Biosafety laboratory airflow risk monitoring methods, equipment, media and products

Abstract

The present invention discloses a method, equipment, medium and product for monitoring the airflow risk of a biosafety laboratory, and relates to the field of data processing. The method comprises: obtaining airflow sensor data and facility status data, calculating a containment integrity index based on these two types of data, dynamically time-regularizing and matching the real-time change curve of the index with a preset baseline database, outputting a matching degree, and when the matching degree is lower than a preset threshold, predicting that there is a risk of rupture of the airflow containment barrier, and locating the source of the risk based on the data source that causes the matching degree abnormality, using the source as the release point of a virtual aerosol pollutant, and performing fluid dynamics simulation in combination with real-time environmental data to generate a three-dimensional spatiotemporal diffusion trajectory of pollutants in the laboratory, and extracting risk quantification indicators, and generating accurate laboratory airflow risk warning information based on the quantitative indicators. The present invention can provide early warning before the risk occurs, alleviating the problem of safety lag in airflow risk monitoring in biosafety laboratories.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, equipment, medium and product for monitoring airflow risk in a biosafety laboratory. Background Art

[0002] The core safety of biosafety laboratories (BSLs), particularly BSL-3 and BSL-4 laboratories handling highly pathogenic pathogens, relies on a rigorous physical containment system. The cornerstone of this physical containment system is the maintenance of a stable, reliable, unidirectional airflow and a precise, multi-stage negative pressure gradient. This ensures that airflow consistently flows from the clean areas of the biosafety laboratory to potentially contaminated areas, then to the core contaminated areas, ultimately passing through high-efficiency particulate matter (HEPA) filters before exiting. This creates an invisible yet crucial airflow containment barrier, preventing internal microbial aerosols from escaping to the external environment. Therefore, real-time, uninterrupted monitoring of the integrity and effectiveness of this containment barrier is paramount to ensuring laboratory safety and public health.

[0003] Related technologies typically employ sensors (such as differential pressure gauges and anemometers) deployed at key locations within the physical containment system. These sensors continuously collect isolated environmental parameters and compare them against pre-set safety thresholds to determine if airflow conditions are normal. If a parameter deviates from the threshold, the system triggers an alarm, alerting management to a potential containment risk.

[0004] However, using this approach, the relevant technology can only issue an alarm after the airflow containment barrier has been substantially damaged, and cannot provide early warning before the risk occurs, so there is a safety lag. Summary of the Invention

[0005] In response to the above-mentioned technical problems and defects, the purpose of the present invention is to provide a method, equipment, medium and product for monitoring the airflow risk of a biosafety laboratory, which can provide early warning before the risk occurs and alleviate the problem of safety lag in the airflow risk monitoring of a biosafety laboratory.

[0006] To achieve the above-mentioned purpose, in a first aspect, the present invention provides a method for monitoring the airflow risk of a biosafety laboratory, comprising: acquiring airflow sensor data and facility status data, the airflow sensor data being used to characterize the pressure gradient and directional airflow in key channels between the clean area, potential contaminated area and core contaminated area in the biosafety laboratory, and the facility status data being used to characterize the operating state of the physical containment system; calculating a containment integrity index based on the airflow sensor data and the facility status data, the containment integrity index being used to dynamically quantify the effectiveness and integrity of the airflow containment barrier of the biosafety laboratory; dynamically time-regularizing the real-time change curve of the containment integrity index with a preset baseline database to obtain a matching degree; and when the matching degree is lower than a preset When the threshold is reached, it is predicted that there is a potential risk of substantial damage to the airflow containment barrier, and the source of the risk is located based on the data source that causes the matching degree to be lower than the preset threshold; the risk source is used as the initial release point of the virtual aerosol pollutants, and real-time environmental data is used as the boundary condition to perform fluid dynamics simulation processing to obtain the three-dimensional spatiotemporal diffusion trajectory of the virtual aerosol pollutants in the biosafety laboratory; risk quantification indicators for guiding emergency response are extracted from the three-dimensional spatiotemporal diffusion trajectory, which include the shortest breakthrough time for pollutants escaping from the risk source to reach the next clean area or exit, and the expected exposure dose calculated for personnel in different locations; laboratory airflow risk warning information is generated based on the risk quantification indicators.

[0007] The present invention employs the aforementioned method and steps, creating a complete technical closed loop from dynamic risk assessment to forward-looking simulation and early warning, to overcome the safety lag inherent in related technologies, which can only provide passive, delayed "post-event warnings." Rather than relying on simple threshold judgments, the present invention calculates a containment integrity index that integrates multi-source data in real time, predictively identifying potential risks and locating their sources before substantial damage to the airflow containment barrier occurs. Subsequently, a scenario-based simulation centered on the risk source is immediately initiated, ultimately outputting quantitative risk indicators (such as minimum breakthrough time and estimated exposure dose) that can directly guide emergency response. This transforms the traditional passive monitoring model into an active, predictive risk management approach, effectively addressing the safety lag inherent in related technologies.

[0008] Optionally, in some embodiments, a containment integrity index is calculated based on airflow sensor data and facility status data, including: deconstructing the physical containment system of the biosafety laboratory into a macro-environmental barrier layer and a key node barrier layer; based on the airflow sensor data, calculating a first health index of the macro-environmental barrier layer using a preset fluid mechanics algorithm; and based on the facility status data, calculating a second health index of the key node barrier layer using a preset operating safety procedure; using the first health index and the second health index as inputs to a preset Bayesian network model for probabilistic reasoning to obtain a containment integrity index, where the containment integrity index is used to characterize the risk of failure of the physical containment system, and the Bayesian network model is used to describe the causal relationship and risk propagation path between the macro-environmental barrier layer and the key node barrier layer.

[0009] Optionally, in some embodiments, the physical containment system of the biosafety laboratory is deconstructed into a macro-environmental barrier layer and a key node barrier layer, including: extracting a set of partitioned physical isolation components between the clean area, the potential contaminated area and the core contaminated area; in the set of partitioned physical isolation components, the enclosure structure and sealing components used to ensure the stability of the pressure difference between the partitions are classified as macro-environmental barrier elements, and the dynamic opening and closing devices or equipment through-structures used to connect the partitions are classified as key node barrier elements; a topological mapping table of macro-environmental barrier elements and key node barrier elements is established; the airflow blocking dependency between the macro-environmental barrier elements and the key node barrier elements is marked in the topological mapping table; based on the marked topological mapping table, a macro-environmental barrier layer is constructed through the macro-environmental barrier elements, and a key node barrier layer is constructed through the key node barrier elements.

[0010] Optionally, in some embodiments, a containment integrity index is calculated based on the airflow sensor data and the facility status data, including: determining a flow field disturbance factor based on the real-time dynamic change of the airflow sensor data, and determining a barrier operation disturbance factor based on the real-time dynamic change of the facility status data; generating a containment disturbance factor sequence in time sequence based on the flow field disturbance factor and the barrier operation disturbance factor; performing nonlinear weighting on the containment disturbance factor sequence, and performing an integral operation with a time attenuation effect to obtain a containment integrity index, which is used to reflect the cumulative impact of historical states and the current impact intensity.

[0011] Optionally, in some embodiments, fluid dynamics simulation processing is performed, including: obtaining the transmission characteristic parameters of the target pathogen currently involved in the biosafety laboratory, the transmission characteristic parameters including at least one of sedimentation velocity, half-life, inactivation rate, and minimum infection dose; based on the transmission characteristic parameters, adjusting the pollutant diffusion model parameters used in the fluid dynamics simulation processing; performing fluid dynamics simulation processing using the adjusted pollutant diffusion model parameters to obtain an optimized three-dimensional space-time diffusion trajectory.

[0012] Optionally, in some embodiments, based on the propagation characteristic parameters, the pollutant diffusion model parameters used in the fluid dynamics simulation processing are adjusted, including: calculating the equivalent gravity sedimentation rate parameters of the corresponding suspended particulate matter in the fluid dynamics simulation model according to the sedimentation velocity of the target pathogen; calculating the concentration attenuation coefficient parameters of the corresponding suspended particulate matter in the fluid dynamics simulation model according to the half-life and / or inactivation rate of the target pathogen; and adjusting the configuration of the pollutant diffusion model parameters based on the equivalent gravity sedimentation rate parameters and the concentration attenuation coefficient parameters.

[0013] Optionally, in some embodiments, the real-time change curve of the containment integrity index is dynamically time-warped matched with a preset baseline database to obtain a matching degree, including: extracting multiple representative historical change curve samples of the containment integrity index from the preset baseline database; performing dynamic time warping calculations on the historical change curve samples to obtain the regularized path cumulative distance between the real-time change curve of the containment integrity index and each historical change curve sample; based on the regularized path cumulative distance, a preset distance matching degree conversion function is used to calculate the matching degree between the real-time change curve and the baseline database.

[0014] In a second aspect, an embodiment of the present invention provides an electronic device comprising: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the electronic device to execute the method described in the first aspect, and any possible implementation method of the first aspect.

[0015] In a third aspect, the present invention provides a computer-readable storage medium comprising instructions, which, when executed on the electronic device, enable the electronic device to execute the method described in the first aspect and any possible implementation of the first aspect.

[0016] In a fourth aspect, the present invention provides a computer program product comprising instructions, which, when the computer program product is run on the electronic device, enables the electronic device to execute the method described in the first aspect and any possible implementation of the first aspect.

[0017] It is understood that the electronic device provided in the second aspect, the storage medium provided in the third aspect, and the computer program product provided in the fourth aspect are all used to execute the method provided by the present invention. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects of the corresponding methods and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1This is a flow chart of a method for monitoring airflow risk in a biosafety laboratory according to an embodiment of the present invention;

[0019] Figure 2 It is a schematic diagram of the architecture of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In a biosafety laboratory (BSL-3), multiple differential pressure gauges and anemometers are typically installed between the core area and the outside. Conventionally, these sensors continuously monitor, for example, the pressure differential between the core lab and the antechamber, the pressure differential between the antechamber and the corridor, and the air velocity in the exhaust duct. The system sets a preset safety threshold, such as a pressure differential between the core lab and the antechamber, which must be maintained between -30 Pa and -50 Pa. If a researcher accidentally opens the door to the core lab, or if the exhaust system suddenly malfunctions, causing the pressure differential to drop to -10 Pa, the system will immediately trigger an alarm, indicating a pressure anomaly. At this point, management will receive the alert and begin investigating whether the door is open or the equipment is malfunctioning, but by then the containment barrier has already been compromised, and potential contaminants may have already escaped. This approach only issues alerts after a problem has occurred, making it impossible to predict it in advance.

[0021] The present invention provides a biosafety laboratory airflow risk monitoring technology. In addition to monitoring the aforementioned pressure differential and air velocity data, the system also collects facility status data, such as access records for core labs (frequent door openings and duration of door openings), HEPA (High Efficiency Particulate Air) filter operating hours, and exhaust fan power. This data is integrated to calculate a containment integrity index.

[0022] Suppose that during a certain period, while the pressure differential between the core laboratory and the antechamber remains within a safe threshold (e.g., -35 Pa), the system detects that the door has been frequently opened within a short period of time, with each opening duration slightly longer than usual. Simultaneously, the exhaust fan power exhibits slight but persistent fluctuations. Although none of these individual parameters triggers a regular alarm, the system dynamically time-warps these subtle variations against the CII (Containment Integrity Index) curves from a historical database of events such as "poor door seal leading to unstable airflow" or "early exhaust system failure." If the match falls below a preset threshold, the system immediately issues a warning: "Potential risk exists in the airflow containment barrier, possibly originating from the core laboratory door or exhaust system." The system then performs a fluid dynamics simulation using the door or exhaust vent as a virtual contaminant release point, predicting the contaminant's potential diffusion path and time to reach the clean area. It also calculates the expected exposure dose for nearby personnel, providing laboratory managers with detailed early warning information and emergency response guidance before an actual hazard occurs.

[0023] This invention overcomes the safety lag inherent in related technologies, which typically provide passive, delayed "after-the-fact warnings" through a complete technical closed loop from dynamic risk assessment to proactive simulation and early warning. Rather than relying on simple threshold judgments, this invention calculates a containment integrity index that integrates multi-source data in real time to predictively identify potential risks and locate their source before substantial damage to the containment barrier occurs. Subsequently, a scenario-based simulation centered on the risk source is immediately initiated, ultimately outputting quantitative risk indicators (such as minimum breakthrough time and estimated exposure dose) that can directly guide emergency response. This transforms the traditional passive monitoring model into an active, predictive risk management approach, effectively addressing the safety lag inherent in related technologies.

[0024] Based on the above technical concept, the embodiment of the present invention provides a method for monitoring airflow risk in a biosafety laboratory, such as Figure 1 As shown, the specific steps include:

[0025] Step 201: Acquire airflow sensor data and facility status data.

[0026] Among them, airflow sensor data is used to characterize the pressure gradient and directional airflow in key channels between clean areas, potential contaminated areas and core contaminated areas in the biosafety laboratory, and facility status data is used to characterize the operating status of the physical containment system.

[0027] First, to monitor airflow sensor data, highly sensitive sensors need to be deployed in key areas of the biosafety laboratory. This includes installing differential pressure sensors (differential pressure gauges) between clean areas, potentially contaminated areas, and core contaminated areas, as well as between adjacent areas, to continuously monitor and obtain accurate pressure gradient data, ensuring that airflow always flows from high-pressure (clean) areas to low-pressure (contaminated) areas. At the same time, wind speed sensors (anemometers) are installed in key channels such as exhaust ducts, air outlets, and transfer windows to monitor the speed and direction of directional airflow in real time and confirm whether the airflow meets unidirectional flow requirements. These sensors are typically connected to a central data acquisition system or the laboratory's building management system (BMS) via a wired or wireless network, enabling automated and continuous data collection and time-stamping to ensure data synchronization and traceability.

[0028] Secondly, for facility status data, a more diverse range of sensors and information sources need to be integrated. For example, switch status sensors should be installed on all physical barriers, such as airtight doors, transfer windows, and inspection hatches, to record their open / close status and duration. The pressure differential across high-efficiency air filters (HEPA) should be monitored to assess their clogging level and service life. Furthermore, the operating status of supply and exhaust fans and air conditioning units (such as start / stop, speed, and power consumption) as well as the opening of valves and dampers should be captured. This data is typically obtained through interface integration with laboratory automation control systems, access control systems, and fire protection systems.

[0029] All collected raw data will be transmitted to the Airflow Risk Monitoring System (hereinafter referred to as the System) to provide comprehensive and accurate input for the subsequent Containment Integrity Index calculation.

[0030] Step 202 : Calculate a containment integrity index based on airflow sensor data and facility status data.

[0031] Among them, the Containment Integrity Index (CII) is used to dynamically quantify the effectiveness and integrity of the airflow containment barrier in the biosafety laboratory.

[0032] Specifically, the system first needs to preprocess these massive data, including data cleaning (removing outliers and noise), standardization (unifying data of different dimensions into a comparable range, such as 0-100 scores) and feature engineering (extracting more meaningful features from the raw data, such as the fluctuation amplitude of the pressure difference, the frequency and cumulative duration of door opening, etc.).

[0033] Next, based on the biosafety laboratory's operating specifications and expert experience, different weights are assigned to each data indicator to reflect its importance to the integrity of airflow containment. For example, the stability of the differential pressure in the core area may be given a higher weight, while the fluctuation of the operating status of a non-critical equipment may be given a lower weight.

[0034] Then, a multivariate mathematical model or algorithm is constructed to integrate these weighted indicators and calculate the final containment integrity index. This model can be a rule-based expert system (e.g., if the pressure differential deviates from threshold X and the door is open for more than Y time, then the CII score is deducted by Z points), or it can be a more complex statistical model (such as principal component analysis, multivariate statistical process control), or a machine learning model (such as support vector machines, neural networks). By learning data patterns from historical data showing normal operation and various abnormal conditions, it can assess the current containment status in real time. The CII calculation result is typically a single numerical value, such as between 0 and 100, with higher values ​​indicating greater integrity and effectiveness of the airflow containment barrier.

[0035] This calculation process is continuous and real-time, ensuring that the CII can dynamically reflect the latest status of the laboratory's airflow containment barrier and form a continuous containment integrity index change curve, providing a basis for subsequent risk prediction and matching.

[0036] In some embodiments, this step may specifically include the following steps:

[0037] S2021, determining a flow field disturbance factor based on the real-time dynamic change of the airflow sensor data, and determining a barrier operation disturbance factor based on the real-time dynamic change of the facility status data.

[0038] The core of this step is to convert the raw data monitored by the sensors into a quantitative "disturbance" indicator. The flow field disturbance factor refers to factors or events that cause the airflow organization (flow field) within the biosafety laboratory to deviate from its normal stable state, such as doors and windows opening, equipment failure, or human activity.

[0039] For the flow disturbance factor, the system continuously monitors airflow sensor data within the biosafety laboratory and between it and the external environment. When these data fluctuate abnormally, such as when the pressure differential exceeds the preset safety range (whether due to excessive positive pressure or insufficient negative pressure), when the supply and exhaust air volumes deviate significantly from the set values, or when the airflow direction reverses, the system calculates a numerical flow disturbance factor based on the magnitude and duration of these real-time dynamic changes. The flow disturbance factor can be a normalized value; for example, the greater the pressure differential deviates from the set value, the higher the flow disturbance factor. For the barrier operation disturbance factor, the system monitors facility status data such as access control status (open, abnormal opening), biosafety cabinet (BSC) operating status (alarms active, fan functioning properly), interlock system operation, and filter blockage or damage.

[0040] When abnormal conditions such as prolonged door openings, BSC alarms, critical fan shutdowns, or interlock failures are detected, a barrier operation disturbance factor is determined based on these real-time dynamic changes. This factor can also be numerically quantified. For example, a BSC alarm might generate a higher disturbance factor than a brief door opening to reflect its potential impact on containment integrity. This step is crucial for data preprocessing and feature extraction, laying the foundation for subsequent comprehensive evaluation.

[0041] S2022: Generate a containment disturbance factor sequence in time sequence based on the flow field disturbance factor and the barrier operation disturbance factor.

[0042] This step is to integrate and record the above-mentioned flow field disturbance factors and barrier operation disturbance factors in the order of their occurrence to form a continuous time series data. At each set time interval (for example, every second, every minute), the system will obtain the flow field disturbance factor and barrier operation disturbance factor at the current moment. Then, these independent disturbance factors will be combined into a comprehensive "containment disturbance factor" and added to the sequence together with the corresponding timestamp. This combination can be a simple summation, weighted average, or judgment based on preset logical rules. For example, if severe flow field disturbances and barrier operation abnormalities occur at the same time, the current containment disturbance factor may be set to a higher value to reflect the combined effect. This sequence not only records the size of the disturbance, but more importantly, it retains the timing information of the disturbance.

[0043] In this way, the system can clearly track the dynamic evolution of the biosafety laboratory's containment status over time, accurately capturing and recording both short-term transient disturbances and long-term persistent issues. This time series serves as a direct input for subsequent historical cumulative impact analysis and time decay integral calculations.

[0044] S2023: Nonlinearly weight the containment disturbance factor sequence and perform an integration operation with a time decay effect to obtain a containment integrity index.

[0045] Among them, the containment integrity index can be used to reflect the cumulative impact of historical status and current impact intensity.

[0046] First, each containment perturbation factor in the series is "nonlinearly weighted." This means that not all perturbations have a linear effect on containment integrity. For example, a mild perturbation might barely affect the index, while a severe perturbation might cause the index to rise sharply. This can be achieved by applying a nonlinear function (such as an exponential function or an S-shaped curve) so that larger perturbations receive higher weights.

[0047] Next, an "integration with time decay effect" is performed. This means that the older the disturbance, the less impact it has on the current containment integrity index, while the most recent disturbance has the greatest impact. This is usually achieved by introducing an exponential decay factor into the integration (or discrete summation) process, i.e., e −λΔt , where λ is the decay constant and Δt is the time interval between the disturbance and the present. In this way, the system can accumulate the effects of all disturbances in history, but at the same time ensure that the impact of historical events will gradually weaken over time.

[0048] Ultimately, the result of the integration operation is the "Containment Integrity Index." This index is a dynamically changing value that not only reflects the current disturbance intensity (reflected by the most recent, least attenuated disturbance factor) but, more importantly, it integrates the cumulative effects of all disturbances over a period of time, providing a comprehensive, real-time, and historically accurate assessment of the containment status.

[0049] Step 203 : Dynamically time-warp matching is performed on the real-time change curve of the containment integrity index with a preset baseline database to obtain a matching degree.

[0050] This step utilizes the Dynamic Time Warping (DTW) algorithm to identify potential risk patterns within the containment barrier. Simultaneously, a baseline database stores CII curves corresponding to a large number of known safe and normal operating conditions in biosafety laboratories, serving as "templates" or "fingerprints." These templates represent typical, healthy CII behavior patterns for laboratories under varying loads and operating modes. The key benefit of the DTW algorithm lies in its ability to effectively handle variations in the speed and duration of time series data. Even if the current CII variation is faster or slower than the baseline template, DTW can find the optimal time alignment path by "stretching" or "compressing" the timeline. This allows DTW to calculate the distance between the real-time CII curve and each healthy template in the baseline database. A smaller distance indicates a higher degree of similarity and a correspondingly higher degree of match, meaning the current CII variation pattern is more similar to a known normal operating pattern.

[0051] The system calculates a match score for each baseline template, which serves as the basis for subsequent risk assessments. This real-time, dynamic matching allows the system to continuously assess the degree to which the current containment barrier's operating status conforms to normal conditions. If the real-time CII curve begins to significantly degrade from its match with all baseline templates, even if the absolute value of the CII has not yet reached the traditional alarm threshold, the system will be able to detect this early sign of deviation from normal, providing early warning of potential anomalies or risks.

[0052] In this example, constructing a baseline database is a critical and rigorous process, designed to comprehensively and accurately capture the airflow containment characteristics of a biosafety laboratory under various safety operating scenarios. First, long-term, continuous data collection is required while one or more biosafety laboratories are operating normally, stably, and in compliance with regulations. This involves continuously recording all airflow sensor data (such as pressure differentials between areas and wind speeds in critical passages) and facility status data (such as door and window opening and closing status, HEPA filter pressure differentials, fan operating parameters, valve openings, etc.) in different seasons, loads (such as personnel entry and exit, equipment operation), and operating modes (such as normal operation and cleaning and maintenance).

[0053] Data collection should cover a variety of typical, healthy operating conditions that a laboratory may experience to ensure that the database is as comprehensive as possible. During the data collection process, any known abnormal or faulty data must be strictly excluded. The collected raw data will undergo preprocessing, including data cleaning (removing noise and occasional outliers), standardization, and feature extraction to form a series of representative "normal state CII change curve templates." These templates can be generated by clustering a large amount of normal data (for example, grouping similar normal operating modes into one category) or by experts manually selecting representative typical normal operating curves. Each template should contain a detailed fingerprint of its corresponding airflow sensor and facility status data.

[0054] Database construction isn't a one-time effort; it requires regular updates and maintenance. For example, after a major laboratory renovation, equipment upgrade, or operational strategy adjustment, data should be recollected and the database updated to ensure it always reflects the laboratory's most up-to-date, true-to-normal operating status. The baseline database is the foundation of dynamic time-warping matching, and its quality directly impacts the accuracy and sensitivity of risk prediction.

[0055] In step 204 , when the matching degree is lower than a preset threshold, it is predicted that there is a potential risk of substantial damage to the airflow containment barrier, and the source of the risk is located based on the data source causing the matching degree to be lower than the preset threshold.

[0056] Specifically, the system presets a "matching threshold" based on experience and tolerance for deviations from normal conditions. Unlike before, if the real-time CII curve matches the corresponding template in the baseline database below this threshold, it means that the current CII variation pattern has significantly deviated from the known normal operating pattern.

[0057] At this point, the system determines that the containment barrier is at risk of substantial damage because its behavior no longer conforms to a safe, stable, and normal state. Once this potential deviation from normal is identified, the system further analyzes the "data source" that causes the current CII change pattern to deviate from normal.

[0058] For example, the system will trace back the original airflow sensor data and facility status data that the current CII calculation relies on. By comparing the differences between the current data and the corresponding data in the normal status database, it can identify which changes in sensor readings (such as the pressure difference in a certain area, the wind speed in a specific channel) or equipment status (such as the open and close status of a door, the operating parameters of a fan) contribute most to the deviation of CII.

[0059] Through this comparative analysis, the system can accurately locate the specific risk sources that cause CII to deviate from its normal pattern, such as "a continuous decrease in the pressure differential between the core area and the potential contamination area" or "abnormal fluctuations in the speed of a certain exhaust fan," thereby providing managers with clear early warning information and intervention directions before actual danger occurs.

[0060] In step 205 , the risk source is used as the initial release point of the virtual aerosol pollutant, and the real-time environmental data is used as the boundary condition to perform fluid dynamics simulation processing to obtain the three-dimensional spatiotemporal diffusion trajectory of the virtual aerosol pollutant in the biosafety laboratory.

[0061] The system uses computational fluid dynamics (CFD) technology to accurately simulate the three-dimensional spatiotemporal diffusion behavior of virtual aerosol pollutants in the specific controlled environment of a biosafety laboratory.

[0062] First, it's important to clarify that "the risk source serves as the initial release point for virtual aerosol pollutants." This means that in the simulation model, the actual potential contamination source (for example, a broken culture dish, an operating bioreactor, a leaking biological sample container, or even the moment of a sneeze or cough) must be precisely located and defined as the starting point of aerosol release. This isn't just a geometric point; its release characteristics must also be defined, including but not limited to the initial aerosol particle size distribution (e.g., PM2.5, PM10), initial release velocity, release duration, and initial concentration or mass flow rate. The accuracy of these parameters directly impacts the realism of subsequent diffusion simulations.

[0063] Secondly, "using real-time environmental data as boundary conditions" is the key to ensuring that simulation results are highly consistent with the actual operating conditions in the laboratory. Real-time environmental data generally includes but is not limited to:

[0064] HVAC (Heating, Ventilation, and Air Conditioning) system parameters: laboratory air supply volume, return air volume, fresh air ratio, supply air temperature, humidity, and air speed and direction at each supply and exhaust vent. These data directly determine the overall airflow distribution within the laboratory;

[0065] Indoor environmental parameters: real-time temperature, humidity, pressure differential within the laboratory (especially the pressure differential with adjacent areas; biosafety laboratories typically maintain negative pressure), and possible local heat sources (such as heating from equipment) or cold sources;

[0066] Human activity: Although difficult to accurately model in real time, in some advanced simulations, the impact of human movement on local airflow can be considered, or humans can be considered as potential secondary pollution sources;

[0067] Door and window status: Whether the laboratory door is open and how long it is open will instantly change the airflow pattern in the laboratory.

[0068] This real-time data is fed into the CFD model as boundary conditions for the computational domain. For example, air inlets are defined as velocity inlet boundaries, exhaust vents as pressure or velocity outlet boundaries, and walls as no-slip wall boundaries. The laboratory's geometry (including internal lab benches, equipment, ventilation ducts, and so on) also needs to be accurately modeled and meshed, discretizing the continuous physical space into millions of computational cells.

[0069] During the simulation phase, CFD software (such as ANSYS Fluent, OpenFOAM, COMSOL Multiphysics, etc.) solves a series of complex partial differential equations based on the set initial and boundary conditions, including the mass conservation equation, momentum conservation equation (Navier-Stokes equation), energy conservation equation, and the transport equation of pollutant components.

[0070] Aerosol diffusion often involves Lagrangian tracking of the particle phase or Eulerian-Eulerian two-phase flow models to accurately describe the particle inertia, gravitational settling, and coupling with the airflow. Because aerosol diffusion is a dynamic process, simulations typically employ unsteady (transient) calculations, which iterate the solution within each time step to capture the temporal changes in pollutant concentration.

[0071] Ultimately, the simulation results are presented in visualizations such as concentration isosurfaces, particle trajectories, and streamlines, clearly demonstrating how aerosol pollutants, starting from the release point, diffuse, dilute, and settle in all directions under the influence of airflow within the laboratory over time, ultimately being expelled or retained in a specific area. This trajectory is three-dimensional, as it accounts for distribution in height, width, and depth; it is also spatiotemporal, as it illustrates the dynamic evolution of pollutant concentration over time.

[0072] By analyzing these results, we can identify potential high-risk areas, assess personnel exposure risks, optimize ventilation system design, and provide a scientific basis for emergency response.

[0073] In one embodiment, to further improve the real-time performance of CFD calculations, this embodiment can construct and deploy an artificial intelligence-based agent model (AI Surrogate Model), also known as a digital twin agent. The core of this technical concept lies in the "offline training, online prediction" mode.

[0074] In the early stages of system deployment, a high-performance computing cluster is first used to perform thousands or even tens of thousands of high-precision, non-steady-state CFD simulations for a large number of potential risk scenarios within the biosafety laboratory (for example, leak sources at hundreds or thousands of different grid points, different ventilation system operating conditions, different door and window opening and closing combinations, etc.), and to build a massive "input-output" dataset.

[0075] This dataset is then used to train a deep learning model, such as a graph neural network (GNN) or Fourier neural operator (FNO) suitable for physical field prediction, so that it can accurately learn and reproduce the complex nonlinear mapping relationship from the initial release conditions to the final three-dimensional spatiotemporal diffusion trajectory of pollutants.

[0076] In actual operation, once the system predicts a potential risk and locates its source, it bypasses time-consuming traditional CFD calculations. Instead, it uses the real-time location of the risk source and environmental parameters as input, instantly feeding them into a pre-trained AI agent model. Leveraging its powerful reasoning capabilities, the model rapidly outputs key quantitative risk indicators, including predicted pollutant diffusion trajectories, minimum breakthrough times, and personnel exposure doses.

[0077] This embodiment adopts this method to compress the physical simulation process, which originally takes a long time, into the scope of real-time response, thereby ensuring the timeliness of early warning information, making early response and precise intervention based on quantitative deduction possible, and ensuring the real-time requirements of simulation.

[0078] This embodiment also provides a solution to ensure real-time CFD simulation processing. The technical concept is to build a "pre-computed scenario-consequence database" and combine it with a fast interpolation matching algorithm. The core of this solution is to completely move computationally intensive work offline.

[0079] Before the system is deployed, the three-dimensional space of the laboratory is gridded to preset hundreds or even thousands of potential virtual pollutant release points. Combined with several key facility operating states (for example, normal ventilation mode, emergency exhaust mode, a door open, etc.), a huge scenario combination matrix is ​​formed.

[0080] Subsequently, using high-performance computing resources, a complete, high-precision CFD simulation is performed for each scenario combination in this matrix, and core risk quantification indicators (such as the shortest breakthrough time, exposure dose distribution map of each area, etc.) are extracted from the simulation results. These "scenario-consequence" key-value pairs are stored in an optimized database.

[0081] When the online monitoring system locates the source of a risk in real time, it doesn't initiate a new CFD calculation. Instead, it uses the current risk source's location and facility status as query criteria, performing a high-speed search within a pre-calculated database. Using advanced fast interpolation algorithms, the system instantly interpolates the results of several pre-calculated results closest to the current scenario, calculating the precise consequences for the specific scenario.

[0082] This embodiment moves most of the computing burden to the offline stage, so that the online response stage only needs to perform lightweight database queries and interpolation operations, thereby transforming the original time-consuming simulation process into a sub-second retrieval and calculation task, ensuring the absolute timeliness of emergency warnings.

[0083] Step 206: extract risk quantification indicators for guiding emergency response from the three-dimensional spatiotemporal diffusion trajectory.

[0084] Among them, risk quantification indicators include the shortest breakthrough time for pollutants escaping from the risk source to reach the next clean area or exit, as well as the expected exposure dose calculated for personnel in different locations.

[0085] The purpose of this step is to transform complex fluid dynamics simulation results into actionable, quantifiable risk indicators to guide emergency response.

[0086] Regarding the extraction of the “shortest breakthrough time for pollutants escaping from the risk source to reach the next clean area or exit,” this requires a time series analysis of the pollutant concentration field in the simulation results.

[0087] Clean areas generally refer to areas within or adjacent to a biosafety laboratory that require high cleanliness or personnel safety from contamination, such as antechambers, buffer rooms, or emergency exits. Exits are passages connecting the laboratory to the outside environment.

[0088] To determine the minimum breakthrough time, the analyst sets an alert threshold concentration (e.g., a trace contaminant concentration below the ambient background concentration but above the detectable limit, or a fraction of the safe exposure limit established based on toxicology data).

[0089] Then, in a simulation animation or time-stepping data, track the pollutant cloud escaping from the release point and monitor when its front first reaches and exceeds this threshold concentration at the boundary of the clean zone or exit. The time from the start of the pollutant release to the first reaching of this boundary is recorded, which is the shortest breakthrough time. This time metric is crucial for emergency response. It provides valuable "warning time" to inform operators and emergency managers how much time they have to take protective measures, initiate emergency procedures, or evacuate personnel before the pollutants spread to critical areas. For example, if the breakthrough time is short, it means that immediate action is needed; if it is longer, there is more buffer time to develop a more thorough response strategy.

[0090] Regarding the "expected exposure dose calculated for personnel at different locations," this requires combining the simulated data on the time-varying pollutant concentration with the personnel's activity trajectory or assumed location within the laboratory. The exposure dose is typically defined as the integral product of the pollutant concentration and the exposure time. The specific calculation method is as follows:

[0091] Determine personnel locations: This can be fixed key workstation locations (such as in front of a biosafety cabinet, next to an incubator, in front of a microscope), or personnel movement paths simulated based on preset scenarios.

[0092] Extracting local concentration: For each selected personnel location, extract the instantaneous concentration value of the pollutant at each time step in the entire simulation time at the location from the simulation results of step 205 .

[0093] Calculate the exposure dose: If a person is assumed to stay at a specific location, then during the simulation time, the instantaneous concentration at that location is multiplied by the corresponding time step, and then the products are accumulated over all time steps. Mathematically, this can be expressed as , where D is the exposure dose, C(t) is the pollutant concentration at the location at time t, and t0 and t f are the start and end time of exposure, respectively. If human movement is considered, it is necessary to match the movement path of the person with the pollutant concentration field in real time, calculate the instantaneous exposure at each point on the path, and accumulate it.

[0094] Exposure dose calculations can help assess potential health risks in different areas or for different operators. High-dose areas or individuals require higher levels of personal protective equipment (PPE), stricter operating procedures, or faster evacuation. These quantitative indicators provide concrete, measurable data for emergency response, enabling the shift from "potential risk" to "how significant the risk is, when, and where it will occur."

[0095] Step 207: Generate laboratory airflow risk warning information based on the risk quantification index.

[0096] Generating early warning information requires a system of warning levels or thresholds. This system is typically established based on biosafety risk assessment standards, toxicology data, national or industry regulations, and the laboratory's own risk tolerance. For example, multiple warning levels can be established, such as "low risk," "medium risk," "high risk," and "emergency evacuation."

[0097] Low risk: The contaminant has a long breakthrough time and the expected exposure dose is well below the safety limit. Routine monitoring or a warning may be required.

[0098] Medium Risk: The contaminant has a moderate breakthrough time and the exposure dose is expected to be close to or slightly above the long-term exposure safety limit. It may be necessary to activate local ventilation, strengthen personal protection, or restrict access to non-essential personnel.

[0099] High risk: The pollutant's breakthrough time is short, and the estimated exposure dose may reach or exceed the short-term exposure safety limit. Immediately initiate emergency response plans, such as activating emergency exhaust ventilation, wearing high-level PPE, and preparing for local isolation.

[0100] Emergency evacuation: The contaminant's breakthrough time is extremely short, and the exposure dose is expected to quickly reach life-threatening levels. A full evacuation must be initiated immediately, and external emergency response forces must be notified.

[0101] Based on these warning levels, the system determines the current risk level based on the "minimum breakthrough time" and "estimated exposure dose" calculated in step 206. For example, if the minimum breakthrough time is less than a set threshold (e.g., 5 minutes) and the estimated exposure dose at a key location exceeds a preset "high risk" threshold, the system will trigger a high-risk warning.

[0102] Warning information can be generated in a variety of forms to ensure that the information can be conveyed to relevant personnel in a timely and accurate manner:

[0103] Visual interface: Risk areas, diffusion paths of contaminants, and current warning levels are displayed graphically (e.g., color-coded areas on the laboratory floor plan, flashing icons) on the laboratory's central monitoring screen or personal terminal.

[0104] Sound / visual alarm: When the risk reaches a certain level, the sound and light alarm will be triggered to alert all personnel in the laboratory.

[0105] SMS / email notification: Automatically send detailed warning information to preset emergency response personnel, laboratory managers, and safety supervisors, including the source of risk, current risk level, and recommended initial measures.

[0106] Prescriptive recommendations: Warning information should not only inform about risks but also include specific action recommendations. For example, when a high risk is detected, the system can prompt instructions such as "Immediately check the seal of the biosafety cabinet," "Activate the emergency exhaust system," or "Wear an N95 mask and evacuate to a safe area."

[0107] Furthermore, the warning information can include an explanation of the airflow risk, such as which ventilation vent is malfunctioning or which area has insufficient negative pressure, helping managers quickly locate the problem and take corrective measures. In this way, step 207 transforms complex CFD simulation and risk quantification results into intelligence information, enabling the laboratory to perceive, assess, and respond to biosafety risks related to internal airflow in real time, thereby maximizing the protection of personnel safety and the environment from contamination.

[0108] The method of this embodiment adopts the above steps, combines the fluid dynamics simulation of virtual aerosol pollutant diffusion with real-time environmental data, and further extracts quantitative risk indicators, thereby achieving accurate prediction and intelligent early warning of airflow contamination risks in biosafety laboratories.

[0109] First, this embodiment can accurately simulate and visualize the diffusion trajectory of pollutants in three-dimensional space and time, revealing potential contamination paths and high-risk areas, which far exceeds the limitations of traditional empirical judgment. Second, by extracting quantitative risk indicators such as "shortest breakthrough time" and "estimated exposure dose", abstract risks are converted into specific and measurable values, providing a clear basis for risk assessment. For example, the shortest breakthrough time provides valuable warning time for emergency response, while the expected exposure dose is directly related to personnel health risks, making the formulation of protective measures more targeted. Most importantly, based on these quantitative indicators, the system can generate timely and instructive laboratory airflow risk warning information. This means that before a potential contamination incident occurs or spreads, laboratory managers and operators can receive accurate warnings, allowing them to quickly initiate emergency plans, such as adjusting the ventilation system, wearing personal protective equipment, restricting personnel activities, or conducting emergency evacuation.

[0110] This embodiment significantly improves the overall safety level of the biosafety laboratory through this early warning intervention, proactive, and data-driven airflow risk management capability, effectively reduces the risk of personnel exposure, avoids cross-contamination, and provides a scientific basis for accident response, thereby maximizing the protection of the life and health of laboratory personnel and environmental safety.

[0111] The embodiment of the present invention also provides another biosafety laboratory airflow risk monitoring method, which specifically includes the following steps:

[0112] Step 301: Acquire airflow sensor data and facility status data.

[0113] This step can refer to the description of the above embodiment and will not be repeated here.

[0114] Step 302: Deconstruct the physical containment system of the biosafety laboratory into a macro-environmental barrier layer and a key node barrier layer.

[0115] The purpose of this step is to provide a structured and hierarchical abstraction of the laboratory's physical safety system. The physical containment of a biosafety laboratory is multi-layered and aims to prevent biohazards from escaping from the laboratory to the external environment.

[0116] The macro-environmental barrier refers to the laboratory's overall, large-scale physical structure and environmental control system, which constitutes the first and broadest line of defense for containment. This typically includes the laboratory's enclosure (the airtightness of the walls, ceiling, and floor), the overall ventilation system (HVAC system, responsible for maintaining negative pressure, controlling airflow direction and air changes), and the air filtration system (such as the installation and maintenance of HEPA filters at exhaust vents).

[0117] The macro-environmental barrier layer focuses on the airflow organization, pressure difference maintenance and air purification capacity of the entire laboratory space to ensure that pollutants do not escape through large-scale airflow diffusion or structural gaps.

[0118] The critical node barrier layer focuses on key equipment and operating areas within the laboratory that come into direct contact with biohazards or where leakage may occur. This includes biosafety cabinets (BSCs, local negative pressure fume hoods used for handling microorganisms), autoclaves (for inactivating waste), centrifuges (which may generate aerosols), incubators, and the seals of various containers and pipelines. Furthermore, access routes for personnel entering and exiting the laboratory (such as airtight doors and interlocking doors) and waste and sample transfer systems are also considered critical nodes.

[0119] The barriers in the key node barrier layer emphasize local containment capabilities and safety guarantees during operations.

[0120] Through this deconstruction, the complex physical containment system can be split into two independent but interrelated components that are easier to monitor and evaluate, laying the foundation for subsequent quantitative evaluation and enabling risk analysis to comprehensively cover the laboratory's physical containment system from macro to micro.

[0121] In some embodiments, step 302 may specifically include S3021-S3025:

[0122] S3021, extract the set of physical isolation components between the clean area, potential contaminated area and core contaminated area.

[0123] The core of this step is a comprehensive and detailed survey and identification of the biosafety laboratory's physical layout. This requires dividing the laboratory's interior space into distinct zones based on their biohazard levels and functions: clean areas (such as offices, corridors, and locker rooms, which carry the lowest biohazard risk); potentially contaminated areas (such as sample processing areas, incubation rooms, and general laboratory operation areas, which carry a certain biohazard risk); and core contaminated areas (such as high-level biosafety cabinet operation rooms, animal infection rooms, and virus culture rooms, which carry the highest biohazard risk).

[0124] Once the zones are delineated, the next step is to systematically identify and list all physical structures that separate these different risk-level areas. This includes, but is not limited to: building materials that make up walls, floors, and ceilings, and how they are joined; all types of doors (e.g., airtight, interlocking, fire-rated, etc.); windows (if present and as part of the separation); and any dedicated passageways or devices used for the entry and exit of materials or personnel, such as transfer windows, autoclaves (as pass-throughs), or dedicated material airlocks.

[0125] This process requires drawing a detailed laboratory floor plan and accurately marking the location, type and area separated by each isolation component on the map to form a complete physical isolation component list or database, laying the foundation for subsequent classification and analysis.

[0126] S3022. In the set of partitioned physical isolation components, the enclosure structures and sealing components used to ensure the stability of the pressure difference between partitions will be classified as macro-environmental barrier elements, and the dynamic opening and closing devices or equipment through-structures used to connect the partitions will be classified as key node barrier elements.

[0127] This step is to functionally classify all physically isolated components identified in the previous step.

[0128] Macro-environmental barrier elements primarily refer to those static or semi-static components that comprise the laboratory's overall enclosure and play a crucial role in maintaining stable pressure differentials between areas. This includes the structural integrity and airtightness of the laboratory's walls, floors, and ceilings, as well as all sealing materials and construction techniques (such as sealants, fireproofing putty, and casing seals) used to fill gaps, pipe penetrations, and cable holes. Together, these elements form the laboratory's large-scale, holistic airtight barrier, fundamental to maintaining negative pressure and preventing the widespread spread of contaminants. They focus on surface airtightness.

[0129] In contrast, critical node barrier elements specifically refer to localized, variable components that connect different areas, require dynamic opening and closing functions, or serve as equipment access pathways. This primarily includes all types of doors (especially those that require frequent opening and closing, require interlocking or airtight features), transfer windows, autoclaves (when used as material access pathways), and any other equipment connection points that penetrate the enclosure and could affect airtightness (such as exhaust connections in biosafety cabinets and pipe valves). These elements represent "point" controls, and due to their dynamic nature or operational requirements, they present potential weaknesses and require more sophisticated monitoring and management.

[0130] This classification allows complex physical containment systems to be simplified into two broad categories of functional elements, facilitating subsequent independent assessment and management.

[0131] S3023, establish a topological mapping table of macro-environmental barrier elements and key node barrier elements.

[0132] This step aims to build a clear, structured data model that shows the spatial relationships and connections between the various elements in the laboratory's physical containment system. This topological map is more than just a list; it is a map that reflects the geographical and logical relationships between elements, showing who is connected to whom and who contains whom.

[0133] For example, it can be a two-dimensional plan with the locations of all walls, doors, windows, and equipment marked on it, and the connections between them represented by lines or symbols. A more advanced implementation can be a three-dimensional model or database structure, in which each feature has a unique ID and records its type, location coordinates, the area it belongs to, and the IDs of other features directly adjacent to or connected to it. For example, a door (critical node barrier feature) will be mapped to the wall (macro-environmental barrier feature) in which it is embedded, and it will be clearly indicated which clean areas, potential contaminated areas, or core contaminated areas the door connects to. Similarly, an autoclave (critical node) will be mapped to the wall it penetrates, and it will be indicated which contaminated areas and non-contaminated areas it connects to.

[0134] The establishment of this topology mapping table is the basis for subsequent analysis of airflow blockage dependencies. It provides a visual and queryable framework, making the structure of the entire physical containment system clear at a glance and easy to understand and manage.

[0135] S3024, mark the airflow blocking dependency between the macro-environment barrier elements and the key node barrier elements in the topology mapping table.

[0136] This step builds upon the established topology map to further understand the functional interactions between the various elements. This process requires analyzing each element’s contribution to airflow blockage and their interdependencies.

[0137] For example, the failure of a macro-environmental barrier element (such as the sealing of a wall) may directly affect the stability of the pressure differential in the area it encloses, thereby affecting the normal operation and airflow control effectiveness of all key node barrier elements (such as biosafety cabinets and transfer windows) within the area. Conversely, the failure of a key node barrier element (such as the aging of the sealing strips on an airtight door) may cause air leakage in a local area, further affecting the regional pressure differential maintained by the entire macro-environmental barrier layer. This dependency can be direct (such as a door embedded in the wall) or indirect (such as overall negative pressure affecting local equipment).

[0138] These dependencies can be marked by adding connecting lines, arrows, and text descriptions on the topology diagram, or by adding "depends on" or "affects" fields to each element in the database and linking them to other related elements. For example, you can mark "The sealing of door A depends on the flatness of wall B" and "The stability of the inflow airflow of the BSC depends on the maintenance of the negative pressure of the entire laboratory."

[0139] This step reveals the “weak links” and “chain reaction” paths in the containment system, which is crucial for subsequent risk assessment and fault diagnosis.

[0140] S3025, based on the annotated topology mapping table, construct a macro-environment barrier layer through macro-environment barrier elements, and construct a key node barrier layer through key node barrier elements.

[0141] This step involves conceptually layering and integrating the laboratory's physical containment system. Once all physical isolation components have been identified, categorized, mapped, and their airflow blocking dependencies annotated, the two logical layers of macro-environmental barriers and critical node barriers can be clearly defined and constructed.

[0142] The macro-environmental barrier layer is constructed as the collective function of all elements categorized as "macro-environmental barrier elements." It represents the laboratory's overall, large-scale airtightness and pressure differential maintenance capabilities. This level of health assessment focuses on the integrity of the entire enclosure, such as the presence of microcracks invisible to the naked eye and the complete sealing of duct penetrations.

[0143] The critical node barrier layer is constructed as the collective function achieved by all elements categorized as "critical node barrier elements." It represents the local containment capability of each key entrance, exit, and equipment connection point within the laboratory. The health assessment at this level will focus on door sealing, transfer window interlocking function, autoclave operation status, and biosafety cabinet airflow performance.

[0144] Through this construction, the complex physical containment system is abstracted into two interrelated but independently evaluable levels, providing a clear input structure for subsequent health index calculation and Bayesian network reasoning, making the assessment of the failure risk of the entire containment system more systematic and operational.

[0145] Step 303: Based on the airflow sensor data, a preset fluid mechanics algorithm is used to calculate the first health index of the macro-environmental barrier layer; and based on the facility status data, a preset operational safety procedure is used to calculate the second health index of the key node barrier layer.

[0146] Specifically, the calculation of the first health index of the macro-environmental barrier layer relies on real-time airflow sensor data to assess the effectiveness of the laboratory's overall airflow control. Airflow sensors are typically deployed at the laboratory's supply, return, and exhaust vents, as well as pressure differential monitoring points between different areas. Based on this real-time data, a fluid dynamics algorithm calculates key parameters, such as:

[0147] Pressure differential maintenance: Monitor the pressure differential between the laboratory and the external environment or adjacent areas to ensure it remains negative and within a preset safety range. Insufficient negative pressure or excessive fluctuations indicate a decrease in containment capacity.

[0148] Air exchange rate: Calculate the actual air exchange rate based on the air supply volume and laboratory volume to ensure that it meets the biosafety level requirements.

[0149] Airflow direction: Analyze airflow sensor data to confirm that airflow always flows from the clean area to the contaminated area to avoid backflow.

[0150] Filter pressure drop: This monitors the pressure difference across the HEPA filter to determine if the filter is clogged or failing. The real-time values ​​of these parameters are compared with preset safety thresholds. Using methods such as weighted averaging or fuzzy logic, an index is calculated to reflect the health of the macro-environmental barrier layer. For example, a score from 0 to 100 is generated, with higher scores indicating better health.

[0151] The calculation of the second health index of the key node barrier layer mainly depends on the facility status data and operational safety procedures. Facility status data includes:

[0152] Equipment operating parameters: For example, the wind speed (downflow and inflow) of the biosafety cabinet, filter status, and alarm information; the temperature, pressure, and whether the sterilization cycle of the autoclave is completed; whether the centrifuge is balanced and the lid is locked, etc.

[0153] Maintenance records: records of regular calibration, inspection, filter replacement, etc. of equipment.

[0154] Operational compliance: Through sensors (such as door magnetic sensors, infrared sensors) or manual input, key operations are monitored to ensure they comply with SOPs (standard operating procedures). For example, whether the doors and windows of the biosafety cabinet are opened too high, whether the operator is wearing PPE correctly, and whether waste is disposed of in a timely manner. The preset operational safety procedures define the acceptable ranges for these parameters and the standardization of operational behaviors. When equipment operating parameters deviate from the safe range, or when operational behaviors that do not comply with safety regulations are detected, the system will deduct points or lower the health of the key node based on the preset weights and logical rules. Ultimately, the health of all key nodes is combined to calculate a second health index that represents the overall health status of the barrier layer of the key nodes.

[0155] In this embodiment, the calculation of the first health index and the second health index is dynamic and can reflect the current status of the laboratory physical containment system in real time.

[0156] In step 304 , the first health index and the second health index are used as inputs to a preset Bayesian network model for probabilistic reasoning to obtain a containment integrity index.

[0157] Among them, the containment integrity index is specifically used to characterize the risk of failure of the physical containment system, and the Bayesian network model is used to describe the causal relationship and risk propagation path between the macro-environmental barrier layer and the key node barrier layer.

[0158] In this embodiment, a Bayesian network is a probabilistic graphical model that uses a directed acyclic graph (DAG) to represent conditional dependencies between variables and quantifies these dependencies using conditional probability distributions. The Bayesian network model is constructed using the following process:

[0159] 1) Node definition: The nodes in the network will include:

[0160] Nodes representing the health of the macro-environmental barrier layer (input).

[0161] Node representing the health of the barrier layer at the critical node (input).

[0162] Represents the intermediate nodes of various potential physical containment failure events (such as "negative pressure failure", "BSC airflow abnormality", "door not closed tightly", "filter blockage", etc.).

[0163] The final output node, the “containment integrity index,” represents the probability or risk of overall failure of the physical containment system.

[0164] 2) Causal Links and Risk Propagation Paths: The edges (directed connections) of the Bayesian network depict the causal relationships between macro- and key-node barrier layers. For example, a decline in the health of the macro-environmental barrier layer (e.g., insufficient overall negative pressure) could lead to functional impairments at multiple key-node barriers (e.g., turbulent airflow into the BSC), increasing the risk of overall system failure. Conversely, a failure at a key node (e.g., a BSC) could affect airflow in a local area, further compromising the effectiveness of the macro-environmental barrier layer. The network illustrates these interactions and how risks propagate across different layers.

[0165] 3) Establishing a Conditional Probability Table (CPT): Each non-root node requires a CPT, which defines the probability of the node being in different states given the state of its parent node. These probabilities can be set and trained based on historical data, expert experience, or simulation results. For example, when the health of the macro-environmental barrier layer is "poor," the probability of airflow anomalies at a key node (such as the BSC) increases significantly.

[0166] 4) Probabilistic reasoning: Once the Bayesian network model is established and parameterized, the first health index and the second health index calculated in real time can be input into the model as evidence (i.e., setting the specific states of these input nodes).

[0167] The Bayesian network model utilizes its internal probabilistic inference algorithms (such as belief propagation and variational inference) to update the probability distributions of all other nodes in the network based on this input evidence, ultimately calculating the posterior probability of the node's "containment integrity index." This posterior probability quantifies the risk of failure of the current physical containment system. A higher index indicates a greater risk of physical containment failure and more urgent intervention is needed.

[0168] Through this probabilistic reasoning, the system can comprehensively consider the mutual influence of various factors and provide a more comprehensive and robust risk assessment result.

[0169] Step 305 : extracting a plurality of representative historical change curve samples of the containment integrity index from a preset baseline database.

[0170] The core of this step is to construct a reference set that includes the change characteristics of the containment integrity index under normal laboratory operation and various possible abnormal or failure modes.

[0171] The establishment of this baseline database is a process of continuous accumulation and optimization. It usually includes the following types of historical curve samples:

[0172] The first type is the normal operating baseline, which is a curve showing the change in the containment integrity index over time when all physical containment systems in the laboratory (including macro-environmental barriers and critical node barriers) are in good, stable, and in compliance with specifications. These curves reflect the "standard pattern" of system health, which may include daily fluctuations but generally maintain a high level of integrity.

[0173] The second type is a baseline of known abnormal patterns. For example, the change curve of the containment integrity index recorded when a biosafety cabinet fan fails, a door seal deteriorates, or the overall negative pressure of the laboratory fluctuates slightly. These curves represent the change pattern of the index caused by specific, known systemic problems.

[0174] The third category could be a maintenance or calibration mode baseline, which is a temporary drop or fluctuation in the index that may occur when regular maintenance, calibration, or system upgrades are performed.

[0175] By extracting these representative historical samples, we can provide a rich reference for comparison with the real-time monitoring containment integrity index curve. Subsequent matching calculations can identify whether the current system state is normal, a specific abnormality, or another unknown state, thus providing a basis for risk warning and fault diagnosis. These curve samples need to be time-aligned and standardized to ensure that they are comparable in shape and trend, although they may not be consistent on the absolute time axis.

[0176] Step 306 , performing dynamic time warping calculation on the historical change curve samples to obtain the cumulative distance of the warping path between the real-time change curve of the containment integrity index and each historical change curve sample.

[0177] The core idea of ​​this step is to allow the points in the time series to be aligned “non-linearly” to minimize the distance between them.

[0178] The DTW algorithm first pairs the real-time containment integrity index curve (the query curve) with each historical change curve sample (the reference curve) in the baseline database. The DTW algorithm constructs a distance matrix, where each element represents the distance between corresponding points in the query and reference curves.

[0179] The DTW algorithm then searches for a "regular path" from the lower left corner of the matrix to the upper right corner, where the sum of the distances between the points is minimized. This path allows one point in one time series to be aligned with multiple points in another time series, or multiple points in one time series to be aligned with one point in another time series, thus compensating for differences in the time axis.

[0180] The resulting cumulative distance for the regularized path is the sum of the distances between all pairs of points on the best matching path. The smaller this cumulative distance, the more similar the shape of the real-time curve is to that of the historical sample curve, even if their speed of change or phase differs. By performing DTW calculations on all historical sample curves, a distance list is generated, reflecting the degree of similarity between the real-time curve and each known historical pattern.

[0181] Step 307 : Based on the regular path cumulative distance, a preset distance matching conversion function is used to calculate the matching degree between the real-time change curve and the baseline database.

[0182] The purpose of this step is to convert the original "regular path cumulative distance" output by the DTW algorithm into a more intuitive and easier to understand "matching degree" indicator.

[0183] Because the cumulative distance itself is a non-negative value, its magnitude depends on the length and numerical range of the time series, making direct use difficult. Therefore, a pre-defined conversion function is required to map it to a standardized matching metric, such as between 0 and 1 (or 0% to 100%). The design of this conversion function is crucial; it is typically a monotonically decreasing function, meaning that smaller distances indicate higher matching.

[0184] Common conversion function forms include: exponential decay function (such as e -k∙distance ), a reciprocal function (such as 1 / (1+k∙distance), or a threshold-based linear mapping. The calibration parameter k needs to be calibrated and optimized based on the actual application scenario and expert experience to ensure that the matching degree accurately reflects the actual similarity. Distance represents the cumulative distance of the regularized path.

[0185] When the real-time curve matches a historical normal operation baseline sample very well, it indicates that the current system state is likely to be normal operation; if it matches a known abnormal pattern baseline sample even more closely, it may indicate that the system is in or is about to enter that abnormal state.

[0186] Ultimately, the degree of match between the real-time curve and the baseline database can be taken as the value with the highest match among all historical samples, or a weighted average of all matching degrees, thereby providing a comprehensive and quantitative containment integrity assessment result.

[0187] In step 308 , when the matching degree is lower than the preset threshold, it is predicted that there is a potential risk of substantial damage to the airflow containment barrier, and the source of the risk is located based on the data source causing the matching degree to be lower than the preset threshold.

[0188] This step can refer to the description of the above embodiment and will not be repeated here.

[0189] In step 309 , the risk source is used as the initial release point of the virtual aerosol pollutant, and the real-time environmental data is used as the boundary condition to perform fluid dynamics simulation processing to obtain the three-dimensional spatiotemporal diffusion trajectory of the virtual aerosol pollutant in the biosafety laboratory.

[0190] In this step, fluid dynamics simulation processing is performed, which specifically includes the following steps:

[0191] S3091, obtain the transmission characteristic parameters of the target pathogens currently involved in the biosafety laboratory.

[0192] Transmission characteristic parameters include at least one of sedimentation velocity, half-life, inactivation rate, and minimum infectious dose. Sedimentation velocity refers to the steady downward velocity of pathogen particles in air or liquid due to gravity. Half-life refers to the time required for the number or infectious activity of a pathogen in a specific environment to decrease to half of its initial value. Inactivation rate refers to the rate at which a pathogen loses its ability to infect under specific environmental conditions (such as temperature, humidity, and ultraviolet radiation). The minimum infectious dose refers to the minimum amount of pathogen required to cause infection in a susceptible host.

[0193] Specifically, obtaining these transmission characteristic parameters is usually a multifaceted process. First, the most important and authoritative sources are scientific literature and professional databases, such as academic journal databases like PubMed and Web of Science, as well as specialized pathogen information repositories (such as biosafety guidelines and pathogen characteristic data published by the CDC and WHO). These data typically contain a large amount of published research data on the settlement, survival, and infectivity of specific pathogens under different environmental conditions.

[0194] Secondly, based on the knowledge and experience of experts in fields such as microbiology, epidemiology, or biosafety, interpretations of existing data can be obtained or key research directions can be pointed out. In some cases, if existing literature data are insufficient to meet the needs of specific laboratory conditions or new pathogens, direct measurements may need to be made through laboratory experiments, such as aerosol stability experiments to determine half-life and inactivation rate, or animal model studies to estimate the minimum infectious dose, but this usually requires high-level biosafety laboratory conditions and professional experimental design.

[0195] Ultimately, the collected data needs to be evaluated and screened to ensure its accuracy, applicability, and compatibility with the current laboratory environment and pathogen strains.

[0196] S3092: Adjust the pollutant diffusion model parameters used in the fluid dynamics simulation process based on the propagation characteristic parameters.

[0197] In this step, the pathogen-specific data obtained in the previous step is input into a mathematical model used in computational fluid dynamics (CFD) software to simulate pollutant dispersion. Fluid dynamics simulations typically use pollutant dispersion models to predict the movement and concentration distribution of particles or gases in the air. These models contain a range of adjustable parameters to adapt to the characteristics of different pollutants.

[0198] For example, if a Lagrangian particle tracking model is used to simulate the movement of pathogen particles, the pathogen's sedimentation velocity will be directly used as an input parameter for the particle's motion in the gravity field. If an Eulerian model is used to simulate the pathogen concentration field, the pathogen's inactivation rate or half-life will be used as a parameter in the pollutant attenuation term in the model to reflect its loss of activity during transmission.

[0199] Furthermore, other model parameters, such as the turbulence model, boundary conditions (e.g., wall adsorption rate), and aerodynamic diameter distribution, may also require fine-tuning based on the characteristics of the pathogen. While the minimum infectious dose does not directly adjust diffusion model parameters, it can influence the interpretation of simulation results for delineating risk zones or assessing exposure levels.

[0200] This process is the key to converting biological characteristics into mathematical language that can be recognized and processed by physical models, ensuring that the simulation results can more accurately reflect the spread behavior of pathogens in the real world.

[0201] In some embodiments, this step may specifically include:

[0202] 1) Based on the settling velocity of the target pathogen, calculate the equivalent gravitational settling rate parameters of the corresponding suspended particles in the fluid dynamics simulation model.

[0203] The core of this step is to convert the actual physical settling characteristics of pathogens into numerical parameters that the simulation model can recognize and process. The settling velocity of pathogens is often affected by their size, density, and shape, as well as the viscosity of the surrounding fluid (air). In fluid dynamics simulations, particularly those based on Lagrangian particle tracking, each simulated "particle" represents a certain number of pathogens, and their motion trajectory is affected by gravity. Therefore, it is necessary to convert the actual measured or retrieved settling velocity of the pathogens into the vertical downward velocity component of the particle under the influence of gravity in the model. For example, if the pathogen is considered a spherical particle, Stokes' Law can be used to relate the particle diameter, density, and fluid viscosity to the settling velocity. For non-spherical particles, a shape factor may need to be introduced. In simulation software, a parameter typically defines the behavior of simulated particles in a gravitational field: the equivalent gravitational settling rate.

[0204] By inputting the pathogen's true settling velocity into this parameter, the simulation model can accurately simulate the process of pathogen particles sinking in the air due to gravity, which is crucial for assessing the vertical distribution of pathogens in space and the risk of deposition on surfaces.

[0205] 2) Based on the half-life and / or inactivation rate of the target pathogen, calculate the corresponding concentration decay coefficient parameters of suspended particulate matter in the fluid dynamics simulation model.

[0206] The purpose of this step is to convert the biological characteristics of pathogens losing their infectious activity in the environment into a mathematical expression of the decay of pollutant concentration over time in the simulation model.

[0207] Both half-life and inactivation rate describe the decay of pathogen activity, and there is a direct mathematical relationship between them. Typically, the inactivation of a pathogen can be approximated as first-order kinetics, meaning that its inactivation rate is proportional to its current concentration. In this case, the concentration decay coefficient (or decay constant) λ can be calculated by the half-life t 1 / 2 The calculation results show that the formula is λ=ln(2) / t 1 / 2 If the inactivation rate is known, it can be directly converted to a decay coefficient.

[0208] Fluid dynamics simulation models, particularly those based on the Euler method, often include a source or sink term to simulate the generation or attenuation of pollutants. The concentration decay coefficient λ is used as a parameter for this sink term, representing the proportion of the pathogen concentration that decreases per unit time due to inactivation.

[0209] By incorporating the biological inactivation characteristics of pathogens into the model, the simulation results can more realistically reflect the dynamic changes in the pathogen's infectious ability during the transmission process, which is of great significance for assessing the risk of long-term exposure and determining the safe operation time.

[0210] 3) Adjust the configuration of pollutant diffusion model parameters based on the equivalent gravity sedimentation rate parameters and concentration attenuation coefficient parameters.

[0211] This step is to systematically integrate the pathogen-specific physical (sedimentation) and biological (inactivation) parameters calculated previously into the pollutant diffusion model settings of the fluid dynamics simulation software. This is not just a simple input of numerical values, but also includes selecting the appropriate model type and configuring the relevant sub-models. For example, if the pathogen is mainly transmitted through aerosols and has significant sedimentation, it may be necessary to select a Lagrangian particle tracking model that can accurately simulate the motion of particles and use the equivalent gravitational sedimentation rate as the velocity component of each particle in the gravity field. If the focus is on the overall distribution and attenuation of the pathogen concentration field, the Euler model may be more appropriate. In this case, the concentration attenuation coefficient will be applied to the attenuation term of the concentration transport equation.

[0212] In addition, other relevant model parameters need to be adjusted according to the characteristics of the pathogen and the simulation purpose, such as turbulence models (such as k-epsilon, k-omega, etc., which affect the speed of air mixing and diffusion), boundary conditions (such as the adsorption or deposition rate of pathogens on the wall), and the degree of meshing.

[0213] This process ensures that the simulation model can comprehensively and accurately reflect the complex transmission behavior of the target pathogen in a specific biosafety laboratory environment, thereby providing a reliable scientific basis for subsequent risk assessment and intervention measures.

[0214] S3093, performing fluid dynamics simulation processing using the adjusted pollutant diffusion model parameters to obtain an optimized three-dimensional spatiotemporal diffusion trajectory.

[0215] After fine-tuning the model parameters, this step involves running computational fluid dynamics (CFD) simulation software to simulate the spread of pathogens within a biosafety laboratory. This process requires input of a detailed 3D geometric model of the laboratory, HVAC operating parameters (such as supply and exhaust air volume, and vent locations), indoor environmental conditions such as temperature and humidity, and the initial pathogen release conditions (such as the release source location, release volume, and release duration).

[0216] The simulation software performs complex numerical calculations on a computer based on fundamental equations of fluid dynamics (such as the Navier-Stokes equations) and contaminant transport equations, combined with adjusted parameters for pathogen transmission characteristics. The resulting output is a time-varying concentration distribution or particle trajectories of the pathogen within the laboratory space. These results are optimized because they fully account for the unique biophysical properties of the target pathogen, making the simulation predictions more realistic.

[0217] The resulting "three-dimensional spatiotemporal diffusion trajectory" is not a simple line; rather, it represents the dynamic process of the pathogen's spread from the release source to various areas within the three-dimensional laboratory space. This includes changes in concentration over time, the formation and dissipation of high-concentration areas, and possible transmission paths. These visual and quantitative results are extremely important for assessing potential infection risks, optimizing ventilation system design, developing emergency response plans, and guiding personnel protection strategies.

[0218] Step 310: extract risk quantification indicators for guiding emergency response from the three-dimensional spatiotemporal diffusion trajectory.

[0219] This step can refer to the description of the above embodiment and will not be repeated here.

[0220] Step 311: Generate laboratory airflow risk warning information based on the risk quantification index.

[0221] This step can refer to the description of the above embodiment and will not be repeated here.

[0222] The method provided in the above embodiment can be executed by an airflow risk monitoring system, which is composed of an electronic device. The following describes the electronic device in the embodiment of the present invention from the perspective of hardware processing. Figure 2 , which is a schematic diagram of a physical device structure of an electronic device in an embodiment of the present invention.

[0223] It should be noted that Figure 2 The structure of the electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0224] like Figure 2 As shown, the electronic device includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes, such as the methods described in the above embodiments, based on programs stored in a read-only memory (ROM) 402 or programs loaded from a storage unit 408 into a random access memory (RAM) 403. RAM 403 also stores various programs and data required for system operation. CPU 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to bus 404.

[0225] The following components are connected to the input / output (I / O) interface 405: an input section 406 including an audio input device, push button switches, and the like; an output section 407 including a display, an audio output device, indicator lights, and the like; a storage section 408 including a hard disk and the like; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card or a modem. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the input / output (I / O) interface 405 as needed. Removable media 411, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 410 as needed, so that computer programs read from the media can be installed in the storage section 408 as needed.

[0226] In particular, according to an embodiment of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, an embodiment of the present invention includes a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 409 and / or installed from removable media 411. When executed by the central processing unit (CPU) 401, the computer program performs the various functions defined in the present invention.

[0227] It should be noted that specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0228] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram can represent a module, program segment, or part of the code, and the above-mentioned module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings.

[0229] Specifically, the electronic device of this embodiment includes a processor and a memory, the memory is coupled to one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and one or more processors call the computer instructions to enable the electronic device to execute the method provided by the above embodiment.

[0230] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments, or may exist independently and not incorporated into the electronic device. The storage medium carries one or more computer programs, and when executed by a processor of the electronic device, the electronic device implements the methods provided in the above embodiments.

[0231] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention.

[0232] As used in the above embodiments, the term “when” may be interpreted to mean “if” or “after” or “in response to determining that” or “in response to detecting that”, depending on the context. Similarly, the phrases “upon determining that” or “if (stated condition or event) is detected” may be interpreted to mean “if determining that” or “in response to determining that” or “upon detecting (stated condition or event)” or “in response to detecting (stated condition or event)”, depending on the context.

[0233] Those skilled in the art will appreciate that all or part of the process steps in the above-described method embodiments can be implemented by a computer program instructing the relevant hardware. The program can be stored in a computer-readable storage medium, and when executed, the program can include the process steps in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for monitoring airflow risk in a biosafety laboratory, characterized in that: include: Acquiring airflow sensor data and facility status data, wherein the airflow sensor data is used to characterize the pressure gradient and directional airflow in critical passages between clean areas, potential contaminated areas, and core contaminated areas within the biosafety laboratory, and the facility status data is used to characterize the operational status of the physical containment system; Calculating a containment integrity index based on the airflow sensor data and the facility status data, wherein the containment integrity index is used to dynamically quantify the effectiveness and integrity of the airflow containment barrier of the biosafety laboratory; Performing dynamic time-warping matching on the real-time change curve of the containment integrity index with a preset baseline database to obtain a matching degree; When the matching degree is lower than a preset threshold, it is predicted that there is a potential risk of substantial damage to the airflow containment barrier, and the source of the risk is located based on the data source causing the matching degree to be lower than the preset threshold; Taking the risk source as the initial release point of the virtual aerosol pollutant and using real-time environmental data as boundary conditions, performing fluid dynamics simulation processing to obtain a three-dimensional spatiotemporal diffusion trajectory of the virtual aerosol pollutant in the biosafety laboratory; Extracting risk quantification indicators for guiding emergency response from the three-dimensional spatiotemporal diffusion trajectory, the risk quantification indicators including the shortest breakthrough time for pollutants escaping from the risk source to reach the next clean area or exit, and the estimated exposure dose calculated for personnel at different locations; Laboratory airflow risk warning information is generated based on the risk quantification index.

2. The method according to claim 1, characterized in that Calculating a containment integrity index based on the airflow sensor data and the facility status data includes: Deconstructing the physical containment system of the biosafety laboratory into a macro-environmental barrier layer and a key node barrier layer; Based on the airflow sensor data, a first health index of the macro-environmental barrier layer is calculated using a preset fluid dynamics algorithm; and based on the facility status data, a second health index of the key node barrier layer is calculated using a preset operational safety procedure; The first health index and the second health index are used as inputs to a preset Bayesian network model for probabilistic reasoning to obtain a containment integrity index. The containment integrity index is used to characterize the risk of failure of the physical containment system. The Bayesian network model is used to describe the causal relationship and risk propagation path between the macro-environmental barrier layer and the key node barrier layer.

3. The method according to claim 2, characterized in that The physical containment system of the biosafety laboratory is deconstructed into a macro-environmental barrier layer and a key node barrier layer, including: Extracting a set of partition physical isolation components between the clean area, the potential contaminated area, and the core contaminated area; In the set of partition physical isolation components, the enclosure structure and sealing components used to ensure the stability of the pressure difference between partitions are classified as macro-environmental barrier elements, and the dynamic opening and closing devices or equipment through-structures used to connect the partitions are classified as key node barrier elements; Establishing a topological mapping table of the macro-environmental barrier elements and key node barrier elements; Marking the airflow blocking dependency relationship between the macro-environment barrier elements and the key node barrier elements in the topology mapping table; Based on the annotated topology mapping table, a macro-environment barrier layer is constructed through the macro-environment barrier elements, and a key node barrier layer is constructed through the key node barrier elements.

4. The method according to claim 1, wherein Calculating a containment integrity index based on the airflow sensor data and the facility status data includes: Determining a flow field disturbance factor based on the real-time dynamic variation of the airflow sensor data, and determining a barrier operation disturbance factor based on the real-time dynamic variation of the facility status data; generating a containment disturbance factor sequence in time sequence based on the flow field disturbance factor and the barrier operation disturbance factor; The containment disturbance factor sequence is nonlinearly weighted and an integral operation with a time decay effect is performed to obtain a containment integrity index, which is used to reflect the cumulative impact of historical states and current impact intensity.

5. The method according to any one of claims 1 to 4, characterized in that The performing of fluid dynamics simulation processing includes: Obtaining transmission characteristic parameters of the target pathogen currently involved in the biosafety laboratory, wherein the transmission characteristic parameters include at least one of sedimentation velocity, half-life, inactivation rate, and minimum infectious dose; Adjusting the pollutant diffusion model parameters used in the fluid dynamics simulation process based on the propagation characteristic parameters; The optimized three-dimensional space-time diffusion trajectory is obtained by performing fluid dynamics simulation on the adjusted pollutant diffusion model parameters.

6. The method according to claim 5, characterized in that The adjusting, based on the propagation characteristic parameters, the pollutant diffusion model parameters used in the fluid dynamics simulation process includes: Calculating the equivalent gravity sedimentation rate parameters of the corresponding suspended particles in the fluid dynamics simulation model according to the sedimentation velocity of the target pathogen; Calculating a concentration decay coefficient parameter of the corresponding suspended particulate matter in the fluid dynamics simulation model according to the half-life and / or inactivation rate of the target pathogen; Based on the equivalent gravitational settling rate parameter and the concentration attenuation coefficient parameter, the configuration of the pollutant diffusion model parameters is adjusted.

7. The method according to claim 1, characterized in that The real-time change curve of the containment integrity index is dynamically time-warped matched with a preset baseline database to obtain a matching degree, including: Extract multiple representative historical change curve samples of the containment integrity index from the preset baseline database; Performing dynamic time warping calculation on the historical change curve samples to obtain a warping path cumulative distance between the real-time change curve of the containment integrity index and each of the historical change curve samples; Based on the regular path cumulative distance, the matching degree between the real-time change curve and the baseline database is calculated by a preset distance matching degree conversion function.

8. An electronic device, characterized in that: including one or more processors and memory; The memory is coupled to the one or more processors, and is configured to store computer program codes, where the computer program codes include computer instructions. The one or more processors call the computer instructions to enable the electronic device to execute the method according to any one of claims 1 to 7.

9. A computer-readable storage medium storing computer instructions, characterized in that: When the computer instructions are executed on an electronic device, the electronic device is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product is run on an electronic device, the electronic device is enabled to perform the method according to any one of claims 1 to 7.

Images