Smoke detection method based on multi-sensor fusion analysis

Abstract

The smoke detection method based on multi-sensor fusion analysis provided by the invention comprises the following steps: when a smoke sensor detects that the smoke concentration in a detection range reaches a first threshold value, acquiring a temperature rising speed in the detection range detected by a temperature sensor, acquiring the environmental heat radiation uniformity in the detection range detected by the human body sensor; generating a corresponding correction coefficient to correct the smoke concentration and the first threshold value to obtain a corresponding corrected smoke concentration and a corrected threshold value; if the corrected smoke concentration reaches a corrected threshold value, a comprehensive risk value is calculated based on the corrected smoke concentration and the temperature rising speed; and if the comprehensive risk value reaches a risk threshold, determining that the fire is a real fire. According to the invention, through the fusion processing of the smoke sensor, the temperature sensor and the human body sensor, the defect that fire detection through smoke is easily interfered by the environment and misinformation is generated at present is overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a smoke detection method based on multi-sensor fusion analysis. Background Art

[0002] Traditional fire detectors, such as those based on temperature, smoke, and light sensing, have numerous limitations. These detectors can only detect specific, single information features and are susceptible to environmental and spatial influences. In open areas or harsh environments, interference from factors like temperature and smoke concentration can slow sensor signal transmission, weaken signal strength, or even cause it to fail, resulting in poor recognition and high false alarm rates. For example, in dusty industrial plants, standard smoke sensors often trigger false alarms due to dust interference.

[0003] With technological advancements, video-based fire detection technology has gradually gained popularity. This technology overcomes the drawback of traditional fire detectors, which require proximity to a fire source to operate. It is suitable for complex and extensive outdoor environments and can provide rich visual information about early-stage fires. However, this technology also faces challenges. Due to the complex and changing scenes and uncertain environmental factors, it is prone to misjudgment when distinguishing smoke from similar interference objects such as haze, clouds, and mist. For example, in foggy weather, a video-based smoke detection algorithm may misidentify fog as smoke, resulting in a false alarm.

[0004] Among the many factors that determine fire detection, smoke, as an early sign of a fire, plays a crucial role in the reliability of fire detection systems. Existing technologies for smoke detection have shortcomings. Some algorithms rely solely on smoke energy at a specific moment, ignoring potential temporary interference, such as drifting white pollutants, which can cause energy attenuation and affect detection accuracy. Furthermore, existing algorithms struggle to accurately distinguish interference objects with characteristics very similar to smoke, further reducing detection reliability.

[0005] Furthermore, traditional fire detection methods lack the ability to adapt to different environments and scenarios. In some special locations, such as kitchens and factories with high dust levels, or areas with significant fluctuations in light intensity between day and night, fixed detection thresholds and sensitivity settings cannot meet the requirements for accurate detection, easily resulting in missed or false alarms. Summary of the Invention

[0006] The main purpose of the present invention is to provide a smoke detection method based on multi-sensor fusion analysis, aiming to overcome the defect that current fire detection through smoke is easily affected by environmental interference and produces false alarms.

[0007] To achieve the above objectives, the present invention provides a smoke detection method based on multi-sensor fusion analysis, comprising the following steps: When the smoke sensor detects that the smoke concentration within the detection range reaches a first threshold, the temperature sensor detects a temperature rise rate within the detection range, and the human body sensor detects a uniformity of ambient thermal radiation within the detection range; Based on the temperature rise rate and the uniformity of the ambient thermal radiation, a corresponding correction coefficient is generated to correct the smoke concentration and the first threshold value respectively, to obtain a corresponding corrected smoke concentration and a corrected threshold value; If the corrected smoke concentration reaches a correction threshold, a comprehensive risk value is calculated based on the corrected smoke concentration and the temperature rise rate; If the comprehensive risk value reaches the risk threshold, it is determined to be a real fire.

[0008] Furthermore, after it is determined to be a real fire, the following steps are taken: The risk level is determined based on the predicted heat source diffusion speed of the temperature rise speed, the position of the person located by the human body sensor, and the smoke concentration, and a differentiated alarm is output based on the risk level.

[0009] Furthermore, based on the temperature rise rate and the uniformity of the ambient thermal radiation, corresponding correction coefficients are generated to correct the smoke concentration and the first threshold value respectively, to obtain corresponding corrected smoke concentration and corrected threshold value, including: If the temperature rising rate is less than a second threshold, generating a smoke concentration correction coefficient to correct the smoke concentration to obtain a corrected smoke concentration; If the uniformity of the ambient heat radiation exceeds a preset uniformity, a determination threshold correction coefficient is generated to correct the first threshold to obtain a corrected threshold.

[0010] Furthermore, calculating a comprehensive risk value based on the corrected smoke concentration and temperature rise rate includes: Obtain the smoke concentration rise start time T1 and the temperature rise rate breaking through the speed threshold time T2; based on the time difference between T1 and T2, mark the corresponding causal relationship and assign the corresponding causal coefficient; Obtain the preset smoldering stage characteristic interval and open flame stage characteristic interval. Different stage characteristic intervals correspond to different corrected smoke concentrations and temperature rise rates. After the corrected smoke concentration and the temperature rise rate are matched to the corresponding stage characteristic interval, a corresponding compensation factor is generated based on the fluctuation of the temperature rise rate; wherein, in different stage characteristic intervals, the generation rules of the compensation factor are different; Determining weights corresponding to the modified smoke concentration and temperature rise rate based on the marked causal association and the matched stage characteristic interval; Based on the corresponding weights, the corrected smoke concentration and temperature rise rate are fused and calculated to obtain a basic risk score; the basic risk score is multiplied by the causal coefficient and the compensation factor in sequence to obtain the comprehensive risk value.

[0011] Furthermore, based on the marked causal relationship and the matched stage characteristic interval, weights corresponding to the modified smoke concentration and temperature rise rate are determined, including: Obtaining a preset weight mapping relationship; wherein the weight mapping relationship includes weight combinations corresponding to different causal relationships and stage feature intervals; Based on the causal relationship and the matched stage characteristic interval, the weights corresponding to the corrected smoke concentration and the temperature rising rate are determined through the weight mapping relationship.

[0012] Furthermore, before performing fusion calculation on the corrected smoke concentration and temperature rising rate based on the corresponding weights, the following steps are included: The corrected smoke concentration and temperature rising rate are normalized to the range of 0-1 respectively.

[0013] Furthermore, obtaining the temperature rising rate within the detection range detected by the temperature sensor includes: The temperature sensor collects temperature data at multiple different points within the detection range at a preset sampling frequency; Calculate the average temperature of each point at the same time, and calculate the temperature rise rate based on the difference between the corresponding temperature averages at adjacent times. If the fluctuation of the temperature rise rate of multiple samples is less than the fluctuation threshold, the average of the temperature rise rates of multiple samples is used as the final temperature rise rate; If the temperature rise rate fluctuation of multiple samples reaches the fluctuation threshold, the preset sampling frequency of the temperature sensor is increased, and after adding multiple samples, the abnormal sampling values ​​are eliminated; the average value of the remaining temperature rise rates is used as the final temperature rise rate.

[0014] Furthermore, obtaining the uniformity of the ambient thermal radiation within the detection range detected by the human body sensor includes: The detection range is divided into a core area and a peripheral area with a preset radius; the core area is the core area in the detection range where the smoke concentration reaches the first threshold; The thermal radiation intensity data of multiple sampling points in the core area and the peripheral area are collected through human body sensors; Calculate the standard deviation of the thermal radiation intensity in the core area, the standard deviation of the thermal radiation intensity in the peripheral area, and the difference in thermal radiation intensity between the core area and the peripheral area; Based on the standard deviation of the thermal radiation intensity of the core area, the standard deviation of the thermal radiation intensity of the peripheral area, and the difference in thermal radiation intensity between the core area and the peripheral area, the uniformity of the ambient thermal radiation is calculated using a preset algorithm.

[0015] Furthermore, when calculating the comprehensive risk value based on the corrected smoke concentration, temperature rise rate, and ambient thermal radiation uniformity, the method further includes: Obtaining a duration during which the corrected smoke concentration reaches a correction threshold, and generating a duration correction factor based on the duration; The comprehensive risk value is calculated based on the duration correction factor, the corrected smoke concentration, and the temperature rise rate.

[0016] The smoke detection method based on multi-sensor fusion analysis provided by the present invention includes: when the smoke sensor detects that the smoke concentration within the detection range has reached a first threshold, obtaining the temperature rise rate detected by the temperature sensor within the detection range, and obtaining the uniformity of the ambient thermal radiation within the detection range detected by the human body sensor; based on the temperature rise rate and the uniformity of the ambient thermal radiation, generating corresponding correction coefficients to correct the smoke concentration and the first threshold, respectively, to obtain corresponding corrected smoke concentration and corrected threshold; if the corrected smoke concentration reaches the correction threshold, calculating a comprehensive risk value based on the corrected smoke concentration and the temperature rise rate; if the comprehensive risk value reaches the risk threshold, determining that it is a real fire. In this invention, by combining the fusion processing of smoke sensors, temperature sensors, and human body sensors, the defect of current fire detection through smoke, which is easily affected by environmental interference and produces false alarms, is overcome. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 1 is a schematic diagram of the steps of a smoke detection method based on multi-sensor fusion analysis in one embodiment of the present invention; Figure 2 This is a structural block diagram of a smoke detection device based on multi-sensor fusion analysis in one embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0018] The implementation, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] Reference Figure 1 In one embodiment of the present invention, a smoke detection method based on multi-sensor fusion analysis is provided, comprising the following steps: Step S1, when the smoke sensor detects that the smoke concentration within the detection range reaches a first threshold, obtaining the temperature rise rate within the detection range detected by the temperature sensor, and obtaining the uniformity of the ambient thermal radiation within the detection range detected by the human body sensor; Step S2, based on the temperature rise rate and the uniformity of the ambient thermal radiation, generating corresponding correction coefficients to correct the smoke concentration and the first threshold value, respectively, to obtain corresponding corrected smoke concentration and corrected threshold value; Step S3, if the corrected smoke concentration reaches the correction threshold, calculating a comprehensive risk value based on the corrected smoke concentration and the temperature rise rate; Step S4: If the comprehensive risk value reaches the risk threshold, it is determined to be a real fire.

[0021] In this embodiment, as described in step S1 above, the detection result of the smoke sensor is used as the initial trigger condition. When the smoke sensor detects in real time that the smoke concentration within its detection range reaches a preset first threshold (this threshold is the initial smoke concentration threshold for triggering a suspected fire determination), the multi-sensor linkage acquisition process is immediately initiated.

[0022] On the one hand, a data collection command is sent to the temperature sensor, which then monitors temperature changes within the same detection range and calculates the temperature change rate per unit time to derive the core parameter, the temperature rise rate. On the other hand, a mode switching and data collection command is sent to the human body sensor (PIR sensor). The thermal radiation distribution within the human body sensor's detection range is then calculated using a preset algorithm and output as the ambient thermal radiation uniformity parameter, providing data support for subsequent correction logic. The core purpose of this step is to quickly supplement two parameters directly related to fire characteristics after the smoke concentration triggers an initial suspected alarm. In one embodiment, the human body infrared sensor operates based on the infrared energy of thermal radiation and its own thermal properties, detecting temperature changes caused by thermal radiation from a heat source. In a smoke detection scenario, it can be used to detect changes in human body thermal radiation in different areas. By analyzing thermal radiation intensity data from multiple sampling points within the detection range, the standard deviation of the thermal radiation intensity and the difference between areas are calculated to determine the uniformity of the ambient thermal radiation.

[0023] As described in step S2 above, smoke concentration correction is performed by generating a unique smoke concentration correction coefficient based on the temperature rise rate. The temperature rise rate is compared with a preset second threshold (used to distinguish between a fire heat source and a non-fire, normal temperature environment) to determine whether the smoke is driven by a heat source. If the temperature rise rate is lower than the second threshold, indicating no significant heat source, a corresponding correction coefficient is generated to reduce the original smoke concentration weight. If the temperature rise rate is higher than the second threshold, indicating a persistent heat source, a corresponding correction coefficient is generated to maintain or adjust the original smoke concentration weight, ultimately resulting in a corrected smoke concentration.

[0024] To correct the first threshold, a unique judgment threshold correction coefficient is generated based on the uniformity of the ambient thermal radiation. The uniformity is compared with a preset uniformity threshold (this threshold is used to distinguish between localized high temperatures in a fire and uniform thermal environments in a non-fire environment) to determine the correlation between smoke and fire. If the uniformity exceeds the preset uniformity threshold, indicating a balanced thermal distribution with no localized high temperatures, a corresponding correction coefficient is generated to increase the first threshold to reduce false alarms. If the uniformity does not exceed the preset uniformity threshold, indicating the presence of localized high temperatures, consistent with the characteristics of a fire heat source, the system generates a corresponding correction coefficient to maintain or adjust the first threshold, ultimately resulting in a corrected threshold.

[0025] As described in step S3 above, when the corrected smoke concentration value obtained in step S2 reaches the correction threshold, it indicates that the current smoke has passed the preliminary authenticity verification, and then enters step S3 to calculate the comprehensive risk value. The calculation process only revolves around two core parameters: corrected smoke concentration and temperature rise rate.

[0026] First, data preprocessing of the corrected smoke concentration and temperature rise rate is performed through a preset algorithm. For example, the two parameters are normalized to the range of 0-1 to ensure that the two are compatible. Secondly, the weights of the two parameters are determined in combination with preset rules. The weight distribution can be associated with the characteristics of the fire development stage. For example, the weight of the corrected smoke concentration is emphasized in the smoldering stage, and the weight of the temperature rise rate is emphasized in the open flame stage. The specific rules are subject to the subsequent refinement of the plan. Finally, based on the determined weights, the pre-processed corrected smoke concentration and temperature rise rate are fused and calculated to obtain a quantitative comprehensive risk value, which directly reflects the extent to which the current scene conforms to the real fire characteristics.

[0027] As described in step S4 above, the comprehensive risk value calculated in step S3 is compared with a preset risk threshold. If the comprehensive risk value reaches or exceeds the risk threshold, it indicates that the multi-dimensional parameters such as smoke and temperature in the current scene meet the characteristics of a fire, and the risk level has reached a level that requires an early warning. Therefore, it is determined to be a real fire, and the subsequent alarm and emergency linkage mechanism is triggered. If the comprehensive risk value does not reach the risk threshold, it indicates that the current scene may be a suspected situation caused by interference factors (such as temporary smoke, local temperature fluctuations, etc.). It is not determined to be a real fire, and the system returns to the initial monitoring state and continues to monitor the parameters within the detection range. This final judgment step ensures the accuracy of fire warnings, avoiding the disruption of normal life caused by false alarms and preventing the safety hazards caused by missed alarms.

[0028] In one embodiment, after determining that a fire is real, the following steps are performed: The risk level is determined based on the predicted heat source diffusion speed of the temperature rise speed, the position of the person located by the human body sensor, and the smoke concentration, and a differentiated alarm is output based on the risk level.

[0029] In this embodiment, once a fire is identified as a real fire, the system first calculates the heat source diffusion rate using a preset model based on the temperature rise rate (faster temperature rise, generally higher heat source diffusion rate). Furthermore, the system uses a human body sensor to determine the specific location of people within the detection range. Combined with smoke concentration (the higher the concentration, the greater the fire's impact and hazard), these three key pieces of information are used to categorize the risk level (e.g., high, medium, and low). A high risk level is determined if the heat source spreads quickly, people are in the core area, and smoke concentration is high; a low risk level is determined if the heat source spreads slowly, people are in a safe area, and smoke concentration is low). Finally, differentiated alerts are generated based on the risk level. For example, high risk triggers audible and visual alarms, emergency evacuation broadcasts, and fire alarm linkage; medium risk triggers local audible and visual alarms and personnel guidance notifications; and low risk triggers only basic audible and visual alarms, ensuring that the alarm response is appropriate to the actual severity of the fire.

[0030] In one embodiment, based on the temperature rise rate and the uniformity of the ambient thermal radiation, a corresponding correction coefficient is generated to correct the smoke concentration and the first threshold value, respectively, to obtain the corresponding corrected smoke concentration and corrected threshold value, including: If the temperature rising rate is less than a second threshold, generating a smoke concentration correction coefficient to correct the smoke concentration to obtain a corrected smoke concentration; If the uniformity of the ambient heat radiation exceeds a preset uniformity, a determination threshold correction coefficient is generated to correct the first threshold to obtain a corrected threshold.

[0031] In this embodiment, when the temperature rise rate is less than the second threshold, the smoke concentration correction mechanism is activated. The second threshold is a critical value used to distinguish between fire heat sources and non-fire environments. A temperature rise rate less than this threshold means that there is no significantly and continuously increasing heat source within the detection range. The smoke at this time is more likely to be caused by non-fire interference factors such as dust and oil smoke. According to the difference between the temperature rise rate and the second threshold, a corresponding smoke concentration correction coefficient is generated by a preset algorithm. This coefficient is less than 1 and is used to reduce the weight of the original smoke concentration in the judgment. The original smoke concentration is multiplied by the correction coefficient to obtain the corrected smoke concentration, that is, the corrected smoke concentration. The core of this correction process is to filter non-fire smoke without heat source support through temperature parameters to ensure that subsequent judgments are based on smoke intensity data that is closer to the actual fire characteristics.

[0032] When the uniformity of the ambient thermal radiation exceeds the preset uniformity threshold, the first threshold correction mechanism is activated. The preset uniformity threshold is the critical value that distinguishes localized high temperatures from non-fire uniform thermal environments. When the uniformity of the ambient thermal radiation exceeds this threshold, it indicates that the thermal radiation distribution within the detection range is balanced, with no obvious localized high-temperature spots. At this point, the smoke is less likely to be associated with a fire and is more likely to be generated by non-fire scenarios such as cooking or equipment heat dissipation. Based on the degree of deviation between the ambient thermal radiation uniformity and the preset uniformity threshold, a preset algorithm generates a corresponding judgment threshold correction coefficient. This coefficient, greater than 1, is used to improve the judgment criteria for the first threshold. The first threshold is multiplied by this correction coefficient to obtain the corrected judgment threshold, or the corrected threshold. The core of this correction process is to adjust the judgment criteria based on the characteristics of the thermal radiation distribution, reducing false alarms in non-fire scenarios and ensuring that only smoke with fire heat source characteristics passes subsequent judgment.

[0033] In one embodiment, calculating the comprehensive risk value based on the corrected smoke concentration and temperature rise rate includes: Obtain the smoke concentration rise start time T1 and the temperature rise rate breaking through the speed threshold time T2; based on the time difference between T1 and T2, mark the corresponding causal relationship and assign the corresponding causal coefficient; Obtain the preset smoldering stage characteristic interval and open flame stage characteristic interval. Different stage characteristic intervals correspond to different corrected smoke concentrations and temperature rise rates. After the corrected smoke concentration and the temperature rise rate are matched to the corresponding stage characteristic interval, a corresponding compensation factor is generated based on the fluctuation of the temperature rise rate; wherein, in different stage characteristic intervals, the generation rules of the compensation factor are different; Determining weights corresponding to the modified smoke concentration and temperature rise rate based on the marked causal association and the matched stage characteristic interval; Based on the corresponding weights, the corrected smoke concentration and temperature rise rate are fused and calculated to obtain a basic risk score; the basic risk score is multiplied by the causal coefficient and the compensation factor in sequence to obtain the comprehensive risk value.

[0034] In this embodiment, two key time points are first recorded: the start time T1, when smoke density begins to rise, and the time T2, when the temperature rise rate exceeds a preset threshold. By calculating the time difference between T1 and T2, the correlation between the smoke density rise and the temperature rise is analyzed, thereby marking the causal relationship between the two and assigning a corresponding causal coefficient. For example, if the time difference between T1 and T2 is extremely small (e.g., less than 3 seconds), it indicates that the smoke and temperature rises occurred almost synchronously, with a strong causal relationship, and the system assigns a higher causal coefficient (e.g., 1.5). If the time difference is large (e.g., more than 10 seconds), it indicates a weaker correlation, likely caused by different factors, and a lower causal coefficient (e.g., 0.6) is assigned. The core of this step is to strengthen the characteristic weight of the synchronous changes in smoke and temperature in real fires through correlation analysis in the time dimension, while reducing the interference caused by the isolated occurrence of smoke and temperature in non-fire scenarios.

[0035] Next, the two preset stage characteristic intervals are called, namely the smoldering stage characteristic interval and the open flame stage characteristic interval. Among them, the smoldering stage characteristic interval corresponds to a parameter combination in which the corrected smoke concentration is high but rises slowly, and the temperature rise rate is low and stable. The open flame stage characteristic interval corresponds to a parameter combination in which the corrected smoke concentration diffuses rapidly, and the temperature rise rate is high and may be accompanied by fluctuations. The numerical combination of the corrected smoke concentration and the temperature rise rate obtained in real time is compared with the characteristic parameters of these two intervals to determine the current stage of development of the suspected fire. This matching process provides a staged judgment basis for the subsequent compensation factor generation and weight allocation, ensuring that subsequent calculations adapt to the characteristic laws of different fire stages.

[0036] After correcting the smoke concentration and temperature rise rate to match the corresponding stage characteristic interval, the temperature rise rate fluctuation is further analyzed (for example, by calculating the fluctuation amplitude through the standard deviation or range of multiple consecutive samplings) and a corresponding compensation factor is generated. The generation rules vary depending on the stage. In the smoldering stage characteristic interval, the temperature rise rate normally fluctuates slightly. If the actual fluctuation amplitude is less than the preset fluctuation threshold for that stage, it indicates that the heat source is stable and meets the smoldering characteristics, and a compensation factor greater than 1 (for example, 1.1) is generated. If the fluctuation amplitude exceeds the threshold, it indicates abnormal interference (such as local airflow), and a compensation factor less than 1 (for example, 0.9) is generated. In the open flame stage characteristic interval, the temperature rise rate is allowed to fluctuate within a reasonable range. If the fluctuation amplitude is within the preset normal range for that stage, it indicates that it meets the dynamic characteristics of open flame combustion, and a higher compensation factor (for example, 1.2) is generated. If the fluctuation amplitude is excessive, it indicates abnormal heat source stability (such as non-fire high temperature interference), and a significantly lower compensation factor (for example, 0.7) is generated. This differentiated rule can accurately identify the authenticity of heat sources in different stages, reducing the impact of environmental interference on risk assessment.

[0037] Then, based on the marked causal relationship (i.e., the strength of the correlation corresponding to the time difference between T1 and T2) and the matching stage characteristic interval (smoldering or open flame stage), the weights corresponding to the modified smoke concentration and temperature rise rate are retrieved from the preset weight mapping relationship. For example, in the smoldering stage scenario with a strong causal relationship, the modified smoke concentration has a greater impact on fire determination, with a weight of 70%, while the temperature rise rate has a weight of 30%. In the open flame stage scenario with a strong causal relationship, the weight of the temperature rise rate is increased to 60%, and the weight of the modified smoke concentration is reduced to 40%. If the causal relationship is weak, the weight difference between the two parameters is reduced (for example, 50% each) to reduce the excessive influence of a single parameter. This step dynamically adjusts the weights to make the risk calculation more closely reflect the core characteristics of fires at different stages.

[0038] Finally, the corrected smoke concentration and temperature rise rate are normalized (e.g., converted into standardized values ​​in the range of 0-1), and then weighted summed according to the weights determined in the previous step to obtain the basic risk score. Subsequently, the basic risk score is multiplied by the causal coefficient and the compensation factor in sequence to finally obtain the comprehensive risk value. For example, if the basic risk score is 0.6, the causal coefficient is 1.5, and the compensation factor is 1.1, then the comprehensive risk value is 0.6×1.5×1.1=0.99. This calculation process uses multi-level parameter correction (causal association strengthens time logic, compensation factor optimization stage adaptability, and weight distribution highlights core characteristics). The final output comprehensive risk value can objectively quantify the extent to which the current scene conforms to the characteristics of a real fire, providing an accurate quantitative basis for subsequent fire judgments.

[0039] In one embodiment, based on the marked causal relationship and the matched stage characteristic interval, weights corresponding to the modified smoke concentration and temperature rise rate are determined, including: Obtaining a preset weight mapping relationship; wherein the weight mapping relationship includes weight combinations corresponding to different causal relationships and stage feature intervals; Based on the causal relationship and the matched stage characteristic interval, the weights corresponding to the corrected smoke concentration and the temperature rising rate are determined through the weight mapping relationship.

[0040] In this embodiment, a pre-set three-dimensional weight mapping table of causal relationships, stage characteristic intervals, and weight combinations is first retrieved from local storage or a cloud database. This mapping table is not a fixed, single rule, but rather a standardized system formed through calibration of extensive fire scenario experimental data and risk assessment models. Its horizontal dimension covers two types of causal relationships: strong causal relationships and weak causal relationships. Its vertical dimension covers two types of stage characteristic intervals: smoldering and open flame. These two dimensions intersect to form four combined scenarios, each corresponding to a unique set of weight combinations (weights for modified smoke concentration and temperature rise rate). For example, in the strong causal-smoldering scenario, the numerical value of the weight combination is set based on the principle of emphasizing core risk parameters and deemphasizing interfering parameters. For example, in the strong causal scenario, the core stage parameters (smouldering with heavy smoke, open flame with heavy temperature) are prioritized, while in the weak causal scenario, the fundamental judgment parameter, smoke, is prioritized. The core value of this step lies in the fact that the pre-set, standardized mapping relationship avoids randomness in weight assignment and ensures consistent and traceable weight determination logic across different scenarios.

[0041] During matching, two key matching conditions are identified: the causal relationship (strong or weak) marked in the previous causal logic verification phase, and the phase characteristic interval (smoldering or open flame) determined in the previous fire phase matching phase. Subsequently, using these two conditions as search keywords, a cross-search is performed in the weight mapping relationship table obtained in step 1. After the search is complete, the weight values ​​in this combination are directly extracted and assigned to the corrected smoke concentration and temperature rise rate, respectively. This dual-condition precise search ensures that the weight allocation is highly compatible with the causal rationality and fire stage of the current scenario, avoiding risk calculation deviations caused by a disconnect between weights and scenarios.

[0042] In one embodiment, before performing the fusion calculation on the corrected smoke concentration and the temperature rise rate based on the corresponding weights, the following steps are included: The corrected smoke concentration and temperature rising rate are normalized to the range of 0-1 respectively.

[0043] In one embodiment, obtaining the temperature rising rate within the detection range detected by the temperature sensor includes: The temperature sensor collects temperature data at multiple different points within the detection range at a preset sampling frequency; Calculate the average temperature of each point at the same time, and calculate the temperature rise rate based on the difference between the corresponding temperature averages at adjacent times. If the fluctuation of the temperature rise rate of multiple samples is less than the fluctuation threshold, the average of the temperature rise rates of multiple samples is used as the final temperature rise rate; If the temperature rise rate fluctuation of multiple samples reaches the fluctuation threshold, the preset sampling frequency of the temperature sensor is increased, and after adding multiple samples, the abnormal sampling values ​​are eliminated; the average value of the remaining temperature rise rates is used as the final temperature rise rate.

[0044] In this embodiment, multi-point, standardized sampling is first implemented to ensure that temperature data fully reflects the distribution and changing trends of heat sources within the detection range, avoiding the limitations of single-point sampling. Specifically, a preset sampling frequency is first configured for the temperature sensor (based on the environmental stability of the detection scenario, such as 1 second / time in conventional scenarios and 0.5 seconds / time in complex scenarios). This frequency must balance data timeliness with sensor energy consumption. Simultaneously, based on the spatial size of the smoke sensor's detection range, the detection range is divided into multiple different points, and the temperature sensor synchronously collects real-time temperature data at each point at a preset frequency. Through multi-point synchronous sampling, single data deviations caused by local temperature anomalies (such as temporary heating of the device) are avoided, providing a comprehensive and objective raw data foundation for the subsequent calculation of an accurate temperature rise rate.

[0045] Then, through mean calculation and fluctuation verification, a stable temperature rise rate is screened out to ensure that the data can reflect the real temperature change trend. The specific operation is divided into three stages: the first stage is mean aggregation, taking the arithmetic mean of the temperature data of multiple points at the same sampling time point (for example, at a certain moment, the temperatures of 3 points are 25°C, 25.2°C, and 24.8°C, then the temperature average value at that moment is 25°C), converting the multi-point data into single-value data reflecting the overall temperature level, eliminating the interference of point differences on trend judgment; the second stage is speed calculation, based on the temperature average of two adjacent sampling time points (such as 25°C at time t1 and 25.3°C at time t2, with a time interval of 1 second), calculate the single temperature rise rate (such as 0.3°C / second), and continuously calculate the temperature rise rate corresponding to multiple sampling (such as 3 times); the second stage is speed calculation, based on the temperature average of two adjacent sampling time points (such as 25°C at time t1 and 25.3°C at time t2, with a time interval of 1 second), calculate the single temperature rise rate (such as 0.3°C / second), and continuously calculate the temperature rise rate corresponding to multiple sampling (such as 3 times); The third stage involves fluctuation verification. A preset fluctuation threshold (based on temperature sensor accuracy and ambient temperature stability, such as 0.1°C / second) is used to calculate the maximum difference between the temperature rise rates of multiple samples (for example, if the three rates are 0.28°C / second, 0.3°C / second, and 0.31°C / second, the maximum difference is 0.03°C / second). If this difference is less than the fluctuation threshold, the temperature rise trend is stable, with no significant abnormal interference. The arithmetic mean of these multiple rates (for example, 0.297°C / second) is then used as the final temperature rise rate. This process of mean aggregation, rate calculation, and fluctuation verification ensures that the final temperature rise rate accurately reflects the overall stable trend of temperature change, avoiding misjudgments caused by single data fluctuations.

[0046] For scenarios with large temperature rise rate fluctuations, the accuracy and reliability of the final data are ensured by increasing the sampling frequency and removing outliers. The specific operation logic is as follows: First, when the temperature rise rate fluctuations across multiple samples reach or exceed the fluctuation threshold (for example, three samples with a maximum difference of 0.2°C / second and 0.4°C / second, respectively, exceeding the threshold of 0.1°C / second), the current temperature fluctuation is determined to be unstable (such as a local heat source with intermittent strength or airflow interference). At this time, the temperature sensor's preset sampling frequency is automatically increased to twice the original frequency (for example, from 1 second to 0.5 seconds), capturing temperature changes with higher temporal resolution and reducing information omissions. Second, multiple additional samples are taken at the increased frequency to obtain more dimensional temperature rise rate data. Then, outlier samples are removed, and the mean and standard deviation of all sampling rates are calculated. Values ​​outside the range of three standard deviations from the mean are identified as outliers and removed from the dataset. Finally, the arithmetic mean of the remaining non-outlier samples is taken as the final temperature rise rate. For unstable scenarios, a dual calibration strategy of high-frequency sampling + anomaly rejection is adopted to avoid data distortion caused by excessive fluctuations. At the same time, through precise screening, it ensures that the final temperature rise rate can reflect the actual heat source change trend, providing high-quality temperature parameters for subsequent multi-sensor fusion analysis.

[0047] In one embodiment, obtaining the uniformity of ambient thermal radiation within the detection range detected by the human body sensor includes: The detection range is divided into a core area and a peripheral area with a preset radius; the core area is the core area in the detection range where the smoke concentration reaches the first threshold; The thermal radiation intensity data of multiple sampling points in the core area and the peripheral area are collected through human body sensors; Calculate the standard deviation of the thermal radiation intensity in the core area, the standard deviation of the thermal radiation intensity in the peripheral area, and the difference in thermal radiation intensity between the core area and the peripheral area; Based on the standard deviation of the thermal radiation intensity of the core area, the standard deviation of the thermal radiation intensity of the peripheral area, and the difference in thermal radiation intensity between the core area and the peripheral area, the uniformity of the ambient thermal radiation is calculated using a preset algorithm.

[0048] In this embodiment, the overall detection range is first divided into two sub-areas based on a preset radius: a core area and a peripheral area. The core area is defined based directly on smoke concentration. Specifically, it is the core area within the detection range where smoke concentration first reaches the first threshold. This area is the key monitoring area where smoke initially accumulates and the fire risk is high. The peripheral area is the area covered by the preset radius extending outward from the core area. It is used to capture the thermal radiation distribution characteristics around the core area, forming a supplementary monitoring area for the core area. This zoning method enables differentiated monitoring of the initial smoke generation area and its surrounding environment.

[0049] After completing the zone division, the human body sensor switches to thermal radiation intensity acquisition mode, performing multi-point sampling in both the core and peripheral areas. In the core area, the sensor collects thermal radiation intensity data at multiple locations according to a pre-defined sampling point distribution rule (e.g., a uniform grid distribution). In the peripheral area, the same sampling rule is applied, collecting equal or proportional amounts of thermal radiation intensity data within the area. Multi-point sampling aims to avoid accidental errors associated with a single sampling point, ensuring that the acquired thermal radiation intensity data truly reflects the overall thermal distribution characteristics of both areas, providing reliable data support for subsequent uniformity calculations.

[0050] Statistical analysis of the collected thermal radiation intensity data for the core and peripheral areas yielded three key parameters: First, the standard deviation of the core area's thermal radiation intensity, which reflects the degree of dispersion of thermal radiation intensity at different sampling points within the core area. A smaller standard deviation indicates a more uniform thermal radiation distribution within the core area. Second, the standard deviation of the peripheral area's thermal radiation intensity, which serves the same purpose as the core area's standard deviation and measures the uniformity of thermal radiation distribution within the peripheral area. Third, the difference in thermal radiation intensity between the core and peripheral areas, calculated by calculating the absolute difference between the average thermal radiation intensities of the two regions, reflects the degree of disparity in the overall thermal radiation levels of the two regions. A smaller difference indicates a more balanced thermal radiation distribution between the regions. These three parameters collectively characterize the thermal radiation distribution characteristics within the detection range from the perspectives of both intra-regional uniformity and inter-regional consistency.

[0051] The three statistical parameters (core area standard deviation, peripheral area standard deviation, and inter-area difference) are input into a preset algorithm for fusion calculation, ultimately yielding the uniformity of ambient thermal radiation. The algorithm's core logic applies a weighted approach to the three parameters, with the standard deviation and inter-area difference negatively correlated with uniformity (i.e., smaller parameter values ​​indicate higher uniformity). For example, the algorithm first normalizes the three parameters to the 0-1 range. A weighted sum is then calculated using preset weights (e.g., 40% for the core area standard deviation, 30% for the peripheral area standard deviation, and 30% for the inter-area difference). Finally, this weighted sum is subtracted from 1 to yield the uniformity of ambient thermal radiation (ranging from 0 to 1, with values ​​closer to 1 indicating more uniform thermal radiation distribution). This multi-parameter fusion calculation comprehensively reflects the overall uniformity of thermal radiation within the detection range, providing a quantitative basis for subsequent adjustments to the first threshold.

[0052] In one embodiment, when calculating the comprehensive risk value based on the corrected smoke concentration, temperature rise rate, and ambient thermal radiation uniformity, the method further includes: Obtaining a duration during which the corrected smoke concentration reaches a correction threshold, and generating a duration correction factor based on the duration; The comprehensive risk value is calculated based on the duration correction factor, the corrected smoke concentration, and the temperature rise rate.

[0053] In this embodiment, the stability and severity of the risk are first determined by correcting the duration of smoke concentration exceeding the threshold, providing a supplementary temporal basis for the risk value. This process is divided into two phases: the first phase is duration acquisition. This phase begins when the corrected smoke concentration first reaches the threshold, continuously monitoring and recording the duration that the concentration remains at or above the threshold. This duration directly reflects the sustained intensity of the risk. A longer duration indicates a more stable smoke risk and a lower probability of non-instantaneous interference. The second stage involves generating a duration correction factor. A pre-defined duration-based factor mapping rule is used, based on the logic that the longer the risk persists, the higher the fire probability. If the duration is ≤ 3 seconds (a transient threshold exceedance, likely indicating interference), a duration correction factor of 0.7-0.8 is generated (reducing the risk weight and avoiding misjudgments due to transient interference); if the duration is 3 seconds < ≤ 8 seconds (short-term duration, indicating a risk of concern), a duration correction factor of 1.0-1.1 is generated (maintaining or slightly increasing the risk weight); and if the duration is > 8 seconds (long-term duration, indicating a highly stable risk), a duration correction factor of 1.3-1.4 is generated (significantly increasing the risk weight and reinforcing the impact of stable risks). The factor values ​​are calibrated using historical fire data to ensure accurate matching of the actual risk probabilities corresponding to different durations. This overcomes the limitations of traditional risk assessments that rely solely on concentration values. By validating the time dimension, it eliminates risk misjudgments caused by transient interference while highlighting the threat level of stable risks, providing a more comprehensive basis for determining the overall risk value.

[0054] When calculating the comprehensive risk value, the duration correction factor is synergistically calculated with the two original parameters (corrected smoke concentration and temperature rise rate) to generate a comprehensive quantitative result that reflects concentration intensity, temperature trend, personnel risk, and time persistence. Integrating the duration correction factor as an independent dimension into the risk assessment, it forms a three-dimensional synergy with the original parameters: value, trend, and duration. This not only retains the advantages of the original multi-parameter integration, but also strengthens the impact of risk persistence through the time factor. Its technical value lies in ensuring that the comprehensive risk value more accurately matches the actual development patterns of fires (fire risk increases over time), avoiding misjudgments of low risks or omissions of high risks due to ignoring the time dimension, and further improving the scientific nature and reliability of risk assessments.

[0055] In one embodiment, before the smoke sensor detects that the smoke concentration within the detection range reaches a first threshold, the method further includes: After the smoke sensor is activated, it first uses the built-in environment type recognition module to collect the initial particle base number and day and night light intensity within the detection range; Based on the initial particle count, the first threshold is automatically adjusted, and the detection sensitivity level of the smoke sensor is adjusted at the same time; Adjust the sampling interval of the smoke sensor based on the day and night light intensity, and adjust the power consumption mode of the smoke sensor.

[0056] In this embodiment, after the smoke sensor is powered on, it does not directly enter the normal detection mode. Instead, it first activates the built-in environment type recognition module (integrated micro-particle detector and light sensor) to collect two key environmental parameters within the detection range: the initial particle base, that is, the baseline concentration value of particles with a diameter of 0.3-10μm in the air when there is no obvious smoke interference in the detection range (unit: mg / m 3 ), used to determine the degree of dust pollution in the environment (e.g., kitchens and factories have higher initial particle counts, while bedrooms and offices have lower counts). Second, day and night light intensity, or the real-time light intensity within the detection range (unit: lux), is used to distinguish between daytime (light intensity ≥ 100 lux) and nighttime (light intensity < 50 lux), and to determine the frequency of human activity and sensor power requirements (at night, human activity is lower, balancing detection accuracy and power consumption). This initial collection of environmental parameters allows the sensor to accurately perceive the inherent characteristics of the current detection scene, providing a basis for subsequent on-demand adjustments and avoiding the inapplicability of fixed parameters in complex environments.

[0057] According to the initial particle count, the first threshold and detection sensitivity are optimized to reduce the interference of inherent dust in the environment on smoke detection, ensuring that the sensor can accurately identify fire smoke. The specific operation is divided into two parts: the first part is the adjustment of the first threshold, which is based on the preset threshold adjustment rule based on the initial particle count. The rule is based on the logic that the more dust in the environment, the higher the first threshold should be to avoid false alarms: if the initial particle count is less than 0.1mg / m 3 (low dust environment, such as bedroom), determine that the environmental interference is small, and maintain the original first threshold (such as 0.3mg / m 3 ) remains unchanged; if 0.1mg / m 3 ≤ Initial particle count<0.2mg / m 3 (Medium dust environment, such as office), determine that the environment has slight interference, increase the first threshold by 10% (such as 0.33mg / m 3 ); if the initial particle count is ≥ 0.2 mg / m 3 (High dust environment, such as kitchen, workshop), if the environmental interference is large, the first threshold is increased by 20% (such as 0.36mg / m 3). The second part is the adjustment of the detection sensitivity level. The sensitivity level is inversely correlated with the initial particle base: in a low-dust environment, the sensitivity is set to the high-efficiency level (which can identify tiny fire smoke particles of 0.3-0.5μm) to ensure that early weak smoke can be captured; in a medium-dust environment, it is set to the standard level (only identifying characteristic fire particles of 0.3-2μm, filtering out some dust particles); in a high-dust environment, it is set to the anti-interference level (only identifying typical fire smoke particles of 0.3-1μm, further eliminating interference from large-particle dust). The adjustment parameters of the threshold and sensitivity are calibrated through a large number of simulation experiments in different dust environments to ensure that the adjustment can balance the requirements of preventing false alarms and preventing missed alarms. Through environmental dust adaptation adjustment, the first threshold and sensitivity can dynamically match the inherent interference of the environment, avoiding frequent false alarms caused by too low a threshold in a high-dust environment, while ensuring detection accuracy in a low-dust environment and improving the environmental adaptability of the sensor.

[0058] The system distinguishes between daytime and nighttime scenarios based on daylight intensity, optimizing sampling intervals and power consumption modes to ensure both detection requirements and energy efficiency, making it particularly suitable for battery-powered sensor scenarios. The first step involves adjusting the sampling interval, using a preset threshold for determining daylight intensity (e.g., 50 lux as the dividing line between day and night). If the daylight intensity is ≥50 lux, the sampling interval is set to 1 second per time, ensuring real-time capture of smoke changes through high-frequency sampling. If the daylight intensity is <50 lux, the sampling interval is extended to 2 seconds per time, reducing the frequency of data collection while maintaining basic detection capabilities. The second part involves power mode adjustment, which coordinates the power mode with the sampling interval. During high-frequency sampling during the day, the sensor switches to high-performance mode, with all detection modules (particle detection and signal processing) operating at full capacity to ensure real-time data acquisition and signal analysis. During low-frequency sampling at night, the sensor switches to low-power mode, shutting down non-core modules (such as the environmental type recognition module) during sampling intervals. Only the core smoke detection circuit remains in low-power operation, while the signal processing unit's operating frequency is reduced, reducing sensor power consumption to 60%-70% of daytime mode. In the transitional light intensity range of 50-100 lux (such as dawn and dusk), a compromise mode is adopted, with a sampling interval of 1.5 seconds per sample and a balanced power mode (normal operation of the core modules and intermittent operation of the non-core modules). This day-night scenario adaptation achieves an optimal balance between sampling efficiency and power consumption at different times of the day: high-frequency sampling during the day ensures safety, while low power consumption at night extends battery life. This not only meets the detection needs of different scenarios, but also improves the sensor's practical lifespan and cost-effectiveness. It is particularly suitable for standalone installations without a continuous power supply.

[0059] In one embodiment, the method further comprises: Collect usage time data of smoke sensors, temperature sensors, and human body sensors, and build a time data matrix; Taking the detection range as the origin, the three-dimensional coordinates of the three types of sensors are obtained to generate a relative position graph; Retrieving a preset data array including the corrected smoke concentration, the temperature rise rate, and the uniformity of the ambient thermal radiation, and performing a data mutation on the preset data array based on the relative position graph to obtain a first data array; Performing cross mutation on the time data matrix and the first data array to obtain a mutated data array and a mutated data matrix; The relative position graph, the variation data array and the variation data matrix are superimposed according to preset rules, and a coding factor is generated based on the characteristics of the superimposed area, which is used to encode and transmit the data collected by the smoke sensor, the temperature sensor and the human body sensor.

[0060] In this embodiment, the sensor's built-in timing module and status monitoring unit first collect three core types of usage time data for each sensor type: cumulative operating hours, reflecting the overall wear and tear of the sensor; effective operating hours over the past 30 days (actual detection time after excluding faults and dormant states, accurate to the hour), reflecting the sensor's recent operational stability; and single continuous operating time (uninterrupted operating time within the current detection cycle, accurate to the minute), reflecting the sensor's current operating status. Subsequently, a time data matrix is ​​constructed using a two-dimensional structure consisting of sensor type and time data: the row dimension corresponds to the three sensor types, and the column dimension corresponds to the three types of time data (accumulated operating hours, effective operating hours over the past 30 days, and single continuous operating time). Each cell in the matrix is ​​populated with the corresponding time data value for the corresponding sensor. This conversion of sensor usage time, a key status feature, into structured matrix data provides a unified data format for subsequent cross-integration with detection parameter data. Furthermore, the use of time data to reflect sensor wear and tear ensures that subsequent coding factors can be associated with sensor reliability characteristics.

[0061] Next, through 3D coordinate positioning and graphical processing, the spatial positional relationships of the three types of sensors are converted into a visual graph, providing a basis for spatial dimensional variation in the data array. This operation is divided into two parts: the first involves 3D coordinate acquisition. Using the geometric center of the smoke detection range as the origin, the sensors' built-in IoT positioning module collects the 3D coordinates of the smoke sensor, temperature sensor, and human body sensor. The coordinate values ​​reflect the spatial position of each sensor relative to the origin. The second part involves relative position graph generation. The collected 3D coordinates of the three types of sensors are imported into the graph generation module, generating a 3D graph centered at the origin and containing the position markers of the three types of sensors. The graph uses markers of different colors or shapes to distinguish the three types of sensors. The graph also indicates the linear distance between each sensor and the origin, as well as the relative distance between sensors. Furthermore, the graph also indicates the distance between each sensor and the boundary of the detection range to indicate whether the sensor is positioned within the optimal detection area.

[0062] Then, based on the spatial relationships reflected in the sensor relative position graph, the data array of core detection parameters is mutated to reflect the impact of spatial position on the detection data. The specific operation consists of three steps: the first step is to retrieve a preset data array. The core parameter data for smoke detection within a preset time period is retrieved from the database. A preset data array is constructed: the rows correspond to the detection time nodes (each row contains detection data for one time node), and the columns correspond to the three core parameters (corrected smoke concentration, temperature rise rate, and ambient thermal radiation uniformity). Each cell is filled with the corresponding parameter detection value for the corresponding time node. The second step is to determine the spatial mutation rule. Key spatial features are extracted from the relative position graph: the distance between each sensor and the origin, and the distance between each sensor and the detection target area. Data mutation weights are set based on these two distances. The third step is the data mutation operation. Based on the matching relationship between the parameter and the spatial weight, each parameter in the preset data array is mutated: the spatial weight of the smoke sensor corresponds to the corrected smoke concentration, the spatial weight of the temperature sensor corresponds to the temperature rise rate, and the spatial weight of the human body sensor corresponds to the ambient thermal radiation uniformity. Each parameter value is multiplied by its corresponding weight to obtain the mutated parameter value.

[0063] Then, a crossover operation is performed on the temporal sensor state data (the temporal data matrix) and the spatially mutated detection data (the first data array), achieving deep fusion and bidirectional mutation of the two data types, generating structured data with both temporal and detection features. This operation is divided into two parts: the first is data dimensionality adaptation. Since the temporal data matrix and the first data array may have different dimensions, the temporal data matrix is ​​first dimensionally expanded: N copies of the temporal data matrix are made according to the number of time nodes in the first data array and arranged in chronological order to form an expanded temporal matrix. Simultaneously, the first data array is association-expanded by sensor type to form an association detection matrix, ensuring that the two data types fully match in terms of time nodes and sensor type dimensions. The second part is a crossover mutation operation, employing a crossover mutation algorithm that combines point-by-point matrix multiplication with eigenvalue compensation. For each time node, the time data of the corresponding sensor in the expanded time matrix is ​​point-by-point multiplied with the detection data of the corresponding sensor in the association detection matrix to generate basic crossover data. The row mean of the temporal data matrix and the column mean of the first data array are then extracted and added to the basic crossover data as compensation values ​​to enhance data features. After cross-mutation, the detection parameter dimension data and time data dimension data are extracted respectively: the cross-mutation data of the detection parameter dimension are arranged by time nodes to form a mutation data array; the cross-mutation data of the time data dimension are arranged by sensor type and time nodes to form a mutation data matrix.

[0064] Finally, by superimposing spatial patterns, arrays, and matrices in various forms, features are extracted to generate coding factors, enabling secure coded transmission of sensor-collected data. In one embodiment, multiple key features are extracted from the superimposed area, then converted to binary and combined to generate coding factors. During data encoding and transmission, the real-time data collected by the sensor is XORed with the coding factors to generate encrypted coded data. The coding factors generated by superimposing multiple forms of data uniquely associate the sensor's spatial position, usage status, and detection data features, ensuring the security of coded transmission while also enabling data traceability and integrity verification, preventing data tampering or leakage during transmission.

[0065] In one embodiment, the method further comprises: The preset smoke concentration, temperature rise rate, ambient thermal radiation uniformity, smoke sensor signal response time, temperature sensor component aging, and human body sensor detection angle deviation are combined to form six types of coding basic parameters; Construct a preset standard coding table, including the correspondence between the original data and the coding symbols; map the six basic coding parameters to different row / column dimensions of the coding table, so that each row of the coding table corresponds to a type of parameter and each column corresponds to a different value range of the parameter, forming a parameter-associated initial coding table; A multi-parameter time series stacking diagram is generated with time as the horizontal axis and six types of coding basic parameters as the vertical axis to extract the stacking structural characteristics of the diagram; the number of columns and row order of the coding table are adjusted according to the stacking structural characteristics to complete the dimensional structure variation of the coding table; Generate parameter trend curves for each of the six basic coding parameters and extract the curvature variation characteristics of each parameter trend curve; perform content variation on the coding code based on the curvature variation characteristics to obtain the final variation coding table; The real-time data generated during the smoke detection process is encoded based on the variation coding table to generate a data coding string; the stacking structure characteristics and curvature change characteristics are used as coding verification information and embedded at the end of the data coding string.

[0066] In this embodiment, the first stage is encoding preparation. The core focus is on screening and integrating two key parameters: core smoke detection parameters (smoke concentration, temperature rise rate, and ambient thermal radiation uniformity), which reflect the safety status of the detection scenario; and sensor operating parameters (signal response time, component aging, and detection angle deviation), which reflect sensor reliability. The units and value ranges of these parameters are standardized to form a unified set of basic encoding parameters, providing a data foundation for subsequent coding table construction.

[0067] Next, the initial coding framework is built. First, a standard coding table corresponding to the original data and coding symbols is constructed according to industry specifications; then the six basic parameters are bound to the coding table: each row corresponds to a type of parameter, and each column corresponds to a value range of the parameter, forming an initial coding table specifically for the smoke detection scenario.

[0068] Next, adjust the coding table structure. First, generate a multi-parameter time series stacked plot (with time on the horizontal axis and parameter on the vertical axis, with parameter curves arranged in layers by importance). Then, extract the stacking order (reflecting parameter importance) and crossover frequency (reflecting parameter correlation) of the parameters in the plot. Finally, adjust the coding table based on these characteristics: increase the number of columns corresponding to parameters with high crossover frequency (resulting in a narrower value range), and reorder the rows by parameter importance, completing the coding table structure variation. The above coding table structure adjustment rules are customizable and will not be detailed here.

[0069] Next, modify the code table's encoding characters. Generate a trend curve for each parameter type, extracting the curve's curvature peak (the moment of parameter mutation) and curvature frequency (how often the parameter fluctuates). Modify the encoding symbols based on these characteristics: increase the symbol complexity at points of parameter mutation, and adjust the symbol format for parameters with frequent fluctuations. Ultimately, a code table with a modified structure and content is formed. The rules for modifying the code table's encoding characters are customizable and will not be detailed here.

[0070] Finally, the real-time detection data is matched to the variation coding table, and the corresponding coding symbols are extracted and spliced ​​into the initial coding string; then the graphic features extracted in the early stage (stacking order, cross frequency, curvature characteristics) are converted into a check code and embedded at the end of the coding string to form a complete coding string; after transmission, the receiving end verifies the data integrity through the check code, and then decodes and restores the detection data.

[0071] Reference Figure 2 In another embodiment of the present invention, a smoke detection device based on multi-sensor fusion analysis is provided, comprising: a detection unit configured to, when the smoke sensor detects that the smoke concentration within the detection range reaches a first threshold, obtain a temperature rise rate within the detection range detected by the temperature sensor, and obtain a uniformity of ambient thermal radiation within the detection range detected by the human body sensor; a correction unit, configured to generate corresponding correction coefficients based on the temperature rise rate and the uniformity of the ambient thermal radiation, and respectively correct the smoke concentration and the first threshold value to obtain corresponding corrected smoke concentration and corrected threshold value; a calculation unit, configured to calculate a comprehensive risk value based on the corrected smoke concentration and the temperature rising rate if the corrected smoke concentration reaches a correction threshold; The determination unit is configured to determine that the fire is a real fire if the comprehensive risk value reaches a risk threshold.

[0072] In this embodiment, for the specific implementation of each unit in the above device embodiment, please refer to the above method embodiment, which will not be repeated here.

[0073] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, memory, display screen, input device, network interface and database connected via a system bus. The processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the above method is implemented.

[0074] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0075] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which implements the above-described method when executed by a processor. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0076] In summary, the smoke detection method based on multi-sensor fusion analysis provided in the embodiment of the present invention includes: when the smoke sensor detects that the smoke concentration within the detection range reaches a first threshold, obtaining the temperature rise rate detected by the temperature sensor within the detection range, and obtaining the uniformity of the environmental thermal radiation within the detection range detected by the human body sensor; based on the temperature rise rate and the uniformity of the environmental thermal radiation, generating corresponding correction coefficients to correct the smoke concentration and the first threshold respectively, to obtain corresponding corrected smoke concentration and corrected threshold; if the corrected smoke concentration reaches the correction threshold, calculating a comprehensive risk value based on the corrected smoke concentration and the temperature rise rate; if the comprehensive risk value reaches the risk threshold, determining that it is a real fire. In the present invention, by combining the fusion processing of smoke sensors, temperature sensors, and human body sensors, the defect that the current fire detection through smoke is easily affected by environmental interference and produces false alarms is overcome.

[0077] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware using a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media provided herein and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM.

[0078] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, apparatus, article, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, apparatus, article, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, apparatus, article, or method comprising the element.

[0079] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A smoke detection method based on multi-sensor fusion analysis, characterized in that: The following steps are involved: When the smoke sensor detects that the smoke concentration within the detection range reaches a first threshold, the temperature sensor detects a temperature rise rate within the detection range, and the human body sensor detects a uniformity of ambient thermal radiation within the detection range; Based on the temperature rise rate and the uniformity of the ambient thermal radiation, a corresponding correction coefficient is generated to correct the smoke concentration and the first threshold value respectively, to obtain a corresponding corrected smoke concentration and a corrected threshold value; If the corrected smoke concentration reaches a correction threshold, a comprehensive risk value is calculated based on the corrected smoke concentration and the temperature rise rate; If the comprehensive risk value reaches the risk threshold, it is determined to be a real fire.

2. The smoke detection method based on multi-sensor fusion analysis according to claim 1 is characterized in that: After it is determined to be a real fire, it includes: The risk level is determined based on the predicted heat source diffusion speed of the temperature rise speed, the position of the person located by the human body sensor, and the smoke concentration, and a differentiated alarm is output based on the risk level.

3. The smoke detection method based on multi-sensor fusion analysis according to claim 1 is characterized in that: Based on the temperature rise rate and the uniformity of the ambient thermal radiation, a corresponding correction coefficient is generated to correct the smoke concentration and the first threshold value respectively, to obtain the corresponding corrected smoke concentration and corrected threshold value, including: If the temperature rising rate is less than a second threshold, generating a smoke concentration correction coefficient to correct the smoke concentration to obtain a corrected smoke concentration; If the uniformity of the ambient heat radiation exceeds a preset uniformity, a determination threshold correction coefficient is generated to correct the first threshold to obtain a corrected threshold.

4. The smoke detection method based on multi-sensor fusion analysis according to claim 1, characterized in that: The comprehensive risk value is calculated based on the corrected smoke concentration and temperature rise rate, including: Obtain the smoke concentration rise start time T1 and the temperature rise rate breaking through the speed threshold time T2; based on the time difference between T1 and T2, mark the corresponding causal relationship and assign the corresponding causal coefficient; Obtain the preset smoldering stage characteristic interval and open flame stage characteristic interval. Different stage characteristic intervals correspond to different corrected smoke concentrations and temperature rise rates. After the corrected smoke concentration and the temperature rise rate are matched to the corresponding stage characteristic interval, a corresponding compensation factor is generated based on the fluctuation of the temperature rise rate; wherein, in different stage characteristic intervals, the generation rules of the compensation factor are different; Determining weights corresponding to the modified smoke concentration and temperature rise rate based on the marked causal association and the matched stage characteristic interval; Based on the corresponding weights, the corrected smoke concentration and temperature rise rate are fused and calculated to obtain a basic risk score; the basic risk score is multiplied by the causal coefficient and the compensation factor in sequence to obtain the comprehensive risk value.

5. The smoke detection method based on multi-sensor fusion analysis according to claim 4 is characterized in that: Based on the marked causal relationship and the matched stage characteristic interval, weights corresponding to the modified smoke concentration and temperature rise rate are determined, including: Obtaining a preset weight mapping relationship; wherein the weight mapping relationship includes weight combinations corresponding to different causal relationships and stage feature intervals; Based on the causal relationship and the matched stage characteristic interval, the weights corresponding to the corrected smoke concentration and the temperature rising rate are determined through the weight mapping relationship.

6. The smoke detection method based on multi-sensor fusion analysis according to claim 4 is characterized in that: Before performing fusion calculation on the corrected smoke concentration and temperature rise rate based on the corresponding weights, the following steps are included: The corrected smoke concentration and temperature rising rate are normalized to the range of 0-1 respectively.

7. The smoke detection method based on multi-sensor fusion analysis according to claim 1, characterized in that: Obtaining the temperature rise rate within the detection range detected by the temperature sensor includes: The temperature sensor collects temperature data at multiple different points within the detection range at a preset sampling frequency; Calculate the average temperature of each point at the same time, and calculate the temperature rise rate based on the difference between the corresponding temperature averages at adjacent times. If the fluctuation of the temperature rise rate of multiple samples is less than the fluctuation threshold, the average of the temperature rise rates of multiple samples is used as the final temperature rise rate; If the temperature rise rate fluctuation of multiple samples reaches the fluctuation threshold, the preset sampling frequency of the temperature sensor is increased, and after adding multiple samples, the abnormal sampling values ​​are eliminated; the average value of the remaining temperature rise rates is used as the final temperature rise rate.

8. The smoke detection method based on multi-sensor fusion analysis according to claim 1, characterized in that: Obtaining the uniformity of ambient thermal radiation within the detection range detected by the human body sensor includes: The detection range is divided into a core area and a peripheral area with a preset radius; the core area is the core area in the detection range where the smoke concentration reaches the first threshold; The thermal radiation intensity data of multiple sampling points in the core area and the peripheral area are collected through human body sensors; Calculate the standard deviation of the thermal radiation intensity in the core area, the standard deviation of the thermal radiation intensity in the peripheral area, and the difference in thermal radiation intensity between the core area and the peripheral area; Based on the standard deviation of the thermal radiation intensity of the core area, the standard deviation of the thermal radiation intensity of the peripheral area, and the difference in thermal radiation intensity between the core area and the peripheral area, the uniformity of the ambient thermal radiation is calculated using a preset algorithm.

Images