Fire control method and fire control system

Abstract

This invention provides a fire control method and a fire protection system. The fire control method includes: real-time acquisition of normalized values ​​of temperature, smoke concentration, VOC concentration, and the highest temperature zone in thermal imaging; calculating a comprehensive fire risk score based on the normalized values ​​of temperature, smoke concentration, VOC concentration, and the highest temperature zone in thermal imaging; and determining whether to execute fire extinguishing operations based on the comprehensive fire risk score. According to the fire control method of this invention, the risk of false triggering can be significantly reduced while ensuring the accuracy of fire response.

Description

Technical Field

[0001] This invention relates to the field of fire protection and liquid cooling control technology, and more specifically, to a fire control method and a fire protection system. Background Technology

[0002] In existing immersion thermal management and fire protection systems, fire detection generally relies on independent threshold judgment mechanisms using a single or a small number of sensors. For example, a fire suppression response is triggered solely by an overheating temperature sensor or excessive smoke concentration. While this method is simple in structure, it is highly susceptible to false alarms due to environmental interference (such as instantaneous temperature rise during equipment startup, suspended dust, or brief release of volatile organic compounds), leading to unnecessary system switching, media waste, or even equipment downtime. Even when some systems attempt to use multi-sensor collaborative judgment, they still lack a dynamic evaluation mechanism for the reliability of sensor signals, failing to effectively distinguish between a real fire and a single-point abnormal disturbance. Especially in high-density, high-electromagnetic-interference environments such as data centers and energy storage battery compartments, the false alarm rate of a single sensor remains high, making it difficult to ensure the accuracy of fire response while reducing the risk of false triggering. Summary of the Invention

[0003] The main objective of this invention is to provide a fire control method and fire protection system that can significantly reduce the risk of false triggering while ensuring the accuracy of fire response.

[0004] To achieve the above objectives, according to one aspect of the present invention, a fire control method is provided, comprising: acquiring in real time temperature, smoke concentration, VOC concentration, and normalized values ​​of the highest temperature zone in thermal imaging; calculating a comprehensive fire risk score based on the normalized values ​​of temperature, smoke concentration, VOC concentration, and the highest temperature zone in thermal imaging; and determining whether to perform fire extinguishing operations based on the comprehensive fire risk score.

[0005] Furthermore, in the step of calculating the comprehensive fire risk score based on temperature, smoke concentration, VOC concentration, and the normalized value of the highest temperature zone in thermal imaging, the calculation method for the comprehensive fire risk score is: Score = (λ1 × T_norm + λ2 × S_norm + λ3 × V_norm + λ4 × I_norm) × K_time × K_duration × K_lock × 100; where λ1-λ4 are weighting coefficients, and satisfy λ1+λ2+λ3+λ4=1, 0≤K_time≤1, 0≤K_duration≤1, 0≤K_lock≤1; T_norm The following parameters are used: 1. Temperature sensor normalization value, calibrated based on the ratio of actual temperature to threshold; 2. S_norm, smoke concentration normalization value, 1 when smoke concentration ≥ set threshold; 3. V_norm, VOC concentration normalization value, 1 when VOC concentration ≥ set threshold; 4. I_norm, the highest temperature zone normalization value output by the thermal imaging module, calibrated based on heat source temperature distribution; 5. K_time, time decay factor, increasing with alarm duration; 6. K_duration, continuous anomaly factor, approaching 1 for consecutive n cycles of exceeding the limit, and approaching 0 for occasional pulses; 7. K_lock, single sensor lockout factor, approaching 0 when only one sensor is abnormal.

[0006] Furthermore, the steps for determining whether to perform fire extinguishing operations based on the comprehensive fire risk score include: performing a single sensor false trigger isolation judgment on the fire protection system; if the warning condition is 1, then activate the warning and increase the monitoring frequency, and do not perform fire extinguishing operations; if the warning condition is 0, and the Score ≥ the set value, then perform fire extinguishing operations; wherein, when the warning condition is 1, the normalized value of any one parameter is 1, and the normalized value of the other three parameters is less than 0.5, and in other cases, the warning condition is 0.

[0007] Furthermore, the fire control method also includes: obtaining the extinguishing agent flow rate when performing fire extinguishing operations; and adjusting the extinguishing agent injection orifice diameter according to the extinguishing agent flow rate, temperature, and heat source area.

[0008] Furthermore, the extinguishing agent injection orifice diameter is calculated using the following formula:

[0009] D_final = min(max(D_base + K_adaptive × (V_current - V_nominal) + Scorr, 0.2), D_max); where D_final is the final injection orifice diameter in mm, D_base is the base orifice diameter in mm, D_max is the maximum orifice diameter, K is in mm, K_adaptive is the adaptive adjustment coefficient, V_current is the current real-time flow velocity in m / s, V_nominal is the rated flow velocity in m / s, and Scorr is the heat source area correction term in mm.

[0010] Furthermore, Scorr is calculated as follows: when S_current > S1, Scorr = A, where A is a positive value and its range is 0.1mm to 0.5mm; when S_current < S2, Scorr = B, where B is a negative value and its range is -0.5mm to -0.1mm.

[0011] Furthermore, the step of adjusting the extinguishing agent injection orifice diameter according to the extinguishing agent flow rate, temperature, and heat source area also includes: based on the current liquid level L_curren and the liquid level threshold L_limit of the fire-fighting space; when L_curren < L_limit, setting the injection enable flag to 1 to control the normal injection of the extinguishing agent; when L_curren ≥ L_limit, setting the injection enable flag to 0 to stop the injection of the extinguishing agent.

[0012] Furthermore, if the warning condition is 1, the steps to activate the warning and increase the monitoring frequency without performing fire extinguishing operations include: when a single sensor anomaly is detected or Sc1≤Score≤Sc2; increasing the sampling frequency; controlling the cooling flow rate to increase; uploading the warning information and not activating fire spraying; where Sc1 is the first set threshold for the comprehensive fire risk score and Sc2 is the second set threshold for the comprehensive fire risk score.

[0013] Furthermore, if the warning condition is 0 and Score ≥ set value, the steps for performing fire extinguishing operations include: when at least two sensor abnormalities are detected or Sc2 ≤ Score ≤ Sc3; the control valve group enters pre-switching preparation; the spray orifice diameter is adjusted through a thermal-fluid coupling algorithm; and the current liquid level L_curren of the fire-fighting space is monitored in real time; where Sc2 is the second set threshold of the comprehensive fire risk score and Sc3 is the third set threshold of the comprehensive fire risk score.

[0014] Furthermore, if the warning condition is 0 and the Score is greater than or equal to the set value, the steps for performing the fire extinguishing operation include: when the Score is detected to be greater than or equal to Sc3; immediately switch to fire mode; adjust the nozzle diameter to the maximum until the liquid level in the fire-fighting space reaches the liquid level threshold L_limit; and forcibly stop spraying; where Sc3 is the third set threshold of the comprehensive fire risk score.

[0015] Furthermore, the fire control method also includes: monitoring whether the system has entered the fire mode; when the system has not entered the fire mode, controlling the cooling branch to be open and the fire branch to be closed through the control valve, so that the shared coolant is cooled and dissipated from the shared storage tank through the cooling branch; when the system enters the fire mode, switching the pipeline status through the control valve, blocking the cooling branch and opening the fire branch, so that the shared coolant is injected into the fire space from the shared storage tank through the fire branch.

[0016] According to another aspect of the present invention, a fire protection system is provided, applied to the above-described fire control method, comprising: a data acquisition module for real-time acquisition of temperature, smoke concentration, VOC concentration, and normalized values ​​of the highest temperature zone in thermal imaging; a risk score calculation module for calculating a comprehensive fire risk score based on the normalized values ​​of temperature, smoke concentration, VOC concentration, and the highest temperature zone in thermal imaging; and a control decision module for determining whether to perform fire extinguishing operations based on the comprehensive fire risk score.

[0017] Furthermore, the fire control system includes: a shared storage tank for containing shared coolant; a shared circulating water pump connected to the shared storage tank; a cooling branch for forming a cooling circuit; a fire branch with spray nozzles communicating with the fire-fighting space; and control valves, through which the cooling branch and the fire branch are optionally connected to the shared circulating water pump.

[0018] According to embodiments of the present invention, a fire control method achieves accurate fire determination by integrating multi-source sensor data. The method first acquires in real-time normalized values ​​of temperature, smoke concentration, VOC concentration, and the highest temperature zone in thermal imaging. These four dimensions correspond to different physical characteristics during the fire process. Subsequently, the system calculates a comprehensive fire risk score based on the normalized values ​​of these four dimensions. This process transforms discrete single-sensor data into a holistic risk quantification index, thus comprehensively reflecting the current safety status of the environment. Finally, the system determines whether to execute fire extinguishing operations based on the comprehensive fire risk score, achieving closed-loop control from data acquisition to decision execution. Compared to existing technologies that rely solely on a single sensor for fire detection, this solution effectively avoids false alarms caused by fluctuations in a single parameter (such as high temperature without a fire source or smoke not generated by a fire) through the comprehensive representation of multi-dimensional data. This significantly improves the accuracy and reliability of fire determination and ensures the precision of fire extinguishing actions. Attached Figure Description

[0019] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0020] Figure 1 This is a flowchart of a fire control method according to an embodiment of the present invention;

[0021] Figure 2 This is a flowchart of the fire risk scoring and mode switching control method of the fire control method according to an embodiment of the present invention;

[0022] Figure 3 This is a flowchart of the heat flow coupling injection and liquid level interlocking control method of the fire protection control method according to an embodiment of the present invention. Detailed Implementation

[0023] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0024] See also Figures 1 to 3 As shown, according to an embodiment of the present invention, the fire control method includes:

[0025] Real-time acquisition of temperature, smoke concentration, VOC concentration, and normalized values ​​of the highest temperature zone in thermal imaging.

[0026] In this embodiment, the focus is first on the synchronous acquisition and data standardization of multi-dimensional environmental parameters. Through continuous data acquisition, data on temperature, smoke concentration, VOC concentration, and the highest temperature zone in thermal imaging are extracted. Subsequently, the raw measurement results are converted into normalized values ​​at a unified scale to eliminate differences in parameter ranges and measurement standards. These steps enable real-time quantitative characterization of heat source features and combustion products, transforming heterogeneous physical signals into a comparable, unified data format. This provides a direct, synchronous, and dimensionally consistent input for cross-validation and comprehensive analysis of multiple parameters.

[0027] A comprehensive fire risk score is calculated based on temperature, smoke concentration, VOC concentration, and the normalized value of the highest temperature zone in thermal imaging.

[0028] In this embodiment, the system acquires monitoring data for temperature, smoke concentration, VOC concentration, and the highest temperature zone in thermal imaging, and converts the raw values ​​into normalized values ​​with unified dimensions. Subsequently, these four normalized values ​​are used as input parameters for centralized calculation to synthesize a quantitative index characterizing the likelihood of a fire in the current environment—the comprehensive fire risk score. These steps, by standardizing and aggregating multi-dimensional environmental state parameters, achieve a unified quantitative expression of multi-source monitoring data. This allows fire risk assessment to be based on the synergistic changes of multiple key indicators, avoiding interference from fluctuations in a single parameter and providing an objective and accurate basis for risk level judgment.

[0029] Whether or not to carry out firefighting operations is determined based on the comprehensive fire risk assessment.

[0030] In this embodiment, the system directly obtains a comprehensive fire risk score, analyzes the current situation based on the score's numerical status, and then concludes whether fire suppression operations need to be initiated. This score directly characterizes the overall risk level of the current environment. By matching the real-time generated score with the judgment criteria, it can clearly output instructions to execute or not execute operations. This step directly establishes a correspondence between the score value and the system's response action, ensuring that the triggering decision for fire suppression operations strictly depends on the calculation results of the comprehensive fire risk score, thus achieving automated action determination based on quantitative assessment results.

[0031] By applying the technical solution of this embodiment, the fire control method achieves accurate fire determination by integrating multi-source sensor data. The method first acquires the normalized values ​​of temperature, smoke concentration, VOC concentration, and the highest temperature zone in thermal imaging in real time. These four dimensions correspond to different physical characteristics during the fire process. Subsequently, the system calculates a comprehensive fire risk score based on the normalized values ​​of these four dimensions. This process transforms discrete single-sensor data into a holistic risk quantification index, thus comprehensively reflecting the current safety status of the environment. Finally, the system determines whether to execute fire extinguishing operations based on the comprehensive fire risk score, achieving closed-loop control from data acquisition to decision execution. Compared to existing technologies that rely solely on a single sensor for fire detection, this solution effectively avoids false alarms caused by fluctuations in a single parameter (such as high temperature without a fire source or smoke not generated by a fire) through the comprehensive representation of multi-dimensional data. This significantly improves the accuracy and reliability of fire determination and ensures the precision of fire extinguishing actions.

[0032] In one embodiment, the method for calculating the comprehensive fire risk score based on temperature, smoke concentration, VOC concentration, and the normalized value of the highest temperature zone in thermal imaging is as follows:

[0033] Score=(λ1×T_norm+λ2×S_norm+λ3×V_norm+λ4×I_norm)×K_time×K_duration×K_lock×100;

[0034] Where λ1-λ4 are weight coefficients, and satisfy λ1+λ2+λ3+λ4=1, 0≤K_time≤1, 0≤K_duration≤1, 0≤K_lock≤1;

[0035] T_norm is the normalized value of the temperature sensor, calibrated based on the ratio of the actual temperature to the threshold.

[0036] S_norm is the smoke concentration normalization value, which is 1 when the smoke concentration is ≥ the set threshold;

[0037] V_norm is the normalized value of VOC concentration, which is 1 when the VOC concentration is greater than or equal to the set threshold.

[0038] I_norm is the normalized value of the highest temperature range output by the thermal imaging module, which is calibrated based on the temperature distribution of the heat source.

[0039] K_time is the time decay factor, which increases with the duration of the alarm.

[0040] K_duration is a persistent abnormality factor, which approaches 1 for continuous n-cycle exceedances and approaches 0 for occasional pulses.

[0041] K_lock is the single-sensor locking factor, which approaches 0 when only one sensor malfunctions.

[0042] In this embodiment, T_norm is the normalized value of the temperature sensor, ranging from 0 to 1. It is calibrated according to the ratio between the actual monitored temperature and the preset safety threshold (e.g., when the preset safety threshold is 80℃, the actual monitored temperature of 60℃ corresponds to 0.75), used to eliminate the original dimensional differences. S_norm is the normalized value of smoke concentration, ranging from 0 to 1. It is assigned a value of 1 when the real-time monitored smoke concentration reaches or exceeds the set threshold; otherwise, it is converted according to the actual concentration ratio, used to quantify the influence of smoke indicators on fire judgment. V_norm is the normalized value of VOC concentration, ranging from 0 to 1. VOC is an abbreviation for Volatile Organic Compounds. It is assigned a value of 1 when the real-time monitored volatile organic compound concentration reaches or exceeds the set threshold, used to characterize the level of characteristic gases generated by early thermal runaway of equipment or decomposition of insulating materials. I_norm is the normalized value of the highest temperature zone of thermal imaging, ranging from 0 to 1. Based on the spatial temperature field data collected by the thermal imaging module, the highest temperature zone value is selected and calibrated according to the heat source temperature distribution characteristics to reflect the local overheating intensity. K_time is the time decay factor, with a value range of 0 ≤ K_time ≤ 1. This factor increases with the duration of the alarm state and is mainly used to filter transient interference signals, ensuring that only persistent anomalies are accumulated and identified as real fires. K_duration is the continuous anomaly factor, with a value range of 0 ≤ K_duration ≤ 1, used to verify the temporal stability of the abnormal state. When the parameter exceeds the limit for n consecutive control cycles, this factor approaches 1; if it is only an occasional pulse signal, it approaches 0, thus suppressing false alarms. K_lock is the single-sensor interlocking factor, with a value range of 0 ≤ K_lock ≤ 1. K_lock is an anti-false alarm logic parameter. When only one sensor in the system strongly alarms while the readings of the other three sensors are all at low levels, this factor is forced to approach 0, thereby blocking the activation of the fire extinguishing system.

[0043] In this embodiment, when calculating the comprehensive fire risk score, the contribution of each sensor is allocated by weight coefficients. Combined with the time decay factor to filter out transient interference, the continuous abnormal factor to confirm the continuity of fire heat release, and the single sensor blocking factor to forcibly block the single-point false alarm signal path, the mathematical model and logic control are coupled, thereby constructing multiple layers of protection against accidental touches, which significantly improves the anti-interference ability and judgment accuracy of fire determination under complex dynamic working conditions.

[0044] For example, when a data center server rack is operating under high load, the environmental monitoring terminal acquires current spatial parameters according to a fixed sampling period and performs a comprehensive fire risk score calculation. The system first maps the ratio of the actual temperature to a calibrated threshold as T_norm, compares the smoke concentration with a set threshold, and directly assigns a value of 1 to obtain S_norm when the smoke concentration is ≥ the set threshold. It then compares the VOC concentration with a set threshold, and directly assigns a value of 1 to obtain V_norm when the VOC concentration is ≥ the set threshold. Finally, it calibrates the highest temperature zone data output by the thermal imaging module based on the heat source temperature distribution pattern as I_norm. Subsequently, the system loads approximately λ1+λ2+λ3+λ4=1. The weighting coefficients λ1 to λ4 of the bundle are used to perform a weighted summation of the normalized data above. Combined with K_time, which increases with the alarm duration, K_duration, which approaches 1 for continuous n-cycle exceedances and 0 for occasional pulses, and K_lock, which approaches 0 when only one sensor is abnormal, multiple maintenance positive compensation is performed. Finally, all variables are completely substituted into the mathematical model Score=(λ1×T_norm+λ2×S_norm+λ3×V_norm+λ4×I_norm)×K_time×K_duration×K_lock×100 to calculate the comprehensive fire risk score and complete the quantitative assessment of the potential fire risk level under this working condition.

[0045] In one embodiment, the step of determining whether to perform firefighting operations based on a comprehensive fire risk score includes:

[0046] Perform single-sensor false trigger isolation judgment on the fire protection system;

[0047] If the warning condition is 1, the warning will be activated and the monitoring frequency will be increased, but no fire extinguishing operations will be performed;

[0048] If the warning condition is 0 and the Score is greater than or equal to the set value, then fire extinguishing operation will be performed.

[0049] When the warning condition is 1, the normalized value of any one parameter is 1, and the normalized values ​​of the other three parameters are less than 0.5. In other cases, the warning condition is 0.

[0050] In this embodiment, the warning condition has two logical values, 0 and 1. When the normalized value of any one parameter is 1 and the normalized value of the other three parameters is less than 0.5, the warning condition is assigned the value of 1. Otherwise, the warning condition is assigned the value of 0.

[0051] During fire control, if a sudden increase in the normalized value of any single parameter is detected while all other parameters are below the set safe range, the fire extinguishing command is automatically intercepted, the sampling frequency is increased, and an early warning is issued. This fault-tolerant isolation logic effectively avoids the problem of false triggering by a single sensor, maintaining the system's high sensitivity response while eliminating erroneous operations caused by data silos, thus improving the accuracy and reliability of fire assessment.

[0052] When only one sensor alarms strongly while the readings of the other three sensors are low, the system forcibly suppresses the fire extinguishing command by using a single sensor lockout factor (K_lock approaches 0), only increasing the monitoring frequency and issuing an early warning, thereby preventing false triggering caused by a single sensor failure or momentary interference.

[0053] When the warning condition is 1, after the warning is activated and the detection frequency is increased, the monitoring continues for time t1. If the fire judgment threshold is still not reached, it indicates that the fire risk is low. At this time, the warning can be lifted and the fire situation can continue to be monitored.

[0054] The continuous monitoring time t1 ranges from 25s to 35s. In one embodiment, t1 = 30s.

[0055] For example, in the operation monitoring of data center energy storage cabinets, the control unit performs single-sensor false trigger isolation judgments on the fire protection system at fixed intervals. If the normalized value of any currently collected parameter is 1 and the normalized values ​​of the other three parameters are all less than 0.5, the system determines that the warning condition is 1, and then activates the warning and increases the monitoring frequency, without performing any fire extinguishing operations. Conversely, if the single high-value feature does not meet the condition, resulting in a warning condition of 0, and the fire risk comprehensive score Score calculated by the algorithm is greater than or equal to the set value, the system directly determines that the action threshold is met and immediately performs fire extinguishing operations, thereby completing the automated decision-making process from anomaly identification to precise intervention.

[0056] In one embodiment, the fire control method further includes:

[0057] When performing fire extinguishing operations, obtain the extinguishing agent flow rate;

[0058] Adjust the nozzle diameter of the extinguishing agent injection based on the extinguishing agent flow rate, temperature, and heat source area.

[0059] The extinguishing agent is delivered to the fire terminal through the fire protection pipeline network. The nozzle orifice diameter is adjusted according to the three parameters of "flow rate, temperature and heat source area" to spray evenly onto the heat source area. The fire is then continuously fed back through comprehensive scoring calculation to finally complete the fire extinguishing.

[0060] This embodiment introduces a real-time flow velocity feedback loop and uses temperature and heat source area characteristics as auxiliary input variables to establish a closed-loop control link of "operating condition perception - dynamic compensation - actuator response". It adjusts the extinguishing agent injection orifice diameter in real time, so that the nozzle injection pattern matches the actual fire scene scale and hydraulic state in real time, realizing intelligent adjustment of the injection orifice diameter. This avoids coverage blind spots or excessive consumption of media caused by fixed injection patterns, and significantly improves the utilization rate and thermal suppression efficiency of extinguishing agents.

[0061] For example, in the scenario of thermal runaway of battery cells in a closed energy storage compartment, when the system confirms the trigger and executes the fire extinguishing operation, the pipeline monitoring module collects the medium flow status in real time to obtain the fire extinguishing agent flow rate. At the same time, the sensor network synchronously outputs the real-time temperature of the fault location and the combustion range identified by thermal imaging as the heat source area. The controller then performs comprehensive calculations based on the obtained fire extinguishing agent flow rate, current temperature and heat source area, and dynamically generates an opening control signal to adjust the fire extinguishing agent injection aperture, thereby achieving adaptive and precise matching between the fire extinguishing agent dosage and the scale of the fire on site.

[0062] In one embodiment, the extinguishing agent injection orifice diameter is calculated using the following formula:

[0063] D_final=min(max(D_base+K_adaptive×(V_current-V_nominal)+Scorr,0.2),D_max);

[0064] Where D_final is the final injection orifice diameter in mm, D_base is the base orifice diameter in mm, D_max is the maximum orifice diameter, K is in mm, K_adaptive is the adaptive adjustment coefficient, V_current is the current real-time flow rate in m / s, V_nominal is the rated flow rate in m / s, and Scorr is the heat source area correction term in mm.

[0065] In this embodiment, a dynamic calculation model incorporating a reference orifice diameter, linear compensation for flow velocity deviation, and heat source area correction is constructed to achieve refined adaptive control of the extinguishing agent injection orifice diameter. The K_adaptive×(V_current-V_nominal) term is used to offset the effects of pipeline pressure fluctuations or flow anomalies in real time, ensuring automatic orifice diameter adjustment to maintain optimal injection conditions when the flow velocity deviates from the rated value. Simultaneously, the Scorr term dynamically corrects the orifice diameter based on the heat source area, achieving precise matching of "larger orifice for larger fires, smaller orifice for smaller fires," thereby improving extinguishing agent utilization and coverage efficiency. Furthermore, by introducing a dual clamping mechanism with a 0.2 mm lower limit and a D_max upper limit, the risk of blockage due to excessively small orifice diameter or insufficient pressure due to excessively large orifice diameter under extreme conditions is effectively avoided, significantly enhancing the system's stability, safety, and extinguishing efficiency in complex fire scenarios.

[0066] For example, in the application scenario of energy storage batteries, when the energy storage battery needs to be controlled for fire protection, the control system collects the real-time flow rate (V_current) of the extinguishing agent and compares it with the rated flow rate (V_nominal). At the same time, it calculates the heat source area correction term (Scorr) based on the detected heated surface size. Then, the processor uses the base aperture (D_base) as the reference value, adds the product of the adaptive adjustment coefficient (K_adaptive) and the flow rate deviation and the heat source area correction term (Scorr), and limits the accumulated result through the lower limit threshold of 0.2 and the upper limit threshold of the maximum aperture (D_max). Finally, it calculates the final injection aperture (D_final) that adapts to the current flow field and heat distribution, thereby completing the closed-loop dynamic adjustment of the actuator opening.

[0067] In one embodiment, Scorr is calculated as follows:

[0068] When S_current > S1, Scorr = A, where A is a positive value and its range is 0.1mm to 0.5mm;

[0069] When S_current < S2, Scorr = B, where B is a negative value and its range is -0.5mm to -0.1mm;

[0070] Otherwise, Scorr=0.

[0071] In this embodiment, when adjusting the injection aperture, when the heat source area S_current > S1, it indicates a large fire. At this time, a positive correction value A needs to be directly applied to rapidly increase the injection aperture to match the fire extinguishing requirements of large flow rate and large coverage area, and improve the suppression ability for large-scale fires. When the heat source area S_current < S2, it indicates a small fire or no fire. At this time, a correction value B (usually B is negative) needs to be applied to make the injection aperture smaller and reduce the fire extinguishing agent flow rate, so as to achieve fine control. In the middle interval where S2 ≤ S_current ≤ S1, Scorr = 0, which means the system keeps the basic aperture unchanged. This "dead zone" design prevents the system from frequently fine-tuning the aperture during medium fire fluctuations, and avoids control oscillations caused by sensor noise or slight fire fluctuations.

[0072] By adopting the above control method, the control logic can be simplified, and the real-time performance and reliability of the injection aperture adjustment can be improved. Compared with complex continuous functions, this threshold-based piecewise logic has extremely small computational amount and is easy to execute quickly in an embedded controller, ensuring a low-latency response of the control instruction. The parameters A, B, S1, and S2 are all discrete constants, which are convenient for on-site debugging and calibration, and reduce the complexity of system parameter tuning.

[0073] By setting two thresholds S1 and S2 (usually S1 > S2, forming a hysteresis interval or an intermediate stable zone), the small fluctuations in the heat source area measurement are effectively filtered. For example, when the fire is slightly jittering near the critical point, the system will not switch frequently, thus protecting the actuator from wear caused by frequent actions and improving the long-term operation reliability of the system.

[0074] In one embodiment, the steps of adjusting the fire extinguishing agent injection aperture according to the fire extinguishing agent flow rate, temperature, and heat source area further include:

[0075] According to the current liquid level L_current and the liquid level threshold L_limit of the fire protection space;

[0076] When L_current < L_limit, set the injection enable flag to 1 and control the normal injection of the fire extinguishing agent; [[ID=I7]]

[0077] When L_current ≥ L_limit, set the injection enable flag to 0 and stop the injection of the fire extinguishing agent.

[0078] In this embodiment, by establishing a fire extinguishing agent supply strategy based on real-time liquid level monitoring, a safe closed-loop control of the submerged fire protection system is achieved, fundamentally eliminating the risks of submerged container overflow, pressure exceeding limits, and media waste caused by continuous spraying. Specifically, by comparing the current liquid level L_current with the threshold L_limit in real time, the state of the spray enable flag En is dynamically controlled to ensure that adaptive spraying based on the "flow rate-temperature-heat source area" algorithm is maintained when the liquid level has not reached the upper limit, while the spray flow is immediately and forcibly cut off when the liquid level reaches the upper limit. Through the above control strategy, not only is the effectiveness of the fire extinguishing process guaranteed, but the inherent safety and operational reliability of the system under closed submerged conditions are also significantly improved, avoiding the problems of submerged container overflow, pressure exceeding limits, and media waste that may be caused by the lack of liquid level feedback in traditional solutions.

[0079] In one embodiment, if the warning condition is 1, the steps of activating the warning and increasing the monitoring frequency, without performing fire extinguishing operations, include:

[0080] When a single sensor anomaly is detected or Sc1≤Score≤Sc2;

[0081] Increase the sampling frequency;

[0082] Control the cooling flow rate to increase;

[0083] Upload warning information but do not activate fire sprinklers;

[0084] Sc1 is the first set threshold for the comprehensive fire risk score, and Sc2 is the second set threshold for the comprehensive fire risk score.

[0085] In this embodiment, by constructing a "warning-monitoring-heat dissipation" response mechanism, a shift from passive fire suppression to proactive prevention and early intervention is achieved. When the fire risk is in the critical range (Sc1≤Score≤Sc2) or there is a risk of false alarms from a single sensor, the system does not directly trigger costly fire suppression actions. Instead, it increases the sampling frequency to accurately confirm the situation and simultaneously increases the cooling flow rate to enhance heat dissipation capacity, thereby suppressing the tendency for thermal runaway at the source. By adopting the above control strategy, the risk of false triggering caused by noise or transient interference from a single sensor is effectively avoided, fire extinguishing agents are saved, and system downtime and maintenance are reduced. Furthermore, enhanced cooling enables early intervention against potential hazards, demonstrating the system's high fault tolerance and optimal energy efficiency in handling uncertain risks.

[0086] In one embodiment, if the warning condition is 0 and the score is greater than or equal to the set value, the steps for performing the fire extinguishing operation include:

[0087] When at least two sensor anomalies are detected or Sc2≤Score≤Sc3;

[0088] The control valve assembly enters pre-switching preparation;

[0089] The injection orifice diameter is adjusted using a thermal-fluid coupling algorithm;

[0090] Real-time monitoring of the current liquid level L_current in the fire-fighting space;

[0091] Sc2 is the second set threshold for the comprehensive fire risk score, and Sc3 is the third set threshold for the comprehensive fire risk score.

[0092] In this embodiment, a multi-verification mechanism is introduced when the warning condition is 0 and the comprehensive score reaches the minimum threshold. This mechanism addresses dual-sensor anomalies or scores within the range of Sc2≤Score≤Sc3, effectively avoiding false alarms from a single sensor and significantly improving the robustness of fire assessment. Simultaneously, the valve group pre-switching preparation steps ensure execution stability. The thermal-fluid coupling algorithm dynamically adjusts the injection orifice diameter to achieve precise matching and efficient delivery of the extinguishing medium. Real-time liquid level monitoring ensures sufficient resources and closed-loop control, thereby achieving rapid, accurate, and reliable automated handling of early or moderate-risk fire conditions while avoiding overreaction and secondary damage.

[0093] In one embodiment, if the warning condition is 0 and the score is greater than or equal to the set value, the steps for performing the fire extinguishing operation include:

[0094] When a score ≥ Sc3 is detected;

[0095] Immediately switch to fire mode;

[0096] Adjust the spray nozzle diameter to the maximum until the liquid level in the fire-fighting space reaches the liquid level threshold L_limit;

[0097] Forcefully stop spraying;

[0098] Sc3 is the third threshold for the comprehensive fire risk score.

[0099] In this embodiment, for scenarios where the fire risk score meets the requirement of Score≥Sc3, an open-loop control strategy is adopted that immediately switches to fire-fighting mode and adjusts to the maximum spray orifice diameter. This abandons complex real-time algorithm adjustments and achieves millisecond-level response and maximum flow output with extremely simple logic, ensuring that the fire can be suppressed quickly and intensely when the fire is critical. At the same time, by setting a clear liquid level threshold as a forced stop condition, the risk of overflow of the extinguishing medium, excessive pressure, and waste of medium due to continuous spraying is effectively avoided while ensuring that the extinguishing medium fully covers the container to prevent reignition.

[0100] The aforementioned method of cooling and fire control based on early warning conditions and comprehensive scores can form a three-level linkage, realizing full-process autonomy from "early warning → fine-tuning → strong spraying → shutdown spraying". It can take appropriate measures at different stages of the disaster, which can not only effectively suppress the occurrence of the disaster, but also avoid the waste of media and improve fire safety and reliability.

[0101] During fire extinguishing using a heat-fluid coupling method, continuous feedback assessment based on the comprehensive fire risk score is required. When the comprehensive fire risk score falls below the safety threshold, the fire suppression system is reset, completing the fire extinguishing operation. If the comprehensive fire risk score remains above the safety threshold, it indicates that fire extinguishing is not yet complete, and fire extinguishing operations must continue, using extinguishing agents to further eliminate fire risks.

[0102] In one embodiment, Sc1 = 60°C, Sc2 = 80°C, and Sc3 = 90°C.

[0103] In one embodiment, the values ​​of Sc1, Sc2, and Sc3 can also be adjusted as needed.

[0104] In one embodiment, the fire control method further includes:

[0105] Whether the monitoring system has entered fire protection mode;

[0106] When the system is not in fire-fighting mode, the cooling branch is opened by the control valve and the fire-fighting branch is blocked, so that the shared coolant is cooled and dissipated from the shared storage tank through the cooling branch.

[0107] When the system enters fire protection mode, the control valve switches the pipeline status, blocking the cooling branch and opening the fire protection branch, so that the shared coolant is injected into the fire protection space from the shared storage tank through the fire protection branch.

[0108] In this embodiment, by constructing a shared "cooling-firefighting" pipeline and an intelligent valve-controlled switching mechanism, resource reuse and system simplification of firefighting and cooling media are achieved, significantly reducing hardware costs and space occupation. The firefighting system has dual-mode adaptive capability. In non-fire conditions, it uses shared coolant for regular heat dissipation to ensure efficient equipment operation. When switching to firefighting mode in an emergency, it can quickly block the cooling path and open the firefighting path, ensuring that the shared coolant (or special fire extinguishing agent) can be injected into the fire scene with zero delay and precise direction. This avoids the redundant construction brought about by traditional independent firefighting systems, and the control valve ensures rapid and reliable switching between cooling mode and firefighting mode.

[0109] According to an embodiment of the present invention, a fire protection system is applied to the above-described fire control method, comprising:

[0110] The data acquisition module is used to acquire normalized values ​​of temperature, smoke concentration, VOC concentration, and the highest temperature zone in thermal imaging in real time.

[0111] The risk scoring calculation module is used to calculate a comprehensive fire risk score based on the temperature, smoke concentration, VOC concentration, and the normalized value of the highest temperature zone in thermal imaging.

[0112] The control decision module is used to determine whether to perform fire extinguishing operations based on the comprehensive fire risk score.

[0113] In this embodiment, by constructing a closed-loop feedback link consisting of a data acquisition module, a risk scoring calculation module, and a control decision module, a technological leap from single threshold triggering to multi-dimensional parameter weighted fusion in fire judgment is achieved. By utilizing the normalization and weighted calculation of four-dimensional data—temperature, smoke, VOC, and thermal imaging—the fire characteristics can be comprehensively and objectively reflected, effectively overcoming the problem of false alarms or missed alarms caused by the susceptibility of single sensors to environmental interference. At the same time, control decisions based on comprehensive scoring enable the system to make refined judgments based on changes in fire risk levels, providing accurate data support for subsequent differentiated responses such as early warning, graded spraying, or full-volume fire suppression based on scoring thresholds. This significantly improves the accuracy and reliability of the system's judgment while ensuring the timeliness of fire suppression.

[0114] In one embodiment, the fire protection system further includes:

[0115] Shared storage tank for holding shared coolant;

[0116] A shared circulating water pump is connected to the shared storage tank;

[0117] Cooling branch circuits are used to form a cooling loop.

[0118] Fire branch lines have spray nozzles that connect to the fire-fighting space;

[0119] The control valve allows the cooling branch and the fire-fighting branch to be selectively connected to the shared circulating water pump.

[0120] In this embodiment, by constructing an integrated fluid network consisting of a shared storage tank, circulating water pump, cooling branch, and fire-fighting branch, the reuse and pipeline sharing of fire-fighting and cooling media are realized, which greatly simplifies the system hardware architecture, eliminates the pipeline redundancy and equipment duplication of the two independent systems of fire-fighting and heat dissipation in the traditional solution, and significantly reduces the initial investment cost, space occupation, and subsequent operation and maintenance complexity. At the same time, by using the selective connection mechanism of control valves for cooling branch and fire-fighting branch, the basis for fluid path switching of the system under different operating conditions is established, which provides the necessary hardware support for efficient heat dissipation under normal conditions and rapid switching to fire spray mode in emergency situations, resulting in high system integration and optimized resource utilization.

[0121] In one embodiment, the control valve includes a three-way valve and a two-way isolation valve.

[0122] In this embodiment, the fire protection system adopts a parallel architecture controlled by a three-way valve, including a fire branch, a cooling branch, a shared circulating water pump, a shared storage tank, and a two-way isolation valve. Both AC and DC systems use DC water pumps, allowing emergency power to directly start the pumps for submersion fire suppression. During normal operation, the system is in cooling mode. The three-way valve guides the shared coolant through the circulating water pump in the cooling branch and shared piping network, while the two-way isolation valve keeps the cooling branch open and the fire branch closed. When a fire occurs, the three-way valve quickly switches the flow direction, and simultaneously, the two-way isolation valve, with its electromagnetically driven two-way sealing structure, rapidly reverses its state (closing the cooling branch and opening the fire branch), allowing the medium in the same storage tank to be immediately injected into the fire scene as a extinguishing agent. This ensures rapid and accurate delivery of the extinguishing agent, realizing the reuse of coolant and extinguishing agent media and the sharing of the piping network. The coordinated action of the valves also ensures rapid and reliable switching between modes.

[0123] After the control valve completes the pipeline switching, the AC liquid chiller immediately switches to DC emergency power supply mode, while the DC liquid chiller power supply remains unchanged and the DC-powered water pump continues to run, providing the pressure required for fire fighting. At the same time, the controller pushes information such as fire location, heat source area, and switching status to the fire control room. The extinguishing agent is delivered to specific points through the shared pipeline network to achieve accurate and rapid fire extinguishing at the fire source location.

[0124] In one embodiment, the fire protection system further includes a temperature sensor, a smoke sensor, a VOC sensor, and a thermal imaging module. The temperature sensor detects the temperature of targets in the fire protection area, the smoke sensor detects the smoke concentration in the fire protection area, the VOC sensor detects the VOC concentration in the fire protection area, and the thermal imaging module detects the temperature distribution of heat sources in the fire protection area.

[0125] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0126] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein.

[0127] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A fire control method, characterized in that, include: Real-time acquisition of temperature, smoke concentration, VOC concentration, and normalized values ​​of the highest temperature zone in thermal imaging; A comprehensive fire risk score is calculated based on temperature, smoke concentration, VOC concentration, and the normalized value of the highest temperature zone in thermal imaging. Whether or not to carry out firefighting operations is determined based on the comprehensive fire risk assessment.

2. The fire control method according to claim 1, characterized in that, In the steps of calculating the comprehensive fire risk score based on temperature, smoke concentration, VOC concentration, and the normalized value of the highest temperature zone in thermal imaging, the calculation method for the comprehensive fire risk score is as follows: Score=(λ1×T_norm+λ2×S_norm+λ3×V_norm+λ4×I_norm)×K_time×K_duration×K_lock×100; Where λ1-λ4 are weight coefficients, and satisfy λ1+λ2+λ3+λ4=1, 0≤K_time≤1, 0≤K_duration≤1, 0≤K_lock≤1; T_norm is the normalized value of the temperature sensor, calibrated based on the ratio of the actual temperature to the threshold. S_norm is the smoke concentration normalization value, which is 1 when the smoke concentration is ≥ the set threshold; V_norm is the normalized value of VOC concentration, which is 1 when the VOC concentration is greater than or equal to the set threshold. I_norm is the normalized value of the highest temperature range output by the thermal imaging module, which is calibrated based on the temperature distribution of the heat source. K_time is the time decay factor, which increases with the duration of the alarm. K_duration is a persistent abnormality factor, which approaches 1 for continuous n-cycle exceedances and approaches 0 for occasional pulses. K_lock is the single-sensor locking factor, which approaches 0 when only one sensor malfunctions.

3. The fire control method according to claim 2, characterized in that, The steps for determining whether to conduct firefighting operations based on a comprehensive fire risk assessment include: Perform single-sensor false trigger isolation judgment on the fire protection system; If the warning condition is 1, the warning will be activated and the monitoring frequency will be increased, but no fire extinguishing operations will be performed; If the warning condition is 0 and the Score is greater than or equal to the set value, then fire extinguishing operation will be performed. When the warning condition is 1, the normalized value of any one parameter is 1, and the normalized values ​​of the other three parameters are less than 0.

5. In other cases, the warning condition is 0.

4. The fire control method according to claim 1, characterized in that, Fire control methods also include: When performing fire extinguishing operations, obtain the extinguishing agent flow rate; Adjust the nozzle diameter of the extinguishing agent injection based on the extinguishing agent flow rate, temperature, and heat source area.

5. The fire control method according to claim 4, characterized in that, The nozzle diameter for extinguishing agent injection is calculated using the following formula: D_final=min(max(D_base+K_adaptive×(V_current-V_nominal)+Scorr,0.2),D_max); Where D_final is the final injection orifice diameter in mm, D_base is the base orifice diameter in mm, D_max is the maximum orifice diameter, K is in mm, K_adaptive is the adaptive adjustment coefficient, V_current is the current real-time flow rate in m / s, V_nominal is the rated flow rate in m / s, and Scorr is the heat source area correction term in mm.

6. The fire control method according to claim 5, characterized in that, Scorr is calculated as follows: When S_current > S1, Scorr = A, where A is a positive value and its range is 0.1mm to 0.5mm; When S_current < S2, Scorr = B, where B is a negative value and its range is -0.5mm to -0.1mm; Otherwise, Scorr=0, where S_current is the current area of ​​the heat source.

7. The fire control method according to claim 4, characterized in that, The steps of adjusting the extinguishing agent injection orifice diameter based on the extinguishing agent flow rate, temperature, and heat source area also include: Based on the current liquid level L_current and the liquid level threshold L_limit in the fire-fighting space; When L_current < L_limit, the spray enable flag is set to 1 to control the normal spraying of the extinguishing agent; When L_current ≥ L_limit, the spray enable flag is set to 0, and the spraying of extinguishing agent is stopped.

8. The fire control method according to claim 3, characterized in that, If the warning condition is 1, the following steps will be taken to activate the warning and increase the monitoring frequency, without performing fire extinguishing operations: When a single sensor anomaly is detected or Sc1≤Score≤Sc2; Increase the sampling frequency; Control the cooling flow rate to increase; Upload warning information but do not activate fire sprinklers; Sc1 is the first set threshold for the comprehensive fire risk score, and Sc2 is the second set threshold for the comprehensive fire risk score.

9. The fire control method according to claim 3, characterized in that, If the warning condition is 0 and the Score is greater than or equal to the set value, the steps for performing fire extinguishing operations include: When at least two sensor anomalies are detected or Sc2≤Score≤Sc3; The control valve assembly enters pre-switching preparation; The injection orifice diameter is adjusted using a thermal-fluid coupling algorithm; Real-time monitoring of the current liquid level L_current in the fire-fighting space; Sc2 is the second set threshold for the comprehensive fire risk score, and Sc3 is the third set threshold for the comprehensive fire risk score.

10. The fire control method according to claim 3, characterized in that, If the warning condition is 0 and the Score is greater than or equal to the set value, the steps for performing fire extinguishing operations include: When a score ≥ Sc3 is detected; Immediately switch to fire mode; Adjust the spray nozzle diameter to the maximum until the liquid level in the fire-fighting space reaches the liquid level threshold L_limit; Forcefully stop spraying; Sc3 is the third threshold for the comprehensive fire risk score.

11. The fire control method according to claim 1, characterized in that, Fire control methods also include: Whether the monitoring system has entered fire protection mode; When the system is not in fire-fighting mode, the cooling branch is opened by the control valve and the fire-fighting branch is blocked, so that the shared coolant is cooled and dissipated from the shared storage tank through the cooling branch. When the system enters fire protection mode, the control valve switches the pipeline status, blocking the cooling branch and opening the fire protection branch, so that the shared coolant is injected into the fire protection space from the shared storage tank through the fire protection branch.

12. A fire protection system, applied to the fire control method according to any one of claims 1 to 11, characterized in that, include: The data acquisition module is used to acquire normalized values ​​of temperature, smoke concentration, VOC concentration, and the highest temperature zone in thermal imaging in real time. The risk scoring calculation module is used to calculate a comprehensive fire risk score based on the temperature, smoke concentration, VOC concentration, and the normalized value of the highest temperature zone in thermal imaging. The control decision module is used to determine whether to perform fire extinguishing operations based on the comprehensive fire risk score.

13. The fire protection system according to claim 12, characterized in that, The fire protection system also includes: Shared storage tank for holding shared coolant; A shared circulating water pump is connected to the shared storage tank; Cooling branch circuits are used to form a cooling loop. Fire branch lines have spray nozzles that connect to the fire-fighting space; The control valve allows the cooling branch and the fire-fighting branch to be selectively connected to the shared circulating water pump.

Images