Intelligent fire extinguishing system for building

Abstract

The invention relates to the technical field of fire fighting and extinguishing, in particular to an intelligent fire fighting and extinguishing system for a building. The system comprises a multi-source information acquisition module, a feature fusion and situation generation module, an intelligent decision-making and fire extinguishing termination judgment module and an execution and feedback control module. The system firstly collects a surface temperature sequence, a characteristic gas concentration and an electric signal parameter respectively, generates a multi-source monitoring data set, extracts a characteristic parameter based on the data set and generates a battery stability situation index, and executes a multi-stage decision according to the index: when the index is lower than a safety threshold, water spray cooling is maintained; when the index exceeds a threshold, starting a stable observation period and converting into a tentative spray mode; and a fire extinguishing termination instruction or a water spraying recovery instruction is generated through trend stability analysis in the observation period. The technical problem of serious water stain loss caused by continuous water spraying when a traditional fire extinguishing system deals with a lithium battery fire disaster is solved, and accurate fire extinguishing control on the premise of ensuring no re-combustion is achieved.

Description

Technical Field

[0001] This invention relates to the field of fire extinguishing technology, and more specifically, to an intelligent fire extinguishing system for buildings. Background Technology

[0002] With the increasing popularity of electric vehicles, lithium battery fires in large underground parking lots have become a new challenge for intelligent fire protection systems. Traditional fire protection systems mainly target ordinary solid fires, triggering sprinkler systems through heat or smoke detectors. However, lithium battery fires have deep chemical reaction characteristics, making them highly susceptible to reignition after the open flame is extinguished, requiring continuous cooling.

[0003] To address the aforementioned issues, existing technologies typically employ continuous water spraying until the battery temperature drops to a safe threshold. Specifically, this approach involves deploying temperature sensing devices such as infrared thermal imagers to monitor the battery pack surface temperature in real time, using this as the sole criterion: when the monitored temperature exceeds a preset safety threshold, the system will continue to perform water spraying for cooling.

[0004] While this method can control battery thermal risks to some extent, its judgment logic based on a single temperature parameter has significant limitations. Prolonged, continuous water spraying for safety purposes can lead to large amounts of accumulated fire-fighting water, causing severe water damage and secondary harm.

[0005] Therefore, there is an urgent need for an intelligent fire protection system that can accurately determine when to terminate a fire and effectively reduce water damage to solve the above problems. Summary of the Invention

[0006] This invention provides an intelligent fire extinguishing system for buildings. It constructs a monitoring system based on multi-source information acquisition and feature fusion to generate a stability state index characterizing the battery status. Based on this index, it introduces an intelligent decision-making mechanism that includes a stable observation period, dynamically adjusting the fire extinguishing operation status according to the trend stability analysis results. This solves the problems mentioned in the background art, namely: The fire involving lithium batteries of electric vehicles in an underground parking lot resulted in severe water damage after continuous water spraying to cool them down following the initial firefighting efforts.

[0007] To achieve the above objectives, the intelligent fire extinguishing system includes a multi-source information acquisition module and a feature fusion and situation generation module. The multi-source information acquisition module is used to acquire surface temperature sequences, characteristic gas concentrations, and electrical signal parameters, and generate a multi-source monitoring dataset. The feature fusion and situation generation module performs feature extraction and fusion processing based on the multi-source monitoring dataset to generate a battery stability situation index. Its distinguishing feature is that it further includes an intelligent decision-making and fire extinguishing termination judgment module, which receives the battery stability index and performs the following processing: When the battery stability index is lower than the safety threshold, continue water spray cooling. When the battery stability index exceeds the safety threshold for the first time, a stability observation period is initiated and the continuous water spray mode is switched to an exploratory spray mode. During the stable observation period, the trend of the battery stability index is analyzed for trend stability and the trend stability analysis results are generated. If the trend stability analysis results indicate that the battery stability index tends to stabilize or continues to improve, a fire extinguishing termination command is generated. If the trend stability analysis results indicate that the battery stability index has deteriorated, a command to resume water spraying will be generated.

[0008] In the aforementioned technical solution, the design concept of the intelligent decision-making and fire extinguishing termination judgment module stems from a deep understanding of the complex characteristics of lithium battery fires. Traditional judgment methods based on a single temperature threshold have inherent flaws, as reaching the target battery surface temperature does not equate to the complete termination of internal electrochemical reactions. This module establishes a multi-stage progressive decision-making process. First, during the continuous cooling phase, it ensures sufficient suppression of violent reactions. Then, by introducing a crucial design element—a stable observation period—it shifts the fire extinguishing strategy from extensive to precise. During the observation period, the system reduces the water spray intensity to maintain a basic level of humidity while simultaneously initiating trend stability analysis. This dynamic monitoring mechanism effectively distinguishes between the temporary stability of the battery state and true safety. If relying solely on the initial threshold judgment, the system may face the risk of reignition due to prematurely stopping water spray, or suffer water damage due to excessive conservatism. However, by continuously tracking the changes in the battery stability index, the system can capture the essential characteristic of the gradual decay of internal reactions, thereby making accurate judgments that conform to the actual state of the battery. This dynamic trend-based decision-making mode ensures both safety redundancy and resource optimization, forming the intelligent core of the entire system.

[0009] Based on this, the feature fusion and situation generation module calculates the gas escape rate by monitoring the decay rate of the feature gas concentration and evaluates the electrochemical activity by tracking the dynamic changes of electrical signal parameters.

[0010] In another technical solution, the trend stability analysis is achieved by calculating the slope of the change of the battery stability index during the observation period, and the trend is determined to be stable when the trend stability coefficient is greater than or equal to zero.

[0011] This technical solution constructs a complete monitoring system from microscopic chemical reactions to macroscopic state trends. The comprehensive analysis of gas escape rate and electrochemical activity reveals the internal state of the battery that temperature parameters cannot reflect. The gas escape rate, by monitoring the dynamic decay of characteristic gas concentrations, directly characterizes the intensity changes of internal chemical reactions; while electrochemical activity, through tracking parameters such as voltage and insulation resistance, reflects the activity level of ion movement within the battery. These two, along with temperature parameters, complement each other. When combined, the system can construct a comprehensive understanding of the battery state. Based on this, trend stability analysis transforms the originally abstract state judgment into a concrete mathematical determination by quantifying the slope of the battery stability index. When the trend stability coefficient is greater than or equal to zero, it indicates that the battery state is stabilizing or continuously improving. This judgment method based on dynamic change rate is more accurate in predicting the future state of the system than comparing absolute values ​​at a single moment. The combination of these two technical solutions enables the system to perceive the internal state of the battery from multiple dimensions and grasp the state evolution law from a time dimension, thereby achieving precise control over the timing of fire extinguishing termination.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention overcomes the limitations of traditional single-parameter temperature monitoring through a collaborative design of multi-source information acquisition and feature fusion. The system constructs an evaluation system that accurately reflects the internal chemical state of the battery by comprehensively analyzing data from three dimensions: surface temperature sequence, characteristic gas concentration, and electrical signal parameters. This multi-parameter fusion design allows the system to cross-verify the battery's true state from different perspectives, effectively avoiding the problems of prematurely stopping water spraying or over-spraying due to misjudgment of a single parameter.

[0013] 2. The intelligent decision-making mechanism of this invention establishes a dynamically variable decision-making model by introducing a stable observation period and trend analysis algorithms. This system not only considers the battery state indicators at the current moment, but more importantly, it achieves a shift from static threshold judgment to dynamic trend prediction by continuously monitoring the changing trend of the battery stability index. This design enables the system to intelligently adjust the fire extinguishing strategy according to the actual evolution of the battery state, ensuring both the reliability of safety assurance and the optimization of water resource utilization. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the overall process structure of the intelligent fire extinguishing system for buildings according to the present invention; Figure 2 This is a schematic diagram of the multi-source information acquisition module of the present invention; Figure 3 This is a flowchart illustrating the intelligent decision-making and fire extinguishing termination judgment module of the present invention.

[0015] The meanings of the labels in the diagram are as follows: 100. Multi-source information acquisition module; 200. Feature fusion and situation generation module; 300. Intelligent decision-making and fire extinguishing termination judgment module; 400. Execution and feedback control module. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Currently, there is a problem of severe water damage caused by continuous water spraying after extinguishing electric vehicle lithium battery fires in underground parking lots. This invention provides an intelligent fire extinguishing system for buildings. (See [link to relevant documentation]). Figure 1 As shown, it includes a multi-source information acquisition module 100, a feature fusion and situation generation module 200, an intelligent decision-making and fire extinguishing termination judgment module 300, and an execution and feedback control module 400. By establishing a multi-dimensional monitoring system based on temperature, gas composition, and electrical signal parameters, and introducing an intelligent decision-making mechanism that includes a stable observation period, it achieves the fundamental goal of accurately terminating fire extinguishing operations while ensuring that the fire does not reignite, thereby minimizing secondary water damage.

[0018] like Figure 2 As shown, the multi-source information acquisition module 100 starts immediately after the open flame is extinguished, and is responsible for building a dynamic monitoring network that comprehensively reflects the internal chemical state of the battery. The module adopts a distributed architecture design and includes three complementary acquisition units: an infrared thermal imaging unit is responsible for acquiring surface temperature sequences, a gas sensing unit is responsible for acquiring characteristic gas concentrations, and an electrical signal monitoring unit is responsible for acquiring electrical signal parameters. The three work together to form a complete monitoring closed loop.

[0019] The infrared thermal imaging unit employs an array deployment scheme, uniformly arranging multiple high-precision infrared temperature probes above the fire suppression area. These probes operate in an alternating scanning mode, ensuring comprehensive coverage of critical areas such as the upper surface, sides, and seams of the power battery pack. Each probe acquires temperature readings at a specific sampling frequency, and all readings are synchronized via timestamps to form a complete surface temperature sequence. This design not only captures the overall temperature distribution of the battery pack but also monitors the changing trends of local hotspots, providing reliable thermal status data for subsequent analysis.

[0020] The gas sensing unit employs a redundant layout strategy, deploying multiple gas sensor arrays at key locations in the space surrounding the battery pack. Each array contains specialized sensors for carbon monoxide, hydrogen, and volatile electrolyte vapors, protected by an explosion-proof housing and equipped with a self-cleaning function to ensure long-term stability. The sensing unit uses an active gas intake method, continuously drawing surrounding gas into the detection chamber via a micro-pump to analyze the concentration changes of each characteristic gas in real time. This layout effectively overcomes the gas diffusion effects caused by airflow in underground parking lots, accurately capturing the concentration changes of characteristic gases released during battery thermal runaway.

[0021] The electrical signal monitoring unit employs a dual-path acquisition mechanism. It prioritizes directly reading the battery pack's total voltage, module voltage, and insulation resistance parameters via a standardized interface connected to the vehicle's battery management system. When vehicle data is unavailable, the unit automatically activates an external measuring device, indirectly acquiring relevant electrical signal parameters through an insulated clamp meter and a non-contact voltage sensor. The integrated signal conditioning circuitry within the unit filters and amplifies the raw signal to ensure that the acquired electrical signal parameters accurately reflect the battery's internal electrochemical state.

[0022] Each acquisition unit is equipped with an independent data preprocessing chip to perform preliminary cleaning and format standardization of the raw data. The processed surface temperature sequence, characteristic gas concentration, and electrical signal parameters are aggregated through a unified data bus, and after timestamp alignment and data packet encapsulation, a structured multi-source monitoring dataset is formed. This dataset is transmitted in real time to the feature fusion and situation generation module 200 via a dedicated communication protocol, providing complete and reliable data support for subsequent intelligent decision-making.

[0023] The feature fusion and situation generation module 200 receives a multi-source monitoring dataset transmitted from the multi-source information acquisition module 100. This dataset contains a complete surface temperature sequence, characteristic gas concentrations, and electrical signal parameters. The module first initiates a data preprocessing mechanism to perform deep cleaning and standardization on the raw dataset. The preprocessing process uses a moving average filtering algorithm to eliminate random fluctuations in temperature readings, employs an adaptive thresholding method to remove outliers in the gas concentration data, and uses a digital filter to eliminate environmental noise interference in the electrical signal parameters. All data streams undergo precise timestamp alignment to ensure that data from different sources remain synchronized under the same time reference, establishing a reliable data foundation for subsequent feature extraction.

[0024] After data preprocessing, the module initiates a multi-dimensional feature extraction process. For the surface temperature sequence, the feature extraction unit constructs a temperature drop trend curve reflecting the battery cooling rate by calculating the temperature change rate between adjacent time windows. Simultaneously, it analyzes the temperature differences between monitoring points on the battery pack surface to form a temperature field gradient map characterizing the uniformity of heat distribution. For characteristic gas concentration data, the analysis unit monitors the concentration changes of carbon monoxide, hydrogen, and volatile electrolyte vapor to calculate the decay rate of gas concentration per unit time, i.e., the gas escape rate. This parameter accurately reflects the weakening trend of internal chemical reactions within the battery. Regarding electrical signal parameter analysis, the processing unit tracks the dynamic changes in the battery's total voltage and insulation resistance, and, combined with the voltage balance of each module, comprehensively evaluates electrochemical activity indicators characterizing the intensity of internal electrochemical reactions within the battery.

[0025] The lightweight assessment model configured within the module receives feature parameters from three dimensions: temperature drop trend, gas escape rate, and electrochemical activity. This model employs a weighted fusion algorithm, dynamically adjusting the weight coefficients of each parameter based on their impact on the battery's stable state. By analyzing the persistence of the temperature drop trend, the stability of the gas escape rate, and the degree of degradation in electrochemical activity, the model ultimately generates a battery stability index ranging from zero to one. This index is a comprehensive quantitative indicator; the closer the value is to one, the more stable the battery state, while a value closer to zero indicates a higher risk of reignition. The battery stability index calculated by the model is transmitted in real-time to subsequent modules via a data interface, providing accurate state assessment data for intelligent decision-making in fire suppression strategies.

[0026] Although the feature fusion and situation generation module 200 can accurately generate a battery stability index reflecting the overall state of the battery, this single value cannot directly determine the optimal time to terminate the firefighting operation. Due to the special nature of lithium battery fires, even if the battery stability index temporarily reaches the safety threshold, there may still be a risk of reignition; while overly conservative continuous water spraying will cause unnecessary water damage. Based on this core contradiction, this invention introduces an intelligent decision-making and fire extinguishing termination judgment module 300, which establishes a multi-stage decision-making mechanism including a stable observation period to achieve precise control of the fire extinguishing process.

[0027] like Figure 3As shown, the intelligent decision-making and fire extinguishing termination judgment module 300 continuously receives the battery stability status index transmitted from the feature fusion and situation generation module 200, and executes a multi-stage decision-making process based on the dynamic change characteristics of the index. During the continuous cooling phase, upon receiving a fire response command, the module immediately sends a continuous water spray command to the execution system. In this phase, the module sets an initial safety threshold value. As long as the real-time transmitted battery stability status index is below this threshold value, regardless of the battery surface temperature reading, the module will maintain the water spray cooling operation, suppressing the internal chemical reactions of the battery through continuous water flow.

[0028] When the battery stability index is detected to have exceeded the preset safety threshold for the first time, the module does not immediately stop the fire suppression operation. Instead, it initiates a key design innovation—the stability observation period mechanism. The module then sends an instruction to the execution system to switch from continuous water spraying to a trial spraying mode, significantly reducing the spraying frequency and flow rate to maintain only a minimal level of humidity. Simultaneously, the module starts a countdown clock and activates a trend analysis algorithm to closely monitor the trajectory of the battery stability index during the observation period.

[0029] During the trend confirmation phase, the module uses a specific evaluation formula to quantify the changing trend of the battery stability index. In the formula; in the formula, Indicates the trend stability coefficient; The battery stability index represents the state of the battery at the end of the observation period; The battery stability index represents the state of the battery at the beginning of the observation period; This indicates the preset observation period duration. The formula calculates the slope of the battery stability index during the observation period, serving as a key basis for judging battery stability. If the calculated trend stability remains within a range of positive growth or small fluctuations close to zero, it indicates that the internal chemical reaction of the battery is continuously weakening and there are no signs of reignition; the module then generates a fire extinguishing termination command. Conversely, if the trend stability shows negative growth or drastic fluctuations, it indicates that the internal reaction of the battery is still continuing; the module immediately sends a command to the execution system to resume water spraying, re-entering the continuous cooling phase.

[0030] This decision-making mechanism, based on dynamic trend analysis of the battery stability index, effectively overcomes the limitations of single threshold judgment. By introducing a stable observation period and trend quantification analysis, this module can minimize water usage while ensuring safety, thereby significantly reducing water damage. The fire extinguishing termination command or water spraying resumption command generated by the module is transmitted in real time to the execution and feedback control module 400 via the control bus, completing a full closed loop from state perception to action decision-making.

[0031] The execution and feedback control module 400, as the final execution unit of this invention, receives instruction signals transmitted from the intelligent decision-making and fire extinguishing termination judgment module 300, including fire extinguishing termination instructions or water spraying resumption instructions, and converts them into precise field equipment control actions. This module adopts a three-level control architecture, including an instruction parsing unit, a multi-level valve control unit, and a status feedback unit, ensuring that the system can achieve precise water flow control and status indication based on the instructions from the decision-making module.

[0032] When the command parsing unit receives a fire extinguishing termination command, it immediately sends a shut-off signal to the multi-level valve control unit. Following a preset shut-off sequence, the control unit first closes the main control valve of the main pipeline in the area, and then sequentially closes the distribution valves of each sprinkler branch, achieving a complete cutoff of the water supply system. Simultaneously, the status feedback unit switches all on-site audible and visual alarms to a green safety status and displays a "risk cleared" message on the central monitoring interface, informing management personnel that the fire response has been successfully completed.

[0033] If the module receives a command to resume water spraying, the command parsing unit will immediately initiate the emergency response process. The multi-stage valve control unit intelligently adjusts the water supply intensity based on the battery stability index value carried in the command: when the index is within the critical risk range, half of the spray branches are activated to implement half-flow spraying; when the index indicates a high-risk state, all branches are activated to implement full-flow spraying. This tiered control mechanism ensures effective control of the risk of reignition while avoiding unnecessary water waste.

[0034] The status feedback unit operates continuously throughout the entire execution process. Through position sensors and flow monitoring devices deployed on the valve actuator, it collects real-time data on the valve's opening and closing status, pipeline water pressure, and actual flow rate, forming a complete execution status feedback signal. These signals are transmitted back to the system's central processing unit in real time, providing the necessary feedback information for closed-loop control of the entire fire suppression system.

[0035] Through the precise execution control and real-time status feedback described above, this module ultimately achieves the core objective of this invention: minimizing water damage by precisely controlling the timing and intensity of water spraying operations, while ensuring that lithium battery fires do not reignite. The system's closed-loop control mechanism guarantees a complete chain from status perception and intelligent decision-making to precise execution, achieving the optimal balance between safety and economy in building intelligent fire extinguishing systems when dealing with electric vehicle lithium battery fires.

[0036] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent fire extinguishing system for buildings, comprising a multi-source information acquisition module (100) and a feature fusion and situation generation module (200), wherein the multi-source information acquisition module (100) is used to acquire surface temperature sequences, characteristic gas concentrations and electrical signal parameters, and generate a multi-source monitoring dataset; the feature fusion and situation generation module (200) performs feature extraction and fusion processing based on the multi-source monitoring dataset to generate a battery stability situation index; Its features are: It also includes an intelligent decision-making and fire extinguishing termination judgment module (300), which receives the battery stability index and performs the following processing: When the battery stability index is lower than the safety threshold, continue water spray cooling. When the battery stability index exceeds the safety threshold for the first time, a stability observation period is initiated and the continuous water spray mode is switched to an exploratory spray mode. During the stable observation period, the trend of the battery stability index is analyzed for trend stability and the trend stability analysis results are generated. If the trend stability analysis results indicate that the battery stability index tends to stabilize or continues to improve, a fire extinguishing termination command is generated. If the trend stability analysis results indicate that the battery stability index has deteriorated, a command to resume water spraying will be generated.

2. The intelligent fire extinguishing system for buildings according to claim 1, characterized in that: The multi-source information acquisition module (100) includes an infrared thermal imaging unit, which uses multiple infrared temperature probes deployed in an array to collect temperature data at multiple key locations on the surface of the battery pack and form a surface temperature sequence.

3. The intelligent fire extinguishing system for buildings according to claim 1, characterized in that: The multi-source information acquisition module (100) includes a gas sensing unit, which employs a redundant gas sensor array to collect the characteristic gas concentrations of carbon monoxide, hydrogen, and volatile electrolyte vapor.

4. The intelligent fire extinguishing system for buildings according to claim 1, characterized in that: The multi-source information acquisition module (100) includes an electrical signal monitoring unit. The electrical signal monitoring unit adopts a dual-path acquisition mechanism and acquires the total voltage, module voltage and insulation resistance parameters of the battery pack by connecting to the battery management system or an external measuring device.

5. The intelligent fire extinguishing system for buildings according to claim 1, characterized in that: The feature fusion and situation generation module (200) constructs a temperature drop trend by calculating the temperature change rate of adjacent time windows and analyzes the temperature difference between monitoring points on the battery pack surface to form a temperature field gradient map.

6. The intelligent fire extinguishing system for buildings according to claim 1, characterized in that: The feature fusion and situation generation module (200) calculates the gas escape rate by monitoring the decay rate of the feature gas concentration and evaluates the electrochemical activity by tracking the dynamic changes of electrical signal parameters.

7. The intelligent fire extinguishing system for buildings according to claim 1, characterized in that: The trend stability analysis is achieved by calculating the slope of the change in the battery stability index during the observation period. When the trend stability coefficient is greater than or equal to zero, the trend is determined to be stable.

8. The intelligent fire extinguishing system for buildings according to claim 1, characterized in that: The intelligent decision-making and fire extinguishing termination judgment module (300) transmits the generated fire extinguishing termination command or water spraying resumption command to the execution and feedback control module (400).

9. The intelligent fire extinguishing system for buildings according to claim 8, characterized in that: The execution and feedback control module (400) includes a multi-stage valve control unit for adjusting the number of spray branches opened and the water flow rate according to the received instructions.

10. The intelligent fire extinguishing system for buildings according to claim 9, characterized in that: The execution and feedback control module (400) includes a status feedback unit, which is used to collect the valve opening and closing status and pipeline water pressure data in real time, and generate execution status feedback signals.

Images