A high-rise building intelligent fire linkage monitoring system

Abstract

The application relates to the technical field of fire safety, in particular to a high-rise building intelligent fire-fighting linkage monitoring system, which comprises an industrial network communication framework, a fire perception layer, a device state acquisition layer, an edge computing node, a general protocol adaptive gateway, a linkage control unit and a dynamic decision server; the fire perception layer is composed of smoke sensors, temperature sensors, flame sensors, combustible gas detectors and wind pressure sensors, smoke concentration sensors, air flow rate sensors and water pressure sensors which are arranged in the horizontal areas of each floor and the vertical shafts; the three-dimensional perception network covering the horizontal areas of each floor and various vertical shafts is constructed, fire detection data and environmental correlation parameters are simultaneously collected, multi-source data fusion and false alarm filtering are completed at the front end, real fire and interference factors such as air flow disturbance can be effectively distinguished, and the false alarm rate caused by the chimney effect and the change of ventilation working conditions of the high-rise building is reduced.

Description

Technical Field

[0001] This invention relates to the field of fire safety technology, specifically to a smart fire-fighting linkage monitoring system for high-rise buildings. Background Technology

[0002] High-rise buildings exhibit a chimney effect, causing rapid smoke and fire spread and posing significant challenges to fire prevention and control. Traditional fire suppression systems rely solely on single smoke and heat detectors, resulting in insufficient sensing capabilities and susceptibility to environmental interference, leading to false alarms and missed alarms. Furthermore, these systems cannot provide comprehensive monitoring of fire equipment status, leaving significant blind spots in concealed areas of the building.

[0003] Building fire protection systems and security systems are currently independent, with incompatible communication protocols and inability to interconnect data. After a fire breaks out, only individual fire-fighting equipment can be activated, failing to coordinate with security equipment to assist emergency evacuation. This lack of a unified, interconnected system also leads to resource waste and can easily cause panic among residents. According to authorization announcement number CN105530455A, a fire-fighting linkage video monitoring system is disclosed, consisting of fire detectors, area lighting modules, a central processor, and monitoring cameras. Its key feature is that the fire detectors, area lighting modules, and monitoring cameras are all connected to the central processor. The beneficial effect of this invention is that when a fire occurs in a building, the system can automatically activate the lighting equipment in the fire area and simultaneously automatically locate the video cameras in the fire area, providing valuable fire information to night-duty personnel in a timely manner.

[0004] Currently, traditional fire alarm linkage systems use fixed preset control logic, which cannot adapt to dynamic changes in working conditions such as fire, outdoor wind, and building ventilation. Fixed detection thresholds and equipment linkage timing make it difficult to cope with complex fire environments, and the alarm anti-interference capability is weak, resulting in significant delays in emergency response. Summary of the Invention

[0005] The purpose of this invention is to provide a smart fire protection linkage monitoring system for high-rise buildings to solve the problems mentioned in the background art.

[0006] To address the aforementioned issues, this invention provides a technical solution: a smart fire protection linkage monitoring system for high-rise buildings, comprising an industrial-grade network communication architecture, a fire perception layer, an equipment status acquisition layer, edge computing nodes, a general protocol adaptive gateway, a linkage control unit, and a dynamic decision server.

[0007] The fire sensing layer consists of smoke detectors, heat detectors, flame detectors, combustible gas detectors, and sensors for wind pressure, smoke concentration, air velocity, and water pressure, which are installed in the horizontal areas of each floor and in each vertical shaft.

[0008] The equipment status acquisition layer is connected to the status output terminals of fire hydrants, sprinkler systems, emergency power supplies, and fire doors;

[0009] The edge computing node receives data from the fire awareness layer and the device status acquisition layer, performs multi-source data fusion, and outputs the fire alarm confidence level based on the comparison between the fusion result and the dynamic threshold.

[0010] The universal protocol adaptive gateway is compatible with Modbus, TCP / IP, RS485 and proprietary security protocols, and connects to video surveillance, access control, elevators, broadcasting, smoke exhaust fans, turnstiles and audible and visual alarms;

[0011] The linkage control unit generates zone-specific control commands based on the fire alarm confidence level and the location of the fire.

[0012] The dynamic decision server performs fire evolution prediction and outputs dynamic thresholds and linkage strategy adjustment parameters, which are then fed back to the edge computing node and linkage control unit.

[0013] Preferably, the edge computing node performs multi-source data fusion calculations in the following manner:

[0014]

[0015] in, The comprehensive fire situation value for the i-th floor and j-th zone is... For the reading of the k-th type detector, The mean and standard deviation of the historical baseline. This represents the wind pressure value for that zone. These are the weighting coefficients.

[0016] Preferably, the edge computing node construction time window Calculate the variance of the detector reading sequence within the range. and rate of change:

[0017]

[0018] in, and At that time, the current fusion result is marked as noise interference and the alarm output is suppressed;

[0019] when and At the same time, increase the fusion weighting coefficient of the corresponding detector. .

[0020] Preferably, the equipment status acquisition layer acquires the hydrant pressure value and the number of start-stop cycles of the pressure stabilizing pump for fire hydrants, the end test water pressure and water flow indicator action signal for sprinkler systems, the battery voltage and backup power switching time for emergency power supplies, and the door magnetic opening and closing status and door closer feedback signal for fire doors.

[0021] Preferably, the universal protocol adaptive gateway includes a protocol parsing engine. The protocol parsing engine has a built-in data frame format library and address mapping table. It sequentially performs frame header recognition, checksum verification, data field extraction, and register address conversion on the received heterogeneous protocol data frames, and outputs standard data packets in a unified format.

[0022] Preferably, the linkage control unit is based on the floor where the fire occurred. And the fire alarm confidence level L, control commands are generated according to the following partitioning strategy: when L reaches a high level, for the first The first layer activates the sprinkler system and fireproof roller shutters to protect the second layer. The floor activates its smoke exhaust fan and closes the fire doors of the vertical pipe shaft, while issuing audible and visual warnings to other floors.

[0023] Preferably, the linkage control unit also synchronously generates security linkage commands: controlling the first Layer and first The layer video camera turned to the coordinates of the fire and started capturing and recording images, controlling the first... The access control system on floors 1 and below unlocks the access locks in the direction of the escape route, controls the elevator system to descend to the first floor and stop responding to external calls, and controls the public address system to broadcast differentiated evacuation messages according to each floor.

[0024] Preferably, the fire evolution prediction performed by the dynamic decision server includes longitudinal smoke diffusion prediction, and the predicted longitudinal diffusion rate is:

[0025]

[0026] in, This represents the difference in smoke concentration between the upper and lower floors. For the difference in floor height, This represents the average air velocity in each vertical shaft. The fitting coefficients are denoted as .

[0027] Preferably, the dynamic decision server is based on the Predict the time it will take for the flue gas to reach the upper floors:

[0028]

[0029] And accordingly in advance The time generates pre-action instructions, which include starting the upper-level smoke exhaust fan in advance and closing the upper-level vertical pipe shaft fire door in advance, wherein:

[0030]

[0031] in, This is a safety advance factor.

[0032] Preferably, the dynamic threshold output by the dynamic decision server is updated in the following way:

[0033]

[0034] in, For dynamic thresholds, As the baseline threshold, The average wind speed in the building environment. This represents the maximum overall fire situation value across all zones. This is the adjustment coefficient.

[0035] The beneficial effects of this invention are reflected in:

[0036] 1. This invention constructs a three-dimensional sensing network covering the horizontal areas of each floor and various vertical shafts, simultaneously collecting fire detection data and environmental parameters, and completing multi-source data fusion and false alarm filtering at the front end. This effectively distinguishes between real fires and interference factors such as airflow disturbances, thereby reducing the false alarm rate in high-rise buildings caused by the chimney effect and changes in ventilation conditions, and solving the problem of insufficient sensing reliability of traditional single-point detection methods in complex environments.

[0037] 2. This invention deploys a universal protocol adaptive gateway to establish communication protocols between fire-fighting equipment and devices in different systems such as video surveillance, access control, elevators, broadcasting, and perimeter security. It can automatically trigger zone-specific linkage strategies based on the specific floor location of the fire, and perform different fire-fighting and security actions on the fire floor, adjacent floors, and other floors, thereby avoiding the waste of equipment resources and panic caused by the indiscriminate linkage of the entire domain in traditional systems.

[0038] 3. This invention introduces a dynamic fire evolution prediction mechanism to simulate the diffusion path of smoke in vertical shafts and horizontal floors in real time, and dynamically adjusts the detection threshold and linkage execution sequence in combination with environmental conditions, so as to activate smoke exhaust and fire isolation equipment in advance before the fire spreads to the upper floors. Attached Figure Description

[0039] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0040] Figure 1 This is a schematic diagram of the multi-source data fusion and false alarm filtering process of the present invention;

[0041] Figure 2 This is a schematic diagram of the cross-system partition differentiation linkage process of the present invention;

[0042] Figure 3 This is a schematic diagram of the dynamic threshold adaptation and pre-action instruction flow of the present invention;

[0043] Figure 4 This is a schematic diagram illustrating the overall coupling relationship of the three processes in this invention. Detailed Implementation

[0044] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are merely illustrative of the technical solution of the present invention and are therefore intended to limit the scope of protection of the present invention.

[0045] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0046] Example 1

[0047] like Figures 1-4 As shown, a smart fire protection linkage monitoring system for high-rise buildings includes an industrial-grade network communication architecture, a fire perception layer, an equipment status acquisition layer, an edge computing node, a general protocol adaptive gateway, a linkage control unit, and a dynamic decision server. The industrial-grade network communication architecture consists of an industrial Ethernet switch, a fiber optic ring network, and a TCP / IP protocol stack, which are used to achieve highly reliable and low-latency data communication between the various components of the system.

[0048] The fire sensing layer consists of smoke detectors, heat detectors, flame detectors, combustible gas detectors, and sensors for wind pressure, smoke concentration, air velocity, and water pressure, which are installed in the horizontal areas of each floor and in each vertical shaft.

[0049] The equipment status acquisition layer connects to the status output terminals of fire hydrants, sprinkler systems, emergency power supplies, and fire doors;

[0050] Edge computing nodes receive data from the fire perception layer and the equipment status acquisition layer, perform multi-source data fusion, and output the fire alarm confidence level based on the comparison between the fusion result and the dynamic threshold.

[0051] The universal protocol adaptive gateway is compatible with Modbus, TCP / IP, RS485 and proprietary security protocols, and can connect to video surveillance, access control, elevators, broadcasting, smoke exhaust fans, turnstiles and audible and visual alarms;

[0052] The linkage control unit generates zone-specific control commands based on the fire alarm confidence level and the location of the fire;

[0053] The dynamic decision server performs fire evolution prediction and outputs dynamic thresholds and linkage strategy adjustment parameters, which are then fed back to edge computing nodes and linkage control units. The linkage control unit has a built-in command priority arbitration module, which determines the priority of commands when the same device is in a unit of time. When multiple control commands are received from different triggering conditions, they are executed in the following order of priority from highest to lowest: high-level fire alarm linkage command, medium-level fire alarm linkage command, manual remote intervention command, periodic self-test command, and system default state retention command. The time window for determining instruction conflicts is set to 500ms or less. ≤2000ms, the default priority value is 1000ms.

[0054] Edge computing nodes receive data from the fire perception layer and equipment status acquisition layer via RS485 bus or industrial Ethernet. Edge computing nodes and dynamic decision servers establish TCP / IP connections through an industrial-grade network communication architecture.

[0055] The linkage control unit has a built-in command priority arbitration module, which determines the priority of commands when the same device operates within a unit of time. When multiple control commands are received from different triggering conditions, they are executed in the following order of priority from high to low: high-level fire alarm linkage command, medium-level fire alarm linkage command, manual remote intervention command, periodic self-test command, and system default state retention command.

[0056] Furthermore, the edge computing node performs multi-source data fusion computation in the following way:

[0057]

[0058] in, The comprehensive fire situation value for the i-th floor and j-th zone is... For the reading of the k-th type detector, The mean and standard deviation of the historical baseline. This represents the wind pressure value for that zone. These are the weighting coefficients;

[0059] Before performing multi-source data fusion, the edge computing node reads data from each detector. Perform validity verification when Beyond the range of the detector Or remain unchanged in duration When the detector fails, it is marked as faulty and removed from the fusion calculation, and the reading from the previous valid time of the detector is used. Replace the current value in the fusion calculation.

[0060] Furthermore, the time window for edge computing node construction Calculate the variance of the detector reading sequence within the range. and rate of change:

[0061]

[0062] in, and At that time, the current fusion result is marked as noise interference and the alarm output is suppressed;

[0063] when and At the same time, increase the fusion weighting coefficient of the corresponding detector. .

[0064] Fire alarm confidence level classification rules:

[0065] Edge computing nodes according to With dynamic threshold The percentage difference outputs the fire alarm confidence level L, and the judgment rule is as follows:

[0066] when When L=0, the fire alarm is not triggered;

[0067] when When L=1, a primary warning is output;

[0068] When a≤ When L ≤ b, L = 2, and output a medium-level fire alarm;

[0069] when When b ≥ 3, L = 3, and an advanced fire alarm is output;

[0070] Where a and b are preset grading thresholds, 0 < a < b.

[0071] Furthermore, the equipment status acquisition layer collects the pressure value at the hydrant outlet and the number of start-stop cycles of the pressure stabilizing pump for fire hydrants, the end-point test water pressure and water flow indicator action signals for sprinkler systems, the battery voltage and backup power switching time for emergency power supplies, and the door magnetic opening and closing status and door closer feedback signals for fire doors.

[0072] Furthermore, the universal protocol adaptive gateway includes a protocol parsing engine. The protocol parsing engine has a built-in data frame format library and address mapping table. It sequentially performs frame header recognition, checksum verification, data field extraction, and register address conversion on the received heterogeneous protocol data frames, and outputs standard data packets in a unified format.

[0073] Furthermore, the linkage control unit adjusts the system based on the floor where the fire occurred. And the fire alarm confidence level L, control commands are generated according to the following partitioning strategy: when L reaches a high level, for the first The first layer activates the sprinkler system and fireproof roller shutters to protect the second layer. The floor activates its smoke exhaust fan and closes the fire doors of the vertical pipe shaft, while issuing audible and visual warnings to other floors.

[0074] Vertical shafts include cable shafts, pipe shafts, stairwells, and elevator shafts. Each vertical shaft is independently equipped with a smoke concentration sensor and an air velocity sensor.

[0075] Furthermore, the linkage control unit also synchronously generates security linkage commands: controlling the first Layer and first The layer video camera turned to the coordinates of the fire and started capturing and recording images, controlling the first... The access control system on floors 1 and below unlocks the access locks in the direction of the escape route, controls the elevator system to descend to the first floor and stop responding to external calls, and controls the public address system to broadcast differentiated evacuation messages according to each floor.

[0076] Furthermore, the fire evolution prediction performed by the dynamic decision server includes longitudinal smoke diffusion prediction, and its predicted longitudinal diffusion rate is:

[0077]

[0078] in, This represents the difference in smoke concentration between the upper and lower floors. For the difference in floor height, This represents the average air velocity in each vertical shaft. The fitting coefficients are denoted as .

[0079] Furthermore, the dynamic decision server, based on Predict the time it will take for the flue gas to reach the upper floors:

[0080]

[0081] And accordingly in advance The time-based pre-action instructions include early activation of the upper-level smoke exhaust fan and early closure of the upper-level vertical pipe shaft fire doors, wherein:

[0082]

[0083] in, The safety advance factor has a value range of 0.3 ≤ ≤0.6.

[0084] Furthermore, the dynamic threshold output by the dynamic decision server is updated in the following way:

[0085]

[0086] in, For dynamic thresholds, As the baseline threshold, The average wind speed in the building environment. This represents the maximum overall fire situation value across all zones. This is the adjustment coefficient.

[0087] Dynamic decision server in periodic The updated Data is distributed to each edge computing node via an industrial-grade network communication architecture, and each edge computing node uses... Recalculate for period And reassess the confidence level of the fire alarm, among which .

[0088] Example 2

[0089] This example illustrates the high false alarm rate of a single smoke detector due to the chimney effect in the vertical shaft of a high-rise building and the large temperature difference between floors. Specifically, it describes the situation near the weak current shaft on the 12th floor of a high-rise office building. In the summer afternoon, the ambient temperature rises and the air conditioning fresh air system causes disturbances, resulting in fluctuations in the readings of a single smoke detector.

[0090] Initial data acquisition: 12th floor, 2nd zone, i=12, j=2, smoke detector readings. , temperature sensing reading Combined with historical baselines =0.02%obs / m =0.005%obs / m =25℃, =2℃;

[0091] Fusion computing, substitution formula:

[0092]

[0093] Calculate Below the dynamic threshold This will not trigger a fire alarm.

[0094] False Alarm Filtering Validation: Constructing a Time Window Inside Sequence, calculation and The system determined the interference to be airflow disturbance, suppressed false alarms, and verified the effectiveness of the detector reading sequence technology.

[0095] Comparison of real fire trigger scenarios: When a real fire occurs at the same location, , Simultaneously, it rose sharply. and The system automatically improves Weight, Exceed Output L=3 Advanced Fire Alarm.

[0096] Example 3

[0097] This example illustrates how traditional fire protection and security systems operate independently, with single-linkage and indiscriminate triggering across the entire area, leading to resource waste and public panic. Specifically, a fire broke out in the kitchen of the catering area on the 8th floor of Building B in a commercial and residential complex.

[0098] Fire location: The edge computing node outputs the fire location as the 1st partition of the 8th layer. =8, confidence level L=3;

[0099] Execute on partition:

[0100] The linkage control unit generates differentiated instructions according to the zoning strategy:

[0101] 8th floor on fire: Activate the sprinkler system, lower the fireproof roller shutter to 1.5m above the ground, and continuously sound the audible and visual alarm;

[0102] The 7th and 9th floors adjacent to each other: turn on the smoke exhaust fan, close the fire door of the vertical pipe shaft, and the audible and visual alarm will sound intermittently;

[0103] Other floors: Only sound and light warnings will be activated, sounding intermittently at a lower frequency than adjacent floors;

[0104] Cross-system security collaboration:

[0105] Video surveillance: The cameras on the 7th, 8th and 9th floors automatically turn to the coordinates of the fire and start high-definition capture, 2 frames per second;

[0106] Access control system: All access locks to evacuation stairwells on the 8th floor and below are unlocked and remain unlocked until the fire alarm is lifted;

[0107] Elevator system: All elevator cars are forced to land on the ground floor, the hall doors remain open, and external call responses are stopped;

[0108] Public Broadcasting:

[0109] The message playing on the 8th floor reads: "Fire alarm has been activated. Please evacuate immediately and orderly via the nearest evacuation staircase."

[0110] Playing on the 7th and 9th floors: Attention, a fire alarm has been triggered on an adjacent floor. Please remain vigilant and prepare to evacuate.

[0111] Other floors will hear: Fire warning has been issued inside the building. Please remain calm and pay attention to subsequent announcements.

[0112] Verify the granularity of control commands: the access control unlocking range is limited to the 8th floor and below, that is, the access control on the fire floor and above remains locked to prevent personnel from accidentally entering dangerous areas, and the broadcast voice is differentiated according to the floor level.

[0113] Example 4

[0114] This example illustrates a linkage strategy that traditional fixed logic cannot dynamically adjust based on the fire spread speed and high-rise wind pressure. Specifically, it describes a 28-story high-rise hotel where a fire broke out in a guest room on the 15th floor due to an electrical short circuit. The weather conditions that day were strong northerly winds, with a wind speed of 12 m / s according to the weather station data.

[0115] Initial state and threshold adjustment: Average ambient wind speed under strong wind conditions =12m / s, substituting this into the formula for updating the dynamic threshold:

[0116]

[0117] The value is set higher than the baseline to avoid false alarms caused by crosswinds blowing into the floors and interfering with the smoke detector's airflow.

[0118] Longitudinal smoke diffusion prediction: Smoke concentration difference between the 15th and 16th floors after a fire on the 15th floor. =0.15%obs / m, floor height difference =4.2m average air velocity in each vertical shaft =1.8 m / s, substituting into the formula for predicting the longitudinal diffusion velocity:

[0119]

[0120] Pre-action instruction generation:

[0121] Time it takes for the smoke to reach the 16th floor:

[0122]

[0123] Take the safety advance factor =0.4:

[0124]

[0125] The system ignites approximately 3.4 seconds after ignition. 2.27 seconds in advance, generate pre-action instructions: start the smoke exhaust fan on the 16th floor in advance, and close the fire door of the vertical pipe shaft between the 15th and 16th floors in advance;

[0126] Dynamic closed-loop feedback: After the pre-action command is executed, the actual measured rate of change of flue gas concentration on the 16th floor. If the value is lower than the predicted value, the system determines that the pre-action is effective and will... and The fitting coefficients are fine-tuned and used as optimization parameters for subsequent predictions.

[0127] Example 5

[0128] Based on Examples 1 to 4, the workflow in specific operations is as follows:

[0129] 1. After the system is powered on, smoke, heat, flame, and combustible gas detectors, as well as wind pressure, smoke concentration, air velocity, and water pressure sensors deployed in the ceilings, shaft walls, and equipment rooms of each floor, continuously collect field signals according to the set sampling period. The collected analog quantities are converted from analog to digital and then aggregated to the edge computing nodes deployed in the corresponding partitions via RS485 bus or industrial Ethernet. After receiving the raw readings uploaded by each detector, the edge computing nodes first perform data validity verification, discarding faulty channel data that exceeds the range or remains unchanged for too long. Then, they use the built-in multi-source fusion model to analyze the data from each detector. The effective data is comprehensively calculated to obtain the current comprehensive fire situation value of the zone. This value is then compared with the dynamic threshold periodically issued by the dynamic decision server. Based on the proportion of the magnitude exceeding the threshold, the current fire situation of the zone is divided into several confidence levels. At the same time, the fusion result is verified a second time based on the variance and instantaneous change rate of the detector readings within the time window. Signals that meet the interference characteristics are marked as noise and filtered out. Finally, only the effective fire alarm signals and their corresponding floor zone numbers and confidence levels are packaged and uploaded to the linkage control unit and the dynamic decision server to complete the front-end perception workflow from signal acquisition to fire confirmation.

[0130] 2. The linkage control unit continuously monitors the fire alarm data uploaded by each edge computing node. Once a valid fire alarm signal is received, along with its associated floor number and confidence level, it immediately reads the internally stored partition linkage strategy table. Based on the floor number, it determines the location of the fire floor and automatically delineates the control range for adjacent floors and other floors. Then, it determines whether to initiate the complete linkage process based on whether the confidence level reaches a preset high-level threshold. After confirmation, the linkage control unit generates multiple differentiated control commands through the universal protocol adaptive gateway and sends them to the corresponding devices: For the fire floor, the command activates the sprinkler system, lowers the fireproof roller shutter, and triggers a continuous audible and visual alarm; for the fire floor... Instructions to adjacent floors above and below the fire floor activate the smoke exhaust fans and close the fire doors of the vertical pipe shafts. Instructions to other floors only trigger intermittent audible and visual warnings. Simultaneously, the linkage control unit sends security coordination instructions in parallel to the video monitoring system through the same gateway to control the cameras on the fire floor and adjacent floors to rotate to preset positions and start capturing and recording. It also sends instructions to the access control system to unlock all access locks to the evacuation stairwells on the fire floor and below, to the elevator control system to force the elevator to land on the first floor and cut off the external call response, and to the public address system to play different evacuation voice messages according to the fire floor, adjacent floors, and other floors. This achieves cross-system synchronous and coordinated action of fire protection equipment and security equipment.

[0131] 3. During system operation, the dynamic decision server continuously receives comprehensive fire situation values, wind pressure and smoke concentration data for each zone, and building environment wind speed information uploaded by all edge computing nodes. In the absence of a fire, the server calculates the optimal alarm threshold in real time based on the changing trends of the current average environmental wind speed and the global maximum situation value, according to the dynamic threshold update model. The calculation results are periodically sent to each edge computing node to replace the original threshold, allowing the front-end sensing sensitivity to automatically adjust with environmental conditions. Once a fire is confirmed, the dynamic decision server immediately switches to fire evolution prediction mode. Based on the smoke concentration difference between the fire floor and adjacent floors, the floor height difference, and the real-time measurement values ​​of air velocity in each vertical shaft, it calculates the longitudinal diffusion of smoke along the vertical shaft. The system predicts the speed of the smoke and calculates the estimated time for the smoke to reach the next floor. Then, it calculates the advance action amount according to the preset safety advance coefficient. Before the smoke actually arrives, the system automatically generates an advance action command and sends it to the linkage control unit, ordering it to start the upper-floor smoke exhaust fan in advance and close the fire door of the vertical pipe shaft in advance. After the advance action command is executed, the dynamic decision server continuously compares the deviation between the measured smoke concentration change on the upper floor and the predicted value. Based on the different directions and magnitudes of the deviation, it corrects the fitting coefficients in the diffusion prediction model online and uses the corrected parameters for the next round of prediction calculation. In this way, the timing accuracy of subsequent advance actions is continuously optimized during the development of a single fire until the fire alarm is lifted, completing the closed-loop operation of the entire process of perception, linkage, decision-making, and correction.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A smart fire protection linkage monitoring system for high-rise buildings, characterized in that: It includes an industrial-grade network communication architecture, a fire sensing layer, an equipment status acquisition layer, edge computing nodes, a general protocol adaptive gateway, a linkage control unit, and a dynamic decision server; The fire sensing layer consists of smoke detectors, heat detectors, flame detectors, combustible gas detectors, and sensors for wind pressure, smoke concentration, air velocity, and water pressure, which are installed in the horizontal areas of each floor and in each vertical shaft. The equipment status acquisition layer is connected to the status output terminals of fire hydrants, sprinkler systems, emergency power supplies, and fire doors; The edge computing node receives data from the fire awareness layer and the device status acquisition layer, performs multi-source data fusion, and outputs the fire alarm confidence level based on the comparison between the fusion result and the dynamic threshold. The universal protocol adaptive gateway is compatible with Modbus, TCP / IP, RS485 and proprietary security protocols, and connects to video surveillance, access control, elevators, broadcasting, smoke exhaust fans, turnstiles and audible and visual alarms; The linkage control unit generates zone-specific control commands based on the fire alarm confidence level and the location of the fire. The dynamic decision server performs fire evolution prediction and outputs dynamic thresholds and linkage strategy adjustment parameters, which are then fed back to the edge computing node and linkage control unit.

2. The intelligent fire protection linkage monitoring system for high-rise buildings according to claim 1, characterized in that: The edge computing node performs multi-source data fusion calculations in the following way: in, The comprehensive fire situation value for the i-th floor and j-th zone is... For the reading of the k-th type detector, The mean and standard deviation of the historical baseline. This represents the wind pressure value for that zone. These are the weighting coefficients; Historical baseline mean and standard deviation After the system completes installation and debugging, it enters a baseline learning period of no less than 7 days (default 168 hours). During the learning period, the system records the readings of each detector under normal operating conditions without fire at a sampling frequency of once per minute, and discards readings that significantly exceed the limit. After identifying outliers within the range, the arithmetic mean of the remaining valid data is calculated as... Calculate its sample standard deviation as During the learning period, if extreme weather or building ventilation system debugging or other atypical conditions are encountered, the system will automatically mark the corresponding time period data as non-steady-state samples and exclude them from the baseline calculation.

3. The intelligent fire protection linkage monitoring system for high-rise buildings according to claim 2, characterized in that: The edge computing node construction time window Calculate the variance of the detector reading sequence within the range. and rate of change: in, and At that time, the current fusion result is marked as noise interference and the alarm output is suppressed; when and At the same time, increase the fusion weighting coefficient of the corresponding detector. .

4. The intelligent fire protection linkage monitoring system for high-rise buildings according to claim 1, characterized in that: The equipment status acquisition layer collects the pressure value at the hydrant outlet and the number of start-stop cycles of the pressure stabilizing pump for fire hydrants, the end-point test water pressure and the action signal of the water flow indicator for sprinkler systems, the battery voltage and backup power switching time for emergency power supplies, and the door magnetic opening and closing status and door closer feedback signal for fire doors.

5. The intelligent fire protection linkage monitoring system for high-rise buildings according to claim 1, characterized in that: The general protocol adaptive gateway includes a protocol parsing engine. The protocol parsing engine has a built-in data frame format library and address mapping table. It sequentially performs frame header recognition, checksum verification, data field extraction, and register address conversion on the received heterogeneous protocol data frames, and outputs standard data packets in a unified format.

6. The intelligent fire protection linkage monitoring system for high-rise buildings according to claim 1, characterized in that: The linkage control unit is based on the floor where the fire occurred. And the fire alarm confidence level L, control commands are generated according to the following partitioning strategy: when L reaches a high level, for the first The first layer activates the sprinkler system and fireproof roller shutters to protect the second layer. The floor activates its smoke exhaust fan and closes the fire doors of the vertical pipe shaft, while issuing audible and visual warnings to other floors.

7. The intelligent fire protection linkage monitoring system for high-rise buildings according to claim 1, characterized in that: The linkage control unit also synchronously generates security linkage commands: controlling the first Layer and first The layer video camera turned to the coordinates of the fire and started capturing and recording images, controlling the first... The access control system on floors 1 and below unlocks the access locks in the direction of the escape route, controls the elevator system to descend to the first floor and stop responding to external calls, and controls the public address system to broadcast differentiated evacuation messages according to each floor.

8. The intelligent fire protection linkage monitoring system for high-rise buildings according to claim 1, characterized in that: The fire evolution prediction performed by the dynamic decision server includes longitudinal smoke diffusion prediction, and its predicted longitudinal diffusion velocity value is: in, This represents the difference in smoke concentration between the upper and lower floors. For the difference in floor height, This represents the average air velocity in each vertical shaft. The fitting coefficients are denoted as .

9. A smart fire protection linkage monitoring system for high-rise buildings according to claim 8, characterized in that: The dynamic decision server is based on the Predict the time it will take for the flue gas to reach the upper floors: And accordingly in advance The time generates pre-action instructions, which include starting the upper-level smoke exhaust fan in advance and closing the upper-level vertical pipe shaft fire door in advance, wherein: in, This is a safety advance factor.

10. A smart fire protection linkage monitoring system for high-rise buildings according to claim 1, characterized in that: The dynamic threshold output by the dynamic decision server is updated in the following way: in, For dynamic thresholds, As the baseline threshold, The average wind speed in the building environment. This represents the maximum overall fire situation value across all zones. This is the adjustment coefficient.

Images