A laboratory data analysis method

Abstract

This invention discloses a laboratory data analysis method, relating to the technical field of laboratory data. The method includes: acquiring the identification information of operators in the laboratory and the experimental time information of each operator; acquiring the instrument names of the instruments accessed sequentially by each operator and the sample names of the samples accessed sequentially by each operator; acquiring sensor data of the experimental process based on the experimental time information; inputting the operator identification information, the experimental time information, the instrument names, the sample names, and the sensor data into a pre-trained analysis model to obtain the laboratory's stress value; and outputting alarm information based on the stress value. This invention improves laboratory safety by tracking operator actions and quantifying the potential risks to the laboratory from those actions.

Description

Technical Field

[0001] This invention relates to the technical field of laboratory data, and more particularly to a laboratory data analysis method. Background Technology

[0002] In a laboratory environment, the end of an experiment is far from the end of safety work. The inherent difficulty in determining potential safety hazards remaining after an experiment constitutes a very real and challenging problem in laboratory safety management. This difficulty in determination is not accidental, but rather stems from the interplay of multiple factors.

[0003] Safety hazards at the end of laboratory operations are often difficult to detect in a timely manner due to their concealed and latent nature. On the one hand, trace chemical residues after experiments (such as uncleaned organic solvents) may slowly react under changes in ambient temperature and humidity, generating toxic gases or explosives, but the initial concentration is extremely low and undetectable by the senses. On the other hand, static risks after equipment shutdown are equally insidious, such as residual pressure in high-pressure reactors or microcracks in centrifuge rotors, which may appear static but could suddenly malfunction upon restarting. Even more complex are waste risks—mixed waste liquids (such as solutions containing cyanide and acid) may initially show no signs of problems, but can quietly react during storage to generate highly toxic gases; these hazards are completely unpredictable to the naked eye.

[0004] The difficulty in human judgment stems primarily from the inherent limitations of our senses: colorless and odorless harmful gases (such as carbon monoxide), trace amounts of radiation, or volatile chemical compounds are completely beyond human perception, while conventional detection tools (such as test strips) often give false "safe" signals due to insufficient sensitivity. Secondly, cognitive biases and operational inertia also pose obstacles: lab personnel tend to focus on the main reaction while neglecting the accumulation of byproducts (such as the slow enrichment of organic peroxides), and are more likely to simplify the finishing process when fatigued (such as leaving the fume hood open). More importantly, the complexity of modern laboratories far exceeds individual experience: the dynamic correlation between multiple coupled potential hazards (such as aging circuits combined with ventilation failures) or sudden environmental disturbances (such as vibration causing gas cylinders to tip over) is almost impossible to predict manually.

[0005] Therefore, improving laboratory safety has become an urgent technical problem to be solved. Summary of the Invention

[0006] The technical problem solved by this invention is the difficulty in timely monitoring of safety hazards in the laboratory.

[0007] To address the aforementioned technical problems, the present invention provides the following technical solution: a laboratory data analysis method, comprising: acquiring the identification information of operators in the laboratory and the experimental time information of each operator; acquiring the instrument names of the instruments accessed sequentially by each operator and the sample names of the samples accessed sequentially by each operator; acquiring sensor data of the experimental process based on the experimental time information; inputting the operator identification information, the experimental time information, the instrument names, the sample names, and the sensor data into a pre-trained analysis model to obtain the stress value of the laboratory; and outputting alarm information based on the stress value.

[0008] Preferably, before acquiring the identification information of the operator in the laboratory, the method further includes: in response to a person entering the laboratory, tracking the person's movement trajectory through a behavior recording device and detecting the person's posture information; if the person's posture information changes from standing to sitting, acquiring the workstation identifier at the person's seat; controlling the start of the experimental recording device at the workstation according to the workstation identifier; performing facial recognition on the person through the experimental recording device to acquire the person's identification information; capturing the person's eye focus and hand movements through the experimental recording device; if the person's eyes are focused on the experimental sample at the workstation and the person's hands are in contact with the experimental instrument, then determining that the person is an operator.

[0009] Preferably, the step of tracking the movement trajectory of a person entering the laboratory and detecting the person's posture information via a behavior recording device includes: recording the start time of the person's entry into the laboratory; tracking the movement trajectory of the person and detecting the person's posture information via the behavior recording device; and obtaining the identification information of operators in the laboratory and the experimental time information of each operator, including: if the person is an operator, sending the operator's identification information via the experimental recording device; tracking the operator's movement trajectory via the behavior recording device until the operator's movement trajectory is lost after a preset time, then determining the last time the operator appeared via the behavior recording device; determining the last time as the end time; and determining the operator's experimental time information based on the start time and the end time.

[0010] Preferably, the behavior recording device includes a camera module, a processing module, a communication module, and a radio frequency information module.

[0011] Preferably, the instruments in the laboratory are equipped with electronic tags; the acquisition of the instrument name of each instrument accessed sequentially by each experimenter and the sample name of each sample accessed sequentially by each experimenter includes: if the behavior recording device detects that the operator has entered the instrument area, then controlling the radio frequency signal transmission module to start; reading the electronic tag information of the instrument through multiple radio frequency signal modules; comparing the previously read electronic tag information with the currently read electronic tag information, and determining the electronic tag that exists in the currently read electronic tag information but does not exist in the previously read electronic tag information as the instrument currently accessed by the operator; determining the instrument name based on the electronic tag of the instrument currently accessed by the operator.

[0012] Preferably, the containers of the samples in the laboratory are equipped with electronic tags; the step of obtaining the names of the instruments that each experimenter touches in sequence and the names of the samples that each experimenter touches in sequence further includes: if the behavior recording device detects that the operator has entered the sample area, controlling the radio frequency signal transmission module to start; reading the electronic tag information of the sample through multiple radio frequency signal modules; comparing the previously read electronic tag information with the currently read electronic tag information, and determining the electronic tag that exists in the currently read electronic tag information but does not exist in the previously read electronic tag information as the sample currently being touched by the operator; and determining the sample name based on the electronic tag of the sample currently being touched by the operator.

[0013] Preferably, the analysis model is a neural network model; the method for generating training samples for the analysis model includes: searching the laboratory's monitoring records, hazard incident handling records, and sensor sensing data records; obtaining the monitoring records corresponding to the hazard incident handling records, and extracting the operator identification information, experiment time information, instrument name, and sample name from the monitoring records; obtaining the sensor data records corresponding to the hazard incident handling records, and extracting sensor data from the sensor data records; and determining the stress value corresponding to the hazard incident handling records based on the accident information in the hazard incident handling records.

[0014] Preferably, before outputting alarm information based on the stress value, the method further includes: determining the experimental project for each experimenter based on the instrument name of the instruments accessed sequentially by each experimenter and the sample name of the samples accessed sequentially by each experimenter; obtaining standard procedure specification information from a preset standard procedure library based on the experimental project for each experimenter; scoring the experimenter's operation based on the standard procedure specification information; and outputting alarm information based on the stress value includes: outputting the alarm information based on the score of each experimenter and the stress value.

[0015] Preferably, the step of outputting the alarm information based on each experimenter's score and the stress value includes: generating alarm details based on each experimenter's score, the experiment item, and the stress value; and outputting the alarm details.

[0016] Preferably, the step of outputting the alarm details includes: obtaining the operator's communication address based on the operator's identification information; and pushing the alarm details to the operator based on the communication address.

[0017] The beneficial effects of this invention are as follows: It acquires the identification information of operators in the laboratory and the experimental time information of each operator; it acquires the names of the instruments that each operator sequentially touches and the names of the samples that each operator sequentially touches; it acquires sensor data of the experimental process based on the experimental time information; it inputs the operator's identification information, experimental time information, instrument names, sample names, and sensor data into a pre-trained analysis model to obtain the laboratory's stress value; and it outputs alarm information based on the stress value. By tracking the operators' operations, it quantifies the degree of risk that the operations may bring to the laboratory, thereby improving the safety of the laboratory. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the basic process of a laboratory data analysis method provided in one embodiment of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0020] Example 1, referring to Figure 1 As an embodiment of the present invention, a laboratory data analysis method is provided, including steps S110-S150:

[0021] S110, acquire the identification information of the operators in the laboratory and the experimental time information of each operator.

[0022] S120, obtain the instrument name of each experimenter who operates and touches the instrument in sequence, and the sample name of each experimenter who operates and touches the sample in sequence.

[0023] S130: Obtain sensor data during the experiment based on the experiment time information.

[0024] S140: Input the operator's identification information, experimental time information, instrument name, sample name and sensor data into the pre-trained analysis model to obtain the laboratory stress value.

[0025] S150, output alarm information based on the coercion value.

[0026] Safety hazards in the laboratory are embedded in the operations of laboratory personnel; therefore, data related to safety hazards can be obtained from these operations. For example, failure to promptly clean up organic solvents can cause them to react slowly under changes in ambient temperature and humidity, generating toxic gases or explosives.

[0027] To address this, risk-related factors were identified: number of operators, experiment duration, instrument type, sample type, and environmental data. These factors were used to quantify the potential risks that operators' actions might pose to the laboratory.

[0028] Among them, the stress value, as a pre-trained analysis model, integrates multiple factors to output a quantitative indicator reflecting the real-time risk in the laboratory, so as to transform the ambiguous safety status into quantifiable data and provide early warning; the sensor data are environmental parameters such as temperature, humidity and gas concentration that are collected in real time through the deployment of IoT sensor networks; furthermore, the status of equipment, such as current or equipment temperature data, can also be collected through sensors.

[0029] The stress value of the laboratory is obtained by inputting the operator's identification information, experiment time information, instrument name, sample name, and sensor data into a pre-trained analysis model. In other words, relevant data on personnel, instruments, samples, and the environment are input into a pre-trained neural network model for multimodal fusion analysis. Specifically, the stress value is a dynamic risk score (typically ranging from 0 to 1) generated by the multimodal neural network model after fusing and analyzing heterogeneous data such as laboratory personnel operations, equipment status, and environmental parameters. Its core design includes the following technologies: feature fusion mechanism, such as the VGG19 image style transfer model, which extracts image content and style features through convolutional layers and designs content loss and style loss functions to achieve multi-source feature fusion. In a physical laboratory scenario, for example, personnel operation behavior (video stream, including identification information of operators, experimental time information, and sample name annotation information), equipment current signal (time series data, including annotation information of instrument name and instrument usage), and gas concentration (sensor stream) can be regarded as different modal inputs to construct a fusion loss function; dynamic weight allocation, setting the parameter update logic of a three-layer neural network (ThreeLayerNet), and the model automatically learns the weight ratio of each risk factor through backpropagation (e.g., the weight of not wearing gloves is greater than the weight of exceeding temperature and humidity limits).

[0030] The neural network model outputs stress values ​​to generate alarm information and achieves accurate risk push, avoiding false alarms and missed alarms.

[0031] The methods for generating training samples for the analysis model include S210~S240:

[0032] S210, Locate the laboratory's monitoring records, hazard incident handling records, and sensor sensing data records;

[0033] S220, Obtain the monitoring record corresponding to the hazard incident handling record, and extract the operator identification information, experiment time information, instrument name, and sample name corresponding to the hazard incident handling record from the monitoring record;

[0034] S230, acquire the sensor data record corresponding to the hazardous event handling record, and extract sensor data from the sensor data record;

[0035] S240, determine the stress value corresponding to the hazardous incident handling record based on the accident information in the hazardous incident handling record.

[0036] By recording hazardous events, a historical accident report database is obtained, and sensor data and monitoring video timestamps are correlated to obtain a true risk mapping of hazardous events and risk-related factors.

[0037] Stress value labeling, also known as risk level labeling, can be based on the severity of an accident, such as minor = 30, fire = 90, and explosion = 100.

[0038] When constructing a risk prediction neural network, an LSTM+Attention network can be used to process time-series data (operation sequence + sensor flow) to capture the dynamic evolution of risk.

[0039] Preferably, before S110, the method further includes S101 to S106:

[0040] S101, in response to personnel entering the laboratory, tracks the personnel's movement trajectory and detects the personnel's posture information through behavior recording devices.

[0041] S102, if the person's posture information changes from standing to sitting, then obtain the workstation identifier at the person's seat location.

[0042] S103, control the start of the experimental recording equipment at the workstation according to the workstation identification.

[0043] S104. Facial recognition of personnel is performed using experimental recording equipment to obtain personnel identification information.

[0044] S105 uses experimental recording equipment to capture the eye focus and hand movements of personnel.

[0045] S106. If the personnel's eyes are focused on the experimental sample at the workstation and the personnel's hands are in contact with the experimental instrument, then the personnel are identified as operators.

[0046] Among them, the behavior recording device is an AI camera with posture tracking, which can recognize the switching between standing and sitting postures through the OpenPose algorithm and automatically bind personnel to workstations; the experimental recording device is a terminal that integrates facial recognition, eye movement and hand tracking, and deploys a depth camera at the workstation to capture actions.

[0047] Eye focus detection determines whether the gaze is locked onto the experimental sample. Specifically, the YOLOv5 object detection model can be used to define the sample area and calculate the intersection of gaze points to avoid misjudgments in non-operational states. Hand contact detection can identify the interaction between the human hand and the instrument. Specifically, MediaPipe gesture recognition and instrument contour matching can be used to ensure the validity of the recording.

[0048] Preferably, S101 includes: in response to personnel entering the laboratory, recording the start time of personnel entering the laboratory; tracking the personnel's movement trajectory through a behavior recording device and detecting the personnel's posture information; S110 further includes: if the personnel is an operator, sending the operator's identification information through the experiment recording device; tracking the operator's movement trajectory through the behavior recording device until the operator's movement trajectory is lost after a preset time, then determining the last time the operator appeared through the behavior recording device; determining the last time as the end time; and determining the operator's experiment time information based on the start time and the end time.

[0049] Preferably, the behavior recording device includes a camera module, a processing module, a communication module, and a radio frequency information module; electronic tags are affixed to instruments in the laboratory and electronic tags are affixed to containers of samples in the laboratory; S120 includes: if the behavior recording device detects that an operator has entered the instrument area, it controls the radio frequency signal transmission module to start; reads the electronic tag information of the instrument through multiple radio frequency signal modules; compares the previously read electronic tag information with the currently read electronic tag information, and determines the electronic tag that exists in the currently read electronic tag information but does not exist in the previously read electronic tag information as the instrument currently being contacted by the operator; determines the instrument name based on the electronic tag of the instrument currently being contacted by the operator; similarly, if the behavior recording device detects that an operator has entered the sample area, it controls the radio frequency signal transmission module to start; reads the electronic tag information of the sample through multiple radio frequency signal modules; compares the previously read electronic tag information with the currently read electronic tag information, and determines the electronic tag that exists in the currently read electronic tag information but does not exist in the previously read electronic tag information as the sample currently being contacted by the operator; determines the sample name based on the electronic tag of the sample currently being contacted by the operator.

[0050] The radio frequency signal module can be a UHF RFID reader / writer, which can cover the operating area by deploying a multi-antenna array in the laboratory to accurately identify objects through obstructions; the electronic tag is a passive RFID tag that is affixed to the instrument, and the tag stores a unique ID that maps to the instrument name to achieve zero-contact automatic identification.

[0051] Tag difference comparison can dynamically update the list of instruments in contact. For example, it can scan once every 0.5 seconds and compare the difference between the tag sets before and after to accurately record the operation sequence (such as taking the centrifuge first and then putting it back).

[0052] Preferably, before S150, the method further includes: determining the experimental project of each experimenter based on the instrument name of the instrument that each experimenter operates and the sample name of the sample that each experimenter operates and touches in sequence; obtaining standard procedure specification information from a preset standard procedure library based on the experimental project of each experimenter; and scoring the experimenter's operation based on the standard procedure specification information.

[0053] The Standard Procedure Library is a digital knowledge base for experimental SOPs, used to store operational specifications in a structured manner (such as "balanced weights must be balanced before centrifugation") to provide judgment criteria.

[0054] Operational scoring can compare actual behavior with the standard and is used to deduct points based on the rule engine. For example, 20 points are deducted for unbalanced weights, indicating quantification personnel error.

[0055] S150 includes: generating alarm details based on each experimenter's score, experiment item, and stress value; obtaining the operator's communication address based on the operator's identification information; and pushing alarm details to the operator based on the communication address.

[0056] The alarm details are generated, including merging the stress value and the operation score. For example, the formula is: Risk Index = Stress Value × 0.7 + Operation Deviation × 0.3, where the operation deviation can be determined based on the experimental score. For example, after determining the actual experimental score using the method disclosed in the patent with publication number "CN112036840A", the degree of deviation is determined based on the full score of the actual experimental score and the preset experimental score. The settings of 0.7 and 0.3 in the formula are empirical parameters and can be set according to actual usage needs to distinguish between environmental risks and human errors. For example, if the stress value is 70 and the operation score is 50, then the alarm is "High-risk operation: Unprotected + Ethanol leakage". In another optional method, the stress value and the operation score can be output together directly. For example, the alarm information is output: Stress value 70, suspected "Unprotected ethanol leakage".

[0057] The entire process will be explained as follows:

[0058] Step 1: Experiment starts; Researcher A swipes his card to enter the laboratory, and the behavior recording system tracks his trajectory; A sits down at workstation number 3, triggering posture recognition, automatically activating the workstation camera, facial recognition to confirm his identity, detecting that his gaze is focused on the concentrated sulfuric acid sample and that he is holding a pipette, thus determining that the experiment has started.

[0059] Step 2: Operation tracking; A walks to the centrifuge area, the RFID reader scans the newly added tag Centrifuge-05, records the contact with the instrument, collects environmental data in real time, and the temperature and humidity sensor detects that the humidity rises sharply due to the volatilization of H2SO4.

[0060] Step 3: Risk Analysis; The analysis model inputs include: Personnel: Researcher_A, Operation sequence: [Take H2SO4 sample, using Centrifuge-05], Sensors: Humidity +60%, Temperature +2℃; The model matches historical accident patterns and outputs a stress value of 85.

[0061] Step 4: Alarm Generation; The process library detected that centrifugation operations require wearing safety goggles, but video analysis A did not wear them, resulting in an operation score of 40. Alarm details are generated: "[Emergency Alarm] Stress Value 85 + Operation Violation 40"

[0062] ● Risk point: Centrifuge vibration accelerates acid evaporation

[0063] ● Measures: "Immediate ventilation + wearing goggles", which will be pushed to A's mobile phone via WeChat.

[0064] This application embodiment obtains the identification information of the operators in the laboratory and the experimental time information of each operator, obtains the instrument names of the instruments that each operator touches in sequence, and the sample names of the samples that each operator touches in sequence. Based on the experimental time information, it obtains the sensor data of the experimental process, inputs the operator identification information, experimental time information, instrument names, sample names, and sensor data into a pre-trained analysis model to obtain the laboratory's stress value; and outputs alarm information based on the stress value. By tracking the operator's operations, the degree of risk that the operation may bring to the laboratory can be quantified, thereby improving the laboratory's safety.

[0065] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0066] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A laboratory data analysis method, characterized in that, include: Obtain the identification information of the operators in the laboratory and the experimental time information of each operator; Obtain the names of the instruments that each experimenter operates and touches in sequence, and the names of the samples that each experimenter touches in sequence; Sensor data from the experimental process are obtained based on the experimental time information; The operator's identification information, the experiment time information, the instrument name, the sample name, and the sensor data are input into a pre-trained analysis model to perform multimodal fusion analysis and construct a fusion loss function; dynamic weight allocation is performed, the parameter update logic of a three-layer neural network is set, and the model automatically learns the weight ratio of each risk factor through backpropagation to obtain the stress value of the laboratory; Output alarm information based on the aforementioned stress value; Before outputting alarm information based on the stress value, the method further includes: The experimental project for each experimenter is determined based on the names of the instruments they operate and access in sequence, and the names of the samples they access in sequence. Standard procedure specifications are retrieved from a pre-defined standard procedure library based on each experimenter's experimental project; The experimenters' operations were scored based on the standard procedure specifications. The step of outputting alarm information based on the coercion value includes: The alarm information is output based on each participant's score and the stress value; The step of outputting the alarm information based on each participant's score and the stress value includes: Alarm details are generated based on each experimenter's score, the experimental item, and the stress value. Output the alarm details.

2. The method as described in claim 1, characterized in that, Before acquiring the identification information of operators in the laboratory, the method further includes: In response to personnel entering the laboratory, the movement trajectory of the personnel is tracked by behavior recording devices, and the posture information of the personnel is detected; If the posture information of the person changes from standing to sitting, then obtain the workstation identifier of the person's seat. The experimental recording equipment at the control station is started according to the workstation identifier; The personnel are identified by facial recognition using the experimental recording device. The experimental recording device captures the person's eye focus and hand movements; If the person's eyes are focused on the experimental sample at the workstation and the person's hands are in contact with the experimental instrument, then the person is identified as an operator.

3. The method as described in claim 2, characterized in that, The response to a person entering the laboratory includes tracking the person's movement trajectory and detecting the person's posture information using a behavior recording device, including: In response to personnel entering the laboratory, the start time of personnel entering the laboratory is recorded; The movement trajectory of the person is tracked by a behavior recording device, and the person's posture information is detected. The acquisition of the identification information of the operators in the laboratory and the experimental time information of each operator includes: If the person is an operator, then the operator's identification information is sent through the experimental recording device; The behavior recording device tracks the operator's movement trajectory until the operator's movement trajectory is lost after a preset time period. Then, the behavior recording device determines the last moment when the operator appeared. The last moment is defined as the end moment; The operator's experiment time information is determined based on the start time and the end time.

4. The method as described in claim 3, characterized in that, The behavior recording device includes a camera module, a processing module, a communication module, and a radio frequency information module.

5. The method as described in claim 4, characterized in that, The instruments in the laboratory are equipped with electronic tags; the acquisition of the instrument names of each instrument accessed sequentially by each experimenter and the sample names of each sample accessed sequentially by each experimenter includes: If the behavior recording device detects that the operator has entered the instrument area, it controls the radio frequency signal transmission module to start. The electronic tag information of the instrument is read by multiple radio frequency signal modules; Compare the previously read electronic tag information with the currently read electronic tag information, and identify the electronic tag that exists in the currently read electronic tag information but does not exist in the previously read electronic tag information as the instrument currently being contacted by the operator; The name of the instrument is determined based on the electronic tag of the instrument currently being used by the operator.

6. The method as described in claim 5, characterized in that, The containers for the samples in the laboratory are equipped with electronic tags; the acquisition of the names of the instruments that each experimenter operates and touches in sequence, and the names of the samples that each experimenter operates and touches in sequence, also includes: If the behavior recording device detects that the operator has entered the sample area, it controls the radio frequency signal transmission module to start. The electronic tag information of the sample is read by multiple radio frequency signal modules; Compare the previously read electronic tag information with the currently read electronic tag information, and identify the electronic tag that exists in the currently read electronic tag information but does not exist in the previously read electronic tag information as the current contact sample of the operator; The sample name is determined based on the electronic tag of the sample currently in contact with the operator.

7. The method as described in claim 6, characterized in that, The analysis model is a neural network model; The method for generating training samples for the analysis model includes: Locate the laboratory's monitoring records, hazard incident handling records, and sensor data records; Obtain the monitoring records corresponding to the hazard incident handling records, and extract the operator identification information, experiment time information, instrument name, and sample name from the monitoring records corresponding to the hazard incident handling records; Obtain the sensor data record corresponding to the hazardous event handling record, and extract sensor data from the sensor data record; The stress value corresponding to the hazardous event handling record is determined based on the accident information in the hazardous event handling record.

8. The method as described in claim 1, characterized in that, The output of the alarm details includes: The operator's communication address is obtained based on the operator's identification information; The alarm details are pushed to the operator based on the communication address.

Images