Laboratory shielding door control system and safety monitoring method

Abstract

The invention discloses a laboratory shielding door control system and a safety monitoring method. The laboratory shielding door control system comprises a sensor unit, a data analysis and storage unit, a control unit and a door control and safety equipment assembly. The sensor unit is used for collecting temperature, humidity, gas, radiation, shielding door state and image data of a laboratory environment, and the data analysis and storage unit predicts environment change through a machine learning model based on sensor data and performs equipment fault prediction and behavior analysis by adopting a deep learning model. The control unit is in linkage with the door control and safety equipment assembly according to the analysis result, and the opening and closing speed of the shielding door and operation of safety equipment are dynamically adjusted. Through an intelligent algorithm and self-adaptive control, automatic management of the shielding door and the safety equipment of the laboratory is realized, and the safety and the management efficiency of the laboratory are improved.

Description

Technical Field

[0001] The invention relates to laboratory safety equipment, and in particular to a laboratory shielding door control system and a safety monitoring method. Background Art

[0002] The current laboratory shielding door control system usually adopts a simple mechanical or electrical control method, which cannot monitor and respond to environmental changes in the laboratory in real time and intelligently. In addition, the existing control system cannot effectively link ventilation, fire extinguishing, alarm and other safety equipment in an emergency, resulting in safety hazards. With the complexity of laboratory environment and the increase in intelligent demand, traditional control methods can no longer meet the safety requirements of modern laboratories. Therefore, there is an urgent need for a shielding door control system that combines intelligent control, data analysis and early warning mechanism, which can monitor environmental parameters in real time, predict potential dangers, and automatically take countermeasures when abnormal situations occur. Summary of the invention

[0003] The purpose of the present invention is to provide a laboratory shielding door control system and a safety monitoring method.

[0004] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0005] The laboratory shielding door control system of the present invention comprises a sensor unit, a data analysis and storage unit, a control unit and a door control and safety equipment assembly;

[0006] The sensor unit collects temperature, humidity, air quality, radiation, personnel approach, shielding door opening and closing status, and image data near the laboratory shielding door;

[0007] The data analysis and storage unit analyzes the data collected by the sensor unit, predicts environmental changes, calculates control information and sends it to the control unit, and stores historical data;

[0008] The control unit executes the commands of the data analysis and storage unit to control the door frame and the safety equipment components;

[0009] The door control and safety equipment components control the opening and closing of the shield door, and control the operation of the fire extinguishing, ventilation and alarm equipment;

[0010] The signal output end of the sensor unit is connected to the data input end of the data analysis and storage unit, the data output end of the data analysis and storage unit is connected to the data input end of the control unit, the control signal output end of the control unit is connected to the control signal input end of the safety device component, and the safety device component is connected to the laboratory shielding door.

[0011] The sensor unit includes a temperature sensor, a humidity sensor, a gas sensor, a radiation sensor, an ultrasonic sensor, a shield door switch state sensor, and a camera; the data analysis and storage unit can predict environmental changes in the next few hours and provide early warnings through a machine learning model trained with historical data and real-time sensor data; the data analysis and storage unit adopts an adaptive fuzzy control algorithm to dynamically adjust the opening and closing speed, force, and timing according to changes in the real-time environment and door state; the data analysis and storage unit analyzes sensor and equipment data through a deep learning model, predicts possible equipment failures in advance, and automatically adjusts operating parameters; the data analysis and storage unit stores all sensor data, control logs, alarm data, and operation records as training data sources for machine learning models, and performs long-term analysis and optimization; the data analysis and storage unit identifies abnormal behaviors through a deep learning model based on sensor and camera data, automatically triggers an alarm, or sends data to the control unit to adjust the state of the door; the control unit receives the control signal of the data analysis and storage unit to control the door control and safety equipment components in a linkage manner; the safety equipment components include a shield door drive controller, a ventilation system, a fire extinguishing system, and an alarm system.

[0012] The safety monitoring method of the laboratory shielding door control system of the present invention comprises the following steps:

[0013] S1: The sensor unit collects environmental data, equipment status data, and image data near the laboratory shielding door in real time;

[0014] S2: The data analysis and storage unit receives the real-time data from the sensor unit, inputs the data into the machine learning model, and performs training based on historical data and real-time data to predict future environmental changes;

[0015] S3: Analyze the status data of sensors and equipment through deep learning models to determine whether there is a potential failure in the equipment. If an abnormality is detected, the system adjusts the operating parameters of the equipment in advance to prevent equipment failure.

[0016] S4: Based on the data from the sensor unit and the image information from the camera, a deep learning model is used to perform behavioral analysis to detect whether a person has abnormal behavior. If an abnormality is detected, an alarm is automatically triggered;

[0017] S5: The opening and closing speed, force and timing of the shield door are dynamically adjusted through the adaptive fuzzy control algorithm and real-time environmental data and door status. When the system predicts environmental anomalies or equipment failures, the control unit controls the door control and safety equipment components in a linked manner according to the instructions of the data analysis and storage unit.

[0018] S6: The data analysis and storage unit stores all sensor data, equipment status data, control logs, alarm records, and operation records in a database as a training data source for subsequent machine learning models;

[0019] S7: When the environmental parameters exceed the preset safety threshold, the data analysis and storage unit immediately sends an alarm signal to the control unit, and the control unit simultaneously links the shielding door and the safety equipment components to perform emergency processing;

[0020] S8: The data analysis and storage unit regularly performs self-inspection and maintenance on the equipment and generates inspection reports; it predicts possible equipment loss or failure through deep learning models.

[0021] The beneficial effects of the present invention are:

[0022] The present invention is a laboratory shielding door control system and safety monitoring method. Compared with the prior art, the present invention collects environmental data and equipment status data in real time through sensors, and combines machine learning and deep learning models to predict future environmental changes and equipment operating status, thereby providing early warning and adjusting control parameters. The adaptive fuzzy control algorithm in the system can automatically adjust the opening and closing speed and strength of the shielding door according to the real-time environment to ensure the safe operation of the shielding door. In addition, the system also has the function of linkage control of safety equipment. When abnormal behavior or emergency situations are detected, it can automatically control the operation of ventilation, fire extinguishing, alarm and other equipment, greatly improving the safety and automation level of the laboratory. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a system structure block diagram of the present invention. DETAILED DESCRIPTION

[0024] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0025] Example:

[0026] like Figure 1As shown: The laboratory shielded door control system of the present invention includes a sensor unit, a data analysis and storage unit, a control unit and a door control and safety equipment component; the signal output end of the sensor unit is connected to the data input end of the data analysis and storage unit, the data output end of the data analysis and storage unit is connected to the data input end of the control unit, the control signal output end of the control unit is connected to the control signal input end of the safety equipment component, and the safety equipment component is connected to the laboratory shielded door. The sensor unit is responsible for collecting environmental and equipment data, the data analysis and storage unit performs data processing and decision-making through intelligent algorithms, and the control unit executes the analysis results to control the shielded door and other safety equipment. This modular design enables the system to have a clear logical hierarchy in terms of functional partitioning, and each module can operate independently and work efficiently together.

[0027] The sensor unit collects temperature, humidity, air quality, radiation, personnel approach, shielding door opening and closing status, and image data near the laboratory shielding door;

[0028] Temperature sensor: monitors the temperature inside the laboratory to ensure that experimental equipment does not overheat.

[0029] Humidity sensor: monitors laboratory humidity and protects equipment from moisture.

[0030] Gas sensor: detect harmful gases in the laboratory to ensure air quality.

[0031] Radiation Sensor: Used in high radiation laboratories to ensure that radiation levels are within safe ranges.

[0032] Ultrasonic sensor: monitors the opening and closing of laboratory shielding doors to ensure that the shielding doors will not be closed when someone passes through.

[0033] Camera: used for image monitoring and combined with deep learning algorithms for human behavior analysis.

[0034] Covering important laboratory environmental parameters such as temperature, humidity, and air quality. By monitoring sensor feedback in real time, the system can respond quickly to environmental changes. This multi-sensor fusion technology ensures that the system has comprehensive control over the laboratory environment and provides sufficient basic data for subsequent data analysis.

[0035] The data analysis and storage unit analyzes the data collected by the sensor unit, predicts environmental changes, calculates control information and sends it to the control unit, and stores historical data; for example, when the system detects that the temperature and humidity change trends in the laboratory may cause equipment damage, it adjusts the shielding door status or starts the ventilation system in advance. The machine learning algorithm trained based on historical data can take corresponding measures in advance when potential risks are detected, and the control unit converts these analysis results into actual control commands. This demonstrates the system's ability to achieve intelligent prediction and response through artificial intelligence technology.

[0036] The control unit executes the commands of the data analysis and storage unit to control the door frame and safety equipment components; through the deep learning model, the system analyzes the equipment status and predicts the occurrence of faults. Such as opening and closing speed, opening and closing timing, etc. At the same time, if it is detected that the equipment is about to fail, the control unit can link the maintenance mechanism according to the prediction results of the deep learning model, and perform equipment maintenance or adjust operating parameters in advance. Deep learning can process complex multi-dimensional data, and through long-term equipment status learning, it can detect potential problems in advance, thereby reducing downtime. This improves the efficiency of equipment maintenance and reduces the failure rate of the system.

[0037] The door control and safety equipment components control the opening and closing of the shield door, and control the operation of the fire extinguishing, ventilation, and alarm equipment. The system can automatically close the shield door, start the ventilation or fire extinguishing equipment, and send out sound and light alarms according to the real-time data feedback from the sensor to ensure the safety of the laboratory. The adaptive fuzzy control algorithm is used to dynamically adjust the opening and closing speed, strength, and timing of the shield door to adapt to changes in the laboratory environment. This control method based on real-time feedback can ensure the safety and flexibility of the door operation, and at the same time adjust the control strategy according to different environments, thereby improving the robustness of the system.

[0038] When the system detects that the internal environment of the laboratory exceeds the safety threshold (such as harmful gas leakage, excessive radiation), the system will automatically close the shielding door and start emergency ventilation and fire extinguishing equipment to ensure the safety of personnel and equipment. The system can monitor the behavior of personnel in the laboratory. Once unauthorized entry or abnormal operation is detected, the system will automatically lock the shielding door and trigger an alarm. At the same time, the control unit can link ventilation, fire extinguishing, alarm and other equipment for comprehensive emergency processing, realizing closed-loop management from identification to response. If an unauthorized person attempts to enter, the system will automatically lock the shielding door and notify relevant personnel through the alarm system.

[0039] The sensor unit includes a temperature sensor, a humidity sensor, a gas sensor, a radiation sensor, an ultrasonic sensor, a shield door switch state sensor, and a camera; the data analysis and storage unit can predict environmental changes in the next few hours and provide early warnings through a machine learning model trained with historical data and real-time sensor data; the data analysis and storage unit adopts an adaptive fuzzy control algorithm to dynamically adjust the opening and closing speed, force, and timing according to changes in the real-time environment and door state; the data analysis and storage unit analyzes sensor and equipment data through a deep learning model, predicts possible equipment failures in advance, and automatically adjusts operating parameters; the data analysis and storage unit stores all sensor data, control logs, alarm data, and operation records as training data sources for machine learning models, and performs long-term analysis and optimization; the data analysis and storage unit identifies abnormal behaviors through a deep learning model based on sensor and camera data, automatically triggers an alarm, or sends data to the control unit to adjust the state of the door; the control unit receives the control signal of the data analysis and storage unit to control the door control and safety equipment components in a linkage manner; the safety equipment components include a shield door drive controller, a ventilation system, a fire extinguishing system, and an alarm system.

[0040] This embodiment uses advanced Internet of Things technology and intelligent control algorithms to collect data in real time through sensors and achieve rapid response in combination with model algorithm calculations. The introduction of machine learning and deep learning models enables the system to have self-learning, prediction and optimization capabilities, thereby improving the security, stability and operating efficiency of the system.

[0041] The safety monitoring method of the laboratory shielding door control system of the present invention comprises the following steps:

[0042] S1: The sensor unit collects environmental data near the laboratory shielding door in real time, including temperature, humidity, air quality, and radiation; collects equipment status data, including the opening and closing status of the shielding door and the operating status of the equipment; and collects image data at the same time, and monitors the personnel activities at the laboratory entrance and exit and the situation inside the laboratory through the camera; this step is the basis for the operation of the system. Through the installation of multiple types of sensors at various key locations in the laboratory, environmental parameters (such as temperature, humidity, air quality, etc.) and equipment status information (such as the opening and closing status of the shielding door) are obtained in real time. At the same time, the image data collected by the camera is used to monitor the laboratory entrance and exit and personnel activities. This step ensures that the system obtains a full range of laboratory dynamic information and provides basic data support for subsequent data analysis and control decisions. Multi-sensor fusion technology enables the system to make a full range of monitoring of the laboratory environment.

[0043] S2: The data analysis and storage unit receives the real-time data of the sensor unit, inputs the data into the machine learning model, and performs training based on historical data and real-time data to predict future environmental changes. The machine learning model predicts the changing trend of temperature, humidity, air quality or radiation in the next few hours and provides early warning. The formula of the machine learning model is as follows:

[0044] f(x)=w T φ(x)+b

[0045] Where: x: input vector, i.e. environmental data collected by sensors, including temperature and humidity; φ(x): mapping function that maps input data to high-dimensional feature space; w: weight vector of the model; b: bias term; f(x): output of the model, i.e. predicted value of environmental change;

[0046] Loss function:

[0047]

[0048] Where: n: the number of samples of training data; y i : The true value of the i-th sample, that is, the actual temperature and humidity; f(x i ): the predicted value of the i-th sample; ∈: the threshold of the allowable error, that is, the prediction error range allowed by the model; C: penalty coefficient, which controls the sensitivity of the model to errors beyond the ∈ range; ||w|| 2 : The norm of the weight vector, which penalizes excessive weights to improve the generalization ability of the model;

[0049] The specific process is as follows: the environmental data x collected by the sensor is input into the model for feature transformation, trained using a machine learning model, and the model parameters w, b, and C are optimized through historical data; the model is used to predict changes in temperature and humidity in the next few hours, and early warnings are triggered based on the prediction results.

[0050] After receiving the data collected by the sensor, the system uses a machine learning model to analyze and process the data. By combining historical data with real-time data, the system can predict the changing trends of environmental parameters (such as temperature, humidity, gas concentration, etc.). This machine learning-based prediction method can effectively identify potential dangers, such as environmental anomalies or conditions that are about to reach dangerous thresholds, thereby achieving early warning. This process improves the foresight and intelligence level of laboratory environmental management.

[0051] S3: Analyze the status data of sensors and devices through a deep learning model to determine whether there is a potential fault in the device. If an abnormality is detected, the system adjusts the operating parameters of the device in advance to prevent the device from failing. The deep learning model includes a forget gate, an input gate, a memory update, and an output gate.

[0052] Forget Gate:

[0053] f t =σ(W f ·[h t-1 , x t ]+b f )

[0054] Where: f t : The output of the forget gate, indicating how much memory unit information should be retained at the current moment; σ: is the Sigmoid activation function, ensuring that the output is between 0 and 1; W f : weight matrix of the forget gate; h t-1 : The hidden state at the previous moment; x t : Current input, including sensor data: shield door drive motor temperature, torque; b f : Bias of forget gate;

[0055] Input Gate:

[0056] i t =σ(W i ·[h t-1 , x t ]+b i )

[0057] Where: i t : The output of the input gate determines which new information will be added to the current memory unit; W i and b i : weight and bias of input gate;

[0058] Memory Update:

[0059]

[0060] Where: is the candidate memory value at the current moment; tanh is the hyperbolic tangent activation function, and the output range is between -1 and 1; W C : is the weight matrix of memory update; b C : is the bias term for memory updating;

[0061] The updated memory cell state is:

[0062] C t =f t ·C t-1 +i t ·C t

[0063] Where: C t : is the current state of the memory unit, indicating the long-term memory of the system at the current moment; C t-1: is the memory unit state at the previous moment;

[0064] Output Gate:

[0065] o t =σ(W o ·[h t-1 , x t ]+b o )

[0066] In the formula: o t : is the output of the output gate, which determines the impact of the current hidden state on the final output, W o and b o : are the weight matrix and bias term of the output gate respectively;

[0067] Hidden state h t The calculation is:

[0068] h t =o t tanh(C t )

[0069] Where: h t is the hidden state at the current moment, combined with the memory unit C t and output gate regulation;

[0070] The specific process is: the status data of the equipment, including current, voltage, and temperature, is input into the deep learning model as a time series, the long-term dependencies are captured through the memory gate mechanism, and the predicted equipment status is output. According to the status changes of the equipment, it is judged whether there is a risk of failure. If it is predicted that the equipment may fail, the system automatically adjusts the operating parameters and generates a maintenance report.

[0071] After the equipment status data is input into the deep learning model, the system conducts in-depth analysis of the motor current, temperature, operating status and other data fed back by the sensor to determine whether the equipment has potential failures. The advantage of the deep learning model is that it can process a large amount of time series data and detect small changes in the data, thereby predicting possible damage or failure of the equipment in advance. This predictive maintenance method can reduce downtime, reduce equipment maintenance costs, and improve the overall operational stability of the equipment.

[0072] S4: Based on the data of the sensor unit and the image information of the camera, the deep learning model is used to perform behavioral analysis to detect whether the personnel have abnormal behavior, including unauthorized entry into the laboratory and misoperation. If an abnormality is detected, an alarm is automatically triggered; the system uses the image data captured by the camera and the deep learning algorithm to monitor the behavior of personnel in real time to detect whether unauthorized personnel have entered the laboratory or other abnormal behaviors have occurred. The image recognition capability of deep learning enables the system to accurately identify the identity and behavior of personnel, and determine whether an alarm needs to be triggered based on the set security rules. When abnormal behavior is detected, the system automatically triggers the sound and light alarm device and locks the shielding door to prevent unauthorized personnel from entering the laboratory.

[0073] S5: The opening and closing speed, force and timing of the shield door are dynamically adjusted through the adaptive fuzzy control algorithm and real-time environmental data and door status. When the system predicts environmental anomalies or equipment failures, the control unit controls the door control and safety equipment components in a linked manner according to the instructions of the data analysis and storage unit. The adaptive fuzzy control algorithm is controlled as follows:

[0074] Assume that the input of fuzzy control is u=(u1,u2,...,u n ), the output is y;

[0075] Membership function:

[0076]

[0077] Where: μ A (x): the membership value of input x to fuzzy set A; c A and σ A : are the center and width of the fuzzy set A, respectively, which are dynamically adjusted according to the environmental parameters;

[0078] Fuzzy rule reasoning:

[0079]

[0080] Where: R j represents the jth fuzzy rule; u1, u2, ..., u n : Input variables, representing the environmental parameters collected by the sensor, including temperature and humidity; Fuzzy set of input variables; B j : Output fuzzy set, representing the system control quantity, including the adjustment of door opening and closing speed;

[0081] Defuzzification:

[0082] The fuzzy output y is defuzzified by weighted average method. j Convert to precise control value:

[0083]

[0084] in: Fuzzy output set B j The membership degree of j :The corresponding fuzzy rule R j The output value of; m: the number of fuzzy rules;

[0085] Adaptive process: Dynamically adjust the parameters c of fuzzy rules and membership functions according to the changes in real-time environment and device status A and σ A , so that the system can adaptively respond to different environmental conditions to ensure the system's adaptability to different environmental changes.

[0086] The system uses an adaptive fuzzy control algorithm to dynamically adjust the opening and closing speed and force of the shielding door based on the real-time feedback of environmental data (such as temperature, humidity, gas concentration) and door status (such as opening and closing position, speed, etc.) from sensors. Fuzzy control can handle complex and nonlinear systems and make flexible adjustments based on current input data. This adaptive control method can ensure that the shielding door can operate safely and smoothly in any environment according to different laboratory environmental conditions.

[0087] S6: The data analysis and storage unit stores all sensor data, equipment status data, control logs, alarm records, and operation records in the database as a training data source for subsequent machine learning models for model optimization and long-term analysis;

[0088] The system stores information such as sensor data, equipment status data, control logs, and alarm records in the database. These data are not only the result of real-time monitoring, but also provide a training data source for the machine learning model. By continuously accumulating historical data, the system can regularly update and optimize the machine learning model to ensure that the prediction accuracy of the model continues to improve, thereby improving the intelligence level of the system. This closed-loop structure of data storage and model optimization enables the system to have the ability of self-learning and optimization.

[0089] S7: When the environmental parameters exceed the preset safety threshold, the data analysis and storage unit immediately sends an alarm signal to the control unit, and the control unit links the shielding door and the safety equipment components to perform emergency processing, including closing the shielding door, starting the ventilation system or the fire extinguishing system; if the camera detects unauthorized entry or identifies abnormal behavior, the system will trigger the alarm system and automatically lock the shielding door;

[0090] The system monitors environmental parameters in real time. Once it detects that the temperature, humidity, gas concentration, etc. exceed the preset safety threshold, the data analysis and storage unit immediately sends an alarm signal to the control unit. After receiving the command, the control unit automatically links the shielding door, ventilation system, fire extinguishing system and other equipment to carry out emergency treatment in time. This step demonstrates the system's automated emergency response capability, which can take protective measures before dangerous situations occur to ensure the safety of the laboratory environment and equipment.

[0091] S8: The data analysis and storage unit regularly performs self-inspection and maintenance on the equipment and generates inspection reports; it predicts possible equipment loss or failure through deep learning models, arranges maintenance or replacement in advance, and prevents sudden failures. The system has a regular self-inspection function, which automatically generates maintenance reports through long-term monitoring of the equipment status and analysis of the equipment's operation. Combined with the deep learning model, the system can predict equipment loss or impending failure in advance and remind users to perform maintenance or replace parts. This predictive maintenance can effectively avoid sudden failures, increase the service life of equipment, and reduce unplanned downtime in the laboratory.

[0092] This embodiment realizes intelligent and automated laboratory shield door control and safety management through the combination of multi-sensor fusion, machine learning model and deep learning model. The system can dynamically monitor the laboratory environment, predict potential risks, and provide comprehensive laboratory safety protection by controlling various safety equipment in a linked manner.

[0093] The technical solution of the present invention is not limited to the above-mentioned specific embodiments. All technical variations made according to the technical solution of the present invention fall within the protection scope of the present invention.

Claims

1. A laboratory shielding door control system, characterized in that: It includes sensor units, data analysis and storage units, control units and door control and security equipment components; The sensor unit collects temperature, humidity, air quality, radiation, personnel approach, shielding door opening and closing status, and image data near the laboratory shielding door; The data analysis and storage unit analyzes the data collected by the sensor unit, predicts environmental changes, calculates control information and sends it to the control unit, and stores historical data; The control unit executes the commands of the data analysis and storage unit to control the door frame and the safety equipment components; The door control and safety equipment components control the opening and closing of the shield door, and control the operation of the fire extinguishing, ventilation and alarm equipment; The signal output end of the sensor unit is connected to the data input end of the data analysis and storage unit, the data output end of the data analysis and storage unit is connected to the data input end of the control unit, the control signal output end of the control unit is connected to the control signal input end of the safety device component, and the safety device component is connected to the laboratory shielding door.

2. The laboratory shielding door control system according to claim 1 is characterized in that: The sensor unit includes a temperature sensor, a humidity sensor, a gas sensor, a radiation sensor, an ultrasonic sensor, a shield door switch state sensor, and a camera; the data analysis and storage unit can predict environmental changes in the next few hours and provide early warnings through a machine learning model trained with historical data and real-time sensor data; the data analysis and storage unit adopts an adaptive fuzzy control algorithm to dynamically adjust the opening and closing speed, force, and timing according to changes in the real-time environment and door state; the data analysis and storage unit analyzes sensor and equipment data through a deep learning model, predicts possible equipment failures in advance, and automatically adjusts operating parameters; the data analysis and storage unit stores all sensor data, control logs, alarm data, and operation records as training data sources for machine learning models, and performs long-term analysis and optimization; the data analysis and storage unit identifies abnormal behaviors through a deep learning model based on sensor and camera data, automatically triggers an alarm, or sends data to the control unit to adjust the state of the door; the control unit receives the control signal of the data analysis and storage unit to control the door control and safety equipment components in a linkage manner; the safety equipment components include a shield door drive controller, a ventilation system, a fire extinguishing system, and an alarm system.

3. A safety monitoring method for a laboratory shielding door control system as claimed in claim 2, characterized in that: The following steps are involved: S1: The sensor unit collects environmental data, equipment status data, and image data near the laboratory shielding door in real time; S2: The data analysis and storage unit receives the real-time data from the sensor unit, inputs the data into the machine learning model, and performs training based on historical data and real-time data to predict future environmental changes; S3: Analyze the status data of sensors and equipment through deep learning models to determine whether there is a potential failure in the equipment. If an abnormality is detected, the system adjusts the operating parameters of the equipment in advance to prevent equipment failure. S4: Based on the data from the sensor unit and the image information from the camera, a deep learning model is used to perform behavioral analysis to detect whether a person has abnormal behavior. If an abnormality is detected, an alarm is automatically triggered; S5: The opening and closing speed, force and timing of the shield door are dynamically adjusted through the adaptive fuzzy control algorithm and real-time environmental data and door status. When the system predicts environmental anomalies or equipment failures, the control unit controls the door control and safety equipment components in a linked manner according to the instructions of the data analysis and storage unit. S6: The data analysis and storage unit stores all sensor data, equipment status data, control logs, alarm records, and operation records in a database as a training data source for subsequent machine learning models; S7: When the environmental parameters exceed the preset safety threshold, the data analysis and storage unit immediately sends an alarm signal to the control unit, and the control unit simultaneously links the shielding door and the safety equipment components to perform emergency processing; S8: The data analysis and storage unit regularly performs self-inspection and maintenance on the equipment and generates inspection reports; it predicts possible equipment loss or failure through deep learning models.

4. The safety monitoring method of the laboratory shielding door control system according to claim 3 is characterized in that: The machine learning model formula is as follows: f(x)=w T φ(x)+b Where: x: input vector, i.e. environmental data collected by the sensor, including temperature and humidity; φ(x): mapping function that maps input data to high-dimensional feature space; w: weight vector of the model; b: bias term; f(x): output of the model, i.e. the predicted value of environmental change; Loss function: Where: n: the number of samples of training data; y i : The true value of the i-th sample, that is, the actual temperature and humidity; f(x i ): the predicted value of the i-th sample; ∈: the threshold of the allowable error, that is, the prediction error range allowed by the model; C: penalty coefficient, which controls the model's sensitivity to errors beyond the range ∈; ||w|| 2 : The norm of the weight vector, which penalizes excessive weights to improve the generalization ability of the model; The specific process is as follows: the environmental data x collected by the sensor is input into the model for feature transformation, trained using a machine learning model, and the model parameters w, b, and C are optimized through historical data; the model is used to predict changes in temperature and humidity in the next few hours, and early warnings are triggered based on the prediction results.

5. The safety monitoring method for a laboratory shielding door control system according to claim 4 is characterized in that: The deep learning model includes a forget gate, an input gate, a memory update, and an output gate; Forget Gate: f t =σ(W f ·[h t-1 ,x t ]+b f ) Where: f t : The output of the forget gate indicates how much memory unit information should be retained at the current moment; σ: is the Sigmoid activation function, which ensures that the output is between 0 and 1; W f : weight matrix of the forget gate; h t-1 : The hidden state at the previous moment; x t : Current input, including sensor data: shield door drive motor temperature, torque; b f : Bias of forget gate; Input Gate: i t =σ(W i ·[h t-1 ,x t ]+b i ) Where: i t : The output of the input gate determines which new information will be added to the current memory unit; W i and b i : weight and bias of input gate; Memory Update: Where: is the candidate memory value at the current moment; tanh is the hyperbolic tangent activation function, and the output range is between -1 and 1; W C : is the weight matrix of memory update; b C : is the bias term for memory updating; The updated memory cell state is: Where: C t : is the current state of the memory unit, indicating the long-term memory of the system at the current moment; C t-1 : is the memory unit state at the previous moment; Output Gate: the t =σ(W o ·[h t-1 ,x t ]+b o ) In the formula: o t : is the output of the output gate, which determines the impact of the current hidden state on the final output, W o and b o : are the weight matrix and bias term of the output gate respectively; Hidden state h t The calculation is: h t =o t ·tanh(C t ) Where: h t is the hidden state at the current moment, combined with the memory unit C t and output gate regulation; The specific process is: the status data of the equipment, including current, voltage, and temperature, is input into the deep learning model as a time series, the long-term dependencies are captured through the memory gate mechanism, and the predicted equipment status is output. According to the status changes of the equipment, it is judged whether there is a risk of failure. If it is predicted that the equipment may fail, the system automatically adjusts the operating parameters and generates a maintenance report.

6. The safety monitoring method of the laboratory shielding door control system according to claim 5 is characterized in that: The adaptive fuzzy control algorithm is controlled as follows: Assume that the input of fuzzy control is u=(u1,u2,...,u n ), the output is y; Membership function: Where: μ A (x): the membership value of input x to fuzzy set A; c A and σ A : are the center and width of the fuzzy set A, respectively, which are dynamically adjusted according to the environmental parameters; Fuzzy rule reasoning: R j :If u1 is and u2 is ,...,u n is then y is B j Where: R j represents the jth fuzzy rule; u1, u2, ..., u n : Input variables, representing the environmental parameters collected by the sensor, including temperature and humidity; Fuzzy set of input variables; B j : Output fuzzy set, representing the system control quantity, including the adjustment of door opening and closing speed; Defuzzification: The fuzzy output y is defuzzified by weighted average method. j Convert to precise control value: in: Fuzzy output set B j The membership degree of j :The corresponding fuzzy rule R j The output value of; m: the number of fuzzy rules; Adaptive process: Dynamically adjust the parameters c of fuzzy rules and membership functions according to the changes in real-time environment and device status A and σ A , so that the system can adaptively respond to different environmental conditions to ensure the system's adaptability to different environmental changes.

Images