Laboratory safety intelligent monitoring management system and method based on Internet of Things and data driving technology

Abstract

The invention discloses a laboratory safety intelligent monitoring management system and method based on the Internet of Things and the data driving technology. The monitoring management system comprises a laboratory environment monitoring module, an experimenter behavior monitoring module, an experimental equipment state monitoring module, a hazardous chemical substance management monitoring module, an intelligent potential safety hazard early warning and evaluation module and an emergency response remote cooperative processing module, and the six modules carry out cooperative processing. The monitoring management method comprises five modes of laboratory environment monitoring, experimenter behavior monitoring, experimental equipment state monitoring, hazardous chemical substance management monitoring and emergency response remote cooperative processing. According to the invention, a newly proposed time sequence prediction attention mechanism module prediction unit is adopted, the prediction accuracy is improved, various factors which are easy to cause experimental accidents in a laboratory are comprehensively considered, early warning can be carried out for different abnormal conditions, and effective supervision of laboratory safety risks is ensured.

Description

Technical Field

[0001] The present invention relates to the field of laboratory safety monitoring, and specifically relates to a laboratory safety intelligent monitoring and management system and method based on Internet of Things and data-driven technology. Background Art

[0002] Laboratory safety situation management is the core task of ensuring the safety of the laboratory operating environment, mainly including the systematic monitoring, comprehensive analysis of potential risks and safety hazards in the laboratory, and the implementation of efficient response strategies. With the continuous development of laboratory experiment contents and technical means, the complexity of laboratory safety management has also increased. In order to effectively prevent accidents and reduce potential risks, laboratory safety situation management needs to combine comprehensive data monitoring with a timely response mechanism to ensure the safety of the laboratory.

[0003] Laboratory safety management faces many challenges, including the lack of integration between laboratory access management and environmental parameter monitoring, making it difficult to comprehensively control the safety situation; insufficient management of the usage status and lifespan of experimental equipment, which may lead to accidental failures. In addition, the risk identification and early warning systems for hazardous chemicals and personnel operation behaviors lag behind, and potential safety hazards may not be detected in time. The untimely prediction and response to these hazards may also lead to major accidents. Therefore, it is necessary to strengthen safety monitoring, improve the early warning mechanism, and enhance the response efficiency to ensure the safe operation of the laboratory environment. Summary of the Invention

[0004] To solve the problems in the background art, the present invention provides a laboratory safety intelligent monitoring and management method and system based on Internet of Things and data-driven technology, which realizes the comprehensive monitoring and intelligent management of the laboratory environment, equipment, hazardous chemicals and personnel through multi-dimensional data collection and analysis, and further improves the safety and efficiency of the laboratory.

[0005] The present invention is realized through the following technical solutions:

[0006] I. A laboratory environment monitoring module of a laboratory safety intelligent monitoring and management system based on Internet of Things and data-driven technology, which collects the environmental data of the laboratory.

[0007] An experimental personnel behavior monitoring module, which collects the data of experimental personnel entering and leaving the door, the voice data of experimental personnel, and the experimental operation data.

[0008] An experimental equipment status monitoring module, which collects the operation log data of experimental equipment.

[0009] A hazardous chemicals management monitoring module, which collects the hazardous chemicals data, the hazardous chemicals storage environment data, and the hazardous chemicals operation data.

[0010] The intelligent early warning and assessment safety hazard module performs anomaly detection on the environmental data received from the laboratory environment monitoring module to obtain anomaly detection results. It processes the in-and-out data of experimental personnel, the voice data of experimental personnel, and the experimental operation data received from the experimental personnel behavior monitoring module respectively to obtain the personnel information comparison result, the result of whether there is an anomaly in the emotions of experimental personnel, and the result of whether there is an anomaly in the experimental operation data. It processes the experimental equipment operation log data received from the experimental equipment status monitoring module to obtain the abnormal operation mode, health status, and service life of the experimental equipment. The obtained abnormal operation mode is then transmitted to the experimental equipment status monitoring module for processing. It processes the hazardous chemical data and the hazardous chemical storage environment data received from the hazardous chemical management monitoring module together to obtain the result of whether there is a hazard in the hazardous chemicals, and stores the hazardous chemical operation data received from the hazardous chemical management monitoring module.

[0011] The emergency response remote collaborative processing module performs alarm processing on the anomaly detection result, personnel information comparison result, the result of whether there is an anomaly in the emotions of experimental personnel, the result of whether there is an anomaly in the experimental operation data, and the result of whether there is a hazard in the hazardous chemicals received from the intelligent early warning and assessment safety hazard module respectively, and processes the health status and service life of the received experimental equipment to obtain a resource allocation plan.

[0012] The laboratory environment monitoring module includes a temperature and humidity sensor 1, a thermal imaging sensor, a smoke sensor, and a gas sensor 1 installed in the experimental operation environment; the experimental personnel behavior monitoring module includes a camera acquisition module and a voice acquisition module; the camera acquisition module includes cameras installed at the laboratory entrance and in the laboratory operation environment; the experimental equipment status monitoring module includes an experimental equipment operation log module and an anomaly processing unit; the hazardous chemical management monitoring module includes an electronic tag module, a hazardous chemical storage acquisition module, and a hazardous chemical operation module; the intelligent early warning and assessment safety hazard module includes a prediction unit based on a time series prediction attention mechanism, a personnel information processing unit, a voice processing unit, an anomaly detection unit, and a storage unit; the emergency response remote collaborative processing module includes a resource allocation unit, an emergency alarm unit, and a collaborative processing unit.

[0013] The temperature and humidity sensor 1, thermal imaging sensor, smoke sensor, and gas sensor 1 transmit the detected environmental data to the anomaly detection unit for processing to obtain an anomaly detection result, which is then transmitted to the emergency alarm unit and the collaborative processing unit in sequence; the camera installed at the entrance of the laboratory transmits the data of the experimenters entering and leaving the door collected to the personnel information processing unit for processing to obtain a personnel information comparison result, which is then transmitted to the emergency alarm unit and the collaborative processing unit in sequence. The camera installed in the experimental operation environment transmits the experimental operation data collected to the prediction unit for processing to obtain whether there is an anomaly result in the experimental operation data, and whether there is an anomaly result in the experimental operation data is then transmitted to the emergency alarm unit and the collaborative processing unit in sequence; the voice collection module transmits the voice data of the experimenters collected to the voice processing unit for processing to obtain whether there is an anomaly result in the emotions of the experimenters, and whether there is an anomaly result in the emotions of the experimenters is then transmitted to the emergency alarm unit and the collaborative processing unit in sequence; the experimental equipment operation log module transmits the experimental equipment operation log data collected to the prediction unit for processing to obtain the abnormal operation mode, health status, and service life of the experimental equipment respectively. The abnormal operation mode is transmitted to the anomaly processing unit, and the health status and service life are transmitted to the resource allocation unit for processing to obtain a resource allocation plan; the electronic tag module includes a number of electronic tags, and each hazardous chemical is pasted with an electronic tag, and each electronic tag records the hazardous chemical data; the hazardous chemical storage collection module includes a temperature and humidity sensor 2 and a gas concentration sensor 2 installed in the hazardous chemical storage environment, and the temperature and humidity sensor 2 and the gas concentration sensor 2 collect the hazardous chemical storage environment data; the hazardous chemical data and the hazardous chemical storage environment data are transmitted to the prediction unit for processing to obtain whether there is a danger result for the hazardous chemical, and whether there is a danger result for the hazardous chemical is then transmitted to the emergency alarm unit and the collaborative processing unit in sequence; the hazardous chemical operation module records the hazardous chemical operation data and transmits it to the storage unit.

[0014] The personnel information processing unit uses a convolutional neural network, the voice processing unit uses a gated recurrent unit network, the anomaly detection unit uses a generative adversarial network, and the collaborative processing unit uses a long short-term memory network; the resource allocation unit uses a deep reinforcement learning network, the collaborative processing unit uses a long short-term memory network, the emergency alarm unit uses an alarm sensor, and the storage unit uses a MySQL database.

[0015] The prediction unit based on the time series prediction attention mechanism includes a temporal convolutional module, an encoder module, a time series prediction attention mechanism module, and a decoder module; the temporal convolutional module receives the input data, processes it, and then inputs it into the encoder module for processing. After the encoder module finishes processing, it is input into the time series prediction attention mechanism module for processing. The time series prediction attention mechanism module inputs the processed result into the decoder module for processing, and the result obtained by the decoder module's processing is used as the output of the prediction unit based on the time series prediction attention mechanism.

[0016] The temporal convolutional module uses a temporal convolutional network.

[0017] Both the encoder and the decoder use gated recurrent unit networks.

[0018] The time series prediction attention mechanism module is set according to the following formula:

[0019] Output final=softmax(A CoT )·Conv v (Q' h )+Conv k (Q' h )

[0020] Among them, Output final represents the output of the time series prediction attention mechanism module, softmax() represents converting a real number vector into a probability distribution, A CoT represents the dynamic attention weight, Conv v represents the key convolution, Q h ' represents the result after processing and enhancing the input query feature, Conv k represents the value convolution.

[0021] A CoT and Q h ' in the time series prediction attention mechanism module are set according to the following formula:

[0022] A CoT =mean(Conv att ([k1,Q' h ),dim=2)

[0023] Q' h =CoordAtt(A h V h )·σ(Conv h (x h ))·σ(Conv w (x w ))

[0024]

[0025] Q h = reshape(Q, B, T, h, D / h), K h = reshape(K, B, T, h, D / h), V h = reshape(V, B, T, h, D / h)

[0026] Q = XW Q , K = XW K , V = XW V

[0027] Among them, mean represents the average operation, Conv att represents convolutional attention, k1 represents the key representation obtained by convolution, [k1, Q″ h represents the concatenation of the key and query vectors, dim = 2 represents the second dimension of the tensor, CoordAtt represents the spatial-aware attention mechanism, A h represents the attention weight in the horizontal direction, V h represents the value vector obtained after the reshape() operation, σ represents the Sigmoid activation function, Conv h represents horizontal convolution, x h represents horizontal features, Conv w represents vertical convolution, x w represents vertical features, Q h represents the query vector obtained after the reshape() operation, represents the transpose of K h K h represents the key vector obtained after the reshape() operation, D is the feature dimension of each time step, h represents the number of heads of the attention mechanism, reshape() represents the operation of changing the tensor shape, Q represents the query vector, K represents the key vector, V represents the value vector, B represents the batch size, T represents the sequence length in the input time series data, X represents the input tensor, W Q 、W K and W V respectively represent three linear projection matrices.

[0028] II. A Laboratory Safety Intelligent Monitoring and Management Method Based on Internet of Things and Data-Driven Technology

[0029] The laboratory safety intelligent monitoring and management method conducts monitoring and management according to five methods: laboratory environment monitoring method, experimental personnel behavior monitoring method, experimental equipment status monitoring method, hazardous chemical management monitoring method, and emergency response remote collaborative processing method.

[0030] The described laboratory environment monitoring method is that the laboratory environment monitoring module transmits the collected environmental data to the anomaly detection unit for anomaly detection to obtain the anomaly detection result, and determines whether to transmit it to the emergency alarm unit for alarm processing according to the anomaly detection result. The emergency alarm unit then transmits the received result to the collaborative processing unit.

[0031] The described experimental personnel behavior monitoring method is to process the collected data of experimental personnel entering and leaving, experimental operation data, and experimental personnel voice data through the camera acquisition module and the voice acquisition module respectively, and determine whether to perform alarm processing according to the processing results.

[0032] The described experimental equipment status monitoring method is to monitor the status of experimental equipment by collecting data through the experimental equipment operation log module.

[0033] The described hazardous chemical management monitoring method is to monitor the hazardous chemical management by collecting data through the electronic tag module, the hazardous chemical storage acquisition module, and the hazardous chemical operation module.

[0034] The described emergency response remote collaborative processing method is to respectively receive the data transmitted by the intelligent early warning and assessment of potential safety hazards module, the laboratory environment monitoring module, and the hazardous chemical management monitoring module, and perform emergency response remote collaborative processing according to the received data.

[0035] The described experimental personnel behavior monitoring method specifically includes the following steps:

[0036] S1. The camera at the entrance of the laboratory transmits the collected data of experimental personnel entering and leaving to the personnel information processing unit for processing to obtain the personnel information comparison result, and determines whether to transmit it to the emergency alarm unit for alarm processing according to the personnel information comparison result. The emergency alarm unit then transmits the received result to the collaborative processing unit;

[0037] S2. The camera in the laboratory operation environment transmits the collected experimental operation data to the prediction unit for prediction to obtain whether there is an abnormal result in the experimental operation data, and determines whether to transmit it to the emergency alarm unit for alarm processing according to whether there is an abnormal result in the experimental operation data. The emergency alarm unit then transmits the received result to the collaborative processing unit.

[0038] S3. The voice acquisition module transmits the collected experimental personnel voice data to the voice processing unit for processing to obtain whether there is an abnormal result in the experimental personnel's emotion, and determines whether to transmit it to the emergency alarm unit for alarm processing according to whether there is an abnormal result in the experimental personnel's emotion. The emergency alarm unit then transmits the received result to the collaborative processing unit.

[0039] The described experimental equipment status monitoring method is specifically as follows:

[0040] The experimental equipment operation log module transmits the collected experimental equipment operation log data to the prediction unit for processing, respectively obtaining the abnormal operation mode of the experimental equipment, the health status and service life of the experimental equipment. The obtained abnormal operation mode of the experimental equipment is transmitted to the abnormal handling unit, and the abnormal handling unit performs corresponding abnormal handling according to the abnormal operation mode. The obtained health status and service life are transmitted to the resource allocation unit.

[0041] The hazardous chemical management and monitoring method specifically includes the following steps:

[0042] G1. The electronic tag on the hazardous chemical records the hazardous chemical data and the hazardous chemical storage environment data collected by the hazardous chemical storage acquisition module, and transmits them together to the prediction unit for processing to obtain the result of whether the hazardous chemical is dangerous. According to the result of whether the hazardous chemical is dangerous, it is judged whether to transmit it to the emergency alarm unit for alarm processing, and the emergency alarm unit transmits the received result to the collaborative processing unit again.

[0043] G2. The hazardous chemical operation module records the hazardous chemical operation data and transmits it to the storage unit. The records stored in the storage unit are used to trace and query the use and transfer situation of the hazardous chemical.

[0044] The emergency response remote collaborative processing method specifically includes the following steps:

[0045] K1. The resource allocation unit performs resource allocation processing on the received health status and service life of the experimental equipment, obtains an optimized resource allocation plan, and adjusts the operation layout of the experimental equipment according to the resource allocation plan.

[0046] K2. The emergency alarm unit respectively receives the abnormal detection result, the personnel information comparison result, the result of whether there is an abnormality in the experimenter's emotion, the result of whether there is an abnormality in the experimental operation data, and the result of whether the hazardous chemical is dangerous, and judges whether to perform corresponding alarm processing according to the results.

[0047] K3. When alarm processing is performed, the collaborative processing unit performs collaborative processing on the different results transmitted by the emergency alarm unit to obtain a collaborative processing result, and performs emergency processing according to the collaborative processing result.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. The present invention comprehensively considers various factors that are prone to experimental accidents in the laboratory, conducts intelligent monitoring and analysis on experimental operators, experimental equipment, laboratory environment, hazardous chemicals, etc. When any slight abnormal behavior occurs in the laboratory, the method of the present invention can give early warnings for different abnormal situations to ensure the effective supervision of laboratory safety risks.

[0050] 2. Based on the Internet of Things and data-driven technologies, six modules, namely the laboratory environment monitoring module, the experimental personnel behavior monitoring module, the experimental equipment status monitoring module, the hazardous chemical management monitoring module, the intelligent early warning and safety hazard assessment module, and the emergency response remote collaborative processing module, are integrated to monitor various factors in the laboratory and manage the monitoring results accordingly, effectively improving the safety and emergency response capabilities of the laboratory.

[0051] 3. The present invention proposes a new prediction unit adopting a time series prediction attention mechanism module, achieving ideal effects in predicting various types of data and significantly improving the prediction accuracy. Brief Description of the Drawings

[0052] Figure 1 It is a specific module diagram of the system of the present invention.

[0053] Figure 2 It is a specific module diagram of the laboratory environment monitoring module of the present invention.

[0054] Figure 3 It is a specific module diagram of the experimental personnel behavior monitoring module of the present invention.

[0055] Figure 4 It is a specific module diagram of the experimental equipment status monitoring module of the present invention.

[0056] Figure 5 It is a specific module diagram of the hazardous chemical management monitoring module of the present invention.

[0057] Figure 6 It is a specific module diagram of the intelligent early warning and safety hazard assessment module of the present invention.

[0058] Figure 7 It is a specific module diagram of the emergency response remote collaborative processing module of the present invention. Detailed Embodiment

[0059] The following provides a more detailed description of the present invention in conjunction with the drawings and embodiments. However, the present invention is not limited thereto. For those of ordinary skill in the art in this technical field, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also considered within the protection scope of the present invention. The content not described in detail in this specification belongs to the prior art well-known to those skilled in the art.

[0060] The embodiments of the present invention are carried out in the school's chemistry laboratory. Hazardous chemicals are stored in the storage warehouse in the chemistry laboratory, and the experimental personnel are the operators in the chemistry laboratory. The experimental personnel operate on the experimental supplies such as hazardous chemicals in the laboratory.

[0061] As Figure 1 shown, the laboratory safety intelligent monitoring and management system in the embodiments of the present invention includes:

[0062] The laboratory environmental monitoring module collects the environmental data of the laboratory.

[0063] The experimental personnel behavior monitoring module collects the data of experimental personnel entering and leaving the door, the voice data of experimental personnel, and the experimental operation data.

[0064] The experimental equipment status monitoring module collects the operation log data of experimental equipment.

[0065] The hazardous chemical management monitoring module collects the hazardous chemical data, the storage environment data of hazardous chemicals, and the operation data of hazardous chemicals.

[0066] The intelligent early warning and assessment of potential safety hazards module performs anomaly detection on the environmental data received from the laboratory environmental monitoring module to obtain the anomaly detection results, processes the data of experimental personnel entering and leaving the door received from the experimental personnel behavior monitoring module to obtain the personnel information comparison results, processes the voice data of experimental personnel received from the experimental personnel behavior monitoring module to obtain whether there is an anomaly in the emotions of experimental personnel, processes the experimental operation data received from the experimental personnel behavior monitoring module to obtain whether there is an anomaly in the experimental operation data, processes the operation log data of experimental equipment received from the experimental equipment status monitoring module to obtain the abnormal operation mode, health status, and service life of the experimental equipment respectively. The obtained abnormal operation mode is then transmitted to the experimental equipment status monitoring module for processing. Processes the hazardous chemical data and the storage environment data of hazardous chemicals received from the hazardous chemical management monitoring module together to obtain whether there is a danger in hazardous chemicals, and stores the operation data of hazardous chemicals received from the hazardous chemical management monitoring module.

[0067] The emergency response remote collaborative processing module performs alarm processing on the received anomaly detection results, personnel information comparison results, whether there is an anomaly in the emotions of experimental personnel, whether there is an anomaly in the experimental operation data, and whether there is a danger in hazardous chemicals respectively, and processes the health status and service life of the received experimental equipment to obtain a resource allocation plan.

[0068] The experimental operation data and the operation data of hazardous chemicals are two uncorrelated data.

[0069] The environmental data of the laboratory is the environmental data of the open space in the laboratory, and the storage environment data of hazardous chemicals is the storage environment data of hazardous chemicals stored in the warehouse.

[0070] Such as Figure 2 、 Figure 3 、 Figure 4 、 Figure 5 、 Figure 6 and Figure 7As shown in the figure, the laboratory environment monitoring module includes a temperature and humidity sensor 1, a thermal imaging sensor, a smoke sensor, and a gas sensor 1 installed in the experimental operation environment; the experimental personnel behavior monitoring module includes a camera acquisition module and a voice acquisition module; the camera acquisition module includes cameras installed at the laboratory entrance and in the laboratory operation environment; the experimental equipment status monitoring module includes an experimental equipment operation log module and an exception handling unit; the hazardous chemical management monitoring module includes an electronic tag module, a hazardous chemical storage acquisition module, and a hazardous chemical operation module; the intelligent early warning and safety hazard assessment module includes a prediction unit based on a time series prediction attention mechanism, a personnel information processing unit, a voice processing unit, an anomaly detection unit, and a storage unit; the emergency response remote collaborative processing module includes a resource allocation unit, an emergency alarm unit, and a collaborative processing unit.

[0071] The temperature and humidity sensor 1, the thermal imaging sensor, the smoke sensor, and the gas sensor 1 transmit the detected environmental data to the anomaly detection unit for processing to obtain an anomaly detection result, and the anomaly detection result is then transmitted to the emergency alarm unit and the collaborative processing unit in sequence; the camera installed at the laboratory entrance transmits the collected data of the experimental personnel entering and leaving the laboratory to the personnel information processing unit for processing to obtain a personnel information comparison result, and the personnel information comparison result is then transmitted to the emergency alarm unit and the collaborative processing unit in sequence. The camera installed in the laboratory operation environment transmits the collected experimental operation data to the prediction unit for processing to obtain whether there is an anomaly result in the experimental operation data, and whether there is an anomaly result in the experimental operation data is then transmitted to the emergency alarm unit and the collaborative processing unit in sequence; the voice acquisition module transmits the collected voice data of the experimental personnel to the voice processing unit for processing to obtain whether there is an anomaly result in the emotions of the experimental personnel, and whether there is an anomaly result in the emotions of the experimental personnel is then transmitted to the emergency alarm unit and the collaborative processing unit in sequence; the experimental equipment operation log module transmits the collected experimental equipment operation log data to the prediction unit for processing to obtain the abnormal operation mode, health status, and service life of the experimental equipment respectively. The abnormal operation mode is transmitted to the exception handling unit, and the health status and service life of the experimental equipment are transmitted to the resource allocation unit for processing to obtain a resource allocation plan; the electronic tag module includes several electronic tags, and each hazardous chemical is respectively attached with an electronic tag, and each electronic tag records the hazardous chemical data; the hazardous chemical storage acquisition module includes a temperature and humidity sensor 2 and a gas concentration sensor 2 installed in the hazardous chemical storage environment, and the temperature and humidity sensor 2 and the gas concentration sensor 2 collect the hazardous chemical storage environment data; the hazardous chemical data and the hazardous chemical storage environment data are transmitted to the prediction unit for processing to obtain whether there is a danger result in the hazardous chemical, and whether there is a danger result in the hazardous chemical is then transmitted to the emergency alarm unit and the collaborative processing unit in sequence; the hazardous chemical operation module records the hazardous chemical operation data and transmits it to the storage unit.

[0072] The environmental data includes laboratory temperature, laboratory humidity, temperature distribution of experimental equipment and areas, smoke concentration, and harmful gas concentration; the data of experimenters entering and leaving the door includes the time when the experimenter enters the door, the face information when entering the door, the time when the experimenter leaves the door, and the face information when leaving the door; the experimental operation data includes the operation actions and behavior patterns of the experimenter and the characteristics of experimental items; the operation log data of experimental equipment includes the power-on and power-off records of the equipment, running time, working status, fault records, maintenance and repair conditions, equipment parameter settings, operator information, etc.; the hazardous chemical data includes the storage location, quantity, usage, and expiration date of hazardous chemicals; the storage environment data of hazardous chemicals includes storage temperature, storage humidity, and storage gas concentration; the operation data of hazardous chemicals includes the usage and transfer of hazardous chemicals.

[0073] The personnel information processing unit uses a convolutional neural network, the speech processing unit uses a gated recurrent unit, the anomaly detection unit uses a generative adversarial network, and the collaborative processing unit uses a long short-term memory network; the resource allocation unit uses a deep reinforcement learning network, the collaborative processing unit uses a long short-term memory network, the emergency alarm unit uses an alarm sensor, and the storage unit uses a MySQL database.

[0074] The prediction unit based on the time series prediction attention mechanism includes a temporal convolutional module, an encoder module, a time series prediction attention mechanism module, and a decoder module; the temporal convolutional module receives the input data, processes it, and then inputs it to the encoder module for processing. After the encoder module finishes processing, it inputs it to the time series prediction attention mechanism module for processing. The time series prediction attention mechanism module inputs the processed result to the decoder module for processing, and the result obtained by the decoder module's processing is used as the output of the prediction unit based on the time series prediction attention mechanism.

[0075] The temporal convolutional module uses a temporal convolutional network (TCN);

[0076] Both the encoder and the decoder use a gated recurrent unit network (GRU).

[0077] The time series prediction attention mechanism module is set according to the following formula:

[0078] Output final=softmax(A CoT )·Conv v (Q' h )+Conv k (Q' h )

[0079] Among them, Output final represents the output of the time series prediction attention mechanism module, softmax() represents converting a real number vector into a probability distribution, A CoTrepresents the dynamic attention weight, Conv v represents the key convolution, Q h ' represents the result after processing and enhancing the input query features, Conv k represents the value convolution.

[0080] A in the time series prediction attention mechanism module CoT and Q h ' are set according to the following formula:

[0081] A CoT = mean(Conv att ([k1, Q' h ), dim = 2)

[0082] Q' h = CoordAtt(A h V h ) · σ(Conv h (x h )) · σ(Conv w (x w ))

[0083]

[0084] Q h = reshape(Q, B, T, h, D / h), K h = reshape(K, B, T, h, D / h), V h = reshape(V, B, T, h, D / h)

[0085] Q = XW Q , K = XW K , V = XW V

[0086] Among them, mean represents the average operation, Conv att represents the convolutional attention, k1 represents the key representation obtained by convolution, [k1, Q″ h represents the concatenation of the key and the query vector, dim = 2 represents the second dimension of the tensor, CoordAtt represents the spatial-aware attention mechanism, A h represents the attention weight in the horizontal direction, V h represents the value vector obtained after the reshape() operation, σ represents the Sigmoid activation function, Conv h represents the horizontal convolution, x h represents the horizontal feature, Conv w represents the vertical convolution, x w represents the vertical feature, Q hDenotes the query vector obtained after the reshape() operation. Denotes K h The transpose of, K h Denotes the key vector obtained after the reshape() operation, D is the feature dimension of each time step, h represents the number of heads of the attention mechanism, reshape() represents the operation of changing the tensor shape, Q is denoted as the query vector, K is denoted as the key vector, V is denoted as the value vector, B represents the batch size, T represents the sequence length in the input time series data, X represents the input tensor, W Q 、W K And W V Respectively represent three linear projection matrices.

[0087] The laboratory safety intelligent monitoring and management method of the embodiments of the present invention is monitored and managed according to five methods: laboratory environment monitoring method, experimental personnel behavior monitoring method, experimental equipment status monitoring method, hazardous chemical management monitoring method, and emergency response remote collaborative processing method.

[0088] The laboratory environment monitoring method is that the laboratory environment monitoring module transmits the collected environmental data to the anomaly detection unit for anomaly detection to obtain an anomaly detection result, and judges whether to transmit it to the emergency alarm unit for alarm processing according to the anomaly detection result, and the emergency alarm unit transmits the received result to the collaborative processing unit again.

[0089] In specific implementation, the steps of the laboratory environment monitoring method are as follows:

[0090] When the camera at the laboratory entrance detects that an experimental personnel enters the laboratory, the temperature and humidity sensor 1, thermal imaging sensor, smoke sensor, and gas sensor 1 transmit the collected environmental data to the anomaly detection unit for anomaly detection to obtain an anomaly detection result: if the anomaly detection result is normal, it means the laboratory environment is safe; if the anomaly detection result is abnormal, experimental environment dangerous data is generated and the experimental environment dangerous data is transmitted to the emergency alarm unit for alarm processing, and the emergency alarm unit transmits the received data to the collaborative processing unit again.

[0091] When the camera at the laboratory entrance does not detect that an experimental personnel enters the laboratory or detects that all experimental personnel leave the laboratory, the temperature and humidity sensor 1, thermal imaging sensor, smoke sensor, and gas sensor 1 stop working.

[0092] The experimental personnel behavior monitoring method is to process the collected experimental personnel entry and exit data, experimental operation data, and experimental personnel voice data through the camera acquisition module and voice acquisition module respectively, and judge whether to perform alarm processing according to the processing results.

[0093] S1. The camera at the entrance of the laboratory will transmit the data of the experimenters entering and leaving the laboratory to the personnel information processing unit for processing to obtain the personnel information comparison result. According to the personnel information comparison result, it is judged whether to transmit it to the emergency alarm unit for alarm processing, and the emergency alarm unit will transmit the received result to the collaborative processing unit again.

[0094] The personnel information comparison result includes whether the face matching at the entrance is successful, whether the number of experimenters exceeds the limit, and whether the personnel retention time exceeds the limit.

[0095] S2. The camera in the laboratory operation environment will transmit the collected experimental operation data to the prediction unit for prediction to obtain whether there is an abnormal result in the experimental operation data. According to whether there is an abnormal result in the experimental operation data, it is judged whether to transmit it to the emergency alarm unit for alarm processing, and the emergency alarm unit will transmit the received result to the collaborative processing unit again.

[0096] Whether there is an abnormal result in the experimental operation data includes whether the operation actions are standard, whether the behavior patterns are abnormal, and whether the experimental items are abnormal.

[0097] S3. The voice acquisition module will transmit the collected voice data of the experimenters to the voice processing unit for processing to obtain whether there is an abnormal result in the emotions of the experimenters. According to whether there is an abnormal result in the emotions of the experimenters, it is judged whether to transmit it to the emergency alarm unit for alarm processing, and the emergency alarm unit will transmit the received result to the collaborative processing unit again.

[0098] In specific implementation, the steps of the experimenter behavior monitoring method are as follows:

[0099] D1. The camera at the entrance of the laboratory will transmit the data of the experimenters entering and leaving the laboratory to the personnel information processing unit. The personnel information processing unit will process the data of the experimenters entering and leaving the laboratory in real time to obtain whether the face matching at the entrance is successful in real time, whether the number of experimenters exceeds the limit, and whether the personnel retention time exceeds the limit respectively; if the face matching at the entrance is successful, the recognized personnel are allowed to enter the laboratory; if the face matching at the entrance is not successful, the recognized personnel are not allowed to enter the laboratory; if the number of experimenters exceeds the limit, the data of the number of experimenters exceeding the limit is generated and transmitted to the emergency alarm unit for alarm processing, and the emergency alarm unit will transmit the received data to the collaborative processing unit again; if the number of experimenters does not exceed the limit, personnel can continue to be allowed to enter; if the personnel retention time exceeds the limit, the data of exceeding the safety period is generated and transmitted to the emergency alarm unit for alarm processing, and the emergency alarm unit will transmit the received result to the collaborative processing unit again; if the personnel retention time does not exceed the limit, personnel can continue to be allowed to stay in the laboratory.

[0100] D2. The camera in the laboratory operation environment transmits the experimentally operated data collected in real time to the prediction unit based on the time series prediction attention mechanism. The prediction unit processes the experimentally operated data of the experimental personnel collected in real time to obtain whether the operation actions are standardized, whether the behavior patterns are abnormal, and whether the experimental items are abnormal respectively. If one or more of the operation actions, behavior patterns, and experimental items are abnormal, the operation abnormal data of the experimental personnel is generated and transmitted to the emergency alarm unit for alarm processing. The emergency alarm unit then transmits the received data to the collaborative processing unit. If none of the operation actions, behavior patterns, and experimental items are abnormal, it indicates that the experimental personnel's operation is normal.

[0101] D3. The voice collection module transmits the experimental personnel voice data collected in real time to the voice processing unit. The voice processing unit processes the experimental personnel voice data to obtain the result of whether the experimental personnel's emotion is abnormal: If the experimental personnel's emotion is abnormal, the emergency event data is generated and transmitted to the emergency alarm unit for alarm processing. The emergency alarm unit then transmits the received data to the collaborative processing unit. If the experimental personnel's emotion is not abnormal, it indicates that no emergency event has occurred.

[0102] The experimental equipment status monitoring method is to monitor the experimental equipment status by collecting data through the experimental equipment operation log module.

[0103] The experimental equipment operation log module transmits the collected experimental equipment operation log data to the prediction unit for processing to obtain the abnormal operation mode, health status, and service life of the experimental equipment respectively. The obtained abnormal operation mode of the experimental equipment is transmitted to the abnormal processing unit, and the abnormal processing unit performs corresponding abnormal processing according to the abnormal operation mode. The obtained health status and service life are transmitted to the resource allocation unit.

[0104] The abnormal operation modes include the experimental equipment running in a high-temperature mode and the experimental equipment running in a short-circuit mode, etc. The abnormal processing includes shutting down the experimental equipment running in the high-temperature mode and repairing the experimental equipment running in the short-circuit mode, etc.

[0105] The hazardous chemical management monitoring method is to monitor the hazardous chemical management by collecting data through the electronic tag module, hazardous chemical storage collection module, and hazardous chemical operation module.

[0106] G1. The electronic tag on the hazardous chemical records the hazardous chemical data and transmits it together with the hazardous chemical storage environment data collected by the hazardous chemical storage collection module to the prediction unit for processing to obtain the result of whether the hazardous chemical is dangerous. According to the result of whether the hazardous chemical is dangerous, it is judged whether to transmit it to the emergency alarm unit for alarm processing. The emergency alarm unit then transmits the received result to the collaborative processing unit.

[0107] G2. The hazardous chemical operation module records the hazardous chemical operation data and transmits it to the storage unit. The records stored in the storage unit are used to trace and query the usage and transfer of hazardous chemicals.

[0108] The usage and transfer of hazardous chemicals include the information of incoming and outgoing of hazardous chemicals, etc.

[0109] In specific implementation, the steps of the hazardous chemical management and monitoring method are as follows:

[0110] H1. The electronic tag (such as RFID tag) on the hazardous chemical records the hazardous chemical data in real time and transmits it together with the hazardous chemical storage environment data collected by the hazardous chemical storage acquisition module to the prediction unit for prediction processing. It is predicted whether there is a leakage of hazardous chemicals, whether the hazardous chemicals are expired, and whether the storage of hazardous chemicals is appropriate. If one or more of the following occur: there is a leakage of hazardous chemicals, the hazardous chemicals are expired, and the storage of hazardous chemicals is inappropriate, hazardous chemical danger data is generated and transmitted to the emergency alarm unit for alarm processing. The emergency alarm unit then transmits the received result to the collaborative processing unit. If none of the following occur: there is a leakage of hazardous chemicals, the hazardous chemicals are expired, and the storage of hazardous chemicals, it means that the hazardous chemicals are not dangerous.

[0111] H2. The hazardous chemical operation module records the hazardous chemical operation data and transmits it in real time to the MySQL database for storage. The records stored in the MySQL database are used for subsequent tracing and querying the usage and transfer of hazardous chemicals.

[0112] The humidity and temperature sensor two and the gas concentration sensor two of the hazardous chemical storage acquisition module are always running.

[0113] The emergency response remote collaborative processing method is to receive the data transmitted by the intelligent early warning and assessment of potential safety hazards module, the laboratory environment monitoring module, and the hazardous chemical management and monitoring module respectively, and perform emergency response remote collaborative processing according to the received data.

[0114] K1. The resource allocation unit processes the health status and service life of the received experimental equipment for resource allocation, obtains an optimized resource allocation plan, and adjusts the operation layout of the experimental equipment according to the resource allocation plan. The operation layout includes the daily operation duration of each experimental equipment, when each equipment is turned on and off, etc.

[0115] K2. The emergency alarm unit receives the abnormal detection result, the personnel information comparison result, whether there is an abnormal result of the experimental personnel's emotion, whether there is an abnormal result of the experimental operation data, and whether there is a dangerous result of the hazardous chemicals respectively, and judges whether to perform corresponding alarm processing according to the results.

[0116] If one or more of the results of anomaly detection, personnel information comparison, whether there is an anomaly in the experimenter's mood, whether there is an anomaly in the experimental operation data, and whether there is a danger in hazardous chemicals are abnormal, mismatched, or dangerous, an alarm is issued.

[0117] K3. When alarm handling is performed, the collaborative processing unit collaboratively processes the different received results to obtain a collaborative processing result, and performs emergency handling according to the collaborative processing result.

[0118] The different results include that the anomaly detection result is abnormal, the personnel information comparison result is mismatched, there is an anomaly in whether there is an anomaly in the experimenter's mood, there is an anomaly in whether there is an anomaly in the experimental operation data, and there is a danger in whether there is a danger in hazardous chemicals.

[0119] The collaborative processing unit will simultaneously receive multiple types of data. The multiple types of data are regarded as multiple alarm events. After processing the multiple types of data, the collaborative processing unit will give an emergency handling plan, including classifying the received data by severity level, notifying relevant personnel, and activating the emergency handling mechanism, etc., and arranging personnel to process them in sequence according to the severity level or allocating personnel to handle different alarm events. The emergency handling includes triggering a fire extinguishing device, closing a gas valve, and starting an emergency safety cabinet, etc.

[0120] The present invention uses the newly proposed prediction unit adopting a time series prediction attention mechanism module, which improves the prediction accuracy. The present invention also comprehensively considers all factors that are prone to experimental accidents in the laboratory, and conducts intelligent monitoring and analysis on experimental operators, experimental equipment, laboratory environment, and hazardous chemicals. When any minor abnormal behavior occurs in the laboratory, this method can give early warnings for different abnormal situations to ensure the effective supervision of laboratory safety risks.

Claims

1. A laboratory safety intelligent monitoring and management system based on the Internet of Things and data-driven technology, characterized in that: A laboratory environment monitoring module for collecting environmental data of the laboratory; An experimental personnel behavior monitoring module for collecting data on the entry and exit of experimental personnel, voice data of experimental personnel, and experimental operation data; An experimental equipment status monitoring module for collecting operation log data of experimental equipment; A hazardous chemical management monitoring module for collecting hazardous chemical data, hazardous chemical storage environment data, and hazardous chemical operation data; An intelligent early warning and assessment of potential safety hazards module that performs anomaly detection on the environmental data received from the laboratory environment monitoring module to obtain an anomaly detection result, processes the data on the entry and exit of experimental personnel, voice data of experimental personnel, and experimental operation data received from the experimental personnel behavior monitoring module respectively to obtain a personnel information comparison result, a result indicating whether there is an anomaly in the emotions of experimental personnel, and a result indicating whether there is an anomaly in the experimental operation data, processes the operation log data of experimental equipment received from the experimental equipment status monitoring module to obtain the abnormal operation mode, health status, and service life of the experimental equipment respectively, transmits the obtained abnormal operation mode to the experimental equipment status monitoring module for processing, processes the hazardous chemical data and hazardous chemical storage environment data received from the hazardous chemical management monitoring module together to obtain a result indicating whether there is a danger in the hazardous chemicals, and stores the hazardous chemical operation data received from the hazardous chemical management monitoring module; An emergency response remote collaborative processing module that performs alarm processing on the anomaly detection result, personnel information comparison result, result indicating whether there is an anomaly in the emotions of experimental personnel, result indicating whether there is an anomaly in the experimental operation data, and result indicating whether there is a danger in the hazardous chemicals received from the intelligent early warning and assessment of potential safety hazards module respectively, and processes the health status and service life of the received experimental equipment to obtain a resource allocation plan.

2. The laboratory safety intelligent monitoring and management system based on the Internet of Things and data-driven technology according to claim 1, characterized in that: The laboratory environment monitoring module includes a temperature and humidity sensor 1, a thermal imaging sensor, a smoke sensor, and a gas sensor 1 installed in the experimental operation environment; the experimental personnel behavior monitoring module includes a camera acquisition module and a voice acquisition module; the camera acquisition module is a camera installed at the entrance of the laboratory and in the experimental operation environment; the experimental equipment status monitoring module includes an experimental equipment operation log module and an anomaly processing unit; the hazardous chemical management monitoring module includes an electronic tag module, a hazardous chemical storage acquisition module, and a hazardous chemical operation module; the intelligent early warning and assessment of potential safety hazards module includes a prediction unit based on a time series prediction attention mechanism, a personnel information processing unit, a voice processing unit, an anomaly detection unit, and a storage unit; the emergency response remote collaborative processing module includes a resource allocation unit, an emergency alarm unit, and a collaborative processing unit; The temperature and humidity sensor 1, thermal imaging sensor, smoke sensor, and gas sensor 1 transmit the detected environmental data to the anomaly detection unit for processing to obtain an anomaly detection result, which is then sequentially transmitted to the emergency alarm unit and the collaborative processing unit; the camera installed at the entrance of the laboratory transmits the data of the experimenters entering and leaving the laboratory collected to the personnel information processing unit for processing to obtain a personnel information comparison result, which is then sequentially transmitted to the emergency alarm unit and the collaborative processing unit. The camera installed in the laboratory operation environment transmits the experimental operation data collected to the prediction unit for processing to obtain whether there is an anomaly result in the experimental operation data, and whether there is an anomaly result in the experimental operation data is then sequentially transmitted to the emergency alarm unit and the collaborative processing unit; the voice collection module transmits the voice data of the experimenters collected to the voice processing unit for processing to obtain whether there is an anomaly result in the emotions of the experimenters, and whether there is an anomaly result in the emotions of the experimenters is then sequentially transmitted to the emergency alarm unit and the collaborative processing unit; the experimental equipment operation log module transmits the experimental equipment operation log data collected to the prediction unit for processing to obtain the abnormal operation mode, health status, and service life of the experimental equipment respectively. The abnormal operation mode is transmitted to the anomaly processing unit, and the health status and service life are transmitted to the resource allocation unit for processing to obtain a resource allocation plan; the electronic tag module includes a number of electronic tags, and each hazardous chemical is pasted with an electronic tag, and each electronic tag records the hazardous chemical data; the hazardous chemical storage collection module includes a temperature and humidity sensor 2 and a gas concentration sensor 2 installed in the hazardous chemical storage environment, and the temperature and humidity sensor 2 and the gas concentration sensor 2 collect the hazardous chemical storage environment data; the hazardous chemical data and the hazardous chemical storage environment data are transmitted to the prediction unit for processing to obtain whether there is a danger result for the hazardous chemical, and whether there is a danger result for the hazardous chemical is then sequentially transmitted to the emergency alarm unit and the collaborative processing unit; the hazardous chemical operation module records the hazardous chemical operation data and transmits it to the storage unit.

3. The laboratory safety intelligent monitoring and management system based on the Internet of Things and data-driven technology according to claim 2, characterized in that: The personnel information processing unit uses a convolutional neural network, the voice processing unit uses a gated recurrent unit network, the anomaly detection unit uses a generative adversarial network, and the collaborative processing unit uses a long short-term memory network; the resource allocation unit uses a deep reinforcement learning network, the collaborative processing unit uses a long short-term memory network, the emergency alarm unit uses an alarm sensor, and the storage unit uses a MySQL database; The prediction unit based on the time series prediction attention mechanism includes a temporal convolutional module, an encoder module, a time series prediction attention mechanism module, and a decoder module; the temporal convolutional module receives the input data, processes it, and then inputs it into the encoder module for processing. After the encoder module finishes processing, it is input into the time series prediction attention mechanism module for processing. The time series prediction attention mechanism module inputs the processed result into the decoder module for processing, and the result obtained by the decoder module is used as the output of the prediction unit based on the time series prediction attention mechanism; The temporal convolutional module uses a temporal convolutional network; Both the encoder and the decoder use gated recurrent unit networks.

4. The intelligent laboratory safety monitoring and management system based on the Internet of Things and data-driven technology according to claim 3, wherein: The time series prediction attention mechanism module is set according to the following formula: Output final=softmax(A CoT )·Conv v (Q' h )+Conv k (Q' h ) Among them, Output final represents the output of the temporal prediction attention mechanism module, softmax() represents converting a real number vector into a probability distribution, and A CoT represents the dynamic attention weight, Conv v represents the key convolution, Q h ' represents the result after processing and enhancing the input query features, Conv k represents the value convolution.

5. The intelligent laboratory safety monitoring and management system based on the Internet of Things and data-driven technology according to claim 4, wherein: A in the timing prediction attention mechanism module CoT and Q h are set according to the following formula: A CoT = mean(Conv att ([k1,Q' h ), dim = 2) Q' h = CoordAtt(A h V h )·σ(Conv h (x h ))·σ(Conv w (x w )) Q h = reshape(Q, B, T, h, D / h), K h = reshape(K, B, T, h, D / h), V h = reshape(V, B, T, h, D / h) Q = XW Q , K = XW K , V = XW V Among them, mean represents the average operation, and Conv att represents convolutional attention, k1 represents the key representation obtained by convolution, and [k1, Q' h '] represents the concatenation of the key and query vectors. dim = 2 represents the second dimension of the tensor. CoordAtt represents the spatial-aware attention mechanism. A h represents the attention weight in the horizontal direction, and V h represents the value vector obtained after the reshape() operation. σ represents the Sigmoid activation function. Conv h represents horizontal convolution, and x h represents horizontal features. Conv w represents vertical convolution, and x w represents vertical features. Q h represents the query vector obtained after the reshape() operation. represents the transpose of K h , and K h represents the key vector obtained after the reshape() operation. D is the feature dimension of each time step, h represents the number of heads of the attention mechanism, reshape() represents the operation of changing the shape of the tensor, Q represents the query vector, K represents the key vector, V represents the value vector, B represents the batch size, T represents the sequence length in the input time series data, X represents the input tensor, W Q , W K and W V represent three linear projection matrices respectively.

6. The intelligent laboratory safety monitoring and management method applied to any one of claims 1-5, wherein: The intelligent laboratory safety monitoring and management method performs monitoring and management according to five methods: laboratory environment monitoring method, experimental personnel behavior monitoring method, experimental equipment status monitoring method, hazardous chemical management monitoring method, and emergency response remote collaborative processing method; The laboratory environment monitoring method is that the laboratory environment monitoring module transmits the collected environmental data to the anomaly detection unit for anomaly detection to obtain an anomaly detection result, and judges whether to transmit it to the emergency alarm unit for alarm processing according to the anomaly detection result. The emergency alarm unit transmits the received result to the collaborative processing unit again; The experimental personnel behavior monitoring method is to process the collected data of the experimental personnel entering and leaving the door, experimental operation data, and experimental personnel voice data through the camera acquisition module and the voice acquisition module respectively, and judge whether to perform alarm processing according to the processed results; The experimental equipment status monitoring method is to monitor the status of experimental equipment by collecting data through the experimental equipment operation log module; The hazardous chemical management monitoring method is to monitor the hazardous chemical management by collecting data through the electronic tag module, the hazardous chemical storage acquisition module, and the hazardous chemical operation module; The emergency response remote collaborative processing method is to receive the data transmitted by the intelligent early warning evaluation safety hazard module, the laboratory environment monitoring module, and the hazardous chemical management monitoring module respectively, and perform emergency response remote collaborative processing according to the received data.

7. The laboratory safety intelligent monitoring and management method based on the Internet of Things and data-driven technology according to claim 6, characterized in that, The experimental personnel behavior monitoring method specifically includes the following steps: S1. The camera at the entrance of the laboratory for entering transmits the collected data of the experimental personnel entering and leaving the door to the personnel information processing unit for processing to obtain a personnel information comparison result, and judges whether to transmit it to the emergency alarm unit for alarm processing according to the personnel information comparison result. The emergency alarm unit transmits the received result to the collaborative processing unit again; S2. The camera in the laboratory operation environment transmits the collected experimental operation data to the prediction unit for prediction to obtain whether there is an abnormal result in the experimental operation data. According to whether there is an abnormal result in the experimental operation data, it is judged whether to transmit it to the emergency alarm unit for alarm processing. The emergency alarm unit then transmits the received result to the collaborative processing unit; S3. The voice acquisition module transmits the collected voice data of the experimental personnel to the voice processing unit for processing to obtain whether there is an abnormal result in the emotions of the experimental personnel. According to whether there is an abnormal result in the emotions of the experimental personnel, it is judged whether to transmit it to the emergency alarm unit for alarm processing. The emergency alarm unit then transmits the received result to the collaborative processing unit.

8. The laboratory safety intelligent monitoring and management method based on the Internet of Things and data-driven technology according to claim 6, characterized in that, The specific method for monitoring the state of the experimental equipment is as follows: The experimental equipment operation log module transmits the collected experimental equipment operation log data to the prediction unit for processing to respectively obtain the abnormal operation mode of the experimental equipment, the health status and service life of the experimental equipment. The obtained abnormal operation mode of the experimental equipment is transmitted to the abnormal processing unit, and the abnormal processing unit performs corresponding abnormal processing according to the abnormal operation mode. The obtained health status and service life are transmitted to the resource allocation unit.

9. The laboratory safety intelligent monitoring and management method based on the Internet of Things and data-driven technology according to claim 6, characterized in that, The specific method for monitoring the management of hazardous chemicals includes the following steps: G1. The electronic tag on the hazardous chemical records the hazardous chemical data and the hazardous chemical storage environment data collected by the hazardous chemical storage acquisition module and transmits them to the prediction unit for processing to obtain whether there is a danger result for the hazardous chemical. According to whether there is a danger result for the hazardous chemical, it is judged whether to transmit it to the emergency alarm unit for alarm processing. The emergency alarm unit then transmits the received result to the collaborative processing unit; G2. The hazardous chemical operation module records the hazardous chemical operation data and transmits it to the storage unit. The records stored in the storage unit are used to trace and query the use and transfer of the hazardous chemical.

10. The laboratory safety intelligent monitoring and management method based on the Internet of Things and data-driven technology according to claim 6, characterized in that, The specific method for remote collaborative emergency response includes the following steps: K1. The resource allocation unit performs resource allocation processing on the received health status and service life of the experimental equipment to obtain an optimized resource allocation plan, and adjusts the operation layout of the experimental equipment according to the resource allocation plan; K2. The emergency alarm unit respectively receives the abnormal detection result, the personnel information comparison result, whether there is an abnormal result in the emotions of the experimental personnel, whether there is an abnormal result in the experimental operation data, and whether there is a danger result for the hazardous chemical, and judges whether to perform corresponding alarm processing according to the results; K3. When alarm processing is performed, the collaborative processing unit performs collaborative processing on the different results transmitted by the emergency alarm unit to obtain a collaborative processing result, and performs emergency processing according to the collaborative processing result.

Images