Communication machine room intelligent fire monitoring and fire extinguishing system based on edge calculation

Abstract

The invention relates to the technical field of communication machine room safety protection, and discloses a communication machine room intelligent fire monitoring and fire extinguishing system based on edge computing, which comprises a fire parameter acquisition module, a temperature characteristic analysis module, a smoke dynamic detection module, a multi-source fusion judgment module, a regulation and control parameter generation module and the like. The fire parameter acquisition module acquires parameters such as temperature gradient, smoke concentration and equipment current, judges fire risk and triggers cooperative monitoring; the temperature characteristic analysis module extracts temperature distribution characteristics and quantitatively evaluates hot spot abnormity; the smoke dynamic detection module analyzes the smoke diffusion trend; the multi-source fusion judgment module is combined with multiple parameters to generate a fire extinguishing or optimizing signal; and the regulation and control parameter generation module is matched with the strategy to generate injection and ventilation parameters. The system also corrects a risk index through an equipment coupling analysis module, and forms closed-loop control by using a dynamic feedback execution module. According to the system, multi-source data fusion monitoring and intelligent fire extinguishing are realized, and the fire prevention and control accuracy of the communication machine room is improved.

Description

Technical Field

[0001] The present invention relates to the field of communication room security protection technology, and specifically to an intelligent fire monitoring and extinguishing system for communication rooms based on edge computing. Background Art

[0002] With the rapid development of communications technology, the safety of communications rooms, as core locations for data transmission and equipment operation, is paramount. Fire is a major security threat facing communications rooms. Once a fire occurs, it not only causes communication interruptions and equipment damage, but can also result in significant economic losses and social impact. Traditional fire monitoring and extinguishing systems in communications rooms have numerous flaws and are unable to meet the stringent fire prevention and control requirements of modern communications rooms.

[0003] Traditional fire detection systems mostly use single-point sensors, which can only monitor the temperature or smoke concentration in a single area and cannot comprehensively and dynamically perceive the fire risk of the entire computer room. For example, measuring temperature with only a single temperature sensor cannot accurately reflect the complex temperature gradient distribution and hot spots within the computer room, and it is easy to miss critical areas with abnormal temperatures. In terms of smoke monitoring, traditional systems lack the ability to effectively analyze and predict the fluctuations and diffusion trends of smoke particle concentrations, making it impossible to timely determine the speed and scope of fire spread.

[0004] Traditional fire risk assessment systems typically rely on a single parameter threshold, lacking the ability to integrate and analyze multi-source data. For example, a fire warning based solely on a temperature exceeding a threshold, without considering changes in other parameters such as smoke concentration and equipment operating current, can easily lead to false or missed alarms. Abnormal equipment operating current is often a significant sign of a fire, yet traditional systems fail to incorporate this into their fire risk assessment.

[0005] In terms of firefighting strategies, traditional systems are relatively fixed and cannot be dynamically adjusted based on the actual fire situation. When a fire occurs, the extinguishing agent's injection rate, injection angle, and ventilation system adjustments all need to be optimized based on the fire's stage and specific circumstances, which is difficult for traditional systems to achieve. For example, different fire scenarios may require different extinguishing agent injection angles to improve extinguishing efficiency, but traditional systems can only spray at preset fixed angles.

[0006] Furthermore, traditional systems lack monitoring of the temperature compatibility between communications equipment and the equipment room environment. The heat dissipation rate and thermal stress distribution of communications equipment can affect its operating status and fire risk. Traditional systems fail to effectively monitor and analyze this, preventing early detection of potential fire hazards caused by temperature mismatches. Furthermore, traditional systems lack dynamic feedback mechanisms for data sampling frequency and monitoring processes, preventing timely adjustments to monitoring strategies based on fire extinguishing and regulation results, making it difficult to establish effective closed-loop control.

[0007] The development of edge computing technology has provided new insights for upgrading fire monitoring and extinguishing systems in communications rooms. Edge computing enables data processing and analysis close to the data source, offering advantages such as low latency and high reliability. This allows for real-time collection, analysis, and rapid response to room fire status parameters. Therefore, an intelligent fire monitoring and extinguishing system for communications rooms based on edge computing is urgently needed to address the aforementioned issues with traditional systems and improve fire prevention and control capabilities in communications rooms. Summary of the Invention

[0008] The purpose of the present invention is to provide an intelligent fire monitoring and extinguishing system for communication rooms based on edge computing to solve the problems raised in the above background technology.

[0009] To achieve the above objectives, the present invention provides the following technical solution: an intelligent fire monitoring and extinguishing system for communication rooms based on edge computing, the system comprising:

[0010] The fire parameter acquisition module is used to dynamically collect the temperature gradient, smoke concentration parameters and equipment operating current of the target communication room to obtain the room fire status parameter set. Based on the room fire status parameter set, the fire risk is determined and analyzed, and a fire warning signal is generated. The generated fire warning signal triggers the collaborative monitoring instruction, and the temperature feature analysis module and the smoke dynamic detection module are executed according to the triggered collaborative monitoring instruction.

[0011] The temperature characteristic analysis module is used to extract the temperature distribution characteristic parameters of the key areas of the target communication room, quantitatively evaluate the hot spot distribution of the temperature abnormality area, and obtain the temperature characteristic abnormality index;

[0012] The smoke dynamic detection module is used to extract the smoke particle concentration fluctuation parameters in the target communication room, predict and analyze the spread trend of the smoke diffusion state, and obtain the smoke diffusion assessment value;

[0013] The multi-source fusion judgment module receives the temperature characteristic anomaly index and smoke diffusion assessment value, performs collaborative analysis on the fire risk in the computer room, and generates fire extinguishing activation signals and system optimization signals.

[0014] The control parameter generation module is used to receive the fire extinguishing start signal and the system optimization signal, match the fire extinguishing strategy, and generate the fire extinguishing agent injection parameters and ventilation adjustment parameters.

[0015] Preferably, the fire risk determination and analysis includes:

[0016] By collecting the temperature gradient parameters of the target communication room in real time, the deviation percentage from the ambient temperature is calculated and marked as the temperature anomaly degree;

[0017] Extract the particle density and change rate from the smoke concentration parameters of the target communication room, calculate the geometric composite vector modulus of the two, and mark it as the smoke concentration characteristic value;

[0018] Collect the peak current and fluctuation frequency of the equipment's operating current parameters, perform normalized weighted calculations, and obtain a comprehensive current anomaly index;

[0019] The temperature anomaly, smoke concentration characteristic value and current anomaly comprehensive index are compared with the preset thresholds respectively. When any parameter exceeds the corresponding threshold, a fire warning signal is generated.

[0020] Preferably, the quantitative evaluation of the hotspot distribution in the temperature anomaly area includes:

[0021] By deploying an infrared sensor array to collect temperature distribution data, a temperature thermogram and gradient change map are generated;

[0022] Extract the hotspot area ratio and the maximum temperature value from the temperature heat map, calculate the product of the two and take the inverse to obtain the hotspot distribution coefficient;

[0023] The temperature rise rate and spatial gradient amplitude are extracted from the gradient change map, and the arithmetic mean of the two is calculated and marked as the gradient characteristic index;

[0024] The hotspot distribution coefficient and the gradient characteristic index are weighted and fused to obtain the temperature characteristic anomaly index.

[0025] Preferably, the predictive analysis of the spread trend of the smoke diffusion state includes:

[0026] Collect smoke particle concentration data of the target communication room in real time and extract characteristic parameters of concentration variation and diffusion direction;

[0027] Construct a smoke spread trend prediction model, input the concentration change amplitude into the model for Kalman filtering, and output the smoke spread probability in the future time interval;

[0028] Extract the main diffusion angle and the secondary diffusion angle from the diffusion direction characteristic parameters, calculate the energy ratio between the two and obtain the directional stability factor;

[0029] The smoke diffusion probability is linearly combined with the directional stability factor to obtain the smoke diffusion assessment value.

[0030] Preferably, the collaborative analysis of the fire risk in the computer room includes:

[0031] Retrieve temperature anomaly data, set its correction coefficient, and obtain the temperature impact compensation value through calculation;

[0032] Normalize the values of the temperature characteristic anomaly index, smoke diffusion assessment value, and temperature impact compensation value to generate a fire risk fusion index;

[0033] Set a fire risk determination threshold. If the fusion index is lower than the threshold, a fire extinguishing start signal is generated. If it is higher than the threshold, a system optimization signal is generated.

[0034] Preferably, the fire extinguishing strategy matching includes:

[0035] If the fire extinguishing start signal is captured, the injection control instruction is triggered, and the release rate and injection angle parameters of the fire extinguishing agent are dynamically adjusted according to the instruction to generate the fire extinguishing agent injection parameters;

[0036] If the system optimization signal is captured, the ventilation adjustment instruction is triggered, and the airflow intensity and response time parameters of the ventilation system are optimized according to the instruction to generate ventilation adjustment parameters.

[0037] Preferably, the system further comprises:

[0038] The device coupling analysis module is used to monitor the temperature coordination between communication equipment and the equipment room environment, extract the difference in equipment heat dissipation rate and thermal stress concentration factor, and generate a device coupling assessment value;

[0039] The multi-source fusion determination module further combines the equipment coupling evaluation value to perform a secondary correction on the fire risk fusion index.

[0040] Preferably, the monitoring of the temperature coordination between the communication equipment and the computer room environment includes:

[0041] Thermocouple sensors are used to collect the heat dissipation rate of the device surface and the rate of change of the ambient temperature. The difference between the two rates is calculated and the absolute value is taken, which is marked as the heat dissipation rate difference value.

[0042] Extract the thermal stress distribution data of the area in contact between the equipment surface and the environment, and calculate the ratio of the maximum thermal stress value to the average thermal stress, which is marked as the thermal stress concentration coefficient;

[0043] The heat dissipation rate difference value and the thermal stress concentration factor are weighted and summed to generate the equipment coupling assessment value.

[0044] Preferably, the system further comprises:

[0045] The dynamic feedback execution module is used to adjust the data sampling frequency of the monitoring system in real time according to the generated fire extinguishing agent injection parameters and ventilation adjustment parameters, and feed back the execution results to the fire parameter acquisition module to form a closed-loop control link.

[0046] Preferably, the specific execution process of the dynamic feedback execution module includes:

[0047] If the extinguishing agent injection parameters involve adjustment of the injection angle, the sampling density of the smoke concentration sensor should be increased simultaneously;

[0048] If the ventilation adjustment parameters involve optimizing the airflow intensity, the number of temperature detection channels should be increased simultaneously;

[0049] Input the adjusted parameters into the fire parameter acquisition module and restart the monitoring process.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] In terms of fire monitoring, the fire parameter acquisition module dynamically collects multi-dimensional parameters such as temperature gradient, smoke concentration, and equipment operating current, and constructs a comprehensive set of computer room fire status parameters. By calculating the temperature anomaly, smoke concentration characteristic values, and current anomaly comprehensive index and comparing them with preset thresholds, a preliminary judgment of fire risk is achieved, which changes the limitations of traditional single parameter judgment and effectively reduces false alarm and missed alarm rates. The temperature characteristic analysis module uses an infrared sensor array to collect temperature distribution data, generate temperature thermograms and gradient change maps, and calculate the hotspot distribution coefficient and gradient characteristic index to achieve a quantitative assessment of the hotspot distribution in the temperature anomaly area. It can accurately locate key areas of temperature anomalies and promptly detect potential fire hazards. The dynamic smoke detection module constructs a smoke spread trend prediction model, combines Kalman filter processing and the calculation of directional stability factors, and realizes the prediction and analysis of the spread trend of smoke diffusion status, providing a strong basis for formulating fire extinguishing strategies in advance.

[0052] In terms of fire risk assessment, the multi-source fusion assessment module receives multi-source data, such as the temperature characteristic anomaly index and smoke diffusion assessment value, and generates a fire risk fusion index through normalization calculation. This fusion index is then corrected by combining the device coupling assessment value. This fusion index fully considers the temperature coordination between the communication equipment and the computer room environment, making the fire risk assessment more accurate and comprehensive. This multi-source data fusion analysis method overcomes the shortcomings of traditional single-parameter assessment and can more realistically reflect the fire risk status of the computer room.

[0053] In terms of fire extinguishing strategy generation, the control parameter generation module matches the fire extinguishing strategy based on the fire extinguishing activation signal and system optimization signal generated by the multi-source fusion judgment module. When the fire extinguishing activation signal is captured, the release rate and injection angle of the fire extinguishing agent are dynamically adjusted to generate targeted fire extinguishing agent injection parameters, improving fire extinguishing efficiency. When the system optimization signal is captured, the ventilation system's airflow intensity and response time parameters are optimized to improve the computer room environment and reduce fire risk. This dynamic adjustment of fire extinguishing strategies based on actual fire risk enables the system to better adapt to different fire scenarios.

[0054] In terms of closed-loop system control, the dynamic feedback execution module adjusts the monitoring system's data sampling frequency in real time based on generated fire extinguishing agent injection parameters and ventilation adjustment parameters. It also feeds the execution results back to the fire parameter acquisition module, forming a closed-loop control chain. For example, when fire extinguishing agent injection parameters involve adjusting the injection angle, the sampling density of the smoke concentration sensor is simultaneously increased; when ventilation adjustment parameters involve optimizing airflow intensity, the number of temperature detection channels is simultaneously increased. This closed-loop control mechanism enables the system to promptly adjust its monitoring strategy based on execution results, achieving dynamic tracking and precise prevention and control of fire risks, further improving the system's reliability and effectiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a working principle diagram of the edge computing-based intelligent fire monitoring and extinguishing system for communication rooms according to the present invention;

[0056] Figure 2 Design drawings for fire risk assessment and analysis;

[0057] Figure 3 Design diagram for quantitative assessment of temperature anomaly areas;

[0058] Figure 4 Design diagram for fire extinguishing strategy matching and dynamic feedback. DETAILED DESCRIPTION

[0059] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0060] See also Figures 1-4 The present invention relates to an intelligent fire monitoring and extinguishing system for communication rooms based on edge computing. The specific implementation scheme is as follows:

[0061] The system includes a fire parameter acquisition module, a temperature characteristic analysis module, a smoke dynamic detection module, a multi-source fusion judgment module and a control parameter generation module.

[0062] Fire parameter acquisition module: Through the sensor array deployed in the target communication room, the temperature gradient, smoke concentration parameters and equipment operating current in the room are dynamically collected to form a room fire status parameter set containing multi-dimensional data. Based on this parameter set, the algorithm preset within the module determines and analyzes the fire risk: in real time, the deviation percentage between the temperature gradient parameter and the ambient temperature (i.e., temperature anomaly) is calculated, the particle density and change rate in the smoke concentration parameter are extracted and the geometric synthetic vector modulus (i.e., smoke concentration characteristic value) is calculated, and the peak current and fluctuation frequency in the equipment operating current parameter are collected and normalized weighted calculation is performed (i.e., current anomaly comprehensive index). When any parameter among the temperature anomaly, smoke concentration characteristic value and current anomaly comprehensive index exceeds the preset threshold, a fire warning signal is generated, and the collaborative monitoring instruction is triggered to start the temperature characteristic analysis module and the smoke dynamic detection module.

[0063] The temperature signature analysis module: After receiving collaborative monitoring instructions, it uses an infrared sensor array to collect temperature distribution data from key areas of the target communications room, generating a temperature thermogram and a gradient change map. The module extracts the area percentage and maximum temperature of the hotspot region from the thermogram, calculates the product of the two and takes the inverse to obtain the hotspot distribution coefficient. The module also extracts the temperature rise rate and spatial gradient amplitude from the gradient change map, calculates their arithmetic mean, and obtains the gradient signature index. The hotspot distribution coefficient and the gradient signature index are weighted and fused together to produce the final temperature signature anomaly index.

[0064] The Smoke Dynamic Detection Module synchronously receives collaborative monitoring commands, collects real-time smoke particle concentration data within the target communications room, and extracts concentration variation and diffusion direction characteristic parameters. Using a pre-built smoke spread trend prediction model, it applies Kalman filtering to the concentration variation and outputs the smoke diffusion probability within a future time interval. Simultaneously, it extracts the primary and secondary diffusion angles from the diffusion direction characteristic parameters and calculates their energy ratio to obtain a directional stability factor. A linear combination of the smoke diffusion probability and the directional stability factor is then performed to generate a smoke diffusion assessment value.

[0065] The Multi-Source Fusion Determination Module receives the temperature anomaly index and smoke diffusion assessment value, retrieves the temperature anomaly data stored in the fire parameter acquisition module, and calculates the temperature impact compensation value using a preset correction coefficient. The temperature anomaly index, smoke diffusion assessment value, and temperature impact compensation value are normalized to generate a fire risk fusion index. A fire risk determination threshold is set. If the fusion index is below the threshold, a fire extinguishing activation signal is generated; if it is above the threshold, a system optimization signal is generated.

[0066] Control Parameter Generation Module: This module matches fire extinguishing strategies based on received fire extinguishing activation signals or system optimization signals. If it's a fire extinguishing activation signal, it triggers the injection control command, dynamically adjusting parameters like the extinguishing agent's release rate and injection angle to generate the extinguishing agent injection parameters. If it's a system optimization signal, it triggers the ventilation control command, optimizing the ventilation system's airflow intensity and response time to generate the ventilation control parameters.

[0067] Example 1: The system is equipped with an equipment coupling analysis module. The core function of this module is to monitor the temperature coordination between the communication equipment and the computer room environment to obtain relevant parameters that can reflect the matching status of the equipment and the ambient temperature, and generate equipment coupling evaluation values through specific calculation logic, providing a more comprehensive fire risk analysis basis for the multi-source fusion judgment module.

[0068] During implementation, the device coupling analysis module first uses thermocouple sensors deployed on the surface of communications equipment and in the equipment room environment to collect two key data types in real time: the heat dissipation rate from the equipment surface and the temperature change rate of the equipment room environment. These two types of data, representing the device's own heat generation status and the ambient temperature dynamics, respectively, serve as the basis for evaluating the compatibility between the two. Thermocouple sensors offer fast response and high measurement accuracy, ensuring that the collected data accurately and promptly reflects the actual situation.

[0069] After collecting the device surface heat dissipation rate and the ambient temperature change rate, the module calculates the difference between these two parameters. Specifically, the difference between the two rates is first calculated, and then the absolute value of this difference is taken to obtain the heat dissipation rate difference. This process aims to eliminate the positive and negative effects of the rate difference, focusing only on the magnitude of the difference. For example, if the device surface heat dissipation rate is 5 degrees Celsius per minute and the ambient temperature change rate is 3 degrees Celsius per minute, the difference between the two is 2 degrees Celsius. After taking the absolute value, the heat dissipation rate difference is 2 degrees Celsius. If the device heat dissipation rate is lower than the ambient temperature change rate, the difference is negative, and after taking the absolute value, the difference is also positive. A larger heat dissipation rate difference indicates a more significant mismatch between the device heat dissipation and the ambient temperature change, which may indicate abnormal device operation or problems with ambient temperature control, thereby increasing the risk of fire.

[0070] At the same time, the device coupling analysis module also extracts thermal stress distribution data for the areas where the device surface contacts the environment. Thermal stress is internal stress caused by temperature fluctuations, and its distribution reflects the efficiency of heat transfer and the strength of the interaction between the device and the environment. Using stress sensors or other thermal stress detection devices, the module monitors thermal stress in key areas where the device surface contacts the environment (such as where the device housing meets the chassis and near heat dissipation vents), obtaining thermal stress values at each point. Based on this data, the module calculates the maximum and average thermal stress values within that area and calculates their ratio, which is labeled the thermal stress concentration factor. For example, if the maximum thermal stress value in a region is 100 MPa and the average thermal stress value is 50 MPa, the thermal stress concentration factor is 2. If the maximum and average thermal stress values are close, the thermal stress distribution is relatively uniform and the concentration factor is low. Conversely, a significant concentration of thermal stress in a localized area may lead to localized overheating of the device and increase the risk of fire.

[0071] After obtaining the heat dissipation rate difference and thermal stress concentration factor, the device coupling analysis module integrates these two parameters to generate a comprehensive assessment of the compatibility between the device and the ambient temperature, known as the device coupling assessment. This integration process uses a weighted summation approach. Specifically, the heat dissipation rate difference and the thermal stress concentration factor are each assigned a weight based on their respective impact on fire risk. The weighting factors are determined based on historical data on equipment operation in the communications room and a fire risk analysis model. This determination typically requires extensive experimentation and data analysis to ensure that the device coupling assessment accurately reflects the actual temperature compatibility. For example, if analysis indicates that the heat dissipation rate difference contributes 60% to the fire risk and the thermal stress concentration factor contributes 40%, the calculation formula for the device coupling assessment can be expressed as "heat dissipation rate difference × 0.6 + thermal stress concentration factor × 0.4." This weighted summation approach integrates the two distinct parameters into a single comprehensive indicator, facilitating subsequent fire risk analysis.

[0072] The device coupling assessment value generated by the device coupling analysis module is transmitted to the multi-source fusion determination module, serving as a key basis for secondary corrections to the fire risk fusion index. After receiving the temperature characteristic anomaly index, smoke diffusion assessment value, and temperature impact compensation value, the multi-source fusion determination module first generates a preliminary fire risk fusion index through normalization. At this point, the module retrieves the device coupling assessment value provided by the device coupling analysis module and performs a secondary correction on the preliminary fusion index. This correction can be achieved by adjusting the weight coefficients used in the normalization calculation or by introducing a correction term based on the device coupling assessment value. For example, a high device coupling assessment value indicates poor temperature coordination between the equipment and the environment, potentially posing a potential fire risk. In this case, the multi-source fusion determination module will increase the weight of temperature-related parameters in the fusion index calculation or directly add a correction value positively correlated with the device coupling assessment value to the preliminary fusion index, ensuring that the final fire risk fusion index more comprehensively reflects the actual fire risk status of the computer room. On the contrary, if the device coupling assessment value is low, it means that the device is well coordinated with the ambient temperature. The module may reduce the weight of the temperature parameter or introduce a negative correction term to make the fusion index closer to the actual risk level.

[0073] Through the collaborative work of the device coupling analysis module and the multi-source fusion judgment module, the system can deeply analyze fire risks from the perspective of the interaction between equipment and the environment, avoiding the limitations of judging based solely on direct fire parameters such as temperature and smoke. This multi-dimensional data fusion and risk correction mechanism makes the system's assessment of fire risks more comprehensive and accurate, and can detect potential fire hazards caused by the mismatch between equipment and ambient temperature in advance, providing a more reliable basis for subsequent fire extinguishing strategy formulation and system optimization. For example, when the device coupling assessment value shows that the heat dissipation rate of equipment in a certain area is abnormal and the thermal stress concentration coefficient is high, even if the current direct fire parameters such as temperature anomaly and smoke concentration characteristic value have not yet exceeded the threshold, the system can improve the fire risk fusion index through secondary correction, issue early warning signals, and prompt staff to promptly investigate equipment failures and adjust ambient temperature control strategies, thereby effectively reducing the possibility of fire.

[0074] In summary, the device coupling analysis module provides an additional risk analysis dimension for the system by providing real-time monitoring and quantitative assessment of the temperature compatibility between communications equipment and the computer room environment, enhancing the comprehensiveness and accuracy of fire monitoring. Combined with the multi-source fusion judgment module, this module forms a multi-layered risk assessment system, encompassing both individual devices and the overall environment. This further enhances the reliability and effectiveness of the edge computing-based intelligent fire monitoring and extinguishing system for communications rooms.

[0075] Example 2: The dynamic feedback execution module set up by the system has the core function of adjusting the data sampling frequency of the monitoring system in real time based on the fire extinguishing agent injection parameters and ventilation adjustment parameters output by the control parameter generation module, and feeding back the execution results to the fire parameter acquisition module to form a complete closed-loop control link to achieve dynamic tracking and accurate response to the fire risk in the communication room.

[0076] During operation, the dynamic feedback execution module first receives two types of parameters from the control parameter generation module: fire extinguishing agent injection parameters and ventilation adjustment parameters. These parameters correspond to the execution instructions of the fire extinguishing activation signal and system optimization signal, respectively. Based on the specific content of the parameters, the dynamic feedback execution module triggers the corresponding data sampling and adjustment mechanism.

[0077] When the received fire extinguishing agent injection parameters involve adjustment of the injection angle, the system has determined that the fire extinguishing agent's injection direction needs to be changed to optimize the fire extinguishing effect. At this point, the dynamic feedback execution module simultaneously triggers the smoke density sensor's sampling density adjustment mechanism. Smoke density, a key parameter reflecting the fire's development, is particularly important for monitoring accuracy after adjusting the fire extinguishing agent's injection angle. For example, if the injection angle is adjusted to focus injection on a specific area in the computer room, the smoke diffusion path in that area may change due to the fire extinguishing agent. In this case, more intensive smoke density data collection is required in and around that area to monitor the fire extinguishing effect and dynamic changes in smoke diffusion in real time. Specifically, the module controls the smoke density sensor to increase the sampling frequency, shorten the data collection interval, or temporarily activate backup sensor nodes within the existing sensor layout to increase the amount of data collected per unit time. This adjustment ensures that the system promptly detects subtle changes in smoke density, providing more accurate data support for subsequent adjustments to the fire extinguishing strategy.

[0078] When the received ventilation adjustment parameters involve optimizing airflow intensity, the system needs to adjust the ventilation system's airflow intensity to improve air circulation within the computer room. This may affect the temperature distribution and smoke diffusion rate. In this case, the dynamic feedback execution module will simultaneously increase the number of temperature detection channels. Temperature, another core parameter for fire monitoring, can exhibit complex gradient variations in temperature distribution across different areas of the computer room when airflow intensity changes. The existing number of temperature detection channels may not be sufficient for comprehensive monitoring. For example, increasing the ventilation system's airflow intensity can cause the temperature around a heat source to rapidly diffuse, creating new areas of abnormal temperature. To promptly capture these changes, the module activates more temperature detection channels, such as activating redundant temperature sensors or expanding the monitoring range of existing sensors, to ensure high-density, comprehensive, real-time monitoring of temperature distribution in critical areas within the computer room. Increasing the number of temperature detection channels improves the spatial resolution of temperature data, enabling the system to more accurately identify areas of abnormal temperature and hotspots. This provides richer raw data for the temperature signature analysis module and enhances the accuracy of fire risk assessment.

[0079] After adjusting the smoke concentration sensor sampling density and the number of temperature detection channels, the dynamic feedback execution module will input the adjusted parameter information into the fire parameter acquisition module. Upon receiving the new parameters, the fire parameter acquisition module will restart the monitoring process according to the updated sampling rules. For example, if the smoke concentration sensor sampling frequency is increased from once per minute to three times per minute, the fire parameter acquisition module will collect smoke concentration data in real time based on this new frequency. If the number of temperature detection channels increases from 10 to 15, the module will simultaneously receive temperature data from the newly added channels and incorporate it into the construction of the computer room fire status parameter set. This closed-loop control mechanism enables the system to dynamically adjust monitoring strategies based on current fire extinguishing or ventilation adjustment operations, forming a "monitoring-determination-execution-feedback-re-monitoring" cycle to ensure continuous tracking and response to computer room fire risks.

[0080] The closed-loop control link of the dynamic feedback execution module is also reflected in the real-time feedback of the execution results. After adjusting the sampling parameters and restarting the monitoring process, the module will continue to collect monitoring data output by the fire parameter acquisition module and analyze whether this data reflects the improvement of the fire extinguishing effect or the change of environmental parameters after ventilation adjustment. For example, if the smoke concentration data shows a clear downward trend in concentration after the fire extinguishing agent is injected, it indicates that the current injection parameters are effective; if the temperature data shows that the hot spot in a certain area is not effectively controlled, it may be necessary to further adjust the fire extinguishing agent injection parameters or ventilation adjustment parameters. This feedback mechanism enables the system to promptly identify potential problems and adjust strategies based on the actual execution effect, avoiding monitoring lags or data loss caused by fixed sampling modes, thereby improving the response speed and adaptability of the entire system.

[0081] In addition, the design of the dynamic feedback execution module fully considers the complexity of the communication room environment and the uncertainty of fire development. By adjusting the sampling frequency and the number of detection channels in real time, the system can maximize the use of existing sensor resources without increasing hardware costs, and realize dynamic focused monitoring of key parameters. For example, under normal monitoring conditions, the sensor operates at a standard frequency and number of channels to reduce system energy consumption and data processing pressure; during fire extinguishing or ventilation adjustment operations, the system automatically switches to a high-density sampling mode, concentrating resources on monitoring the parameters most affected by the operation, ensuring that high-quality monitoring data is obtained during critical periods. This differentiated sampling strategy not only ensures the efficient operation of the system, but also improves the accuracy of fire emergency response.

[0082] Example 3: In the fire risk determination and analysis process of the fire parameter acquisition module, the calculation of temperature anomaly, smoke density characteristic value and current anomaly comprehensive index is involved. Each parameter realizes the quantitative assessment of fire risk through specific logic.

[0083] The calculation of temperature anomaly is based on a comparative analysis of the temperature gradient parameters collected in real time within the target communication room and the ambient temperature. Temperature sensors deployed within the room continuously collect temperature gradient data from different areas and simultaneously obtain a preset ambient temperature baseline value. The temperature anomaly calculation formula is:

[0084]

[0085] Among them, T 梯度 It represents the temperature gradient parameter of the target communication room collected in real time, reflecting the temperature distribution difference and change trend of each area in the room; T 环境 This is a preset ambient temperature baseline value, typically determined based on the ambient temperature range during normal equipment room operation, and is used to represent the standard temperature state of the equipment room environment. This formula calculates the relative deviation of the temperature gradient from the ambient temperature and takes its absolute value, quantifying the degree of temperature anomaly as a percentage. A larger value indicates a more significant deviation from normal temperature distribution and a higher fire risk.

[0086] The extraction of smoke concentration characteristic values combines two key parameters: smoke particle density and change rate. Smoke sensors monitor the smoke particle density (number of smoke particles per unit volume) and its rate of change over time in the computer room in real time. In specific calculations, the particle density is denoted as ρ, in particles per cubic meter; the change rate is denoted as v, in particles per (cubic meter per second). The smoke concentration characteristic value is calculated by geometrically synthesizing the vector modulus using the following formula:

[0087]

[0088] This formula treats particle density and rate of change as two orthogonal components of a vector. Using a modulo operation, it yields a value that comprehensively characterizes the dynamic characteristics of smoke concentration. Particle density reflects the current concentration of smoke, while the rate of change reflects the speed of smoke generation or diffusion. The square root of the sum of the squares of these two values highlights the combined impact of high concentrations or high rates of change on fire risk. Larger values indicate more pronounced smoke concentration anomalies and a higher likelihood of fire.

[0089] The calculation of the current anomaly comprehensive index is to perform normalized weighted processing on the peak current and fluctuation frequency in the equipment operating current parameters. The peak current I during the equipment operation is collected by the current sensor. peak and fluctuation frequency f, and obtain the average current I of the device under normal operating conditions avg , Maximum current I max , minimum current I min , and the average value of the fluctuation frequency f avg , maximum value of fluctuation frequency f max , minimum value of fluctuation frequency f min First, the peak current and the fluctuation frequency are normalized and converted into dimensionless values in the interval [0,1]. The formulas are:

[0090]

[0091] The purpose of normalization is to eliminate the influence of different parameter dimensions and value ranges, making the two comparable. Subsequently, according to the degree of influence of peak current and fluctuation frequency on fire risk, weight coefficients α and β are assigned to them respectively (α + β = 1, and α and β are both decimals greater than 0). The current anomaly comprehensive index is obtained by weighted summation, and the formula is:

[0092] Current anomaly comprehensive index = α·normalized peak current + β·normalized fluctuation frequency

[0093] The peak current reflects the instantaneous load state of the equipment during operation. Excessively high peak values may indicate potential faults such as short circuits or overloads. The fluctuation frequency reflects the stability of the current. Abnormal frequency fluctuations may indicate poor contact between internal components or control system anomalies. The weighting coefficients α and β are set based on the equipment type, historical operating data, and the fire risk analysis model. For example, for high-power communications equipment, the peak current weight α may be set to 0.6, and the fluctuation frequency weight β to 0.4 to emphasize the dominant role of peak current anomalies in fire risk. The current anomaly comprehensive index ranges from [0, 1]. Larger values indicate more significant deviations from normal current parameters and a higher risk of fire caused by current anomalies.

[0094] After completing the calculation of the above three parameters, the fire parameter acquisition module compares the temperature anomaly, smoke density characteristic value, and current anomaly comprehensive index with their respective preset thresholds. These thresholds are determined by analyzing historical fire data in the communication room, equipment safety operation parameters, and environmental monitoring standards. For example, the temperature anomaly threshold may be set at 15%, and the smoke density characteristic value threshold may be set at 200 (particles / cubic meter). 2 / Second 2 The square root of the current anomaly comprehensive index threshold is set to 0.7. When any parameter exceeds the corresponding threshold, the module determines that there is a fire risk, generates a fire warning signal, and triggers a collaborative monitoring instruction to activate the temperature signature analysis module and the smoke dynamic detection module for further in-depth analysis of temperature and smoke parameters.

[0095] This multi-parameter collaborative judgment mechanism avoids the limitations of single-parameter monitoring. For example, when the temperature anomaly does not exceed the threshold but the current anomaly index is significantly elevated, the system can still identify potential fire hazards through the abnormal current parameters and issue an early warning. In another example, when the smoke concentration characteristic value suddenly increases but the temperature is not yet significantly abnormal, the system can respond promptly based on the change in smoke parameters to provide an early warning of fire. By organically combining temperature, smoke, and current parameters, the fire parameter acquisition module can comprehensively assess fire risks from multiple dimensions, including equipment operating status and environmental physical characteristics, improving the comprehensiveness and accuracy of monitoring.

[0096] Furthermore, methods such as normalization and weighted summation employed in the parameter calculation process ensure the comparability and integration of parameters of different properties. For example, temperature anomaly is expressed as a percentage, smoke density characteristic values are expressed in composite physical units, and the current anomaly comprehensive index is a dimensionless value. These three are uniformly converted into a "whether the risk threshold is exceeded" decision logic through threshold comparison, enabling the system to quickly and intuitively generate fire warning signals. This standardized parameter processing process not only facilitates the system's algorithm implementation and logical control but also lays the foundation for subsequent multi-source data integration.

[0097] The fire parameter acquisition module utilizes precise parameter definitions, scientific calculation formulas, and rigorous decision logic to dynamically monitor and quantitatively assess fire risks in communications rooms. The calculation of temperature anomaly, smoke density characteristic values, and a comprehensive current anomaly index establishes a comprehensive risk monitoring system from the perspectives of temperature field distribution, smoke diffusion, and equipment electrical safety. The coordinated work of these three elements enables the system to capture fire precursor information promptly and accurately, providing a reliable basis for subsequent fire prevention and control measures.

[0098] Example 4: In the temperature signature analysis module, extracting temperature distribution characteristic parameters and quantitatively assessing hotspot distribution in key areas of a target communications room requires deploying a specific sensor array and executing a multi-level data processing flow. For example, a communications room is divided into key areas such as power supply equipment, server cabinets, and patch panels. Each area is equipped with an infrared sensor array consisting of multiple high-precision infrared temperature sensors distributed in a grid pattern across the target area. These arrays collect temperature data at each point in real time and transmit it to the module backend.

[0099] The infrared sensor array continuously collects temperature distribution data of key areas, and the system generates temperature heat maps and gradient change maps based on the collected data. The temperature heat map uses different color gradients to intuitively display the temperature distribution in the area. For example, blue is set to represent 20℃-25℃ (normal temperature range), yellow represents 25℃-30℃ (warning temperature range), and red represents above 30℃ (high temperature abnormal range). Suppose in the temperature heat map of the server cabinet area, a certain area appears to be a large area of red, indicating that there is a significant high temperature abnormality in the area. The gradient change map reflects the spatial temperature change rate and gradient amplitude in the form of curves or contour lines. For example, the greater the temperature difference between adjacent sensor nodes, the higher the slope of the curve or the density of the contour lines, which means the temperature gradient is more significant.

[0100] Two key parameters are extracted from the temperature heatmap: the hotspot area percentage and the maximum temperature. The hotspot area percentage refers to the ratio of the area where the temperature exceeds a preset high temperature threshold (e.g., 30°C) to the area of the entire critical area. For example, if the total area of a server cabinet area is 20 square meters and the hotspot area is 5 square meters, the hotspot area percentage is 25%. The maximum temperature is the highest temperature detected within the hotspot area, such as if the surface temperature of a cabinet in that area reaches 35°C. The module multiplies these two parameters: 25% × 35°C = 8.75. The inverse of this product yields the hotspot distribution coefficient, 1 / 8.75 ≈ 0.114. This coefficient reflects the combined influence of the concentration of hotspots and the extreme temperature. A smaller value indicates a larger hotspot area percentage and higher temperatures, indicating a higher fire risk.

[0101] The temperature rise rate and spatial gradient amplitude are extracted from the gradient change map. The temperature rise rate is the temperature increase per unit time and is calculated using continuously collected temperature data. For example, if the temperature in a certain area rises from 28°C to 33°C in 5 minutes, the temperature rise rate is (33°C - 28°C) / 5 minutes = 1°C / minute. The spatial gradient amplitude represents the temperature difference within a unit spatial distance. For example, if two adjacent sensor nodes are 0.5 meters apart and their temperatures are 30°C and 25°C, respectively, the spatial gradient amplitude is (30°C - 25°C) / 0.5 meters = 10°C / meter. The module takes the arithmetic mean of these two parameters, i.e., (1°C / minute + 10°C / meter) / 2 = 5.5, to obtain the gradient characteristic index. This index characterizes the dynamic trend and spatial distribution of temperature changes. A larger value indicates a more drastic temperature change and a higher potential fire risk.

[0102] The module performs a weighted fusion of the hotspot distribution coefficient and the gradient characteristic index to generate a temperature characteristic anomaly index. This weighted fusion process assigns different weights to the two parameters, determined based on the fire risk characteristics of each area in the communications room. For example, in the power supply area, due to the high power consumption and concentrated heat generation of the equipment, the hotspot distribution coefficient might be weighted 0.6 and the gradient characteristic index 0.4. In the server cabinet area, due to the high heat dissipation requirements and the more dynamic temperature distribution, the weights might be adjusted to 0.5:0.5. Assuming the weights are set to 0.5:0.5 in the server cabinet area example above, the temperature characteristic anomaly index = 0.114 × 0.5 + 5.5 × 0.5 = 0.057 + 2.75 = 2.807. This value comprehensively reflects the degree of anomaly in the hotspot distribution and temperature gradient. A larger value indicates a more significant temperature characteristic anomaly, requiring further integration with other parameters to assess the fire risk.

[0103] In actual applications, the execution process of the temperature feature analysis module needs to be linked with the fire parameter acquisition module. When the fire parameter acquisition module generates a fire warning signal and triggers the collaborative monitoring instruction, the temperature feature analysis module will start in-depth analysis to avoid invalid calculations caused by daily environmental fluctuations. For example, if the air-conditioning system in the computer room temporarily fails during a certain period of time, causing the overall ambient temperature to rise by 2°C, the temperature anomaly calculated by the fire parameter acquisition module may not exceed the threshold (such as 15%), so the collaborative monitoring instruction will not be triggered, and the temperature feature analysis module will remain dormant, and only the fire parameter acquisition module will continue to monitor the basic temperature data.

[0104] Furthermore, the deployment density of the infrared sensor array needs to be adjusted based on the regional risk level. For high-risk areas (such as corners where power equipment is concentrated), the sensor spacing can be reduced to 0.3 meters to improve the spatial resolution of temperature data. For low-risk areas (such as patch panels), the spacing can be increased to 1 meter to optimize hardware resource allocation. The sensor sampling frequency can also be dynamically adjusted. Under normal monitoring conditions, data is collected once per second. Upon receiving a collaborative monitoring command, the sampling frequency is increased to 5 times per second to ensure that even subtle temperature changes are captured.

[0105] Generating temperature heat maps and gradient change maps requires the local data processing capabilities of edge computing nodes. Edge computing nodes have built-in image processing algorithms that receive sensor data and render graphics in real time, eliminating the latency associated with transmitting raw data to the cloud. For example, if a temperature anomaly is detected in a server cabinet area, the edge node can generate the latest heat map and gradient map within 50 milliseconds and simultaneously transmit them to the multi-source fusion judgment module, supporting rapid fire risk assessment.

[0106] Through the above process, the temperature signature analysis module achieves a quantitative assessment of temperature anomalies in key areas of the communications room. Its core approach is to visualize heat maps and gradient maps, combined with the numerical calculation of hotspot distribution and temperature change rates, to form a multi-dimensional characterization of temperature risk. This analysis method not only locates specific areas of temperature anomalies but also provides standardized input for subsequent multi-source data fusion in the form of characteristic indices, ensuring the logical coherence and assessment accuracy of the entire fire monitoring system.

[0107] Example 5: In the dynamic smoke detection module, predicting and analyzing the spread of smoke in a target communications room requires combining real-time smoke particle concentration data and diffusion direction characteristic parameters to achieve a multi-dimensional assessment through model calculation and logical analysis. For example, a communications room has a rectangular interior space with a long side of 15 meters, a short side of 10 meters, and a ceiling height of 3 meters. The dynamic smoke detection module collects real-time smoke particle concentration data using multiple laser particle counters deployed on the ceiling, while also using a wind direction sensor to monitor the direction of smoke diffusion.

[0108] The laser particle counter collects smoke particle concentration data at different locations within the room once per second, extracting characteristic parameters for concentration variation and diffusion direction. The concentration variation is the absolute difference between the concentrations at two adjacent sampling moments. For example, if the concentration at time t is 500 particles / cubic meter and at time t+1 is 600 particles / cubic meter, the concentration variation is 100 particles / cubic meter. Diffusion direction characteristic parameters are obtained through the wind direction sensor and include the primary diffusion angle and the secondary diffusion angle. The primary diffusion angle is the angle between the main direction of smoke diffusion and due north (e.g., 60°), while the secondary diffusion angle is the angle between the secondary diffusion direction and the primary diffusion angle (e.g., ±30°), reflecting the multidirectional nature of smoke diffusion.

[0109] The smoke spread trend prediction model is built based on the Kalman filter algorithm. Using state and observation equations, the model recursively estimates concentration fluctuations to predict the probability of smoke spread within future time intervals. Assuming the currently collected concentration fluctuation sequence is [80, 100, 120] particles per cubic meter (corresponding to data from the past three seconds), the model calculates the mean and variance of the historical data and predicts that the concentration fluctuation within the next five seconds will likely be between 110 and 130 particles per cubic meter. The model then outputs the smoke spread probability. The spread probability is expressed as a percentage, with higher values indicating a greater likelihood that the smoke will spread over a larger area within the predicted time period. For example, a 75% probability of smoke spread within the next five seconds indicates a 75% chance that smoke will cover more than 50% of the equipment room area.

[0110] The main diffusion angle and secondary diffusion angle in the diffusion direction characteristic parameters are used to calculate the directional stability factor. The main diffusion angle represents the dominant direction of smoke diffusion, and the secondary diffusion angle represents the range of secondary diffusion directions accompanying the main direction. For example, if the main diffusion angle is 60° and the secondary diffusion angle is ±30°, the energy in the main diffusion direction can be calculated by trigonometric functions (such as cos60°=0.5), and the energy in the secondary diffusion direction is the sum of the squares of cos(60°±30°), that is, cos30° 2 +cos90° 2 ≈0.75 + 0 = 0.75. The directional stability factor is the ratio of the main diffusion angle energy to the total energy, i.e., 0.5 / (0.5 + 0.75) = 0.4. This factor ranges from 0 to 1. A larger value indicates a more concentrated smoke diffusion direction and greater stability. A smaller value indicates a more dispersed diffusion direction, which may be affected by factors such as airflow disturbances, resulting in a more complex fire spread path.

[0111] The smoke spread assessment value is generated by a linear combination of the smoke spread probability and the directional stability factor. The weighting of this linear combination is determined based on the characteristics of the equipment room environment. For example, if the ventilation system is operating stably and the smoke spread direction is relatively controllable, the smoke spread probability weight can be set to 0.6, and the directional stability factor weight can be set to 0.4. If the airflow in the equipment room is turbulent (e.g., due to uneven airflow caused by an air conditioning malfunction), the weights can be adjusted to 0.4:0.6 to emphasize the impact of directional stability on the risk assessment. Assuming in the above example, the smoke spread probability is 75% (i.e., 0.75), the directional stability factor is 0.4, and the weights are set to 0.5:0.5, the smoke spread assessment value is = 0.75 × 0.5 + 0.4 × 0.5 = 0.375 + 0.2 = 0.575. This value comprehensively reflects the probability of smoke spread and directional stability. Higher values indicate more significant smoke spread and a higher fire risk.

[0112] In actual applications, the execution of the dynamic smoke detection module needs to be coordinated with the fire parameter acquisition module and the temperature characteristic analysis module. For example, when the fire parameter acquisition module generates a fire warning signal due to an excessive current abnormality comprehensive index, the dynamic smoke detection module initiates high-frequency data acquisition (the sampling frequency is increased from 1 time per second to 5 times per second) to track the dynamic changes of smoke diffusion in real time. If the temperature characteristic analysis module detects a significant decrease in the hot spot distribution coefficient in the server cabinet area (for example, from 0.114 to 0.08), indicating that the high-temperature area has expanded or the temperature has increased, the calculation of the smoke diffusion assessment value may automatically increase the influence weight of the temperature-related parameters to reflect the synergistic effect of temperature and smoke.

[0113] When deploying laser particle counters, consider the obstruction effects of the equipment layout within the computer room. For example, in aisles between server cabinets, where obstructions can create "shadow zones" for smoke diffusion, sensors should be deployed more densely (e.g., with a detection point every 2 meters) to ensure there are no blind spots. Wind direction sensors should be installed in the center of the room's ceiling to avoid obstructions by walls or equipment that could affect monitoring accuracy.

[0114] The parameters of the smoke spread prediction model require regular calibration. For example, the Kalman filter model's initial state vector and covariance matrix are updated monthly based on historical monitoring data to adapt to changes in smoke diffusion characteristics caused by factors such as seasonal changes and equipment load adjustments. If the airflow direction in the computer room's heating mode in winter is opposite to that in cooling mode in summer, the model can automatically adjust the prediction logic for the diffusion direction based on historical data learning.

[0115] The dynamic smoke detection module enables real-time monitoring and trend prediction of smoke diffusion. Its core is to use the Kalman filter algorithm to process concentration change data, combined with a multi-dimensional analysis of the diffusion direction, to form a quantitative assessment of smoke risk. This analysis method not only predicts the spatial range of smoke diffusion, but also assesses the controllability of the diffusion path through the directional stability factor, providing key input for the multi-source fusion judgment module, helping the system formulate targeted fire extinguishing or ventilation strategies. For example, when the smoke diffusion assessment value is high and the directional stability factor is low, the system can prioritize adjusting the airflow direction of the ventilation system to suppress the spread of smoke to high-risk areas. If the assessment value is high and the direction is stable, the directional injection strategy of the fire extinguishing agent is triggered to intercept the predicted diffusion path.

[0116] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0117] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent fire monitoring and extinguishing system for communication rooms based on edge computing, characterized in that: include: The fire parameter acquisition module is used to dynamically collect the temperature gradient, smoke concentration parameters and equipment operating current of the target communication room to obtain the room fire status parameter set. Based on the room fire status parameter set, the fire risk is determined and analyzed, and a fire warning signal is generated. The generated fire warning signal triggers the collaborative monitoring instruction, and the temperature feature analysis module and the smoke dynamic detection module are executed according to the triggered collaborative monitoring instruction. The temperature characteristic analysis module is used to extract the temperature distribution characteristic parameters of the key areas of the target communication room, quantitatively evaluate the hot spot distribution of the temperature abnormality area, and obtain the temperature characteristic abnormality index; The smoke dynamic detection module is used to extract the smoke particle concentration fluctuation parameters in the target communication room, predict and analyze the spread trend of the smoke diffusion state, and obtain the smoke diffusion assessment value; The multi-source fusion judgment module receives the temperature characteristic anomaly index and smoke diffusion assessment value, performs collaborative analysis on the fire risk in the computer room, and generates fire extinguishing activation signals and system optimization signals. The control parameter generation module is used to receive the fire extinguishing start signal and the system optimization signal, match the fire extinguishing strategy, and generate the fire extinguishing agent injection parameters and ventilation adjustment parameters.

2. According to claim 1, an edge computing-based intelligent fire monitoring and extinguishing system for communication rooms is characterized in that: The fire risk determination and analysis includes: By collecting the temperature gradient parameters of the target communication room in real time, the deviation percentage from the ambient temperature is calculated and marked as the temperature anomaly degree; Extract the particle density and change rate from the smoke concentration parameters of the target communication room, calculate the geometric composite vector modulus of the two, and mark it as the smoke concentration characteristic value; Collect the peak current and fluctuation frequency of the equipment's operating current parameters, perform normalized weighted calculations, and obtain a comprehensive current anomaly index; The temperature anomaly, smoke concentration characteristic value and current anomaly comprehensive index are compared with the preset thresholds respectively. When any parameter exceeds the corresponding threshold, a fire warning signal is generated.

3. The edge computing-based intelligent fire monitoring and extinguishing system for communication rooms according to claim 1 is characterized in that: The quantitative evaluation of the hotspot distribution in the temperature anomaly area includes: By deploying an infrared sensor array to collect temperature distribution data, a temperature thermogram and gradient change map are generated; Extract the hotspot area ratio and the maximum temperature value from the temperature heat map, calculate the product of the two and take the inverse to obtain the hotspot distribution coefficient; The temperature rise rate and spatial gradient amplitude are extracted from the gradient change map, and the arithmetic mean of the two is calculated and marked as the gradient characteristic index; The hotspot distribution coefficient and the gradient characteristic index are weighted and fused to obtain the temperature characteristic anomaly index.

4. The edge computing-based intelligent fire monitoring and extinguishing system for communication rooms according to claim 1 is characterized in that: The prediction and analysis of the spread trend of smoke diffusion includes: Collect smoke particle concentration data of the target communication room in real time and extract characteristic parameters of concentration variation and diffusion direction; Construct a smoke spread trend prediction model, input the concentration change amplitude into the model for Kalman filtering, and output the smoke spread probability in the future time interval; Extract the main diffusion angle and the secondary diffusion angle from the diffusion direction characteristic parameters, calculate the energy ratio between the two and obtain the directional stability factor; The smoke diffusion probability is linearly combined with the directional stability factor to obtain the smoke diffusion assessment value.

5. The edge computing-based intelligent fire monitoring and extinguishing system for communication rooms according to claim 1 is characterized in that: The collaborative analysis of the fire risk in the computer room includes: Retrieve temperature anomaly data, set its correction coefficient, and obtain the temperature impact compensation value through calculation; Normalize the values of the temperature characteristic anomaly index, smoke diffusion assessment value, and temperature impact compensation value to generate a fire risk fusion index; Set a fire risk determination threshold. If the fusion index is lower than the threshold, a fire extinguishing start signal is generated. If it is higher than the threshold, a system optimization signal is generated.

6. The edge computing-based intelligent fire monitoring and extinguishing system for communication rooms according to claim 1 is characterized in that: The fire extinguishing strategy matching includes: If the fire extinguishing start signal is captured, the injection control instruction is triggered, and the release rate and injection angle parameters of the fire extinguishing agent are dynamically adjusted according to the instruction to generate the fire extinguishing agent injection parameters; If the system optimization signal is captured, the ventilation adjustment instruction is triggered, and the airflow intensity and response time parameters of the ventilation system are optimized according to the instruction to generate ventilation adjustment parameters.

7. The edge computing-based intelligent fire monitoring and extinguishing system for communication rooms according to claim 1 is characterized in that: Also includes: The device coupling analysis module is used to monitor the temperature coordination between communication equipment and the equipment room environment, extract the difference in equipment heat dissipation rate and thermal stress concentration factor, and generate a device coupling assessment value; The multi-source fusion determination module further combines the equipment coupling evaluation value to perform a secondary correction on the fire risk fusion index.

8. The edge computing-based intelligent fire monitoring and extinguishing system for communication rooms according to claim 7 is characterized in that: The monitoring of the temperature coordination between the communication equipment and the computer room environment includes: Thermocouple sensors are used to collect the heat dissipation rate of the device surface and the rate of change of the ambient temperature. The difference between the two rates is calculated and the absolute value is taken, which is marked as the heat dissipation rate difference value. Extract the thermal stress distribution data of the area in contact between the equipment surface and the environment, and calculate the ratio of the maximum thermal stress value to the average thermal stress, which is marked as the thermal stress concentration coefficient; The heat dissipation rate difference value and the thermal stress concentration factor are weighted and summed to generate the equipment coupling assessment value.

9. The edge computing-based intelligent fire monitoring and extinguishing system for communication rooms according to claim 1 is characterized in that: Also includes: The dynamic feedback execution module is used to adjust the data sampling frequency of the monitoring system in real time according to the generated fire extinguishing agent injection parameters and ventilation adjustment parameters, and feed back the execution results to the fire parameter acquisition module to form a closed-loop control link.

10. The edge computing-based intelligent fire monitoring and extinguishing system for communication rooms according to claim 9 is characterized in that: The specific execution process of the dynamic feedback execution module includes: If the extinguishing agent injection parameters involve adjustment of the injection angle, the sampling density of the smoke concentration sensor should be increased simultaneously; If the ventilation adjustment parameters involve optimizing the airflow intensity, the number of temperature detection channels should be increased simultaneously; Input the adjusted parameters into the fire parameter acquisition module and restart the monitoring process.

Images