Multi-modal fusion fireproof door autonomous decision control method and system

Abstract

The invention belongs to the technical field of fireproof door management and control, and particularly relates to a multi-modal fusion fireproof door autonomous decision control method and system, and the system comprises a multi-modal sensing module, a fire behavior recognition and judgment module, an autonomous decision control module, a fireproof door abnormity alarm module and a supervision terminal. Multi-modal data of the fireproof door and the surrounding environment of the fireproof door are collected through the multi-modal sensing module, the fire behavior recognition and judgment module carries out fusion analysis on the multi-modal data so as to recognize the fire behavior condition, and when the fire behavior is recognized, the fireproof door is controlled to be closed so as to prevent the fire behavior from spreading; real-time monitoring, intelligent analysis and autonomous control of the state of the fireproof door are achieved, the fire safety level of a building is effectively improved, comprehensive monitoring analysis and timely early warning are conducted on the abnormity of the fireproof door through the fireproof door abnormity alarm module, and therefore optimization measures such as inspection and maintenance are conducted on the corresponding fireproof door in time so that stable operation of the fireproof door can be guaranteed; and the building fire safety level is further ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of fire door control, and in particular to a multi-modal fusion fire door autonomous decision-making control method and system. Background Art

[0002] Fire doors refer to doors that can meet the requirements of fire resistance stability, integrity and thermal insulation within a certain period of time. They are mainly distributed between fire compartments, evacuation stairwells and vertical shafts. A Chinese invention patent with publication number CN114673425A discloses an automatic control system for fire doors. The technical solution of this invention automatically closes the fire door when a fire occurs, and automatically opens the fire door when it detects the approach of fire rescue personnel, which is beneficial for the passage of rescue personnel and improves the rescue efficiency.

[0003] However, in actual application, the above-mentioned technical solution mainly focuses on monitoring the fire situation and controlling the opening and closing of the fire door. It is not possible to comprehensively monitor and analyze the abnormal operation of the fire door and issue a timely warning while realizing the automatic control of the fire door. It is also difficult to timely inspect and repair the fire door and take other optimization measures to ensure the stable operation of the fire door, which is not conducive to improving the fire safety level of the building.

[0004] In view of the above technical defects, a solution is now proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a multi-modal fusion fire door autonomous decision-making control method and system, which solves the problem that the prior art is unable to comprehensively monitor and analyze the abnormal operation conditions of the fire door and timely warn while realizing automatic control of the fire door, and it is difficult to timely inspect and repair the fire door and other optimization measures to ensure the stable operation of the fire door, which is not conducive to improving the level of building fire safety.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A multi-modal fusion fire door autonomous decision-making control system, including a multi-modal perception module, a fire situation identification and judgment module, an autonomous decision-making control module, a fire door abnormal alarm module and a monitoring terminal; the multi-modal perception module integrates multiple sensors to collect multiple information of the fire door and its surrounding environment, and sends the collected multi-modal data to the fire situation identification and judgment module;

[0008] The fire identification and judgment module receives the multimodal data collected by the multimodal perception module, fuses and analyzes the multimodal data through deep learning algorithms and convolutional neural networks, identifies the fire situation based on the data, and sends the fire identification and judgment results to the autonomous decision-making control module and the supervision terminal;

[0009] When a fire is identified, the autonomous decision-making control module immediately issues a command to control the fire door to close to prevent the fire from spreading, and sends the control information to the supervision terminal; when the fire door abnormal alarm module receives the appearance alarm signal, collision alarm signal or performance test alarm signal, it generates corresponding abnormal alarm information and sends it to the supervision terminal, and the supervision terminal displays the abnormal alarm information and issues a corresponding warning.

[0010] Furthermore, the fire door abnormality alarm module is communicatively connected to the fire door appearance monitoring module, the fire door collision detection and evaluation module and the fire door test and analysis module. The fire door appearance monitoring module collects images of the appearance of the fire door, analyzes the fire door appearance image to generate an appearance alarm signal or an appearance qualified signal, and sends the appearance alarm signal or the appearance qualified signal to the supervision terminal;

[0011] The fire door collision detection and evaluation module monitors the fire door for collisions, analyzes the collision risk when the fire door is hit, and determines whether to generate a collision alarm signal based on this. When the collision alarm signal is generated, it is sent to the fire door abnormal alarm module; the fire door test and analysis module tests and analyzes the operating performance of the fire door, generates a performance test alarm signal or a performance test pass signal through analysis, and sends the performance test alarm signal or the performance test pass signal to the fire door test and analysis module.

[0012] Furthermore, the specific analysis process of the fire door appearance monitoring module is as follows:

[0013] Based on the appearance image of the fire door, cracks on the fire door are identified, the crack width and crack extension length of the corresponding crack are collected, and the crack width and crack extension length are numerically compared with the preset crack width threshold and the preset crack extension length threshold, respectively. If the crack width or the crack extension length exceeds the corresponding preset threshold, the corresponding crack is marked as a dangerous crack;

[0014] If there are dangerous cracks on the fire door, an appearance alarm signal is generated; if there are no dangerous cracks on the fire door, the concave-convex area on the fire door is captured based on the fire door appearance image, the area and concave-convex depth of the corresponding concave-convex area are numerically calculated to obtain the concave-convex coefficient, and the concave-convex coefficient is numerically compared with the preset concave-convex coefficient threshold. If the concave-convex coefficient exceeds the preset concave-convex coefficient threshold, the corresponding concave-convex area is marked as a dangerous area;

[0015] The number of dangerous areas on the fire door is obtained and marked as concave-convex danger values, and the concave-convex coefficients of all concave-convex areas on the fire door are averaged to obtain the concave-convex expression value, and the concave-convex coefficient with the largest value on the fire door is marked as the concave-convex table amplitude. The concave-convex monitoring value is obtained by numerically calculating the concave-convex danger value, the concave-convex expression value and the concave-convex table amplitude. The concave-convex monitoring value is numerically compared with the preset concave-convex monitoring threshold. If the concave-convex monitoring value exceeds the preset concave-convex monitoring threshold, an appearance alarm signal is generated; if the concave-convex monitoring value does not exceed the preset concave-convex monitoring threshold, an appearance qualified signal is generated.

[0016] Furthermore, the specific analysis process for analyzing the collision risk when the fire door is hit is as follows:

[0017] The moment when the fire door is hit is collected and marked as the first moment, and the moment when the collision ends is collected and marked as the second moment, and the interval between the first moment and the second moment is marked as the characteristic duration;

[0018] The maximum and average values ​​of the impact forces received by the fire door during the corresponding collision process are obtained and marked as force amplitude values ​​and force performance values. The collision risk coefficient is obtained by numerically calculating the characteristic duration, force amplitude value and force performance value. The collision risk coefficient is numerically compared with the preset collision risk coefficient threshold. If the collision risk coefficient exceeds the preset collision risk coefficient threshold, a collision alarm signal is generated.

[0019] Furthermore, the specific analysis process of the fire door test analysis module is as follows:

[0020] Carry out several driving tests on the fire door, collect the driving execution delay time of the driving motor in the fire door, and collect its running speed in the process of the driving motor opening or closing the fire door, calculate the difference between the running speed and the set standard speed and take the absolute value to obtain the driving speed detection value, mark the number of driving speed detection values ​​that exceed the preset driving speed detection threshold in the corresponding driving process as the driving abnormal value, and calculate the average of all driving speed detection values ​​in the corresponding driving process to obtain the driving deviation value;

[0021] The drive performance value is obtained by numerically calculating the drive execution delay time, drive abnormality value and drive deviation value, the drive evaluation value is obtained by averaging the drive performance values ​​of all currently performed drive tests, and the drive evaluation value is numerically compared with the preset drive evaluation threshold. If the drive evaluation value exceeds the preset drive evaluation threshold, a performance test alarm signal is generated.

[0022] Furthermore, if the driving evaluation value does not exceed the preset driving evaluation threshold, when the fire door is closed, pressure detection is performed through a plurality of pressure sensors installed at the contact position between the fire door and the door frame, and a plurality of groups of pressure data are collected accordingly, and the closing pressure detection value is obtained by averaging all the pressure data;

[0023] The closing pressure analysis value is obtained by calculating the average of all closing pressure detection values ​​when the current fire door is closed several times, and the proportion of the number of occurrences of the closing pressure detection value not exceeding the preset closing pressure detection threshold is marked as the closing abnormality value, and the closing pressure analysis value and the closing abnormality value are numerically compared with the preset closing pressure analysis threshold and the preset closing abnormality threshold respectively;

[0024] If the closing pressure analysis value does not exceed the preset closing pressure analysis threshold or the closing abnormal occupancy value exceeds the preset closing abnormal occupancy threshold, a performance test alarm signal is generated; if the closing pressure analysis value exceeds the preset closing pressure analysis threshold and the closing abnormal occupancy value does not exceed the preset closing abnormal occupancy threshold, a performance test qualified signal is generated.

[0025] Furthermore, the fire door abnormal alarm module is communicatively connected to the regional management evaluation module. The regional management evaluation module is used to set a supervision period, obtain all fire doors existing in the corresponding area, analyze the management status of all fire doors in the area during the supervision period, generate a management low obstruction signal or a management alarm signal through analysis, and send the management low obstruction signal or the management alarm signal to the supervision terminal. When the supervision terminal receives the management alarm signal, it issues a corresponding warning.

[0026] Furthermore, the specific analysis process of the regional management assessment module is as follows:

[0027] The total number of abnormal alarm information generated by all fire doors in the corresponding area during the supervision period is obtained and marked as an abnormal alarm value, and the abnormal alarm value is numerically compared with the preset abnormal alarm threshold. If the abnormal alarm value exceeds the preset abnormal alarm threshold, a management alarm signal is generated;

[0028] If the abnormal alarm value does not exceed the preset abnormal alarm threshold, the time when the corresponding abnormal alarm information is generated is marked as the obstacle time, and the time when the administrator goes to check the fire door is marked as the correction time, and the interval between the correction time and the obstacle time is marked as the error correction time;

[0029] All the fault correction times in the supervision period are obtained and their average is calculated to obtain the fault correction detection value, and the fault correction time is numerically compared with the preset fault correction time threshold. If the fault correction time exceeds the preset fault correction time threshold, the corresponding fault correction time is marked as the error correction time, and the number of error correction times in the supervision period is obtained and marked as the error correction detection value;

[0030] The district management evaluation value is obtained by numerically calculating the abnormal alarm value, the fault correction detection value and the error correction detection value, and the district management evaluation value is numerically compared with the preset district management evaluation threshold. If the district management evaluation value exceeds the preset district management evaluation threshold, a management alarm signal is generated; if the district management evaluation value does not exceed the preset district management evaluation threshold, a management low obstacle signal is generated.

[0031] Furthermore, the present invention also proposes a multi-modal fusion fire door autonomous decision-making control method, comprising the following steps:

[0032] Step 1: Collect various information about the fire door and its surrounding environment;

[0033] Step 2: Fusion and analysis of multimodal data to identify fire conditions;

[0034] Step 3: When a fire is detected, close the fire door to prevent the fire from spreading;

[0035] Step 4: Analyze the fire door abnormalities one by one and determine whether to generate abnormal alarm information;

[0036] Step 5: When abnormal alarm information is generated, the monitoring terminal issues a corresponding warning.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] 1. In the present invention, the multimodal data of the fire door and its surrounding environment are collected through the multimodal perception module, and the fire identification and judgment module performs fusion analysis on the multimodal data to identify the fire situation. When the fire situation is identified, the fire door is controlled to be closed to prevent the fire from spreading, thereby realizing real-time monitoring, intelligent analysis and autonomous control of the fire door status. The fire door abnormality alarm module performs comprehensive monitoring and analysis of the fire door abnormality and timely warns, and timely inspects and repairs the corresponding fire door to ensure the stable operation of the fire door, which is conducive to improving the building fire safety level;

[0039] 2. In the present invention, the management status of all fire doors in the area during the supervision period is analyzed through the area management evaluation module, and a management low-obstruction signal or a management alarm signal is generated through the analysis. When the management alarm signal is generated, the management personnel are reminded to strengthen the supervision of the fire doors in the corresponding area in the future, so as to ensure the stable operation of all fire doors in the corresponding area and significantly improve the regional fire safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0041] Figure 1 This is a system block diagram of Embodiment 1 of the present invention;

[0042] Figure 2 This is a system block diagram of Embodiment 2 of the present invention;

[0043] Figure 3 This is a flow chart of the method of Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Embodiment 1: Figure 1 As shown, a multi-modal fusion fire door autonomous decision-making control system proposed in the present invention includes a multi-modal perception module, a fire situation identification and judgment module, an autonomous decision-making control module, a fire door abnormal alarm module and a monitoring terminal;

[0046] The multimodal perception module integrates multiple sensors to collect multiple information about the fire door and its surrounding environment (including temperature sensors to detect changes in ambient temperature; smoke sensors to detect smoke concentration; infrared sensors to detect infrared radiation from human bodies or flames; sound sensors to capture abnormal sounds, such as alarms and explosions; and visual sensors, such as high-definition cameras, to monitor the scene in real time), and sends the collected multimodal data to the fire identification module;

[0047] The fire identification and judgment module receives the multimodal data collected by the multimodal perception module, integrates and analyzes the multimodal data through advanced technologies such as deep learning algorithms and convolutional neural networks, identifies the fire situation based on the data, and sends the fire identification and judgment results to the autonomous decision-making control module and the supervision terminal;

[0048] When a fire is identified, the autonomous decision-making control module immediately issues a command to control the fire door to close to prevent the fire from spreading, and sends the control information to the supervision terminal; in an emergency, the autonomous decision-making control module will also intelligently adjust the opening status of the fire door according to the evacuation situation of personnel to ensure that the evacuation channel is unobstructed; the present invention integrates multiple sensors and intelligent algorithms to achieve real-time monitoring, intelligent analysis and autonomous control of the fire door status, effectively improving the level of building fire safety.

[0049] When the fire door abnormal alarm module receives the appearance alarm signal, collision alarm signal or performance test alarm signal, it generates the corresponding abnormal alarm information and sends it to the supervision terminal. The supervision terminal displays the abnormal alarm information and issues a corresponding warning, thereby realizing comprehensive monitoring and analysis of fire door abnormalities and timely warning, so as to remind management personnel to take optimization measures such as inspection and maintenance of the corresponding fire doors, ensure the stable operation of the fire doors, and further ensure the level of building fire safety.

[0050] It should be noted that the fire door abnormality alarm module is communicatively connected to the fire door appearance monitoring module, the fire door collision detection and evaluation module and the fire door test and analysis module. The fire door appearance monitoring module collects images of the fire door appearance, analyzes the fire door appearance image to generate an appearance alarm signal or an appearance qualified signal, and sends the appearance alarm signal or the appearance qualified signal to the supervision terminal, which can not only reasonably judge the appearance condition of the fire door, but also provide information support for the analysis process of the fire door abnormality alarm module to ensure the accuracy of its analysis results; the specific analysis process of the fire door appearance monitoring module is as follows:

[0051] Based on the appearance image of the fire door, cracks on the fire door are identified, the crack width and crack extension length of the corresponding crack are collected, and the crack width and crack extension length are numerically compared with the preset crack width threshold and the preset crack extension length threshold, respectively. If the crack width or the crack extension length exceeds the corresponding preset threshold, it indicates that the corresponding crack has a greater adverse effect on the fire door, and the corresponding crack is marked as a dangerous crack;

[0052] If there are dangerous cracks on the fire door, it indicates that the fire protection effect of the fire door is greatly affected, and an appearance alarm signal is generated; if there are no dangerous cracks on the fire door, the concave and convex areas on the fire door are captured based on the fire door appearance image;

[0053] By formula The area YP and the concave-convex depth ZF of the corresponding concave-convex area are numerically calculated to obtain a concave-convex coefficient TW, wherein eu and tm are preset proportional coefficients with values ​​greater than zero, and the larger the value of the concave-convex coefficient TW, the more serious the concave-convex condition of the corresponding concave-convex area; the concave-convex coefficient TW is numerically compared with the preset concave-convex coefficient threshold, and if the concave-convex coefficient TW exceeds the preset concave-convex coefficient threshold, it indicates that the concave-convex condition of the corresponding concave-convex area is more serious, and the corresponding concave-convex area is marked as a dangerous area;

[0054] The number of dangerous areas on the fire door is obtained and marked as concave-convex dangerous condition values, and the concave-convex coefficients of all concave-convex areas on the fire door are averaged to obtain the concave-convex expression value, and the concave-convex coefficient with the largest value on the fire door is marked as the concave-convex expression amplitude;

[0055] By formula The concave-convex danger value ZP, the concave-convex performance value QW and the concave-convex table amplitude TF are numerically calculated to obtain the concave-convex monitoring value YX, wherein a1, a2 and a3 are preset weight coefficients, and a1>a2>a3>0; and the larger the value of the concave-convex monitoring value YX, the more serious the deformation of the fire door;

[0056] The concave-convex monitoring value YX is numerically compared with the preset concave-convex monitoring threshold. If the concave-convex monitoring value YX exceeds the preset concave-convex monitoring threshold, it indicates that the deformation of the fire door is serious, which is not conducive to ensuring its fire prevention effect, and an appearance alarm signal is generated; if the concave-convex monitoring value YX does not exceed the preset concave-convex monitoring threshold, it indicates that the safety hazards of the fire door's appearance are generally small, and an appearance qualified signal is generated.

[0057] The fire door collision detection and evaluation module monitors the collision of the fire door, analyzes the collision risk when the fire door is hit, and judges whether to generate a collision alarm signal based on this. When the collision alarm signal is generated, it is sent to the fire door abnormal alarm module. It can not only monitor the collision status of the fire door in real time and accurately evaluate the risk of damage, but also provide information support for the analysis process of the fire door abnormal alarm module to ensure the accuracy of its analysis results; the specific analysis process is as follows:

[0058] The moment when the fire door is hit is collected and marked as the first moment, and the moment when the collision ends is collected and marked as the second moment, and the interval between the first moment and the second moment is marked as the characteristic duration; wherein, the larger the value of the characteristic duration is, the longer the duration of the corresponding collision is;

[0059] The maximum and average impact forces on the fire door during the corresponding collision process are obtained and marked as the force amplitude value and force performance value. The characteristic duration XS, the force amplitude value RF and the force performance value WL are numerically calculated to obtain the collision risk coefficient XN; wherein hu, tp, ey are preset proportional coefficients with values ​​greater than zero, and the larger the value of the collision risk coefficient XN, the greater the damage caused to the fire door by the corresponding collision;

[0060] The collision risk coefficient XN is numerically compared with the preset collision risk coefficient threshold. If the collision risk coefficient XN exceeds the preset collision risk coefficient threshold, it indicates that the corresponding collision has caused great damage to the fire door. It is necessary to inspect the fire door and determine the damage in time, and repair the corresponding fire door in time as needed, then a collision alarm signal is generated.

[0061] The fire door test and analysis module tests and analyzes the operating performance of the fire door, generates a performance test alarm signal or a performance test pass signal through analysis, and sends the performance test alarm signal or the performance test pass signal to the fire door test and analysis module, which can not only accurately analyze the performance of the fire door, but also provide information support for the analysis process of the fire door abnormal alarm module, further ensuring the accuracy of its analysis results; the specific analysis process of the fire door test and analysis module is as follows:

[0062] Perform several drive tests on the fire door, collect the drive execution delay time of the drive motor in the fire door, and collect its running speed in the process of the drive motor opening or closing the fire door, calculate the difference between the running speed and the set standard speed and take the absolute value to obtain the drive speed detection value, and compare the drive speed detection value with the preset drive speed detection threshold, and mark the number of drive speed detection values ​​exceeding the preset drive speed detection threshold in the corresponding drive process as a drive abnormal value, and calculate the average of all drive speed detection values ​​in the corresponding drive process to obtain a drive deviation value;

[0063] By formula The drive execution delay time YW, the drive abnormality value QY and the drive deviation value QP are numerically calculated to obtain the drive performance value FX; wherein b1, b2 and b3 are preset proportional coefficients with values ​​greater than zero, and the larger the value of the drive performance value FX is, the worse the drive operation performance of the fire door is overall;

[0064] The driving performance values ​​of all currently performed driving tests are averaged to obtain the driving evaluation value, and the driving evaluation value is numerically compared with the preset driving evaluation threshold. If the driving evaluation value exceeds the preset driving evaluation threshold, it indicates that the driving operation performance of the fire door is generally poor and its performance is abnormal, and a performance test alarm signal is generated.

[0065] Furthermore, if the driving evaluation value does not exceed the preset driving evaluation threshold, when the fire door is closed, pressure detection is performed through a plurality of pressure sensors installed at the contact position between the fire door and the door frame, and a plurality of groups of pressure data are collected accordingly, and the closing pressure detection value is obtained by averaging all the pressure data; wherein, the larger the closing pressure detection value, the poorer the closing effect of the fire door at that time;

[0066] The closing pressure analysis value is obtained by calculating the average of all closing pressure detection values ​​when the current fire door is closed several times, and the proportion of the number of occurrences of the closing pressure detection value not exceeding the preset closing pressure detection threshold is marked as the closing abnormality value, and the closing pressure analysis value and the closing abnormality value are numerically compared with the preset closing pressure analysis threshold and the preset closing abnormality threshold respectively;

[0067] If the closing pressure analysis value does not exceed the preset closing pressure analysis threshold or the closing abnormal occupancy value exceeds the preset closing abnormal occupancy threshold, it indicates that the closing performance of the fire door is poor and its performance is abnormal, then a performance test alarm signal is generated; if the closing pressure analysis value exceeds the preset closing pressure analysis threshold and the closing abnormal occupancy value does not exceed the preset closing abnormal occupancy threshold, it indicates that the closing performance of the fire door is poor and its performance is generally good, then a performance test qualified signal is generated.

[0068] Embodiment 2: Figure 2 As shown, the difference between this embodiment and the first embodiment is that the fire door abnormal alarm module is connected to the regional management evaluation module in communication, and the regional management evaluation module is used to set the supervision period, obtain all the fire doors in the corresponding area, analyze the management status of all the fire doors in the area during the supervision period, and generate a management low-obstruction signal or a management alarm signal through analysis;

[0069] And send the management low-obstruction signal or management alarm signal to the supervision terminal. When the supervision terminal receives the management alarm signal, it issues a corresponding warning to remind the management personnel to strengthen the supervision of the fire doors in the corresponding area in the future, increase management investment and personnel, and ensure the stable operation of all fire doors in the corresponding area, thereby improving the regional fire safety. The specific analysis process of the regional management evaluation module is as follows:

[0070] The total number of abnormal alarm information generated by all fire doors in the corresponding area during the supervision period is obtained and marked as an abnormal alarm value, and the abnormal alarm value is numerically compared with the preset abnormal alarm threshold. If the abnormal alarm value exceeds the preset abnormal alarm threshold, it indicates that the management of all fire doors in the area during the supervision period is difficult and it is necessary to strengthen the supervision of all fire doors in the area in the future, and a management alarm signal is generated;

[0071] If the abnormal alarm value does not exceed the preset abnormal alarm threshold, the time when the corresponding abnormal alarm information is generated is marked as the obstacle time, and the time when the administrator goes to check the fire door is marked as the correction time, and the interval between the correction time and the obstacle time is marked as the error correction time; wherein, the larger the error correction time is, the less timely the processing of the corresponding abnormal alarm information is;

[0072] All the fault correction times in the supervision period are obtained and their average is calculated to obtain the fault correction detection value, and the fault correction time is numerically compared with the preset fault correction time threshold. If the fault correction time exceeds the preset fault correction time threshold, the corresponding fault correction time is marked as the error correction time, and the number of error correction times in the supervision period is obtained and marked as the error correction detection value;

[0073] By formula The abnormal alarm value SY, the fault correction detection value PM and the error correction detection value TN are numerically calculated to obtain the zone management evaluation value QX, wherein q1, q2 and q3 are preset proportional coefficients with values ​​greater than zero, and the larger the value of the zone management evaluation value QX, the worse the overall management performance of all fire doors in the area during the supervision period;

[0074] The zone management assessment value QX is numerically compared with the preset zone management assessment threshold. If the zone management assessment value QX exceeds the preset zone management assessment threshold, it indicates that the management status of all fire doors in the area during the supervision period is generally poor, and it is necessary to strengthen the supervision of all fire doors in the area in the future, and a management alarm signal is generated; if the zone management assessment value QX does not exceed the preset zone management assessment threshold, it indicates that the management status of all fire doors in the area during the supervision period is generally good, and a management low barrier signal is generated.

[0075] Embodiment 3: Figure 3 As shown, the difference between this embodiment and the first and second embodiments is that the present invention proposes a multi-modal fusion fire door autonomous decision-making control method, comprising the following steps:

[0076] Step 1: Collect various information about the fire door and its surrounding environment;

[0077] Step 2: Fusion and analysis of multimodal data to identify fire conditions;

[0078] Step 3: When a fire is detected, close the fire door to prevent the fire from spreading;

[0079] Step 4: Analyze the fire door abnormalities one by one and determine whether to generate abnormal alarm information;

[0080] Step 5: When abnormal alarm information is generated, the monitoring terminal issues a corresponding warning.

[0081] The working principle of the present invention is as follows: when in use, the multimodal data of the fire door and its surrounding environment are collected through the multimodal perception module, and the fire situation identification and judgment module performs fusion analysis on the multimodal data to identify the fire situation. When the fire situation is identified, the autonomous decision-making control module controls the fire door to be closed to prevent the fire from spreading, thereby realizing real-time monitoring, intelligent analysis and autonomous control of the fire door status, effectively improving the building fire safety level, and comprehensively monitoring and analyzing the fire door anomalies and timely warnings through the fire door abnormality alarm module to remind management personnel to take optimization measures such as inspection and maintenance of the corresponding fire doors, thereby ensuring the stable operation of the fire doors, and further ensuring the building fire safety level, intelligence and automation levels are high.

[0082] The above formulas are all dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions. The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the contents of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that technicians in the relevant technical field can understand and use the present invention well. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A multi-modal fusion fire door autonomous decision-making control system, characterized in that: It includes a multi-modal perception module, a fire identification and judgment module, an autonomous decision-making control module, a fire door abnormal alarm module and a monitoring terminal; the multi-modal perception module collects a variety of information about the fire door and its surrounding environment, and the fire identification and judgment module integrates and analyzes the multi-modal data to identify the fire situation; When a fire is identified, the autonomous decision-making control module controls the fire door to close to prevent the fire from spreading, and sends the control information to the supervision terminal; when the fire door abnormal alarm module receives the appearance alarm signal, collision alarm signal or performance test alarm signal, it generates corresponding abnormal alarm information and sends it to the supervision terminal, and the supervision terminal displays the abnormal alarm information and issues a corresponding warning.

2. According to claim 1, a multi-modal fusion fire door autonomous decision-making control system is characterized in that: The fire door abnormality alarm module is communicatively connected to the fire door appearance monitoring module, the fire door collision detection and evaluation module and the fire door test and analysis module. The fire door appearance monitoring module collects images of the appearance of the fire door, analyzes the fire door appearance image to generate an appearance alarm signal or an appearance qualified signal, and sends the appearance alarm signal or the appearance qualified signal to the supervision terminal; The fire door collision detection and evaluation module monitors the fire door for collisions, analyzes the collision risk when the fire door is hit, and determines whether to generate a collision alarm signal based on this. When the collision alarm signal is generated, it is sent to the fire door abnormal alarm module; the fire door test and analysis module tests and analyzes the operating performance of the fire door, generates a performance test alarm signal or a performance test pass signal through analysis, and sends the performance test alarm signal or the performance test pass signal to the fire door test and analysis module.

3. The multi-modal fusion fire door autonomous decision-making control system according to claim 2 is characterized in that: The specific analysis process of the fire door appearance monitoring module is as follows: Based on the appearance image of the fire door, cracks on the fire door are identified. If there are dangerous cracks on the fire door, an appearance alarm signal is generated; if there are no dangerous cracks on the fire door, the concave and convex areas on the fire door are captured based on the appearance image of the fire door, and the concave and convex monitoring value is obtained by numerically calculating the concave and convex dangerous value, the concave and convex performance value and the concave and convex surface amplitude. If the concave and convex monitoring value exceeds the preset concave and convex monitoring threshold, an appearance alarm signal is generated; If the concave-convex monitoring value does not exceed the preset concave-convex monitoring threshold, a shape qualification signal is generated.

4. The multi-modal fusion fire door autonomous decision-making control system according to claim 2 is characterized in that: The specific analysis process for analyzing the collision hazard when the fire door is hit is as follows: The collision risk coefficient is obtained by numerically calculating the characteristic duration, force amplitude value and force performance value. If the collision risk coefficient exceeds the preset collision risk coefficient threshold, a collision alarm signal is generated.

5. The multi-modal fusion fire door autonomous decision-making control system according to claim 3 is characterized in that: The specific analysis process of the fire door test analysis module is as follows: The fire door is driven several times and the driving performance value is obtained by numerically calculating the driving execution delay time, driving abnormal value and driving deviation value. The driving performance values ​​of all driving tests currently being performed are averaged to obtain the driving evaluation value. If the driving evaluation value exceeds the preset driving evaluation threshold, a performance test alarm signal is generated.

6. The multi-modal fusion fire door autonomous decision-making control system according to claim 3 is characterized in that: If the driving evaluation value does not exceed the preset driving evaluation threshold, the closure pressure analysis value and the closure abnormality value are numerically compared with the preset closure pressure analysis threshold and the preset closure abnormality threshold respectively; If the closing pressure analysis value does not exceed the preset closing pressure analysis threshold or the closing abnormality value exceeds the preset closing abnormality threshold, a performance test alarm signal is generated; otherwise, a performance test qualified signal is generated.

7. The multi-modal fusion fire door autonomous decision-making control system according to claim 1 is characterized in that: The fire door abnormal alarm module is communicated with the regional management evaluation module. The regional management evaluation module analyzes the management status of all fire doors in the area during the supervision period, generates a management low obstruction signal or a management alarm signal through analysis, and sends the management low obstruction signal or the management alarm signal to the supervision terminal. When the supervision terminal receives the management alarm signal, it issues a corresponding warning.

8. The multi-modal fusion fire door autonomous decision-making control system according to claim 7 is characterized in that: The specific analysis process of the regional management assessment module is as follows: The district management evaluation value is obtained by numerically calculating the abnormal alarm value, the fault correction detection value and the error correction detection value. If the district management evaluation value exceeds the preset district management evaluation threshold, a management alarm signal is generated; if the district management evaluation value does not exceed the preset district management evaluation threshold, a management low obstacle signal is generated.

9. A multi-modal fusion fire door autonomous decision-making control method, characterized in that: The following steps are involved: Step 1: Collect various information about the fire door and its surrounding environment; Step 2: Fusion and analysis of multimodal data to identify fire conditions; Step 3: When a fire is detected, close the fire door to prevent the fire from spreading; Step 4: Analyze the fire door abnormalities one by one and determine whether to generate abnormal alarm information; Step 5: When abnormal alarm information is generated, the monitoring terminal issues a corresponding warning.

Images