:

[0111] Obtain the spatial position distribution of the multiple ventilation openings; perform multi-time series hierarchical detection on the multiple smoke gas samples, obtain the time series smoke concentration change of each sampling point, construct multiple smoke concentration distribution maps, use an interpolation algorithm to analyze the multiple smoke concentration distribution maps, and establish a corresponding fitted smoke concentration distribution map based on the spatial position distribution; identify the fitted smoke concentration distribution map to obtain fire location information, and update the fire warning signal according to the fire location information.

[0112] Furthermore, the distributed control module is further configured to perform the following steps:

[0113] Identify a peak concentration area of ​​the fitted smoke concentration distribution map, where the peak concentration area is a concentration area where the smoke concentration is greater than a preset concentration threshold; calculate the fitted smoke concentration distribution map to perform gradient calculation to obtain a concentration gradient vector field; analyze whether a positive gradient vector field exists in the peak concentration area based on the vector direction of the concentration gradient vector field; if a positive gradient vector field exists in the peak concentration area, mark it as a fire candidate area; and output fire location information of a first fire area in the fire candidate area.

[0114] Furthermore, the distributed control module is further configured to perform the following steps:

[0115] Obtain the concentration mean, gradient centrality, and diffusion rate of the fire candidate area, perform a fusion calculation on the concentration mean, gradient centrality, and diffusion rate, and output the location reliability of each fire candidate area; and output the fire location information of the first fire area in the fire candidate area based on the location reliability.

[0116] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0117] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

[0118] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A hierarchical detection method for aspirating smoke fires in complex environments, characterized in that: The method comprises: Identify the ventilation holes in the monitoring area under complex environments, and deploy an aspirating smoke sensor at the ventilation holes. The aspirating smoke sensor includes a micro air pump controlled by a control unit through communication. Monitor the status of the vent to obtain operating mode parameters of the vent, wherein the operating mode parameters include ventilation start / stop status, wind speed, and ventilation direction; Analyzing the working mode parameters to obtain an adaptive strategy that matches the working mode parameters, and the control unit controlling the air pump according to the matched adaptive strategy to drive the aspiration type smoke sensor to collect smoke gas samples; Perform multi-time series hierarchical detection based on the smoke gas samples and output a fire warning signal; wherein, obtaining the spatial position distribution of multiple ventilation openings; Perform multi-time series graded detection on multiple smoke gas samples, obtain the time series smoke concentration changes at each sampling point, and construct multiple smoke concentration distribution maps. Analyzing the plurality of smoke concentration distribution maps using an interpolation algorithm to establish a corresponding fitted smoke concentration distribution map based on the spatial position distribution; Identifying the fitted smoke concentration distribution map to obtain fire location information, and updating the fire warning signal according to the fire location information; Methods for identifying vents in a monitoring area under a complex environment and detecting whether the current environment is a complex environment include: Analyzing spatial structural information of the current environment, wherein the spatial structural information includes building structure distribution and ventilation system distribution; collecting aerodynamic characteristics based on the building structure distribution and ventilation system distribution, performing simulation based on the aerodynamic characteristics and a preset smoke diffusion path, and outputting a smoke diffusion disturbance index; When the smoke diffusion disturbance index is greater than a preset disturbance index, the current environment is identified as a complex environment, and the ventilation holes in the monitoring area under the complex environment are identified; When there are multiple ventilation openings in the monitoring area, air-suction smoke detectors are respectively arranged on the multiple ventilation openings, the multiple air-suction smoke detectors are connected by NB-IOT communication, and distributed control of the multiple air-suction smoke detectors is performed based on an integrated processor; Obtaining a plurality of matching adaptive strategies corresponding to the plurality of aspiration-type smoke detectors, and obtaining a plurality of smoke gas samples according to the plurality of matching adaptive strategies; Perform multi-time series hierarchical detection based on the multiple smoke gas samples and update the fire warning signal; Identifying the fitted smoke concentration distribution map to obtain fire location information, the method includes: Identifying a peak concentration region of the fitted smoke concentration distribution map, wherein the peak concentration region is a concentration region where the smoke concentration is greater than a preset concentration threshold; Performing gradient calculation on the fitted smoke concentration distribution map to obtain a concentration gradient vector field; Analyzing whether there is a positive gradient vector field in the peak concentration region in combination with the vector direction of the concentration gradient vector field, and if there is a positive gradient vector field in the peak concentration region, marking it as a candidate fire region; outputting fire location information of a first fire area in the fire candidate area; Analyzing the operating mode parameters to obtain an adaptive strategy that matches the operating mode parameters, the method comprising: Setting a strategy mapping library containing multiple ventilation scenarios, wherein the strategy mapping library includes a mapping relationship between each working mode parameter and an adaptive strategy for controlling the air extraction pump; Based on the KNN algorithm, the working mode parameters are input into the strategy mapping library for strategy matching to obtain an adaptive strategy that matches the working mode parameters, wherein the adaptive strategy includes a pump speed, a sampling period, a filtering strength, and a voltage stabilization time for controlling the pump.

2. The method according to claim 1, wherein Outputting fire location information of a first fire area in the fire candidate area, the method comprising: Obtaining the concentration mean, gradient centrality, and diffusion rate of the candidate fire area, performing a fusion calculation on the concentration mean, gradient centrality, and diffusion rate, and outputting the location reliability of each candidate fire area; Fire location information of a first fire area in the fire candidate areas is outputted based on the location reliability.

3. The method according to claim 1, wherein Inputting the working mode parameters into the strategy mapping library for strategy matching based on the KNN algorithm, the method comprising: Inputting the working mode parameters into the strategy mapping library, obtaining weighted distances between the working mode parameters and the working mode parameters corresponding to each adaptive strategy under the working mode parameters, and outputting a weighted distance set; An adaptive strategy of selecting the first K nearest neighbors in the weighted distance set based on the KNN algorithm; According to the K historical call probabilities corresponding to the K nearest neighbor adaptive strategies, an adaptive strategy matching the working mode parameters is output.

4. The method according to claim 1, wherein Driving the aspirating smoke sensor to collect smoke gas samples, the method further includes: outputting a smoke diffusion disturbance sub-index of the ventilation opening according to the working mode parameters of the ventilation opening; According to the smoke diffusion disturbance sub-index of the ventilation port, a gas sampling compensation coefficient is obtained, and the gas sampling compensation coefficient is used to compensate for the frequency of collecting the smoke gas sample.

5. A hierarchical detection system for aspirating smoke fires in complex environments, characterized by: The system is used to perform the method according to any one of claims 1 to 4, and the system comprises: A smoke sensor deployment module, which is used to identify ventilation openings in a monitoring area under complex environments and deploy a suction-type smoke sensor on the ventilation openings. The suction-type smoke sensor includes a micro-vacuum pump, which is controlled by a control unit through communication. A status monitoring module, the status monitoring module is used to monitor the status of the vents and obtain working mode parameters of the vents, the working mode parameters including ventilation start and stop status, wind speed and ventilation direction; a smoke gas sample collection module, the smoke gas sample collection module being configured to analyze the operating mode parameters to obtain an adaptive strategy matching the operating mode parameters, the control unit controlling the air pump according to the matching adaptive strategy to drive the aspiration-type smoke sensor to collect smoke gas samples; A multi-time sequence hierarchical detection module is used to perform multi-time sequence hierarchical detection based on the smoke gas sample and output a fire warning signal.

Images