A High-Accuracy Fire Alarm System and Judgment Method Based on Multi-Parameter Fusion

Abstract

This invention belongs to the field of multi-sensor fire detection technology, specifically a high-accuracy fire alarm system and judgment method based on multi-parameter fusion. The system includes: a sampling differential construction module, a causal trend identification module, a light-temperature ratio inference module, a smoke-temperature abrupt change screening module, and a comprehensive trigger decision module. In this invention, the initial disturbance trend is constructed by processing the differential changes in temperature and smoke on a unified time axis; the thermal diffusion behavior path is constructed by the event sequence relationship between pressure changes and temperature changes; the directional consistency between optical attenuation and heat enhancement is extracted by the ratio structure of brightness changes and temperature changes; and the joint abrupt change information of the fire amplification stage is extracted by the synchronous transition of smoke changes and temperature changes. The correlation of multiple parameters in time segments is linked together in a continuous structure to form an overall fire trend chain from weak disturbance to thermal diffusion and then to abrupt change transition, realizing stable capture and accurate judgment of fire status under multi-stage conditions.

Description

Technical Field

[0001] This invention belongs to the field of multi-sensor fire detection technology, specifically, it relates to a high-accuracy fire alarm system and judgment method based on multi-parameter fusion. Background Technology

[0002] Multi-sensor fire detection technology utilizes multiple environmental parameters to collaboratively identify signs of fire. Its core aspects include acquiring, analyzing, and comprehensively judging various types of environmental information such as temperature, temperature difference, smoke particle concentration, changes in air composition, combustible material release characteristics, water pressure and level changes, and optical images. It forms a collaborative identification system through different types of detectors, improving the ability to capture characteristics of the initial stage of a fire. This typically includes multiple basic detection methods such as thermal detection, smoke detection, gas detection, and image monitoring. Based on this, a joint analysis mechanism for the collected parameters is formed, constituting a complete fire monitoring and alarm technology system. Traditional fire alarm systems refer to those designed for fire hazards inside buildings. This type of alarm, which uses a single or limited number of independent parameters to determine the nature of a fire, is primarily aimed at identifying early-stage fire characteristics. Traditional methods involve installing independent temperature detectors to monitor the absolute rise in ambient temperature, using thermistors to sense temperature changes, combining this with smoke detectors to monitor changes in the concentration of smoke particles in the air, relying on photoelectric scattering or ionization structures for smoke identification, and using pressure switches or water flow indicators to identify changes in water pressure or flow in sprinkler pipes to determine whether the fire suppression system has been activated. Alternatively, independent cameras can be used to visually monitor the scene and allow for manual identification, making fire judgments based on a single or limited number of parameters to trigger an alarm for the occurrence of a fire.

[0003] Existing technologies rely on single or limited changes in quantities such as temperature, smoke concentration, pressure fluctuations, or optical imaging for fire detection. The monitoring results are greatly affected by the instantaneous fluctuations of single signals. Environmental disturbances, equipment reading errors, or local non-fire phenomena can all cause abnormal changes in individual monitored quantities. There is a lack of means to identify the correlation between multiple parameters, and it is impossible to determine the common direction of multiple parameters over time. This makes it impossible to accurately determine the initial synchronous changes in temperature and smoke during the weak fire stage, nor can it identify the transmission behavior between changes in air pressure and temperature rise. The directional relationship between optical brightness and heat changes cannot be detected, and the common transition characteristics of different parameters during the sudden fire stage cannot be captured. The entire monitoring chain is in a fragmented state, affecting the stability and continuity of fire determination. Summary of the Invention

[0004] This invention aims to solve the technical problems existing in the prior art by providing a high-accuracy fire alarm system and judgment method based on multi-parameter fusion. It constructs the initial disturbance trend by processing the differential changes in temperature and smoke on a unified time axis, constructs the thermal diffusion behavior path by the event sequence relationship between pressure and temperature changes, extracts the directional consistency between optical attenuation and heat enhancement by the ratio structure of brightness and temperature changes, and extracts the joint mutation information of the fire amplification stage by the synchronous transition of smoke and temperature changes. It connects the correlation of multiple parameters in time segments in a continuous structure to form an overall fire trend chain from weak disturbance to thermal diffusion and then to mutation transition, so as to achieve stable capture and accurate judgment of fire status under multi-stage conditions.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A high-accuracy fire alarm system based on multi-parameter fusion, comprising: The sampling difference construction module acquires environmental monitoring sampling information, analyzes temperature and smoke sampling sequences, performs difference operations on adjacent sampling points and maps them to a unified time axis, compares two types of difference directions, judges the consistency trend in continuous segments, calibrates regional disturbance characteristics, and generates disturbance coupling segment quantity information. Based on the disturbance coupling segment quantity information, the causal trend identification module extracts pressure and temperature sequences, identifies contraction and heating events, analyzes the adjacent structures of the event time sequence and the corresponding trigger positions, establishes the correlation chain of gas contraction and heating behavior, forms thermal action paths, and constructs thermal diffusion and migration information. Based on the thermal diffusion and migration information, the light-temperature ratio inference module extracts the brightness and temperature sequence, calculates the ratio of change, and identifies the consistency of optical attenuation and temperature rise ratio characteristics and directions by comparing the direction of ratio rise and fall in continuous segments. It then constructs linkage features and generates light-temperature drift ratio parameters. Based on the light-temperature drift ratio parameter, the smoke-temperature mutation screening module extracts smoke and temperature sequences, calculates differences and analyzes synchronization features, compares directional correlations in continuous segments, identifies two types of differential synchronous transition behaviors, constructs joint mutation features, and generates smoke-temperature mutation segment records. The integrated triggering decision module, based on the disturbance coupling segment quantity information, heat diffusion migration information, light temperature drift ratio parameters, and smoke temperature change segment records, sequentially organizes multiple types of trend information in a unified time series, analyzes the directional performance and coordination, identifies overlapping trend segments, and constructs fire alarm triggering criteria by comprehensively combining trend performance, generating fire alarm triggering segment information.

[0006] The following are further optimizations of the above technical solution by the present invention: The disturbance coupling segment information includes temperature disturbance segment markers, smoke change segment markers, and synchronous disturbance trend codes. The thermal diffusion migration information includes pressure contraction event sequence labels, temperature rise event sequence labels, and thermal action path indexes. The light-temperature drift ratio parameters include brightness change ratio parameter groups, temperature change ratio parameter groups, and light-temperature ratio offset direction records. The smoke-temperature abrupt change segment records include a set of smoke abrupt change time segments, a set of temperature abrupt change time segments, and a concentrated area number for smoke-temperature jumps. The fire alarm triggering segment information includes a set of multi-parameter overlapping time segments, a fire alarm triggering judgment parameter set, and a global triggering segment number.

[0007] Further optimization: The sampling difference construction module includes: The temperature difference extraction submodule acquires environmental monitoring sampling information, performs differential calculations on adjacent temperature sampling points and records the temperature disturbance direction, arranges the recorded disturbance direction in chronological order and organizes it into a continuous paragraph structure, analyzes the continuity of the disturbance direction in the paragraph based on the continuous paragraph structure, forms the paragraph mark of temperature disturbance by organizing the continuous structure, and generates the temperature disturbance paragraph quantity. The smoke differential extraction submodule acquires the smoke sampling sequence based on the temperature perturbation segment quantity, performs differential operation on adjacent smoke sampling points and records the smoke density change, maps the density change information to the time axis position consistent with the temperature perturbation segment quantity, analyzes the distribution of density change in continuous time segments based on the mapping position, forms the segment structure of smoke change by sorting out the distribution, and generates the smoke change segment quantity. The differential trend comparison submodule, based on the amount of smoke change segments, calls the amount of temperature disturbance segments and the amount of smoke change segments, aligns the two types of segments on a unified time axis and compares the directional relationship between the two types of segments, determines the segments with consistent directions in continuous segments based on the directional relationship, and establishes the disturbance coupling segment amount information by organizing the consistent segments to form a joint structure of regional disturbances.

[0008] Further optimization: The causal trend identification module includes: The event feature extraction submodule extracts the pressure sampling sequence and temperature sampling sequence based on the perturbation coupling segment quantity information, scans the intervals in the pressure sequence where the values ​​decrease and identifies contraction events, screens the intervals in the temperature sequence where the values ​​increase and identifies heating events, and generates a binary event attribute set. The temporal structure association submodule calls the binary event attribute set, arranges the pressure contraction event and the temperature rise event in chronological order, calculates the time interval value of adjacent heterogeneous events and analyzes the close proximity structure of events on the time axis, compares the trigger time of the contraction event with the response position of the temperature rise event, and generates temporal close proximity association quantity. The path chain construction submodule establishes a correlation chain between gas contraction behavior and heating behavior based on the temporally adjacent correlation quantities, forming a thermal action path. It integrates the path extension direction and diffusion rate data, evaluates the migration characteristics and diffusion range of thermal energy in the medium, and constructs thermal diffusion and migration information.

[0009] Further optimization: The process of scanning the intervals where the values ​​in the pressure sequence decrease and identifying contraction events, and screening the intervals where the values ​​in the temperature sequence increase and identifying heating events, specifically involves: The environmental pressure reference set is constructed by collecting pressure sampling data at the initial stage of system startup, and the numerical standard deviation of the environmental pressure reference set is calculated and set as the minimum effective pressure fluctuation threshold. Perform a first-order difference operation on adjacent sampling points in the pressure sampling sequence to generate a pressure change rate sequence. Locate continuous segments in the pressure change rate sequence where the values ​​are continuously less than zero, and mark the range corresponding to the target continuous segments on the time axis as the interval where the values ​​decrease. The pressure differential values ​​within the range of the decrease are summed to obtain the cumulative pressure contraction. When the absolute value of the cumulative pressure contraction exceeds the minimum effective pressure fluctuation threshold, the corresponding range of the decrease is determined as the contraction event. The ambient temperature reference set is constructed by collecting temperature sampling data at the initial stage of system startup, and the numerical standard deviation of the ambient temperature reference set is calculated and set as the minimum effective temperature rise fluctuation threshold. Perform a first-order difference operation on adjacent sampling points in the temperature sampling sequence to generate a temperature change rate sequence. Locate continuous segments in the temperature change rate sequence where the values ​​are continuously greater than zero, and mark the range corresponding to the target continuous segments on the time axis as the interval where the values ​​increase. The temperature difference values ​​within the range of the rising value are summed to obtain the cumulative temperature rise. When the cumulative temperature rise exceeds the minimum effective temperature fluctuation threshold, the corresponding range of the rising value is determined as the temperature rise event.

[0010] Further optimization: The light-temperature ratio inference module includes: The brightness change extraction submodule acquires a brightness sampling sequence based on the thermal diffusion and migration information, detects the brightness change of adjacent sampling points in the brightness sampling sequence and records the direction of change, organizes the brightness change direction into a continuous structure according to time segments, analyzes the attenuation performance of brightness change in segments based on the continuous structure, and forms brightness change information. The temperature change processing submodule obtains a temperature sampling sequence based on the brightness change information, detects the temperature change of adjacent sampling points in the temperature sampling sequence and records the direction of change, maps the temperature change direction to the time segment of the brightness change segment, analyzes the rising performance of temperature change in the segment based on the mapped time segment, and generates the temperature change segment quantity. The proportional linkage construction submodule, based on the temperature change segment quantity, performs proportional calculations on the two types of change quantities within a unified time segment and records the proportional direction. Arranged by time segment, it judges whether the proportional direction is consistent in continuous segments, constructs the linkage characteristics between optical change and temperature change, and establishes optical-temperature drift proportional parameters.

[0011] Further optimization: The flue gas temperature mutation screening module includes: The differential change extraction submodule obtains the smoke sampling sequence and the temperature sampling sequence based on the light-temperature drift ratio parameter. It performs differential analysis on adjacent sampling points in the smoke sampling sequence and records the density change. It performs differential analysis on adjacent sampling points in the temperature sampling sequence and records the temperature change. The two types of differential records are arranged in chronological order to generate a differential fragment sequence. The synchronization performance comparison submodule aligns the two types of differential segments in a unified time series based on the differential segment sequence, compares the directional relationship between the two types of differential segments, determines the synchronization performance of the two types of differential segments in continuous segments, and generates a synchronization direction segment quantity. The transition behavior construction submodule, based on the synchronous direction segment quantity, identifies two types of differential synchronous transition behaviors within continuous segments, constructs joint abrupt change characteristics of smoke change and temperature change, and establishes a record of smoke-temperature abrupt change segments.

[0012] Further optimization: The integrated triggering decision module includes: The trend feature processing submodule collects and processes the disturbance coupling segment quantity information, thermal diffusion migration information, light temperature drift ratio parameter, and smoke temperature sudden change segment record in chronological order into a unified sequence, records the time segment and directional structure of each type of trend information, arranges the time segment in order into a trend segment list, and generates the trend basic structure quantity. The coordination analysis submodule analyzes the directional performance of various trend information in time segments based on the aforementioned trend infrastructure quantity, compares the coordination of multiple trend directions in continuous segments, and generates trend coordination segment quantity. The alarm triggering construction submodule, based on the aforementioned trend coordination segment quantity, constructs the fire alarm triggering basis and establishes fire alarm triggering segment information by comprehensively considering the combined trend performance of multiple trends.

[0013] This invention also provides a high-accuracy fire alarm judgment method based on multi-parameter fusion, which is executed using the aforementioned high-accuracy fire alarm system based on multi-parameter fusion, and includes the following steps: S1: Acquire environmental monitoring sampling information, analyze temperature and smoke sampling sequences, perform differential operations on adjacent sampling points and map them to a unified time axis, compare two types of differential directions, determine the consistency trend in continuous segments, calibrate regional disturbance characteristics, and generate disturbance coupling segment quantity information. S2: Based on the perturbation coupling segment quantity information, extract the pressure and temperature sequence, identify contraction and heating events, analyze the adjacent structures of the event time sequence and the corresponding trigger positions, establish the correlation chain of gas contraction and heating behavior, form the thermal action path, and construct thermal diffusion and migration information. S3: Based on the thermal diffusion and migration information, extract the brightness and temperature sequence, calculate the change ratio, and by comparing the direction of the ratio rise and fall in continuous segments, identify the optical attenuation and temperature rise ratio characteristics and direction consistency, construct linkage features, and generate optical temperature drift ratio parameters. S4: Based on the light-temperature drift ratio parameter, extract the smoke and temperature sequence, calculate the difference and analyze the synchronization characteristics, compare the directional correlation in continuous segments, identify two types of differential synchronization transition behaviors, construct joint mutation features, and generate smoke-temperature mutation segment records. S5: Based on the disturbance coupling segment quantity information, thermal diffusion migration information, light temperature drift ratio parameter, and smoke temperature change segment record, the multiple types of trend information are sequentially organized in a unified time series, the directional performance and coordination are analyzed, overlapping trend segments are identified, and the fire alarm triggering basis is constructed by comprehensively combining the trend performance, and fire alarm triggering segment information is generated.

[0014] The present invention, by adopting the above technical solution, has at least the following beneficial effects: 1. This invention constructs an initial disturbance trend by processing the differential changes in temperature and smoke along a unified time axis, constructs a thermal diffusion behavior path by the sequential relationship between pressure and temperature changes, extracts the directional consistency between optical attenuation and heat enhancement by the ratio structure of brightness and temperature changes, and extracts joint abrupt change information of the fire amplification stage by the synchronous transition of smoke and temperature changes. It connects the correlation of multiple parameters in time segments in a continuous structure to form an overall fire trend chain from weak disturbance to thermal diffusion and then to abrupt transition, thereby achieving stable capture and accurate determination of fire status under multi-stage conditions. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the system modules in an embodiment of the present invention; Figure 2 This is a schematic diagram of the system framework in an embodiment of the present invention; Figure 3 This is a flowchart of the sampling difference construction module in an embodiment of the present invention; Figure 4This is a flowchart of the causal trend identification module in an embodiment of the present invention; Figure 5 This is a flowchart of the light-temperature ratio inference module in an embodiment of the present invention; Figure 6 This is a flowchart of the flue gas temperature sudden change screening module in an embodiment of the present invention; Figure 7 This is a flowchart of the integrated triggering decision module in an embodiment of the present invention; Figure 8 This is a flowchart of the method in an embodiment of the present invention. Detailed Implementation

[0016] In embodiments of the present invention, words such as “example” and “for example” are used to indicate that something is an example, illustration, or description; any embodiment or design described as “example” in the present invention should not be construed as being more preferred or more advantageous than other embodiments or designs; more precisely, the use of the word “example” is intended to present the concept in a specific manner; furthermore, in embodiments of the present invention, the meaning expressed by “and / or” can be both, or either one.

[0017] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that when the distinction between them is not emphasized, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that when the distinction between them is not emphasized, their intended meanings are consistent.

[0018] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0019] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0020] This invention provides a high-accuracy fire alarm system based on multi-parameter fusion, such as... Figure 1-2 The diagram shown illustrates a high-accuracy fire alarm system based on multi-parameter fusion. The system includes: The sampling difference construction module acquires environmental monitoring sampling information, analyzes temperature and smoke sampling sequences, performs difference operations on adjacent sampling points and maps them to a unified time axis, compares two types of difference directions, judges the consistency trend in continuous segments, calibrates regional disturbance characteristics, and generates disturbance coupling segment quantity information. The causal trend identification module extracts pressure and temperature sequences based on perturbation coupling segment quantity information, identifies contraction and heating events, analyzes the adjacent structures of the event time sequence and the corresponding triggering positions, establishes the correlation chain of gas contraction and heating behavior, forms thermal action paths, and constructs thermal diffusion and migration information. The light-temperature ratio inference module extracts brightness and temperature sequences based on thermal diffusion and migration information, calculates the ratio of changes, and identifies the consistency of optical attenuation and temperature rise ratio characteristics and directions by comparing the direction of ratio rise and fall in continuous segments. It then constructs linkage features and generates light-temperature drift ratio parameters. The smoke temperature mutation screening module extracts smoke and temperature sequences based on the photothermal drift ratio parameter, calculates the difference and analyzes the synchronization characteristics, compares the directional correlation in continuous segments, identifies two types of differential synchronous transition behaviors, constructs joint mutation features, and generates smoke temperature mutation segment records. The integrated triggering decision module, based on disturbance coupling segment quantity information, heat diffusion migration information, light temperature drift ratio parameters, and smoke temperature change segment records, sequentially organizes multiple types of trend information in a unified time series, analyzes directional performance and coordination, identifies overlapping trend segments, and constructs fire alarm triggering criteria by comprehensively combining trend performance, generating fire alarm triggering segment information.

[0021] The disturbance coupling segment information includes temperature disturbance segment markers, smoke change segment markers, and synchronous disturbance trend codes; the thermal diffusion migration information includes pressure contraction event sequence labels, temperature rise event sequence labels, and thermal action path indexes; the light and temperature drift ratio parameters include brightness change ratio parameter groups, temperature change ratio parameter groups, and light and temperature ratio offset direction records; the smoke and temperature abrupt change segment records include smoke abrupt change time segment sets, temperature abrupt change time segment sets, and smoke and temperature jump concentrated area numbers; and the fire alarm triggering segment information includes multi-parameter overlapping time segment sets, fire alarm triggering judgment parameter sets, and global triggering segment numbers.

[0022] Specifically, such as Figure 2 , 3 As shown, the sampling difference construction module includes: The temperature difference extraction submodule acquires environmental monitoring sampling information, performs differential calculations on adjacent temperature sampling points and records the temperature disturbance direction, arranges the recorded disturbance direction in chronological order and organizes it into a continuous paragraph structure, analyzes the continuity of the disturbance direction in the paragraph based on the continuous paragraph structure, forms the paragraph mark of temperature disturbance by organizing the continuous structure, and generates the temperature disturbance paragraph quantity. First, high-sensitivity temperature sensors deployed in the monitoring area collect ambient temperature data at a frequency of 10 times per second, obtaining a time series containing 300 consecutive sampling points. The system then performs a first-order difference operation on each pair of adjacent temperature sampling points in the series. Specifically, this operation involves taking the current time... Temperature value minus the previous moment The temperature value is used to obtain the instantaneous temperature change; for example, if The temperature at that moment was 25.05 degrees Celsius. If the time interval is 25.02 degrees Celsius, the calculated difference value is +0.03 degrees Celsius, and the system records the perturbation direction at this point as "positive temperature rise". Conversely, if the difference is negative, it is recorded as "negative temperature fall". The system sets a small noise filtering interval, such as ±0.01 degrees Celsius. Any change with an absolute difference value smaller than this interval is marked as "silent" and not counted as a perturbation direction. Subsequently, the system arranges all non-silent perturbation directions in chronological order of timestamps, scans the entire sequence, and looks for segments where the "positive temperature rise" mark appears consecutively. For example, from the 50th sampling point to the 80th sampling point, if the results of 30 consecutive difference operations are all positive and exceed the noise interval, the system defines this continuous time span as an independent temperature perturbation segment structure. The system further calculates the duration and average rate of change of this segment, for example, lasting 3 seconds with an average increase of 0.05 degrees Celsius per second, packages these attributes, and finally generates specific temperature perturbation segment quantities for subsequent correlation analysis.

[0023] The smoke differential extraction submodule acquires the smoke sampling sequence based on the temperature perturbation segment quantity, performs differential operation on adjacent smoke sampling points and records the smoke density change, maps the density change information to the time axis position consistent with the temperature perturbation segment quantity, analyzes the distribution of density change in continuous time segments based on the mapping position, forms the segment structure of smoke change by sorting out the distribution, and generates the smoke change segment quantity. Upon receiving the aforementioned temperature disturbance segment, the system first locks onto the time range corresponding to that segment (e.g., seconds 50 to 53 on the time axis), and uses this as a reference to retrieve the smoke concentration sampling sequence within the same time period. This sequence is provided by the photoelectric smoke detector at the same sampling frequency, with the unit being the attenuation percentage per meter (% / m). The system performs differential calculations on adjacent smoke sampling points within this time period, calculating by subtracting the concentration value of the previous sampling point from the concentration value of the subsequent sampling point, recording the density change increment at each moment. These density change increments are then mapped to a unified time axis scale that is completely consistent with the temperature data, ensuring that the two are aligned at the microsecond level. The system analyzes the distribution of these density changes over time and counts the percentage of sampling points where the smoke density increases positively within the time window corresponding to the temperature disturbance segment. For example, among the 30 sampling points, 28 points have positive smoke difference values, and the values ​​gradually increase from 0.01% / m to 0.05% / m. This distribution, which is continuous and monotonically increasing over time, is identified by the system as a valid smoke change distribution. The system organizes this series of continuously increasing smoke density change features, removes occasional single-point noise drop, constructs a stable smoke change segment structure, and outputs the number of smoke change segments that include the start and end times of the segment and the cumulative density increment.

[0024] The differential trend comparison submodule, based on the amount of smoke change segments, calls the amount of temperature disturbance segments and the amount of smoke change segments, aligns the two types of segments on a unified time axis and compares the directional relationship between the two types of segments, judges the segments with the same direction in continuous segments based on the directional relationship, and forms a joint structure of regional disturbances by organizing the consistent segments, and establishes disturbance coupling segment amount information. The system retrieves the temperature disturbance segment data and the smoke change segment data, performing a strict time axis alignment operation. Under a unified time coordinate system, the system compares the change direction records of the two segments point-by-point at the same time. Specifically, the system checks each sampling moment to determine whether the temperature difference direction is "positive temperature rise" and whether the smoke difference direction is "positive density increase." The system sets an alignment tolerance window, for example, allowing a time deviation of no more than 0.5 seconds at the starting point. As long as the change direction of the two remains consistent within the main duration (e.g., overlap rate exceeding 80%), it is considered a directional relationship match. A practical example illustrates this: if the temperature rise segment covers a time interval... The smoke increases the time interval of the paragraph coverage. The overlapping area between the two is Furthermore, if both types of parameters show a positive increasing trend within the overlapping interval, the system determines that the direction consistency of the continuous segment is valid; the system then assigns this overlapping and direction-consistent time segment to the system. The data are extracted and defined as the joint structure of regional perturbations. The temperature rise rate and smoke growth rate data within this structure are then integrated to finally establish perturbation coupling segment quantity information containing precise start and end times and coupling characteristics.

[0025] Specifically, such as Figure 2 , 4 As shown, the causal trend identification module includes: The event feature extraction submodule extracts pressure and temperature sampling sequences based on perturbation coupling segment quantity information, scans the intervals where the pressure values ​​decrease and identifies contraction events, screens the intervals where the temperature values ​​increase and identifies heating events, and generates a binary event attribute set. First, pressure and temperature sampling data were collected continuously for 10 minutes during the initial startup phase of the system (in the absence of fire alarms) to construct environmental pressure and environmental temperature benchmark sets, respectively. As shown in Table 1, the system performed statistical analysis on 6000 sampling points in the environmental pressure benchmark set and calculated its numerical standard deviation. Assuming the calculated pressure standard deviation is 2 Pascals (Pa), the system set 3 times this value, i.e., 6 Pascals, as the minimum effective pressure fluctuation threshold. Similarly, if the standard deviation of the temperature benchmark set is 0.05 degrees Celsius, then 0.15 degrees Celsius is set as the minimum effective temperature fluctuation threshold. After entering the real-time monitoring phase, the system performed a first-order difference operation on the pressure sampling sequence to generate a pressure change rate sequence, and scanned and found that from... arrive In 20 consecutive sampling points, if the pressure change rate is consistently negative (e.g., an average decrease of 0.5 Pa per sampling point), the system marks this range as a numerical decrease interval. Subsequently, the system accumulates all pressure difference values ​​within this interval to calculate the cumulative pressure contraction as -10 Pa. The system takes its absolute value of 10 Pa and compares it with the preset minimum effective pressure fluctuation threshold of 6 Pa. Since 10 Pa is greater than 6 Pa, the interval is determined to be an effective contraction event. For the temperature sequence, the system locates a continuous segment with a consistently positive change rate and accumulates it to obtain a cumulative temperature rise of 0.5 degrees Celsius. Since 0.5 degrees Celsius is greater than the minimum effective temperature fluctuation threshold of 0.15 degrees Celsius, the system confirms this interval as a temperature rise event. Finally, a binary event attribute set containing the start time, duration, and cumulative amplitude of these two types of events is generated.

[0026] Table 1: Example Table of Environmental Benchmarks and Threshold Settings As shown in Table 1, the threshold is dynamically generated based on the background noise of the environment using statistical methods.

[0027] The temporal structure association submodule calls the binary event attribute set to arrange the pressure contraction event and the temperature rise event in chronological order, calculates the time interval value of adjacent heterogeneous events and analyzes the close proximity structure of events on the time axis, compares the trigger time of the contraction event with the response position of the temperature rise event, and generates temporal close proximity association quantity. The system calls upon the generated binary event attribute set to arrange all confirmed pressure contraction events and temperature rise events on the timeline according to their timestamps. The system focuses on analyzing the time intervals between adjacent anomaly events (i.e., a contraction event immediately followed by a temperature rise event, or vice versa). Specifically, it extracts the end time of the pressure contraction event. The beginning of the subsequent temperature rise event Calculate the time difference between the two. The system analyzes whether the time difference falls within a preset adjacent window, such as between 0 and 5 seconds. If the actual calculated interval is 1.5 seconds, it indicates that the two events are highly adjacent in time structure. At the same time, the system compares the trigger time (starting point) of the contraction event with the response position of the heating event (inflection point of the heating curve) to check whether there is a causal logical sequence relationship. In the early stage of a fire in a closed space, a slight air pressure contraction caused by oxygen consumption is usually observed first, followed by heat release leading to a temperature increase. The system quantifies this sequence relationship. If the contraction event starts at 10:00:05 and the heating event responds at 10:00:07, and the interval between the two events conforms to the characteristics of thermodynamic response delay, the system generates a time-series adjacent correlation quantity to clearly record that these two events belong to different stages of the same physical process.

[0028] The path chain construction submodule establishes a correlation chain between gas contraction behavior and heating behavior based on temporally adjacent correlation quantities, forming a thermal action path. It integrates the extension direction and diffusion rate data of the path, evaluates the migration characteristics and diffusion range of thermal energy in the medium, and constructs thermal diffusion and migration information. Based on the generated temporally adjacent correlations, the paired gas contraction and heating behaviors are logically connected to establish a complete correlation chain. The system uses this correlation chain to integrate the extension direction and diffusion velocity data of the path. Specifically, the system combines sensor spatial layout information. If the correlation chain shows that the contraction event is first captured by the sensor located in the center of the room, and the subsequent heating event is captured by the center and the north side sensors in turn, the system infers that the thermal path is "extending from the center to the north". The system further calculates the diffusion velocity by dividing the physical distance between the two sensors (e.g., 5 meters) by the response time difference of the heating event on the two sensors (e.g., 2.5 seconds), and obtains that the migration velocity of heat energy in the medium is 2 m / s. The system evaluates whether this velocity conforms to the characteristic range of air thermal convection (e.g., 0.5 m / s to 3 m / s). If it does, the migration characteristic of heat energy in the medium is confirmed as natural convection diffusion, and the diffusion range is estimated by combining the coverage area of ​​the involved sensors (e.g., a circular area with a radius of 5 meters). Finally, the system structures these data, which include path direction, migration velocity, and coverage area, to construct detailed thermal diffusion and migration information.

[0029] Specifically, such as Figure 2 , 5 As shown, the light-temperature ratio inference module includes: The brightness change extraction submodule acquires a brightness sampling sequence based on thermal diffusion and migration information. It detects the brightness change of adjacent sampling points in the brightness sampling sequence and records the direction of change. The brightness change direction is organized into a continuous structure according to time segments. The attenuation performance of brightness change in segments is analyzed based on the continuous structure to form brightness change information. Based on the aforementioned time window and region locked by thermal diffusion and migration information, the system retrieves the brightness sampling sequence collected by the light sensor within that region. The system performs differential operations on adjacent sampling points in the brightness sequence. For example, if the current brightness is 450 lumens (Lux) and the previous brightness was 452 lumens, the change is -2 lumens, and the system records this change direction as "negative decay". The system organizes these change directions by time segment and identifies structures with continuous negative changes. For example, if the brightness value shows a continuous downward trend within 10 consecutive seconds without a significant rebound, the system analyzes the decay performance of brightness changes within the segment based on this continuous structure and calculates its average decay rate. Assuming that the brightness decreases by a total of 100 lumens within 10 seconds, the average decay rate is 10 lumens / second. The system encapsulates the start time, end time, and total decay amount of this decay process to form brightness change information. This step aims to capture the characteristics of reduced ambient light intensity caused by smoke particles blocking light or damage to the light source, eliminating the influence of random fluctuations in light intensity.

[0030] The temperature change processing submodule acquires a temperature sampling sequence based on brightness change information, detects the temperature change of adjacent sampling points in the temperature sampling sequence and records the direction of change, maps the temperature change direction to the time segment of brightness change segment, analyzes the rising performance of temperature change in the segment based on the mapped time segment, and generates temperature change segment quantity. Using the time segment in the generated brightness change information as a benchmark (e.g., the 10-second interval mentioned above), data for the corresponding time period is extracted from the temperature sampling sequence. The system calculates the change in adjacent sampling points in this extracted temperature data and records the direction of change. Unlike full-time scanning, this only focuses on the temperature performance during brightness decay. The system analyzes whether the temperature shows a continuous upward trend within the mapped time segment. For example, within the 10 seconds of brightness decrease, the corresponding temperature sequence shows a process of rising from 25.0 degrees Celsius to 25.5 degrees Celsius, without any downward trend in between. The system calculates the temperature rise amplitude and the slope of the temperature rise curve within this segment. If it is confirmed that the temperature change shows a stable upward trend, the system organizes the temperature change characteristics within this time period (e.g., temperature rise amplitude of 0.5 degrees Celsius, linearity of 0.95) to generate a temperature change segment quantity. This process ensures that the extraction of temperature features is strictly limited to the time domain of optical anomaly occurrence, providing a time-synchronized physical quantity for subsequent proportional analysis.

[0031] The proportional linkage construction submodule is based on the temperature change segment quantity. Within a unified time segment, it performs proportional calculation on two types of change quantities and records the proportional direction. Arranged by time segment, it judges whether the proportional direction is consistent in continuous segments, constructs the linkage characteristics of optical change and temperature change, and establishes optical temperature drift proportional parameters. Based on the aforementioned temperature change segments and brightness change information, a ratio calculation is performed on the two types of changes within a unified time segment. Specifically, the system selects each second as a calculation node and calculates the ratio of the absolute value of the brightness change to the temperature change at that node. For example, if the brightness decreases by 10 lumens and the temperature increases by 0.05 degrees Celsius at a certain moment, the calculated light-temperature ratio is 200. The system arranges this series of ratios in chronological order and judges the directional consistency and stability of the ratio in continuous segments. If the ratio remains within the preset "smoke shading - heat release" characteristic range (e.g., between 150 and 250) in 10 consecutive calculation nodes, and there are no drastic jumps, the system determines that the two have a clear linkage characteristic. This linkage characteristic reflects the physical relationship between the amount of smoke generated and the amount of heat released in a fire, which is different from the case of simply turning on the light (only brightness changes) or heating with an electric heater (only temperature changes). The system constructs the verified ratio value sequence and its corresponding directional consistency identifier to establish the final light-temperature drift ratio parameter.

[0032] Specifically, such as Figure 2 , 6 As shown, the flue gas temperature mutation screening module includes: The differential change extraction submodule obtains the smoke sampling sequence and temperature sampling sequence based on the light-temperature drift ratio parameter. It performs differential analysis on adjacent sampling points in the smoke sampling sequence and records the density change, and performs differential analysis on adjacent sampling points in the temperature sampling sequence and records the temperature change. The two types of differential records are arranged in chronological order to generate a differential fragment sequence. Based on the suspected fire alarm period indicated by the light-temperature drift ratio parameter, high-precision smoke sampling sequences and temperature sampling sequences are extracted again; the system performs differential operations on adjacent points of the smoke sequence and records the density change value ( Simultaneously, differential analysis is performed on the temperature sequence, and the temperature change values ​​are recorded. This section focuses not only on direction but also on the magnitude of change; the system organizes the calculated two types of difference records in chronological order, generating a series of numerical pairs. The composed differential segment sequence; for example, in Time, recorded as ,exist Time record as The system organizes these raw differential numerical streams into continuous time segments to prepare for subsequent identification of abrupt changes. The purpose of this step is to obtain the lowest-level instantaneous rate of change data without smoothing, so as to preserve the high-frequency abrupt change features in the signal.

[0033] The synchronization performance comparison submodule is based on the differential segment sequence. It aligns two types of differential segments in a unified time series, compares the directional relationship between the two types of differential segments, determines the synchronization performance of the two types of differential segments in continuous segments, and generates the synchronization direction segment quantity. The system receives a sequence of differential segments and strictly aligns the smoke differential segments and temperature differential segments within a unified time series. It compares the numerical trends of the two types of differential segments at the same time, focusing on their synchronization within consecutive segments. Specifically, the system checks for situations where both differential values ​​increase significantly simultaneously. For example, the system sets a synchronization judgment logic: if, within three consecutive seconds, the smoke differential value at each moment is greater than the previous moment (acceleration greater than 0), and the corresponding temperature differential value is also greater than the previous moment (temperature acceleration), then the two are determined to have synchronous acceleration characteristics. The system records the duration and intensity of this synchronization enhancement, generating a synchronization direction segment quantity. Taking actual data as an example, if from the 40th to the 43rd second, the smoke increment jumps from 0.1 to 0.5, and the temperature increment jumps from 0.02 to 0.1, both showing a synchronous explosive growth, the system locks this segment as a significant area of ​​synchronization.

[0034] The transition behavior construction submodule is based on the synchronous direction segment quantity. By identifying two types of differential synchronous transition behaviors within continuous segments, it constructs the joint abrupt change characteristics of smoke change and temperature change, and establishes a record of smoke-temperature abrupt change segments. Based on the synchronous direction segment quantity, the system focuses on identifying synchronous transition behaviors of two types of differences within continuous segments. The system defines "transition" as an anomaly in the second derivative of the rate of change, i.e., a sudden change in acceleration. The system calculates the acceleration of smoke and temperature changes. When it detects that both of them simultaneously exhibit transitions exceeding a preset threshold within the same time window (e.g., within 0.5 seconds) (e.g., the smoke rate of change doubles instantly, and the temperature rate of change doubles instantly), the system constructs a joint abrupt change feature of smoke and temperature changes. This feature corresponds to the flashover or rapid spread stage in the fire development process. The system records the time of occurrence of the identified abrupt change, the numerical span before and after the abrupt change, and the duration of the abrupt change, establishing a smoke-temperature abrupt change segment record. As shown in Table 2, the system may record that at 10:01:30, the smoke rate of change suddenly changes from a slow linear increase to an exponential increase, while the temperature rate of change also undergoes a simultaneous abrupt change in the same direction. This joint abrupt change feature is strong evidence that distinguishes it from slow smoldering or environmental interference.

[0035] Table 2: Example Table of Combined Fluctuation Characteristics of Smoke Temperature As shown in Table 2, When the rates of change of both change simultaneously undergo a jump of orders of magnitude, it is confirmed as a joint mutation.

[0036] Specifically, such as Figure 2 , 7 As shown, the integrated trigger decision module includes: The trend feature processing submodule collects and processes information on disturbance coupling segment quantity, thermal diffusion and migration information, light and temperature drift ratio parameters, and smoke and temperature change segment records in chronological order into a unified sequence. It records the time segment and directional structure of each type of trend information, arranges the time segments in order into a trend segment list, and generates the basic trend structure quantity. Based on information on perturbation coupling segments, thermal diffusion and migration, light-temperature drift ratio parameters, and records of fluctuating smoke-temperature zones, a comprehensive time-series processing task is performed. The system establishes a unified timeline with millisecond-level resolution. The system collects marked effective time segments from various types of information in chronological order, such as perturbation coupling segments. Thermal diffusion migration period Light temperature drift period Moment of sudden change in smoke temperature The system records the directional structure of each type of trend information within the corresponding time segment (such as "positive enhancement" and "negative decay"). Then, the system sorts these time segments by their start time, organizing them into a multi-dimensional list of trend segments. In this list, each row represents a time step, and each column represents the state of a physical trend. The system generates a trend infrastructure quantity, which is essentially a time series matrix that clearly shows whether the various physical characteristics detected by the system exist and their specific manifestations at each time point, providing a structured data foundation for subsequent comprehensive and coordinated analysis.

[0037] The coordination analysis submodule analyzes the directional performance of various trend information in time segments based on the trend infrastructure quantity, compares the coordination of multiple trend directions in continuous segments, and generates trend coordination segment quantity. Based on the trend infrastructure, the system deeply analyzes the directional performance and coordination of various trend information in time segments. The system employs a sliding window algorithm to check whether there is logical mutual verification between multiple trend directions within any given time window (e.g., 5 seconds wide). Specifically, the system compares the coordination of multiple trend directions in consecutive segments: if within the same window, temperature shows an upward trend, smoke shows an increasing trend, air pressure shows a contraction followed by recovery trend, and light shows a decreasing trend, and the directions of these trends conform to the "positive coupling" relationship in combustion physics, the system calculates its coordination score. For example, if four of the four characteristics conform to fire logic, the coordination score is 100%; if only three conform, the coordination score is 75%. The system pays special attention to consecutive segments where the coordination score remains consistently high (e.g., greater than 75%), generating trend coordination segment quantities. These segment quantities identify specific sections where multi-source sensor data achieve a high degree of consensus in the time dimension, excluding the possibility of occasional faults of a single sensor or single-factor environmental interference (e.g., only smoke without temperature rise).

[0038] The alarm triggering construction submodule is based on the trend coordination segment quantity. By comprehensively considering the joint trend performance of multiple trends, it constructs the basis for fire alarm triggering and establishes fire alarm triggering segment information. Based on the trend coordination segment quantity, the final fire alarm determination logic is executed; the system integrates the joint trend performance of multiple trends to check whether there is a "full feature overlap area"; for example, the system finds that within a time interval... Within the data, not only are the trends of the four physical quantities highly coordinated (100% coordination), but it also includes the high-risk characteristic of "smoke-temperature abrupt change." Based on this, the system constructs a fire alarm triggering basis, assuming that the data pattern of this section fully conforms to the physical model of a real fire. The system then establishes fire alarm triggering section information, which includes the exact time of the alarm trigger, the sensor area number involved, and a description of the core characteristics that led to the trigger (such as "strong coupling of multiple parameters accompanied by abrupt changes"). Finally, the system outputs this triggering information to the alarm controller, activating the audible and visual alarms and linkage control devices. Through this progressive logical verification, the system ensures that the alarm signal is generated based on a comprehensive, coordinated data chain that conforms to physical laws, thereby greatly improving the accuracy and reliability of the alarm.

[0039] Please see Figure 8 The present invention also provides a high-accuracy fire alarm judgment method based on multi-parameter fusion, which is executed by the above-mentioned high-accuracy fire alarm system based on multi-parameter fusion, and includes the following steps: S1: Acquire environmental monitoring sampling information, analyze temperature and smoke sampling sequences, perform differential operations on adjacent sampling points and map them to a unified time axis, compare two types of differential directions, determine the consistency trend in continuous segments, calibrate regional disturbance characteristics, and generate disturbance coupling segment quantity information. S2: Based on the perturbation coupling segment quantity information, extract pressure and temperature sequences, identify contraction and heating events, analyze the adjacent structures of the event time sequence and the corresponding triggering positions, establish the correlation chain of gas contraction and heating behavior, form thermal action paths, and construct thermal diffusion and migration information; S3: Based on thermal diffusion and migration information, extract brightness and temperature sequences, calculate the proportion of change, and by comparing the direction of the increase and decrease of the proportion in continuous segments, identify the optical attenuation and temperature rise ratio characteristics and direction consistency, construct linkage features, and generate optical temperature drift ratio parameters. S4: Based on the light-temperature drift ratio parameter, extract the smoke and temperature sequences, calculate the difference and analyze the synchronization features, compare the directional correlation in continuous segments, identify two types of differential synchronization transition behaviors, construct joint mutation features, and generate smoke-temperature mutation segment records. S5: Based on disturbance coupling segment quantity information, thermal diffusion migration information, light temperature drift ratio parameters, and smoke temperature change segment records, multiple types of trend information are sequentially organized in a unified time series, the directional performance and coordination are analyzed, overlapping trend segments are identified, and the fire alarm triggering basis is constructed by comprehensively combining trend performance, generating fire alarm triggering segment information.

[0040] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention; therefore, the scope of protection of the present invention should be determined by the scope of protection of the claims.

Claims

1. A high-accuracy fire alarm system based on multi-parameter fusion, characterized in that, The system includes: The sampling difference construction module acquires environmental monitoring sampling information, analyzes temperature and smoke sampling sequences, performs difference operations on adjacent sampling points and maps them to a unified time axis, compares two types of difference directions, judges the consistency trend in continuous segments, calibrates regional disturbance characteristics, and generates disturbance coupling segment quantity information. Based on the disturbance coupling segment quantity information, the causal trend identification module extracts pressure and temperature sequences, identifies contraction and heating events, analyzes the adjacent structures of the event time sequence and the corresponding trigger positions, establishes the correlation chain of gas contraction and heating behavior, forms thermal action paths, and constructs thermal diffusion and migration information. Based on the thermal diffusion and migration information, the light-temperature ratio inference module extracts the brightness and temperature sequence, calculates the ratio of change, and identifies the consistency of optical attenuation and temperature rise ratio characteristics and directions by comparing the direction of ratio rise and fall in continuous segments. It then constructs linkage features and generates light-temperature drift ratio parameters. The smoke temperature mutation screening module extracts smoke and temperature sequences based on the light-temperature drift ratio parameter, calculates the difference and analyzes the synchronization characteristics, compares the directional correlation in continuous segments, identifies two types of differential synchronization transition behaviors, constructs joint mutation features, and generates smoke temperature mutation segment records.

2. The high-accuracy fire alarm system based on multi-parameter fusion according to claim 1, characterized in that, The disturbance coupling segment information includes temperature disturbance segment markers, smoke change segment markers, and synchronous disturbance trend codes. The thermal diffusion migration information includes pressure contraction event sequence labels, temperature rise event sequence labels, and thermal action path indexes. The light-temperature drift ratio parameters include brightness change ratio parameter groups, temperature change ratio parameter groups, and light-temperature ratio offset direction records. The smoke-temperature abrupt change segment records include smoke abrupt change time segment sets, temperature abrupt change time segment sets, and smoke-temperature jump concentrated area numbers.

3. The high-accuracy fire alarm system based on multi-parameter fusion according to claim 1, characterized in that, The sampling difference construction module includes: The temperature difference extraction submodule acquires environmental monitoring sampling information, performs differential calculations on adjacent temperature sampling points and records the temperature disturbance direction, arranges the recorded disturbance direction in chronological order and organizes it into a continuous paragraph structure, analyzes the continuity of the disturbance direction in the paragraph based on the continuous paragraph structure, forms the paragraph mark of temperature disturbance by organizing the continuous structure, and generates the temperature disturbance paragraph quantity. The smoke differential extraction submodule acquires the smoke sampling sequence based on the temperature perturbation segment quantity, performs differential operation on adjacent smoke sampling points and records the smoke density change, maps the density change information to the time axis position consistent with the temperature perturbation segment quantity, analyzes the distribution of density change in continuous time segments based on the mapping position, forms the segment structure of smoke change by sorting out the distribution, and generates the smoke change segment quantity. The differential trend comparison submodule, based on the amount of smoke change segments, calls the amount of temperature disturbance segments and the amount of smoke change segments, aligns the two types of segments on a unified time axis and compares the directional relationship between the two types of segments, determines the segments with consistent directions in continuous segments based on the directional relationship, and establishes the disturbance coupling segment amount information by organizing the consistent segments to form a joint structure of regional disturbances.

4. The high-accuracy fire alarm system based on multi-parameter fusion according to claim 3, characterized in that, The causal trend identification module includes: The event feature extraction submodule extracts the pressure sampling sequence and temperature sampling sequence based on the perturbation coupling segment quantity information, scans the intervals in the pressure sequence where the values ​​decrease and identifies contraction events, screens the intervals in the temperature sequence where the values ​​increase and identifies heating events, and generates a binary event attribute set. The temporal structure association submodule calls the binary event attribute set, arranges the pressure contraction event and the temperature rise event in chronological order, calculates the time interval value of adjacent heterogeneous events and analyzes the close proximity structure of events on the time axis, compares the trigger time of the contraction event with the response position of the temperature rise event, and generates temporal close proximity association quantity. The path chain construction submodule establishes a correlation chain between gas contraction behavior and heating behavior based on the temporally adjacent correlation quantities, forming a thermal action path. It integrates the path extension direction and diffusion rate data, evaluates the migration characteristics and diffusion range of thermal energy in the medium, and constructs thermal diffusion and migration information.

5. The high-accuracy fire alarm system based on multi-parameter fusion according to claim 4, characterized in that, The process of scanning the intervals of decreasing values ​​in the pressure sequence and identifying contraction events, and screening the intervals of increasing values ​​in the temperature sequence and identifying heating events, specifically involves: The environmental pressure reference set is constructed by collecting pressure sampling data at the initial stage of system startup, and the numerical standard deviation of the environmental pressure reference set is calculated and set as the minimum effective pressure fluctuation threshold. Perform a first-order difference operation on adjacent sampling points in the pressure sampling sequence to generate a pressure change rate sequence. Locate continuous segments in the pressure change rate sequence where the values ​​are continuously less than zero, and mark the range corresponding to the target continuous segments on the time axis as the interval where the values ​​decrease. The pressure differential values ​​within the range of the decrease are summed to obtain the cumulative pressure contraction. When the absolute value of the cumulative pressure contraction exceeds the minimum effective pressure fluctuation threshold, the corresponding range of the decrease is determined as the contraction event. The ambient temperature reference set is constructed by collecting temperature sampling data at the initial stage of system startup, and the numerical standard deviation of the ambient temperature reference set is calculated and set as the minimum effective temperature rise fluctuation threshold. Perform a first-order difference operation on adjacent sampling points in the temperature sampling sequence to generate a temperature change rate sequence. Locate continuous segments in the temperature change rate sequence where the values ​​are continuously greater than zero, and mark the range corresponding to the target continuous segments on the time axis as the interval where the values ​​increase. The temperature difference values ​​within the range of the rising value are summed to obtain the cumulative temperature rise. When the cumulative temperature rise exceeds the minimum effective temperature fluctuation threshold, the corresponding range of the rising value is determined as the temperature rise event.

6. The high-accuracy fire alarm system based on multi-parameter fusion according to claim 4, characterized in that, The light-temperature ratio inference module includes: The brightness change extraction submodule acquires a brightness sampling sequence based on the thermal diffusion and migration information, detects the brightness change of adjacent sampling points in the brightness sampling sequence and records the direction of change, organizes the brightness change direction into a continuous structure according to time segments, analyzes the attenuation performance of brightness change in segments based on the continuous structure, and forms brightness change information. The temperature change processing submodule obtains a temperature sampling sequence based on the brightness change information, detects the temperature change of adjacent sampling points in the temperature sampling sequence and records the direction of change, maps the temperature change direction to the time segment of the brightness change segment, analyzes the rising performance of temperature change in the segment based on the mapped time segment, and generates the temperature change segment quantity. The proportional linkage construction submodule, based on the temperature change segment quantity, performs proportional calculations on the two types of change quantities within a unified time segment and records the proportional direction. Arranged by time segment, it judges whether the proportional direction is consistent in continuous segments, constructs the linkage characteristics between optical change and temperature change, and establishes optical-temperature drift proportional parameters.

7. The high-accuracy fire alarm system based on multi-parameter fusion according to claim 6, characterized in that, The flue gas temperature mutation screening module includes: The differential change extraction submodule obtains the smoke sampling sequence and the temperature sampling sequence based on the light-temperature drift ratio parameter. It performs differential analysis on adjacent sampling points in the smoke sampling sequence and records the density change. It performs differential analysis on adjacent sampling points in the temperature sampling sequence and records the temperature change. The two types of differential records are arranged in chronological order to generate a differential fragment sequence. The synchronization performance comparison submodule aligns the two types of differential segments in a unified time series based on the differential segment sequence, compares the directional relationship between the two types of differential segments, determines the synchronization performance of the two types of differential segments in continuous segments, and generates a synchronization direction segment quantity. The transition behavior construction submodule, based on the synchronous direction segment quantity, identifies two types of differential synchronous transition behaviors within continuous segments, constructs joint abrupt change characteristics of smoke change and temperature change, and establishes a record of smoke-temperature abrupt change segments.

8. The high-accuracy fire alarm system based on multi-parameter fusion according to claim 1, characterized in that, The system also includes: The integrated triggering decision module, based on the disturbance coupling segment quantity information, heat diffusion migration information, light temperature drift ratio parameter, and smoke temperature sudden change segment record, sequentially organizes multiple types of trend information in a unified time series, analyzes the directional performance and coordination, identifies overlapping trend segments, and constructs fire alarm triggering basis by comprehensively combining trend performance, generating fire alarm triggering segment information. The fire alarm triggering section information includes a set of multi-parameter overlapping time segments, a set of fire alarm triggering judgment parameters, and a global number of the triggering segment.

9. The high-accuracy fire alarm system based on multi-parameter fusion according to claim 8, characterized in that, The integrated trigger decision module includes: The trend feature processing submodule collects and processes the disturbance coupling segment quantity information, thermal diffusion migration information, light temperature drift ratio parameter, and smoke temperature sudden change segment record in chronological order into a unified sequence, records the time segment and directional structure of each type of trend information, arranges the time segment in order into a trend segment list, and generates the trend basic structure quantity. The coordination analysis submodule analyzes the directional performance of various trend information in time segments based on the aforementioned trend infrastructure quantity, compares the coordination of multiple trend directions in continuous segments, and generates trend coordination segment quantity. The alarm triggering construction submodule, based on the aforementioned trend coordination segment quantity, constructs the fire alarm triggering basis and establishes fire alarm triggering segment information by comprehensively considering the combined trend performance of multiple trends.

10. A high-accuracy fire alarm judgment method based on multi-parameter fusion, characterized in that, The high-accuracy fire alarm system based on multi-parameter fusion according to any one of claims 1-9 is executed by including the following steps: S1: Acquire environmental monitoring sampling information, analyze temperature and smoke sampling sequences, perform differential operations on adjacent sampling points and map them to a unified time axis, compare two types of differential directions, determine the consistency trend in continuous segments, calibrate regional disturbance characteristics, and generate disturbance coupling segment quantity information. S2: Based on the perturbation coupling segment quantity information, extract the pressure and temperature sequence, identify contraction and heating events, analyze the adjacent structures of the event time sequence and the corresponding trigger positions, establish the correlation chain of gas contraction and heating behavior, form the thermal action path, and construct thermal diffusion and migration information. S3: Based on the thermal diffusion and migration information, extract the brightness and temperature sequence, calculate the change ratio, and by comparing the direction of the ratio rise and fall in continuous segments, identify the optical attenuation and temperature rise ratio characteristics and direction consistency, construct linkage features, and generate optical temperature drift ratio parameters. S4: Based on the light-temperature drift ratio parameter, extract the smoke and temperature sequence, calculate the difference and analyze the synchronization characteristics, compare the directional correlation in continuous segments, identify two types of differential synchronization transition behaviors, construct joint mutation features, and generate smoke-temperature mutation segment records. S5: Based on the disturbance coupling segment quantity information, thermal diffusion migration information, light temperature drift ratio parameter, and smoke temperature change segment record, the multiple types of trend information are sequentially organized in a unified time series, the directional performance and coordination are analyzed, overlapping trend segments are identified, and the fire alarm triggering basis is constructed by comprehensively combining the trend performance, and fire alarm triggering segment information is generated.

Images