CN-121724417-B - Remote duty monitoring method and system

CN121724417BCN 121724417 BCN121724417 BCN 121724417BCN-121724417-B

Abstract

The application provides a remote duty monitoring method and a remote duty monitoring system, wherein the method comprises the steps of carrying out geometric analysis on an included angle of a door body orientation angle and an airflow speed vector according to environment interaction parameters, combining a gap width interval to determine potential smoke diffusion risk levels, extracting feature data of similar scenes from a pre-established historical case matching database according to a judging result of high-risk combination to generate corresponding risk scene tags, determining treatment priority, merging airflow interaction intensity and smoke concentration readings according to a judging result of an accelerating smoke diffusion channel to determine a high-risk scene judging result, if the high-risk scene is judged, retrieving a treatment protocol sequence corresponding to the treatment priority and the risk scene tags from the database to generate a real-time response instruction, and automatically adjusting a fireproof door state to obtain an optimized safety configuration state.

Inventors

XU LEI
CAO NAN
FENG WEIBIAO

Assignees

北京华祺永安消防工程有限公司

Dates

Publication Date: 20260508
Application Date: 20251212

Claims (9)

1. A method for remote duty monitoring, the method comprising: Acquiring door body orientation angles, airflow velocity vectors, airflow interaction strength, gap width interval data, temperature distribution and smoke concentration readings in real time through a multipoint sensor array arranged on the periphery of the fireproof door, and denoising the original data to obtain environment interaction parameters; Performing geometric analysis on the included angle between the door body orientation angle and the airflow speed vector according to the environment interaction parameters, and determining potential smoke diffusion risk level by combining the gap width interval; Tracking and analyzing environmental interaction parameter fluctuation by combining potential smoke diffusion risk levels with dynamic change frequency and temperature distribution and adopting a time sequence comparison method to judge whether a high-risk combination of windward risk marks and small gap characteristic value superposition exists or not; Extracting similar scene characteristic data from a historical case matching database according to the high-risk combination judgment result to generate a risk scene label and determining priority; acquiring real-time image data of a remote monitoring platform based on a high-risk combination judgment result, carrying out boundary extraction analysis on a door seam area to identify a small change trend of the width of the seam, and judging whether an accelerating flue gas diffusion channel exists or not; Determining a high-risk scene judgment result according to the accelerated smoke diffusion channel judgment result by fusing the airflow interaction strength and the smoke concentration reading; When the high risk scene is judged, retrieving a disposal protocol sequence corresponding to the priority and the risk scene label from a database to generate a real-time response instruction, and automatically adjusting the fireproof door state to obtain an optimized safety configuration state; Updating the dynamic change frequency record in the monitoring log according to the optimized safety configuration state, transmitting the adjusted environment interaction parameters to the central processing unit, circularly verifying the airflow interaction intensity attenuation trend, and judging whether the overall risk is reduced to a controllable range; The method for judging whether a high-risk combination of windward risk marks and small gap characteristic value superposition exists by combining potential smoke diffusion risk grades with dynamic change frequency and temperature distribution and adopting a time sequence comparison method to track and analyze environmental interaction parameter fluctuation comprises the following steps: continuously acquiring environment interaction parameters according to potential smoke diffusion risk levels, and constructing a time sequence array; counting the ratio of the change times of each parameter to the length of the time window by adopting a sliding time window method to obtain dynamic change frequency, generating a windward risk mark when the included angle between the door body angle and the airflow speed vector is smaller than a preset value, and generating a small gap characteristic value when the gap width is in a micro-interval; obtaining temperature distribution variation through differential operation, calculating temperature distribution variation rate, carrying out weighted summation on dynamic variation frequency and the temperature distribution variation rate, normalizing to obtain an environmental parameter fluctuation index, and comparing the environmental parameter fluctuation index with a historical contemporaneous data sequence to obtain fluctuation anomaly degree; And when the windward risk mark and the small gap characteristic value exist simultaneously and the environmental parameter fluctuation index and the fluctuation anomaly degree exceed the corresponding threshold values, judging that high-risk combination exists.
2. The method for remote on-duty monitoring according to claim 1, wherein the acquiring the door body orientation angle, the airflow velocity vector, the airflow interaction intensity, the gap width interval data, the temperature distribution and the smoke concentration reading in real time through the multi-point sensor array deployed around the fireproof door, denoising the original data to obtain the environment interaction parameters comprises: arranging a sensing array around the fireproof door according to a regular hexagonal grid, and performing triangular positioning through known distances between adjacent nodes and readings of angle sensors to obtain a sensing network topological structure and space coordinates; synchronously acquiring inclination angle data, an airflow speed vector, airflow interaction intensity, temperature distribution data, gap width interval data and smoke concentration readings, carrying out coordinate transformation on the airflow speed vector to calculate an included angle cosine value of a door body normal vector and the airflow speed vector, and judging whether the door body normal vector and the airflow speed vector are in a windward state or not; and if the state is judged to be the windward side state, carrying out noise suppression processing on each item of data by adopting a Kalman filter, and fusing the filtered data by a weighted average method to obtain environment interaction parameters.
3. The method for remote on-duty monitoring according to claim 1, wherein the geometric analysis of the angle between the door body orientation angle and the airflow velocity vector according to the environmental interaction parameter, and the determination of the potential smoke diffusion risk level in combination with the gap width interval, comprises: extracting a door body orientation angle and an airflow speed vector from environment interaction parameters, obtaining an included angle cosine value through vector dot product operation, and calculating an actual included angle value by using an inverse cosine function; and inquiring a corresponding permeability coefficient value from a pre-established flue gas permeability coefficient database, and determining a potential flue gas diffusion risk level through the flue gas permeability coefficient and a preset risk threshold value set comparison.
4. The method of claim 1, wherein the extracting similar scene feature data from the historical case matching database according to the high-risk combination judgment result to generate a risk scene tag and determining a priority comprises: Extracting a current environmental parameter feature vector, and calculating a similarity value with the historical case feature vector through a cosine similarity algorithm; reading records of the fire development stage, the smoke diffusion rate and the door damage degree of the historical case with the similarity exceeding the similarity threshold value, and generating a risk scene label; And inquiring a preset hazard level table and a response time table through the risk scene tag, calculating a comprehensive score, and determining the treatment priority.
5. The method for remote on-duty monitoring according to claim 1, wherein the obtaining the real-time image data of the remote monitoring platform based on the high-risk combination judgment result, performing boundary extraction analysis on the door seam area to identify the small change trend of the width of the seam, and judging whether the accelerated flue gas diffusion channel exists comprises: acquiring a real-time video stream of a fireproof door area from a remote monitoring platform, extracting continuous image frames, and carrying out graying and histogram equalization to obtain a preprocessed image; Detecting linear characteristics of a door frame through Hough transformation to determine a door gap interested region, and extracting a door gap outline in the interested region by adopting Canny edge detection to obtain a boundary pixel point set; Calculating the distance between left and right boundary pixels, converting the distance into physical width values, constructing a time sequence width array according to time sequence, traversing a sliding window to calculate the change rate of the width values in the window, and marking the window as an accelerating flue gas diffusion channel when the change rate of the width is in an increasing trend in a plurality of continuous windows.
6. The method for remote on-duty monitoring according to claim 1, wherein the determining the high risk scene determination result according to the accelerated smoke diffusion channel determination result by integrating the airflow interaction intensity and the smoke concentration reading comprises: Acquiring an airflow interaction intensity value and smoke concentration reading combination from a sensor network to form a real-time monitoring data pair, and matching with a pre-established two-dimensional risk judgment table; and determining a high-risk scene judgment result when the airflow interaction intensity exceeds a preset intensity threshold value and the smoke concentration exceeds a preset concentration threshold value.
7. The method for remote on-duty monitoring according to claim 1, wherein retrieving a treatment protocol sequence corresponding to a priority and risk scene tag from a database to generate a real-time response instruction when the high risk scene is determined, automatically adjusting the fire door state to obtain an optimized security configuration state, comprises: if the high risk scene is judged, constructing a query index according to the priority and the risk scene label to retrieve a disposal protocol database to acquire a protocol sequence; Converting each parameter value in the protocol sequence to form a real-time response instruction set; and sending the real-time response instruction set to an execution component through a preset communication interface, monitoring a return signal, and confirming the optimized safety configuration state.
8. The remote on-duty monitoring method according to claim 1, wherein updating the dynamic change frequency record in the monitoring log according to the optimized security configuration state and transmitting the adjusted environment interaction parameter to the central processing unit, and circularly verifying the airflow interaction intensity attenuation trend comprises: writing the current door body angle and the sealing pressure into a monitoring log to calculate the time change rate according to the optimized safety configuration state so as to obtain a dynamic change frequency record; And packaging and transmitting the dynamic change frequency record and the environmental interaction parameters subjected to the re-acquisition and adjustment to a central processing unit, circularly reading the air flow interaction intensity numerical value to calculate a difference value, and judging that the overall risk is reduced to a controllable range when the intensity value is continuously reduced and falls below a safety threshold.
9. A remote duty monitoring system implemented based on the remote duty monitoring method of claim 1, the system comprising: the data acquisition and preprocessing module is used for acquiring preliminary environment interaction parameters in real time through a multipoint sensor array around the fireproof door, and generating environment interaction parameters after denoising; The risk level analysis module is used for analyzing the included angle between the door body and the airflow and the width of the gap according to the environment interaction parameters and determining the smoke diffusion risk level; the high-risk combination judging module is used for judging whether high-risk combination overlapped by a windward side and a small gap exists or not by combining the risk level, the dynamic change frequency and the temperature data and comparing and tracking environmental parameter fluctuation through time sequence; The treatment priority determining module is used for extracting similar scene features from the historical case database based on the high-risk combination result, generating a risk scene label and determining the treatment priority; The image analysis module is used for acquiring a remote monitoring real-time image based on the high-risk combination result, carrying out boundary extraction analysis on the gate seam area, identifying the tiny change trend of the width of the seam, and judging whether an accelerating flue gas diffusion channel exists or not; the high-risk scene judging module is used for fusing the air flow intensity and the smoke concentration according to the judging result of the accelerated smoke diffusion channel to determine a high-risk scene; the response execution module is used for searching a disposal protocol corresponding to the priority and the risk scene label under the high risk scene, generating a real-time instruction and automatically adjusting the fireproof door to a safety configuration state; And the state updating and verifying module is used for updating the dynamic change frequency of the monitoring log according to the safety configuration state, transmitting the adjusted environment interaction parameters, circularly verifying the airflow intensity attenuation trend and judging whether the overall risk is controllable.

Description

Remote duty monitoring method and system Technical Field The invention relates to the technical field of information, in particular to a remote on-duty monitoring method and a system. Background Fireproof doors are key facilities for preventing and controlling building fires, and the smoke and fire isolating effect is directly related to personnel evacuation safety. In a practical scene, when the fireproof door is positioned on the windward side and a small gap exists, the air flow accelerates the invasion of smoke or flame through the slit effect, and the risk brought by the combination is even higher than that of a large gap, but the core problem is difficult to effectively solve by the traditional remote duty monitoring method. First, traditional monitoring cannot accurately identify the high risk combination of "windward side + minor gap". The door body orientation needs to be subjected to angle data acquisition through a sensor, the airflow direction is frequently changed under the influence of environmental wind and a building ventilation system, the dynamic fluctuation can interfere the judgment of a windward side, meanwhile, the small gap of a fireproof door is difficult to capture by a conventional monitoring means due to small size, and particularly, the risk characteristic can be completely omitted in a complex light or shielding environment, so that a high-risk scene cannot be recognized in time, and subsequent treatment is more difficult to talk. Second, even though some risk features are barely perceived, there is a lack of dynamic tracking and comprehensive decision mechanisms for risk combinations. The data such as the air flow intensity, the smoke concentration and the temperature related to the fire are in dynamic changes, the traditional method cannot continuously track the fluctuation trend of the data, and the multi-dimensional data cannot be fused to judge whether the risk is continuously updated, so that the corresponding treatment priority and response instruction cannot be matched accurately, invalid operation is triggered by mistake, critical emergency adjustment is delayed, and finally effective management and control of a high-risk scene cannot be realized. In summary, the failure of high risk combination recognition is a fundamental obstacle, and the lack of dynamic tracking and comprehensive determination makes the recognized risk unable to be effectively handled, so that a set of coherent technical schemes is needed to comprehensively solve from recognition to determination, and ensure that the high risk scene can be accurately captured and responded in time. Disclosure of Invention Aiming at the defects and defects in the prior art, the invention provides a remote duty monitoring method, which mainly comprises the following steps: Acquiring door body orientation angles, airflow velocity vectors, airflow interaction strength, gap width interval data, temperature distribution and smoke concentration readings in real time through a multipoint sensor array arranged on the periphery of the fireproof door, and denoising the original data to obtain environment interaction parameters; Performing geometric analysis on the included angle between the door body orientation angle and the airflow speed vector according to the environment interaction parameters, and determining potential smoke diffusion risk level by combining the gap width interval; Tracking and analyzing environmental interaction parameter fluctuation by combining potential smoke diffusion risk levels with dynamic change frequency and temperature distribution and adopting a time sequence comparison method to judge whether a high-risk combination of windward risk marks and small gap characteristic value superposition exists or not; Extracting similar scene characteristic data from a historical case matching database according to the high-risk combination judgment result to generate a risk scene label and determining priority; acquiring real-time image data of a remote monitoring platform based on a high-risk combination judgment result, carrying out boundary extraction analysis on a door seam area to identify a small change trend of the width of the seam, and judging whether an accelerating flue gas diffusion channel exists or not; Determining a high-risk scene judgment result according to the accelerated smoke diffusion channel judgment result by fusing the airflow interaction strength and the smoke concentration reading; When the high risk scene is judged, retrieving a disposal protocol sequence corresponding to the priority and the risk scene label from a database to generate a real-time response instruction, and automatically adjusting the fireproof door state to obtain an optimized safety configuration state; and updating the dynamic change frequency record in the monitoring log according to the optimized safety configuration state, transmitting the adjusted environment interaction parameters to the central processing