Laboratory environment safety monitoring system and method

Abstract

The invention relates to a laboratory environment safety monitoring system and method. The system comprises: an acquisition module for acquiring environmental data of a laboratory; the extraction module is used for preprocessing the environment data and identifying at least one temperature abnormal area from the preprocessed data; the analysis module is used for analyzing a temperature time sequence change rule of each temperature abnormal region; the calculation module is used for determining at least one target temperature anomaly region meeting a preset association condition with a smoke source under the condition that the smoke source is identified based on each temperature time sequence change rule, and calculating a dynamic thermal trend of each target temperature anomaly region under the influence of the smoke source, and the fusion module is used for executing an early warning decision based on the dynamic thermal trend. Therefore, the problems of high false alarm rate and delayed instant abnormal response caused by simple data superposition in related technologies are solved, and the reliability and emergency response efficiency of a monitoring system are remarkably improved.

Description

Technical Field

[0001] This application relates to the field of environmental safety monitoring technology, and in particular to a laboratory environmental safety monitoring system and method. Background Technology

[0002] As the core location for scientific research and technological development, the safety of the laboratory environment directly affects valuable experimental equipment, critical research data, and the lives of researchers. Among the major risks faced by laboratories are heat-related safety accidents (such as fires and explosions) caused by equipment overheating, short circuits, or uncontrolled chemical reactions. Therefore, achieving 24 / 7, precise, and intelligent monitoring of the laboratory environment, especially temperature anomalies and early signs of fire, is a critical issue that urgently needs to be addressed in laboratory safety management.

[0003] In related technologies, laboratory environmental safety monitoring mainly relies on multi-sensor fusion technology. For example, one approach is to fuse vibration, heat source and fiber optic signals for perimeter security; another approach is to fuse visible light, infrared thermal imaging, gas and vibration data through spatiotemporal reference alignment, and introduce an attention mechanism for feature weighting.

[0004] However, most of the above methods only involve simple overlaying at the data layer or rudimentary logical fusion at the decision layer, resulting in high false alarm rates and delayed responses to transient anomalies. These issues urgently need to be addressed. Summary of the Invention

[0005] This application provides a laboratory environmental safety monitoring system and method to solve the problems of high false alarm rate and delayed response to instantaneous anomalies caused by simple data superposition in related technologies, and significantly improves the reliability and emergency response efficiency of the monitoring system.

[0006] To achieve the above objectives, a first aspect of this application provides a laboratory environmental safety monitoring system, comprising: The acquisition module is used to acquire environmental data of the laboratory. An extraction module is used to preprocess the environmental data and identify at least one temperature anomaly area from the preprocessed data. Analysis module, the analysis module is used to analyze the temperature time series variation pattern of at least one of the temperature anomaly regions; The fusion module is used to identify at least one target temperature anomaly region that meets a preset association condition with the smoke source when a smoke source is identified based on at least one of the temperature time-series change patterns, and to calculate the dynamic thermal trend of at least one of the target temperature anomaly regions under the influence of the smoke source, so as to perform early warning decision based on the dynamic thermal trend.

[0007] According to one embodiment of this application, the fusion module includes: The identification unit is used to evaluate the smoke particle signal in the environmental data based on at least one of the temperature time series variation patterns, and to determine the smoke source based on the evaluation results; The determining unit is used to obtain the spatiotemporal location of the smoke source and determine at least one target temperature anomaly region that satisfies the preset association condition with the smoke source based on the spatiotemporal location. The calculation unit is used to calculate the temperature change rate and temperature-smoke correlation degree of at least one of the target temperature anomaly areas, and to obtain the dynamic thermal trend based on the temperature change rate and the temperature-smoke correlation degree; The early warning unit is used to make an early warning judgment based on the dynamic heat trend, and to trigger a safety early warning decision when the judgment result meets the preset early warning conditions.

[0008] According to one embodiment of this application, the determining unit is specifically used for: Obtain the spatiotemporal location of the smoke source, and determine the target space search range centered on the spatiotemporal location; All temperature anomaly regions within the target space search range are taken as preliminary candidate regions, and temperature feature vectors of temperature time-series change patterns of each preliminary candidate region are extracted. The temperature feature vectors of each candidate region are taken as the temperature change patterns of the corresponding preliminary candidate regions. Acquire the concentration time series data of the smoke source, determine the time when the smoke source concentration rises based on the concentration time series data, and extract the concentration feature vector of the concentration time series data, and use the concentration feature vector as the smoke concentration change pattern; At least one final candidate region is determined from all the preliminary candidate regions, wherein the final candidate region is the region where the start time or abrupt change time of the temperature time series change pattern is no later than the time when the smoke source concentration rises; The temporal coupling degree between the temperature change pattern and the smoke source concentration change pattern of each of the final candidate regions is calculated, and at least one region whose temporal coupling degree meets the preset condition is determined as the target temperature anomaly region.

[0009] According to one embodiment of this application, the computing unit is specifically used for: Calculate the rate of temperature change of at least one of the target temperature anomaly regions within a preset time window; Determine the temperature time series data of the corresponding target temperature anomaly region based on the temperature time series variation pattern of at least one of the target temperature anomaly regions; Calculate the temperature-smoke correlation degree between the temperature time series data of at least one of the target temperature anomaly regions and the concentration time series data of the smoke source within the preset time window; The dynamic thermal trend is obtained based on the temperature change rate and the temperature-smoke correlation.

[0010] According to one embodiment of this application, the rate of temperature change is:

[0011] in, The temperature change rate is... The temperature value at the end of the preset time window. The temperature value at the start of the preset time window. The preset time window is defined as follows.

[0012] According to one embodiment of this application, the temperature-smoke correlation is:

[0013] in, This represents the correlation coefficient between temperature and the concentration of the smoke source. Let be the covariance between temperature and the concentration of the smoke source. For temperature, The concentration of the smoke source. The standard deviation of temperature, This represents the standard deviation of the concentration of the smoke source.

[0014] According to one embodiment of this application, the analysis module is specifically used for: Obtain a heat map of at least one of the temperature anomaly regions; Based on the area of ​​each heat map, each heat map is divided into a central heat map point and an edge heat map point according to a preset division ratio; The temperature changes of the central thermal power station and the edge thermal power stations are monitored separately, and the temperature anomaly areas where the temperature changes of the central thermal power station and the edge thermal power stations meet the preset spatial orientation change conditions are determined to have an expanding trend.

[0015] According to one embodiment of this application, the analysis module is further configured to: Acquire temperature data for each of the temperature anomaly regions at different time points; Calculate the temperature difference between adjacent time points for each of the temperature anomaly regions, and determine the temperature anomaly regions whose temperature changes meet the preset frequency conditions as core regions based on the temperature difference values. At least one adjacent temperature anomaly region adjacent to the core region is obtained, and the thermal characteristic changes of the boundary edge between each adjacent temperature anomaly region and the core region are monitored. The core region whose thermal characteristic changes at the boundary edge meet the preset enhancement conditions is determined to have an expansion trend.

[0016] According to one embodiment of this application, the fusion module further includes: A positioning unit is used to acquire location information of at least one of the target temperature anomaly areas, and to locate potential faults in the target based on the dynamic thermal trend and the location information.

[0017] The laboratory environmental safety monitoring system proposed in this application acquires laboratory environmental data through an acquisition module; preprocesses the environmental data through an extraction module and identifies at least one temperature anomaly area from the preprocessed data; analyzes the temperature time-series variation pattern of each temperature anomaly area through an analysis module; and, in the case of identifying a smoke source based on the temperature time-series variation pattern, determines at least one target temperature anomaly area that meets preset correlation conditions with the smoke source through a fusion module, and calculates the dynamic thermal trend of each target temperature anomaly area under the influence of the smoke source, so as to perform early warning decisions based on the dynamic thermal trend. This solves the problems of high false alarm rate and delayed response to instantaneous anomalies caused by simple data superposition in related technologies, significantly improving the reliability and emergency response efficiency of the monitoring system.

[0018] To achieve the above objectives, a second aspect of this application provides a method for monitoring laboratory environmental safety, comprising the following steps: Acquire environmental data of the laboratory, including temperature distribution images, gas concentration, temperature and humidity, and smoke particle concentration; The environmental data is preprocessed, and at least one temperature anomaly area is identified from the preprocessed data; Analyze the temporal variation patterns of temperature in each of the aforementioned temperature anomaly regions; When a smoke source is identified based on each of the temperature time-series change patterns, at least one target temperature anomaly region that meets the preset association conditions with the smoke source is determined, and the dynamic thermal trend of each target temperature anomaly region under the influence of the smoke source is calculated, so as to perform early warning decision based on the dynamic thermal trend.

[0019] The laboratory environmental safety monitoring method proposed in the embodiments of this application solves the problems of high false alarm rate and delayed response to instantaneous anomalies caused by simple data superposition in related technologies through the laboratory environmental safety monitoring system, and significantly improves the reliability and emergency response efficiency of the monitoring system.

[0020] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0021] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a block diagram of a laboratory environmental safety monitoring system according to an embodiment of this application; Figure 2 This is a flowchart of a laboratory environmental safety monitoring method provided according to an embodiment of this application. Detailed Implementation

[0022] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0023] The laboratory environmental safety monitoring system and method proposed according to the embodiments of this application will be described below with reference to the accompanying drawings. First, the laboratory environmental safety monitoring system proposed according to the embodiments of this application will be described with reference to the accompanying drawings.

[0024] Figure 1 This is a block diagram of a laboratory environmental safety monitoring system according to an embodiment of this application.

[0025] For example, such as Figure 1 As shown, the laboratory environmental safety monitoring system 10 includes: an acquisition module 100, an extraction module 200, an analysis module 300, and a fusion module 400. The acquisition module 100 acquires environmental data from the laboratory; the extraction module 200 preprocesses the environmental data and identifies at least one temperature anomaly region from the preprocessed data; the analysis module 300 analyzes the temporal temperature variation patterns of each temperature anomaly region; and the fusion module 400, when a smoke source is identified based on the temporal temperature variation patterns of each region, determines at least one target temperature anomaly region that meets preset correlation conditions with the smoke source, and calculates the dynamic thermal trend of each target temperature anomaly region under the influence of the smoke source, so as to execute early warning decisions based on the dynamic thermal trends.

[0026] It is understood that laboratory environmental data includes, but is not limited to, temperature distribution images, gas concentrations, temperature and humidity, and smoke particles. A temperature anomaly area refers to a localized spatial range identified through thermal imaging where the temperature is significantly higher or lower than the average level of the laboratory environment. In this embodiment, the temperature anomaly area primarily refers to a localized spatial range where the temperature is significantly higher than the average level of the laboratory environment. The temporal variation pattern of temperature refers to the dynamic pattern and statistical characteristics (such as heating rate, fluctuation period, diffusion trend, etc.) of the temperature anomaly area over a period of time, which is the basis for judging the evolution of risk. Dynamic thermal trend refers to the quantitative prediction and comprehensive assessment of the future temperature change direction and rate of the target temperature anomaly area after fusing smoke source information.

[0027] Specifically, the laboratory environmental safety monitoring system 10 proposed in this application mainly consists of an acquisition module 100, an extraction module 200, an analysis module 300, and a fusion module 400. The acquisition module 100, acting as the system's sensor, can include thermal imaging sensors, smoke sensors, gas sensors, and temperature and humidity sensors, comprehensively collecting multi-dimensional environmental data such as thermal imaging, smoke concentration, gas levels, and temperature and humidity. Specifically, the thermal imaging sensor can be installed on the ceiling or walls of the laboratory to fully cover the laboratory area; the gas sensor can be installed near locations where gas leaks may occur, such as ventilation ducts or next to experimental equipment; the temperature and humidity sensor can be installed in the center of the laboratory to accurately reflect the environmental temperature and humidity; the smoke sensor and flame sensor can both be installed on the ceiling or at a high location for timely detection of smoke and flames. Through simultaneous acquisition of multiple parameters, the system can obtain complete state information of the laboratory environment, avoiding the limitations of data from a single sensor; furthermore, the temperature distribution image generated based on thermal imaging technology can intuitively locate potential risk sources such as smoke particles, providing basic data support for subsequent analysis. The extraction module 200 preprocesses the raw data acquired by the acquisition module 100, performing tasks such as filtering and denoising to reduce the impact of noise on the analysis results and improve the accuracy of subsequent modules. Then, image segmentation technology is used to locate and partition the preprocessed data, resulting in individual temperature anomaly regions. This provides clear objects for subsequent analysis and reduces computational complexity. The analysis module 300 further performs continuous time-series analysis on each temperature anomaly region, transforming static "hot spots" into dynamic descriptions of "thermal behavior."

[0028] Ultimately, the fusion module 400 can first verify and trace the smoke signal based on the dynamic patterns (i.e., temperature time-series change patterns) provided by the analysis module 300, identifying the real smoke source; then, through spatiotemporal and pattern matching, it can lock the target area with the strongest correlation to the smoke source (i.e., preset correlation conditions) from among many temperature anomaly areas; next, it calculates the dynamic thermal trend of each target temperature anomaly area under the influence of the smoke source; finally, the system strictly triggers graded warnings based on this trend (e.g., only when the temperature shows an upward trend and is strongly positively correlated with smoke), thereby achieving a leap from multi-source information perception to accurate risk decision-making, significantly improving the reliability and intelligence level of laboratory safety monitoring.

[0029] It should be noted that, in the embodiments of this application, the analysis module 300 can deeply interpret the propagation pattern and range change law of heat in the temperature anomaly area by constructing analysis mechanisms such as diffusion type and region transformation type, which will be described in detail below.

[0030] Optionally, in some embodiments, the analysis module 300 is specifically used to: acquire a heat map of each temperature anomaly region; divide each heat map into a central heat map and an edge heat map according to a preset division ratio based on the area of ​​each heat map; monitor the temperature changes of the central heat map and the edge heat map respectively, and determine the temperature anomaly region whose temperature changes of the central heat map and the edge heat map satisfy a preset spatial orientation change condition as having an expanding range trend.

[0031] As can be understood, a heatmap is a visual representation of the temperature levels and spatial distribution within an area of ​​temperature anomalies, presented in pseudo-color image form, where color intensity directly corresponds to temperature levels. A thermal data point refers to a representative sub-region manually divided within the heatmap according to specific rules (such as spatial location or area proportion) for ease of analysis; it is the basic unit for quantitative monitoring by the system. Preset spatial orientation change conditions refer to quantitative or logical criteria pre-defined by the system to determine whether temperature changes in the central and peripheral regions exhibit correlated diffusion characteristics.

[0032] Specifically, for the diffusion-type analysis mechanism, the system does not treat the temperature anomaly area as a homogeneous whole. Instead, it first calls the temperature data of each temperature anomaly area identified and segmented by the extraction module 200 and generates its corresponding heat map. Then, the heat map of each temperature anomaly area is divided into two key observation areas, the central thermal point and the edge thermal point, according to the area according to a preset ratio (for example, the center occupies one-fifth and the edge occupies two-fifths). This division is intended to represent the "core heat generation area" of the heat source and its "heat influence frontier" respectively. Subsequently, the system can synchronously monitor the color changes of the central and peripheral thermal points on the heat map (identifying the synergy between their changes). Specifically, at least four monitoring points can be deployed within the central thermal point. Only when a temperature rise is detected at the central point (the overall color of the central thermal point shows a darkening trend) is the darkest monitoring point recorded as a reference point. If, at this time or later, the color of the peripheral thermal points darkens, exhibiting a pattern without a clear directional correlation to the central reference point—for example, the entire peripheral point darkens uniformly without a clear direction; multiple discontinuous darkening areas appear within the peripheral points; or the darkening area appears on the opposite side away from the central reference point; or the darkening area forms a ring around the center within the peripheral points—these patterns indicate that heat propagation is not controlled directionally by a single central heat source. Instead, it may be due to air convection, radiative heat transfer, or other factors causing heat to be widely absorbed or redistributed within the peripheral region. This suggests that the area of ​​abnormal heat influence is expanding and may be out of control. Based on this, the system can determine that the temperature anomaly area is expanding. If the darkening areas of edge thermal points show a trend towards the reference point—for example, if the darkening area forms a distinct band-shaped or wedge-shaped high-temperature zone within the edge thermal point and points towards the central reference point—this morphology indicates that heat is being conducted directionally and directly from the central core to its nearest edge. This represents a controlled, localized heat transfer, allowing the system to determine that widespread, uncontrolled heat diffusion has not occurred. Therefore, by quantifying the color changes of thermal points, the system can distinguish between localized overheating and regional risk diffusion, avoiding false alarms. Furthermore, it only needs to monitor the color changes of key points, eliminating the need for full-map analysis and improving processing efficiency.

[0033] Optionally, in some embodiments, the analysis module 300 is further configured to: acquire temperature data of each temperature anomaly region at different time points; calculate the temperature difference between adjacent time points for each temperature anomaly region, and determine the temperature anomaly region whose temperature change meets a preset frequency condition as the core region based on the temperature difference; acquire at least one adjacent temperature anomaly region adjacent to the core region, and monitor the thermal characteristic changes at the boundary edge between each adjacent temperature anomaly region and the core region; and determine the core region whose thermal characteristic changes at the boundary edge meet a preset enhancement condition as having an expansion trend.

[0034] Understandably, preset frequency conditions refer to dynamic thresholds or filtering rules set by the system for temperature differences, used to identify areas where temperature is changing rapidly, continuously, or significantly from among numerous temperature anomaly areas. Changes in the thermal characteristics of the boundary edge refer to the physical properties exhibited on the shared boundary between the core region and its neighboring regions (adjacent temperature anomaly areas), which may include the average temperature of the boundary region, the temperature gradient (spatial rate of change), or its corresponding color distribution and changes on a heatmap. Preset enhancement conditions refer to quantitative criteria used to determine whether the thermal state of the boundary edge is "upgrading," for example, a continuous increase in the average temperature of the boundary edge.

[0035] Specifically, for the regional transformation analysis mechanism, the system can first perform time-series insight by acquiring temperature data for each temperature anomaly region at different time points and calculating the temperature difference between adjacent time points for each temperature anomaly region. This allows the system to filter out the regions with the most active and drastic temperature changes (i.e., the highest temperature difference) and identify them as the core region (i.e., the primary dynamic heat source) within the current monitoring period. Subsequently, the system expands its analysis perspective from a single region to a spatial correlation network, actively exploring at least one adjacent temperature anomaly region spatially adjacent to the core region, and synchronously collecting and comparing the color changes of the shared boundaries (i.e., the junction edges) between adjacent regions. If the shared boundary color of any pair of adjacent temperature anomaly regions shows a trend of continuous deepening or expansion of the high-temperature color gamut, the system can determine that the influence range of the core anomaly region with higher temperature in this adjacent region group is expanding outwards. To further quantify the dynamic characteristics of this diffusion process, the system can immediately preset a short monitoring period (e.g., less than 10 seconds) between these adjacent regions and perform rapid feature calculations within this high-frequency observation window. The core computational features include the rate and gradient of temperature change from the center of the temperature anomaly region toward the edge, thereby accurately capturing the instantaneous intensity and pattern of heat transfer from the core to the edge.

[0036] The fast feature calculation performed within this high-frequency observation window can be represented by the following formula:

[0037] in, For time Time, location Temperature gradient at that location, For time Time, location Temperature value at that location, , For spatial step size, Radial distance;

[0038] in, The temperature diffusivity is... For thermal diffusivity, The change in temperature For spatial distance, For time step.

[0039] Conversely, if the colors of adjacent boundaries do not show a consistent darkening trend, it can be determined that the current thermal anomaly is still in a local stable stage and has not yet formed a significant range expansion. Therefore, it is not necessary to start the above short-term feature calculation.

[0040] It should be noted that this analysis method uses time series analysis, allowing the system to track the evolution trend of abnormal temperature areas in real time and provide early warnings of escalating risks. At the same time, by combining the color changes at the edges of adjacent areas, the system can identify independent events and chain reactions, and optimize early warning strategies.

[0041] Optionally, in some embodiments, the fusion module 400 includes: an identification unit, a determination unit, a calculation unit, and an early warning unit. The identification unit is used to evaluate the smoke particle signal in the environmental data based on the temporal variation pattern of each temperature, and determine the smoke source based on the evaluation results. The determination unit is used to obtain the spatiotemporal location of the smoke source, and determine at least one target temperature anomaly region that meets preset correlation conditions with the smoke source based on the spatiotemporal location. The calculation unit is used to calculate the temperature change rate and temperature-smoke correlation degree of each target temperature anomaly region, and obtain a dynamic thermal trend based on the temperature change rate and temperature-smoke correlation degree. The early warning unit is used to make an early warning judgment based on the dynamic thermal trend, and trigger a safety early warning decision when the judgment result meets preset early warning conditions.

[0042] Understandably, the temperature-smoke correlation refers to a statistical indicator (such as the Pearson correlation coefficient) used to quantify the degree of linear correlation between time-series temperature data and time-series smoke source concentration data in a target temperature anomaly area. A value close to +1 indicates a strong positive correlation (temperature and smoke are rising synchronously), close to 0 indicates no significant correlation, and close to -1 indicates a strong negative correlation. Preset warning conditions refer to the final decision threshold set by the system to trigger an alarm, based on dynamic thermal trends. For example, warning conditions are only met when the rate of temperature change is positive (temperature is rising) and the temperature-smoke correlation is greater than a high positive threshold (strong positive correlation).

[0043] Specifically, the fusion module 400 mainly consists of an identification unit, a determination unit, a calculation unit, and an early warning unit. The identification unit does not simply confirm the smoke sensor readings; instead, it utilizes the temperature time-series change patterns provided by the analysis module 300 to cross-validate and assess the risk of the original smoke particle signal (detecting the presence of smoke particles through pattern recognition technology). This identifies genuine and reliable smoke sources and pinpoints their spatiotemporal coordinates. By combining temperature and smoke data, the system can distinguish between real fires and false positives such as normal equipment cooling, improving early warning reliability. Furthermore, the dual verification mechanism avoids the limitations of relying solely on smoke sensors, reducing invalid alarms. The determination unit, after the identification unit identifies a smoke source, uses the spatiotemporal location of the smoke source as an anchor point. By analyzing the temporal coupling degree of spatial proximity and change patterns (e.g., whether the temperature rise starts earlier than or simultaneously with smoke generation), it filters out the target temperature anomaly region from all temperature anomaly regions that is most likely causally related to the smoke source (i.e., meets preset correlation conditions). The calculation unit can then perform relevant calculations on the identified target temperature anomaly area: on the one hand, it calculates the rate of temperature change to quantify the speed of thermal risk evolution; on the other hand, it calculates the temperature-smoke correlation to statistically confirm the causal strength between thermal changes and smoke generation. Integrating these two calculations generates a dynamic thermal trend characterizing the risk momentum. Finally, the early warning unit can strictly enforce preset early warning conditions based on the dynamic thermal trend. The system will only trigger an early warning when it confirms that the temperature in the target temperature anomaly area is showing a continuous upward trend, and that this trend is strongly correlated with the smoke particle signal. Conversely, if the system identifies smoke particles, but the associated target temperature anomaly area does not show a clear upward temperature trend, the system will remain silent and will not trigger an early warning.

[0044] Understandably, this design aims to significantly improve the accuracy and effectiveness of early warning systems through strict, causal-based triggering conditions, thereby filtering out a large number of non-fire-related smoke interference events caused by the complex laboratory environment. These interference events are diverse in reality, such as: smoldering of paper and fabrics in poorly ventilated environments (low-temperature, flameless incomplete combustion), releasing smoke but with minimal temperature rise; or the low-temperature oxidation and decomposition of certain chemicals (metal powders, organic peroxides). Localized electric sparks may ignite trace amounts of dust or insulating materials, producing smoke, but due to low energy and insufficient heat conduction, this smoke is not captured by macroscopic temperature monitoring. Smoke from corridors, ventilation ducts, or outdoors (such as during construction or incineration) enters the laboratory through airflow and has no causal relationship with the indoor thermal environment. Aerosols generated by physical or chemical changes such as solvent evaporation, acid mist generation, and catalyst decomposition are often non-exothermic or endothermic processes. Volatile organic compounds or bioaerosols produced by microbial culture also do not involve significant heat release.

[0045] Therefore, the no-temperature-rise, no-warning strategy adopted by this system is not a functional defect, but rather an intelligent design specifically for the unique laboratory environment. It deeply understands the fundamental physical differences between fires (significant heat release) and large-scale non-fire smoke events (no significant heat release), thus accurately identifying genuine thermal safety risks in complex laboratory environments, minimizing false alarms, and ensuring the high value and reliability of warning signals.

[0046] Optionally, in some embodiments, the determining unit is specifically used for: acquiring the spatiotemporal location of the smoke source, and determining the target space search range centered on the spatiotemporal location; taking all temperature anomaly regions within the target space search range as preliminary candidate regions, and extracting the temperature feature vector of the temperature time series change law of each preliminary candidate region, and taking the temperature feature vector of each candidate region as the temperature change pattern of the corresponding preliminary candidate region; acquiring the concentration time series data of the smoke source, determining the smoke source concentration rise time based on the concentration time series data, and extracting the concentration feature vector of the concentration time series data, and taking the concentration feature vector as the smoke concentration change pattern; determining at least one final candidate region from all preliminary candidate regions, wherein the final candidate region is a region where the start time or abrupt change time of the temperature time series change law is not later than the smoke source concentration rise time; calculating the temporal coupling degree between the temperature change pattern and the smoke source concentration change pattern of each final candidate region, and determining at least one region whose temporal coupling degree meets the preset condition as the target temperature anomaly region.

[0047] It is understandable that the temperature feature vector refers to a set of quantitative feature parameters extracted from the temporal variation pattern of a certain temperature anomaly region, used to mathematically characterize its variation pattern, and may include initial temperature, heating rate, fluctuation frequency, slope of the trend line, etc. The concentration feature vector refers to a set of quantitative feature parameters extracted from the time-series data of smoke source concentration, used to mathematically characterize its variation pattern, and may include concentration rise rate, peak time, fluctuation amplitude, shape of the growth curve, etc. Temporal coupling degree refers to a metric used to quantify the synchronicity and morphological similarity of two time series (temperature change pattern and smoke concentration change pattern in this embodiment) on the time axis. It can not only determine whether they change in the same direction, but also assess the degree of matching in the timing, rhythm, and shape of the changes, for example, it can be calculated through dynamic time warping distance or a specific similarity algorithm.

[0048] Specifically, the determination unit can define a target spatial search range based on the identified smoke source location, performing initial spatial screening and including all temperature anomaly regions within this range as preliminary candidates. Next, the system can extract key parameters from the temporal temperature change patterns of each preliminary candidate region, constructing a standardized temperature feature vector (i.e., temperature change pattern); simultaneously, it also extracts a concentration feature vector (i.e., smoke concentration change pattern) from the smoke source concentration data. Subsequently, the system can determine the precise moment when the smoke begins to rise significantly based on the concentration data, retaining only candidate regions whose own temperature anomaly start or abrupt change is no later than the smoke rise moment as final candidate regions. This step is crucial, ensuring that the temperature change has the temporal potential to cause smoke, conforming to the physical logic of fire where heat precedes smoke. Finally, the determination unit can calculate the temporal coupling degree between the temperature change pattern and the smoke concentration change pattern of each final candidate region. Only regions that are highly synchronized with the smoke concentration change pattern in terms of change rhythm and morphology, and whose coupling degree meets preset conditions, will be ultimately confirmed as target temperature anomaly regions strongly correlated with the smoke source.

[0049] Optionally, in some embodiments, the calculation unit is specifically used for: calculating the temperature change rate of each target temperature anomaly region within a preset time window; determining the temperature time series data of the corresponding target temperature anomaly region based on the temperature time series change law of each target temperature anomaly region; calculating the temperature-smoke correlation degree between the temperature time series data of each target temperature anomaly region and the concentration time series data of the smoke source within a preset time window; and obtaining the dynamic thermal trend based on the temperature change rate and the temperature-smoke correlation degree.

[0050] Understandably, a preset time window refers to a fixed or dynamic time period (such as the past 30 seconds) set for trend analysis. The system extracts data within this time period for calculation to ensure the timeliness and consistency of the analysis. Temperature time series data refers to the sequence of temperature observations recorded at fixed sampling intervals (such as per second) for a specific target temperature anomaly area within the preset time window, arranged in chronological order. It serves as the original data basis for calculating the rate of change and correlation.

[0051] Specifically, the calculation unit first calculates the rate of temperature change for each identified target temperature anomaly region within a preset time window, based on its temperature time-series data, using a difference or fitting algorithm. This indicator can intuitively quantify the severity and direction of the thermal state change within the target temperature anomaly region (e.g., a rate of increase of 0.5℃ / second). Next, within the same time window, the calculation unit synchronously aligns and statistically calculates the temperature time-series data of the target temperature anomaly region with the concentration time-series data of the smoke source, deriving the temperature-smoke correlation. This indicator reveals the strength and statistical reliability of the causal relationship between temperature change and smoke generation. Finally, the calculation unit integrates these two independent quantitative indicators, representing the rate of evolution and the strength of causality respectively, to generate a high-order dynamic thermal trend containing vector information. For example, a trend of "high temperature change rate and high positive correlation" can represent a high-risk pattern of "rapid temperature rise and strong correlation with smoke generation"; while a trend of "high temperature change rate and low correlation" can indicate another risk of "equipment overheating but no ignition of materials." Through this structured computation, the system can transform vague trends into decision-making criteria that can be described and compared by clear mathematical indicators, laying a solid foundation for subsequent accurate early warning.

[0052] Optionally, in some embodiments, the rate of temperature change is:

[0053] in, For the rate of temperature change, The temperature value at the end of the preset time window. The temperature value at the start of the preset time window. This is a preset time window.

[0054] The temperature-smoke correlation is:

[0055] in, Let be the correlation coefficient between temperature and the concentration of the smoke source, ranging from [-1, 1]. It is positively correlated. It is negatively correlated. Unrelated Let be the covariance between temperature and the concentration of the smoke source. For temperature, The concentration of the smoke source. The standard deviation of temperature, This represents the standard deviation of the concentration of the smoke source.

[0056] Furthermore, after completing the dynamic thermal trend analysis, the computing unit can also be used to perform self-learning benchmark construction and adaptive judgment functions. That is, the system can extract the temperature anomaly region with the strongest correlation to the smoke particle signal based on historical analysis data and mark it as the reference temperature anomaly region. Subsequently, it extracts the 3-5 peak concentration values ​​with the highest smoke particle concentration corresponding to the reference temperature anomaly region and simultaneously obtains the temperature rise of the reference temperature anomaly region relative to the benchmark at these peak times, which is recorded as the reference temperature difference. In this way, a set of temperature-smoke feature pairs characterizing typical risk events is established as an adaptive benchmark for subsequent judgment.

[0057] During subsequent real-time monitoring, when the system identifies a temperature change in a certain area on the new heat map that matches the reference temperature difference, it will initiate intelligent comparison: if the actual smoke concentration detected at this time is lower than the peak concentration value in the historical typical feature pair, the system can determine that the current temperature change is not related to smoke particles and belongs to a non-fire-related thermal event; conversely, if the actual smoke concentration reaches or exceeds the historical peak concentration value, the system can determine that the current temperature change is highly correlated with smoke particles and conforms to the typical fire development characteristics, thereby increasing the warning level or confidence level. Through this dynamic adaptive comparison mechanism, the system can effectively distinguish between events of different natures with similar temperature rise patterns, significantly reducing the false alarm rate caused by environmental interference or non-fire heat sources, while ensuring sensitive identification of real fire hazards.

[0058] Optionally, in some embodiments, the fusion module 400 further includes: a positioning unit, which is used to acquire the location information of at least one target temperature anomaly area and locate the target potential fault based on the dynamic thermal trend and the location information.

[0059] Specifically, the positioning unit can acquire the precise location information (i.e., spatial coordinates) of at least one target temperature anomaly area selected by the determining unit, serving as the physical basis for positioning. Subsequently, the positioning unit can integrate the dynamic thermal trend generated by the computing unit. In the spatial dimension, it can pinpoint the physical range of the thermal anomaly based on the location information of the target temperature anomaly area. In the temporal and causal dimensions, it can further identify the area most synchronous in time and statistically most relevant to the smoke source from multiple possible target areas based on the dynamic thermal trend, especially the temperature-smoke correlation. For example, the target temperature anomaly area with a sustained and rapid temperature increase (high rate of change) and a strong positive correlation with changes in smoke concentration (high correlation) can be prioritized as the most likely initial fault point or fire origin.

[0060] By combining spatial coordinate locking with time-varying causal strength verification, the system can move beyond simple hotspot marking and achieve a leap from anomaly detection to root cause localization. This significantly reduces the response time from alarm to cause identification, providing clear operational targets for subsequent precise emergency response (such as targeted ventilation, fire extinguishing, or equipment power outages), thereby effectively improving the proactive intervention capability and reliability of the entire safety monitoring system.

[0061] In summary, the laboratory environmental safety monitoring system proposed in this application has at least the following beneficial effects: (1) The system acquires temperature distribution images of the entire laboratory area through thermal imaging sensors and automatically identifies and segments abnormal temperature areas using an extraction module. Based on this, a diffusion-type analysis mechanism and a region transformation-type analysis mechanism are introduced to perform dynamic evolution analysis on abnormal areas. For example, by monitoring the color coordination changes of the central thermal data point and the edge thermal data points, the system can determine whether the thermal anomaly is in a local stable state or has a tendency to spread to the surrounding areas, thereby achieving real-time assessment of the risk range and effectively reducing false alarms and missed alarms.

[0062] (2) The system uses a fusion module to verify and trace the source of smoke particle signals by combining the temperature time-series change pattern, avoiding the limitations of relying solely on smoke sensors. Furthermore, it quantifies the temperature change rate and temperature-smoke correlation in the target temperature anomaly area, constructing a dynamic thermal trend describing the development momentum of the risk. This mechanism can effectively distinguish between equipment overheating scenarios with rising temperatures but no smoke, and non-fire interference (such as chemical aerosols) with high smoke concentration but no significant temperature change, significantly improving the accuracy and reliability of early warnings.

[0063] (3) The positioning unit in the fusion module quickly locates potential fault sources based on dynamic thermal trends and spatial location information. The early warning unit adopts a dynamic threshold strategy, triggering a high-level warning only when the temperature shows a continuous upward trend and is strongly positively correlated with the smoke signal. This strategy enables the system to intelligently distinguish between normal heat dissipation and overheating faults, as well as between harmless experimental smoke and dangerous combustion smoke, thereby improving the pertinence and efficiency of emergency response.

[0064] (4) After identifying an abnormal diffusion trend, the system can initiate a refined analysis within a very short monitoring period (e.g., less than 10 seconds) to quantify the spatial and temporal propagation characteristics of heat through physical quantities such as temperature gradient and temperature diffusion coefficient. For example, a sudden increase in temperature gradient may indicate a local short circuit, while an abnormal diffusion coefficient may reflect a ventilation failure, providing real-time basis for on-site intervention.

[0065] (5) The system has self-learning capabilities and can extract reference temperature differences and corresponding peak smoke concentrations based on historical high-risk events to construct a dynamic early warning benchmark. In subsequent monitoring, if a similar temperature rise pattern is detected but the smoke concentration is significantly lower than the historical benchmark, it is determined to be a low-risk event; if both the temperature rise and smoke concentration match or exceed historical characteristics, it is determined to be a high-risk event and an early warning is triggered. This adaptive mechanism can further improve the system's adaptability to complex environmental changes and reduce false alarms caused by atypical interferences.

[0066] The laboratory environmental safety monitoring system proposed in this application acquires laboratory environmental data through an acquisition module; preprocesses the environmental data through an extraction module and identifies at least one temperature anomaly area from the preprocessed data; analyzes the temperature time-series variation pattern of each temperature anomaly area through an analysis module; and, in the case of identifying a smoke source based on the temperature time-series variation pattern, determines at least one target temperature anomaly area that meets preset correlation conditions with the smoke source through a fusion module, and calculates the dynamic thermal trend of each target temperature anomaly area under the influence of the smoke source, so as to perform early warning decisions based on the dynamic thermal trend. This solves the problems of high false alarm rate and delayed response to instantaneous anomalies caused by simple data superposition in related technologies, significantly improving the reliability and emergency response efficiency of the monitoring system.

[0067] Next, referring to the accompanying drawings, a laboratory environmental safety monitoring method according to an embodiment of this application is described. This method employs... Figure 1 Laboratory environmental safety monitoring system as described in this example.

[0068] Figure 2 This is a flowchart of a laboratory environmental safety monitoring method according to an embodiment of this application.

[0069] like Figure 2 As shown, the laboratory environmental safety monitoring method includes the following steps: In step S201, environmental data of the laboratory is acquired, including temperature distribution images, gas concentration, temperature and humidity, and smoke particle concentration.

[0070] In step S202, the environmental data is preprocessed, and at least one temperature anomaly area is identified from the preprocessed data.

[0071] In step S203, the temporal variation pattern of temperature in each temperature anomaly region is analyzed.

[0072] In step S204, if a smoke source is identified based on the temperature time-series change pattern, at least one target temperature anomaly region that meets the preset association conditions with the smoke source is determined, and the dynamic thermal trend of each target temperature anomaly region under the influence of the smoke source is calculated, so as to perform early warning decision based on the dynamic thermal trend.

[0073] It should be noted that the foregoing explanation of the laboratory environmental safety monitoring system embodiment also applies to the laboratory environmental safety monitoring method of this embodiment, and will not be repeated here.

[0074] The laboratory environmental safety monitoring method proposed in the embodiments of this application solves the problems of high false alarm rate and delayed response to instantaneous anomalies caused by simple data superposition in related technologies through the laboratory environmental safety monitoring system, and significantly improves the reliability and emergency response efficiency of the monitoring system.

[0075] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0076] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0077] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A laboratory environmental safety monitoring system, characterized in that, include: The acquisition module is used to acquire environmental data of the laboratory. An extraction module is used to preprocess the environmental data and identify at least one temperature anomaly area from the preprocessed data. Analysis module, which is used to analyze the temporal variation pattern of temperature in each of the temperature anomaly regions; The fusion module is used to identify at least one target temperature anomaly region that meets preset association conditions with the smoke source when a smoke source is identified based on each of the temperature time-series change patterns, and to calculate the dynamic thermal trend of each target temperature anomaly region under the influence of the smoke source, so as to perform early warning decisions based on the dynamic thermal trend.

2. The system according to claim 1, characterized in that, The fusion module includes: The identification unit is used to evaluate the smoke particle signal in the environmental data based on the temperature time series variation pattern, and determine the smoke source based on the evaluation result; The determining unit is used to obtain the spatiotemporal location of the smoke source and determine at least one target temperature anomaly region that satisfies the preset association condition with the smoke source based on the spatiotemporal location. The calculation unit is used to calculate the temperature change rate and temperature-smoke correlation degree for each of the target temperature anomaly regions, and to obtain the dynamic thermal trend based on the temperature change rate and the temperature-smoke correlation degree. The early warning unit is used to make an early warning judgment based on the dynamic heat trend, and to trigger a safety early warning decision when the judgment result meets the preset early warning conditions.

3. The system according to claim 2, characterized in that, The determining unit is specifically used for: Obtain the spatiotemporal location of the smoke source, and determine the target space search range centered on the spatiotemporal location; All temperature anomaly regions within the target space search range are taken as preliminary candidate regions, and temperature feature vectors of temperature time-series change patterns of each preliminary candidate region are extracted. The temperature feature vectors of each candidate region are taken as the temperature change patterns of the corresponding preliminary candidate regions. Acquire the concentration time series data of the smoke source, determine the time when the smoke source concentration rises based on the concentration time series data, and extract the concentration feature vector of the concentration time series data, and use the concentration feature vector as the smoke concentration change pattern; At least one final candidate region is determined from all the preliminary candidate regions, wherein the final candidate region is the region where the start time or abrupt change time of the temperature time series change pattern is no later than the time when the smoke source concentration rises; The temporal coupling degree between the temperature change pattern and the smoke source concentration change pattern of each of the final candidate regions is calculated, and at least one region whose temporal coupling degree meets the preset conditions is determined as the target temperature anomaly region.

4. The system according to claim 3, characterized in that, The computing unit is specifically used for: Calculate the rate of temperature change for each of the target temperature anomaly regions within a preset time window; The temperature time series data of the corresponding target temperature anomaly region are determined based on the temperature time series variation pattern of each of the target temperature anomaly regions. Calculate the temperature-smoke correlation degree between the time-series data of the temperature of each target temperature anomaly region and the time-series data of the concentration of the smoke source within the preset time window; The dynamic thermal trend is obtained based on the temperature change rate and the temperature-smoke correlation.

5. The system according to claim 4, characterized in that, The rate of temperature change is: in, The temperature change rate is... The temperature value at the end of the preset time window. The temperature value at the start of the preset time window. The preset time window is defined as follows.

6. The system according to claim 5, characterized in that, The temperature-smoke correlation is: in, This represents the correlation coefficient between temperature and the concentration of the smoke source. Let be the covariance between temperature and the concentration of the smoke source. For temperature, The concentration of the smoke source. The standard deviation of temperature, This represents the standard deviation of the concentration of the smoke source.

7. The system according to claim 1, characterized in that, The analysis module is specifically used for: Obtain a heat map for each of the temperature anomaly regions; Based on the area of ​​each heat map, each heat map is divided into a central heat map point and an edge heat map point according to a preset division ratio; The temperature changes of the central thermal power station and the edge thermal power stations are monitored separately, and the temperature anomaly areas where the temperature changes of the central thermal power station and the temperature changes of the edge thermal power stations meet the preset spatial orientation change conditions are determined to have an expanding trend.

8. The system according to claim 7, characterized in that, The analysis module is also used for: Acquire temperature data for each of the temperature anomaly regions at different time points; Calculate the temperature difference between adjacent time points for each of the temperature anomaly regions, and determine the temperature anomaly regions whose temperature changes meet the preset frequency conditions as core regions based on the temperature difference values. At least one adjacent temperature anomaly region adjacent to the core region is obtained, and the thermal characteristic changes of the boundary edge between each adjacent temperature anomaly region and the core region are monitored. The core region whose thermal characteristic changes at the boundary edge meet the preset enhancement conditions is determined to have an expansion trend.

9. The system according to claim 2, characterized in that, The fusion module further includes: A positioning unit is used to acquire location information of at least one of the target temperature anomaly areas, and to locate potential faults in the target based on the dynamic thermal trend and the location information.

10. A method for monitoring laboratory environmental safety, characterized in that, The method employs the laboratory environmental safety monitoring system as described in any one of claims 1-9, and the method includes the following steps: Acquire environmental data of the laboratory, including temperature distribution images, gas concentration, temperature and humidity, and smoke particle concentration; The environmental data is preprocessed, and at least one temperature anomaly area is identified from the preprocessed data; Analyze the temporal variation patterns of temperature in each of the aforementioned temperature anomaly regions; When a smoke source is identified based on each of the temperature time-series change patterns, at least one target temperature anomaly region that meets the preset association conditions with the smoke source is determined, and the dynamic thermal trend of each target temperature anomaly region under the influence of the smoke source is calculated, so as to perform early warning decision based on the dynamic thermal trend.

Images