Behavior analysis control method for safe operation of ventilation cabinet

Abstract

The invention discloses a behavior analysis control method for safe operation of a fume hood, and relates to the technical field of behavior analysis control, and the method comprises the following steps: defining a safe operation reference and a real-time monitoring system of the fume hood; performing behavior pattern analysis according to the reference parameters and the real-time monitoring data; performing safety early warning according to a behavior analysis result; automatic control measures are taken according to the early warning signal; performing energy efficiency optimization according to a control result; performing long-term operation data feedback according to the optimization scheme; further optimizing the behavior mode and the energy efficiency according to the operation data; and carrying out automatic upgrading and self-adjustment on the system according to an optimization result. The operation safety, the energy efficiency and the operation efficiency of the fume hood are effectively improved, the stability and the high efficiency of a system in a complex environment are ensured, the flexibility of the system and the adaptability of the fume hood in different environments are improved, and efficient and safe operation can be kept under frequently-changing operation conditions in a laboratory or an industrial environment.

Description

Technical Field

[0001] The present invention relates to the technical field of behavior analysis and control, and specifically provides a behavior analysis and control method for the safe operation of a fume hood. Background Art

[0002] With the continuous improvement of laboratory safety management requirements, as an important device to ensure the safety of the laboratory environment and the health of operators, the safety and energy efficiency of the operation of the fume hood have received increasing attention. The control system of the fume hood usually relies on rule-based settings, such as sensor data of air flow velocity, vibration range, temperature and humidity, etc. However, these traditional control methods often show certain limitations when facing complex and dynamic experimental environments. With the rapid development of intelligent technologies, reinforcement learning, deep learning, and transfer learning have been gradually introduced into the control system of the fume hood to achieve adaptive optimization regulation. Nevertheless, when applying these advanced methods, the existing technologies still face the following two main problems: 1. In the existing technology, there are deficiencies in multi-factor comprehensive optimization: The existing reinforcement learning methods usually focus on the optimization of a single target, ignoring the multiple complex relationships among safety, energy efficiency, and equipment stability during the control process of the fume hood. Traditional methods rely more on rule settings, and it is difficult for the system to perform real-time and effective adjustment in the face of complex air flow fluctuations, vibration changes, and abnormal temperature and humidity.

[0003] 2. In the existing technology, there are deficiencies in environmental adaptability: The existing deep learning and transfer learning technologies have weak adaptability in new environments. When the device is transferred to different experimental environments, or faces unknown operating states, the existing learning models often require a large amount of new data for training, resulting in a slow adaptation process and an inability to quickly adjust the control strategy. Summary of the Invention

[0004] In view of the deficiencies of the existing technology, the present invention provides a behavior analysis and control method for the safe operation of a fume hood to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions: In the first aspect, an embodiment of the present invention provides a behavior analysis and control method for the safe operation of a fume hood, including the following steps: S1. Define the safety operation benchmark and real-time monitoring system of the fume hood to obtain the operation data of the fume hood; S2. Use a machine learning algorithm to perform behavior pattern analysis based on the operation data of the fume hood to obtain a behavior analysis result; S3. Perform safety warnings based on the behavior analysis result to obtain a warning signal; S4. Perform automatic control measures based on the warning signal to obtain a control result; S5. Optimize energy efficiency according to the control results, adopt an optimization algorithm to optimize the control parameters, and obtain an optimization plan; S6. Conduct long-term operation data feedback according to the optimization plan to obtain optimized operation data; S7. Further optimize the behavior pattern and energy efficiency according to the operation data to obtain an optimization result; S8. Conduct automatic upgrade and self-adjustment of the system according to the optimization result.

[0006] To further optimize this technical solution, the steps of defining the safety operation benchmark and real-time monitoring system for the fume hood in S1 include: Define the safety operation benchmark of the fume hood to obtain benchmark parameters; Real-time monitor the status of the fume hood through a sensor network to obtain the operation data of the fume hood; Use Internet of Things technology to transmit the real-time monitoring data to the data processing center; Judge whether the fume hood is within the safety benchmark range; If it exceeds the safety benchmark range, record the abnormal event and generate an alarm signal.

[0007] To further optimize this technical solution, the steps of conducting safety warning in S3 include: Based on the behavior analysis result, the system judges whether there is an abnormal behavior pattern; If there is an abnormal behavior pattern, generate a warning signal based on the pattern recognition algorithm; Compare the behavior pattern with historical data to judge whether it belongs to a known dangerous pattern; If it belongs to a known dangerous pattern, issue an emergency alarm and give suggestions on taking corresponding safety measures.

[0008] To further optimize this technical solution, the steps of comparing the behavior pattern with historical data include: Data collection and preprocessing: Obtain real-time monitoring data through a sensor network, preprocess the data, and obtain preprocessed data; Behavior pattern recognition: Through machine learning algorithms, the system analyzes the preprocessed data and recognizes the behavior pattern; Storage and management of historical data: Store the historical operation data and the labeled behavior patterns in the historical database of the system, and calibrate the historical data as a comparison standard; Pattern comparison and matching: After identifying the behavior pattern at the current moment, the system compares it with historical data, calculates the similarity between the current pattern and the historical pattern, and matches it with the historical pattern according to the similarity; Decision-making and feedback: If the similarity between the new behavior pattern and the abnormal pattern is high, a safety warning or automatic adjustment measure is triggered; if the similarity with the normal pattern is high, the current operating state is maintained and the data changes are continuously monitored.

[0009] To further optimize this technical solution, the steps for taking automatic control measures in S4 include: Real-time detection of warning signals; After detecting a warning signal, the system automatically activates the preset safety control measures; Using adaptive control technology, adjust the strategy according to the feedback results of the safety control measures to improve the system's response ability to abnormal situations.

[0010] To further optimize this technical solution, the steps for providing long-term operation data feedback in S6 include: The system continuously monitors the optimized operation data; The system continuously evaluates the persistence and reliability of the optimization effect according to the changes in real-time data; The system makes adjustments according to the evaluation results to ensure that the equipment operates in a safe and efficient state.

[0011] To further optimize this technical solution, the steps for further optimizing the behavior pattern and energy efficiency in S7 include: Based on historical data and real-time feedback, the system uses deep reinforcement learning technology with multi-factor dynamic adjustment, combines the results of behavior analysis and energy efficiency optimization, continuously learns, and improves the control algorithm by modifying the algorithm weight coefficients to further optimize the system performance.

[0012] To further optimize this technical solution, the deep reinforcement learning technology with multi-factor dynamic adjustment includes: Model construction: Construct the following reward function for the training and optimization of reinforcement learning: ; ; Where: : Airflow velocity, : Vibration frequency change difference, : Temperature and humidity change, obtained through the sensor network, where Calculated by the method of weighted average: is the temperature, is the humidity, is the temperature weight coefficient, is the humidity weight coefficient, ; , , : Benchmark parameter, defined in step S1; : Deviation of air flow velocity, the smaller the deviation, the greater the reward; : Deviation of vibration frequency, the smaller the deviation, the greater the reward; : Deviation of temperature and humidity, the smaller the deviation, the greater the reward; : Energy efficiency score, the larger the score, the more energy-efficient; : Ideal power consumption; : Actual power consumption; , , , : Weight coefficient, adjusted according to the actual effect; Usage of the model: Input data; calculate the energy efficiency score; through the calculation of the reward function, the system obtains the control strategy and adjusts it according to the actual execution result to obtain the optimal control strategy; as the environment changes or the equipment operation time increases, the system continues to adjust the control strategy to adaptively respond to the new operating state.

[0013] To further optimize this technical solution, the steps for the system to perform automatic upgrade and self-adjustment in S8 include: Based on the optimization results, the system continuously updates its control strategy through deep learning and transfer learning technologies, thereby performing self-adjustment and upgrade.

[0014] To further optimize this technical solution, the deep learning and transfer learning technologies include: Model construction: Combining deep learning and transfer learning, construct the following model: ; Wherein: : Control output at time ; : Weight coefficient between deep learning and transfer learning, controlling the relative influence in the final outputs of both, and adjusted according to the actual effect; : Environmental difference, obtained by comparing the data acquired by the sensor network in the new environment and the old environment; : Deep learning function, outputting a control strategy based on the input data and continuously optimizing it; : Transfer learning function to adjust the control strategy to adapt to the new environment; Usage of the model: Input data into the deep learning model to obtain the initial control strategy; optimize the initial control strategy through the deep learning model to obtain the optimized control strategy; adjust the optimized control strategy through transfer learning to obtain the adjusted control strategy; continuously adjust and adapt to the environment and feedback by adjusting the weight coefficients, and finally output the optimal control strategy.

[0015] In a second aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program instructions are executed by the processor, the steps of a method for analyzing and controlling the behavior of a fume hood for safe operation as described in the first aspect of the present invention are implemented.

[0016] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program instructions are executed by the processor, the steps of a method for analyzing and controlling the behavior of a fume hood for safe operation as described in the first aspect of the present invention are implemented.

[0017] Compared with the prior art, the present invention provides a method for analyzing and controlling the behavior of a fume hood for safe operation, having the following beneficial effects: This method for analyzing and controlling the behavior of a fume hood for safe operation successfully solves the deficiencies in the prior art by combining multi-factor comprehensive optimization with the problem of environmental adaptability. Traditional fume hood control methods usually rely on single factors or static rules for operation decision-making, and cannot comprehensively consider the complex relationships among multi-dimensional data such as air flow velocity, vibration range, and temperature and humidity changes. By introducing a multi-factor comprehensive optimization method that combines reinforcement learning and deep learning to achieve the purpose of enhanced learning, the system can dynamically adjust the device operation state, comprehensively evaluate the real-time changes of multiple factors such as air flow, vibration, and temperature and humidity, thereby maximizing energy efficiency and operation efficiency while ensuring the safety of the device. Reinforcement learning optimizes the control strategy through multiple explorations and feedbacks, and deep learning improves the accuracy of multi-factor decision-making and the adaptive ability of the system by automatically extracting and integrating key features. This comprehensive optimization method effectively improves the safety of fume hood operation while avoiding the risk of ignoring certain key factors in traditional methods, ensuring the stability and high efficiency of the system in a complex environment.

[0018] In addition, the system also addresses the limitations of traditional equipment in the face of insufficient environmental adaptability. Traditional control methods often require re - debugging or re - training when the environment changes, lacking the ability to quickly adapt to new environments. By introducing transfer learning, the system can rapidly transfer existing knowledge between different experimental environments, achieving quick adaptation to new environments. Transfer learning fine - tunes deep - learning models, utilizes the similarity between historical data and the current environment, and quickly updates control strategies, reducing the time and cost of re - training. This rapid adaptation ability greatly enhances the flexibility of the system, enabling it to cope with frequently changing operating conditions in laboratory or industrial environments and ensuring that the equipment always operates at an optimal state. In this way, the system not only improves the adaptability of the fume hood in different environments but also maintains efficient and safe operation under changing experimental requirements and environmental conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following - described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0020] Figure 1 It is a schematic flowchart of a method for analyzing and controlling the safety operation of a fume hood proposed by the present invention; Figure 2 It is a schematic flowchart of comparing behavioral patterns and historical data of a method for analyzing and controlling the safety operation of a fume hood proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] To make the above - mentioned objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention in conjunction with the drawings in the specification.

[0022] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0023] Secondly, the so - called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation of the present invention. The appearances of "in one embodiment" in different parts of this specification do not all refer to the same embodiment, nor are they separate or selectively exclusive embodiments from other embodiments.

[0024] Embodiment 1: Refer toFigures 1 to 2 , which is the first embodiment of the present invention. This embodiment provides a method for analyzing and controlling the safety operation of a fume hood, including the following steps: S1. Define the safety operation benchmark and real-time monitoring system of the fume hood to obtain the operation data of the fume hood.

[0025] In this embodiment, the steps of defining the safety operation benchmark and real-time monitoring system of the fume hood include: Define the safety operation benchmark of the fume hood: According to industry standards, historical maintenance records, and the suggestions of equipment manufacturers, set the safety operation benchmark of the fume hood, including key parameters such as air flow velocity, vibration range, temperature and humidity changes, etc.; Real-time monitor the status of the fume hood through a sensor network: Use sensors such as air flow sensors, vibration sensors, temperature and humidity sensors, etc. to real-time monitor the status of the fume hood and compare it with the reference value; Utilize the Internet of Things technology to real-time transmit the operation data of the fume hood collected by the sensors to the data processing center; Judge whether the fume hood is within the safe working range; After exceeding the safety benchmark range, record the abnormal event and generate an alarm signal.

[0026] S2. Conduct behavior pattern analysis based on the operation data of the fume hood, and use machine learning algorithms to analyze the real-time monitoring data to identify potential safety hazards in advance and obtain the behavior analysis result.

[0027] In this embodiment, use machine learning algorithms (such as support vector machine SVM or deep neural network DNN) to analyze these real-time data to identify whether there are abnormal behavior patterns. By long-term analyzing the trends of data such as air flow, vibration, temperature and humidity, the system identifies normal and abnormal behavior patterns, and identifies potential safety hazards and failure risks in advance.

[0028] S3. Conduct safety warning based on the behavior analysis result to obtain a warning signal.

[0029] In this embodiment, the steps of conducting safety warning include: Based on the behavior analysis result, the system judges whether there are abnormal behavior patterns; If there are abnormal behavior patterns, generate a warning signal based on the pattern recognition algorithm; Compare the behavior pattern with historical data to judge whether it belongs to a known dangerous pattern; If it belongs to a known dangerous pattern, issue an emergency alarm and give suggestions on taking corresponding safety measures.

[0030] Furthermore, the step of comparing the behavior pattern with historical data includes: Data collection and preprocessing: Collect real-time data of the current fume hood, including key parameters such as air flow velocity, vibration range, temperature and humidity. To ensure the accuracy and consistency of the data, the system preprocesses this data, including denoising, standardization, and filling in missing data; Behavior pattern recognition: Through pattern recognition algorithms (such as K-means clustering or random forest algorithms), the system analyzes the current sensor data and identifies possible behavior patterns. For example, the system will detect fluctuations in air flow, abnormal amplitudes of vibrations, and abnormal changes in temperature and humidity. Based on these features, the system can identify whether the equipment is in a normal operating state or whether there are potential safety hazards; Storage and management of historical data: Store historical operation data and labeled behavior patterns in the system's historical database and calibrate the historical data. These data include behavior patterns under normal operation, fault patterns, and corresponding equipment status, environmental conditions, etc. After long-term accumulation and analysis, they have been clearly calibrated and used as a comparison standard for the system to judge whether an abnormality occurs; Pattern comparison and matching: After identifying the behavior pattern at the current moment, the system compares it with the historical data, including matching based on feature similarity, time series analysis, or classification matching through machine learning methods, calculates the similarity between the current pattern and the historical pattern, and matches it with the historical pattern to determine whether it is a known abnormal behavior; Decision-making and feedback: Based on the comparison results, the system makes corresponding decisions. If the new behavior pattern highly matches the abnormal pattern in the historical data, the system will trigger a safety warning or automatic adjustment measures. If the behavior pattern is similar to the normal pattern in the historical data, the system will maintain the current operating state and continue to monitor data changes; The comparison between the behavior pattern and the historical data can help the system quickly identify and respond to abnormal situations, improve the safety and stability of the equipment, and at the same time provide strong support for subsequent intelligent optimization decisions.

[0031] S4. Perform automatic control measures according to the warning signal to obtain a control result.

[0032] In this embodiment, the steps of performing automatic control measures include: Real-time detection of warning signals; After detecting the warning signal, the system automatically activates preset safety control measures to prevent the fume hood from being in an unsafe operating state. The control measures include but are not limited to automatically adjusting the air flow rate, temperature and humidity, or equipment status of the fume hood, or temporarily closing the fume hood for inspection according to the actual situation; Using adaptive control technology (i.e., the control method can automatically adjust its control strategy or parameters according to the dynamic changes of the controlled object or environment to achieve a better control effect), the control parameters of the fume hood are automatically adjusted according to the feedback results of the safety control measures to ensure that the equipment always operates within a safe range. The automatic control method can quickly respond to safety warnings and reduce the risks brought by human misoperation and response delays.

[0033] S5. Optimize energy efficiency according to the control results, adopt an optimization algorithm to optimize the control parameters, reduce energy consumption costs, improve production efficiency, and obtain an optimized solution.

[0034] In this embodiment, by using an optimization algorithm (such as Particle Swarm Optimization PSO or Genetic Algorithm GA), the system finely adjusts control parameters such as air flow, temperature, and humidity to ensure that, on the premise of ensuring equipment safety, energy consumption is minimized to the greatest extent. The optimization goal is to find a balance point so that the equipment can not only maintain safe operation but also maximize energy efficiency. This process is applicable to fume hoods operating for a long time, and energy efficiency optimization helps reduce energy consumption costs and improve the overall economy of the equipment. Through intelligent energy efficiency control, the system can dynamically adjust the operation of the fume hood according to the needs of the laboratory or industrial environment, avoid energy waste, and improve production efficiency.

[0035] S6. Conduct long-term operation data feedback according to the optimized solution to obtain optimized operation data.

[0036] In this embodiment, the steps of conducting long-term operation data feedback include: The system continuously monitors the optimized operation data, including the energy consumption, temperature, and humidity of the fume hood, etc.; The system continuously evaluates the persistence and reliability of the optimization effect according to the changes in real-time data; The system makes adjustments according to the evaluation results to ensure that the equipment operates in a safe and efficient state; The long-term feedback mechanism enables the system to continuously optimize the control scheme according to factors such as environmental changes and equipment aging, and maintain the best performance of the equipment. This process helps the system continuously adapt to the needs of different experimental environments and ensures the long-term stable and efficient operation of the fume hood.

[0037] S7. Further optimize the behavior pattern and energy efficiency according to the operation data to obtain an optimized result.

[0038] In this embodiment, the steps of further optimizing the behavior pattern and energy efficiency include: Based on historical data and real-time feedback, through deep reinforcement learning technology with multi-factor dynamic adjustment, combined with behavior pattern recognition and energy efficiency optimization, the system continuously learns and improves the control algorithm by modifying the algorithm weight coefficients. During the continuous adjustment process, the system not only optimizes the safety of equipment operation but also achieves the maximum energy efficiency optimization while ensuring safety. Deep reinforcement learning enables the system to gradually learn the optimal operation mode in an uncertain environment, thereby improving the accuracy and response ability of control, making the operation of the fume hood more intelligent and efficient.

[0039] Further, the deep reinforcement learning technology with multi-factor dynamic adjustment includes: Model construction: Construct the following reward function for the training and optimization of reinforcement learning: ; ; Where: : Airflow velocity, : Difference in vibration frequency change, : Temperature and humidity change, obtained through the sensor network, where Calculated by the weighted average method: is the temperature, is the humidity, is the temperature weight coefficient, is the humidity weight coefficient, ; , , : Benchmark parameter, defined in step S1; : Deviation of airflow velocity, the smaller the reward, the greater; : Deviation of vibration frequency, the smaller the reward, the greater; : Deviation of temperature and humidity, the smaller the reward, the greater; : Energy efficiency score, the larger the more energy-efficient; : Ideal power consumption; : Actual power consumption; , , , : Weight coefficient; Model usage: Input data: Input data and calculate the deviation; Calculate the energy efficiency score: The system simultaneously calculates the energy efficiency score of the current device, evaluates the energy efficiency status of the device through the ratio between the actual power consumption and the ideal power consumption. This score will play a positive promoting role in the reward function, thus driving the system to adjust parameters such as air flow, temperature, and humidity according to the actual situation to optimize energy efficiency; Obtain the optimal control strategy: Based on the current state and the reward function, the system selects the best control actions through reinforcement learning, such as adjusting the air flow speed, vibration frequency, temperature, and humidity settings. Through long-term interaction and feedback (i.e., adjusting the strategy according to the rewards of the actual execution results), it gradually learns an optimal control strategy, enabling the device to maintain the best safety and energy efficiency under different operating conditions; Adaptive learning: As the environment changes or the device runs for a longer time, the system continues to adjust the control actions, enabling the device to adaptively respond to new operating states (for example, fluctuations in air flow speed, changes in vibration frequency, sudden changes in temperature and humidity, etc.); Select the best control action through reinforcement learning: ; Among them, represents the state, ; represents the action, including adjusting the air flow speed, adjusting the vibration frequency, adjusting the temperature and humidity; represents the learning rate, which controls the Q-value update speed; represents the discount factor, which is the weight of future rewards; represents in the new state select the maximum Q-value among all possible actions at this time, which represents the optimal future reward, and the action at this time is the best control action.

[0040] This method designs a dynamically adjusted reward function based on the air flow speed, vibration range, and temperature and humidity change data, combined with the deviation from the reference value. During the long-term operation process, the reinforcement learning system will further optimize and improve its control strategy through the accumulated rewards and feedback, so that the device can maintain good safety and stability while operating efficiently.

[0041] S8. Perform automatic upgrade and self-adjustment of the system according to the optimization results.

[0042] In this embodiment, the steps for performing automatic upgrade and self-adjustment of the system include: Based on the optimization results, the system continuously updates its control strategy through deep learning and transfer learning techniques, thereby performing self - adjustment and upgrading. Over time, the system will gradually improve the recognition and control accuracy and be able to automatically adapt to changes in different environments. This ability of self - adjustment and upgrading ensures that the system can flexibly respond when facing new situations or problems, improving the safety and energy efficiency of the device. When the operating environment changes significantly, it can still provide the optimal control strategy.

[0043] Furthermore, the deep learning and transfer learning techniques include: Model construction: Combining deep learning and transfer learning, the following model is constructed: ; Where: represents the control output or decision result at time moment. For example, may represent an adjustment instruction, such as control instructions like "the air flow speed should be increased by 10% and the vibration amplitude should be reduced by 5%"; : is the weight coefficient between deep learning and transfer learning, controlling the relative influence in the final outputs of both. By adjusting this parameter, the system can balance the roles of deep learning and transfer learning, ensuring that the model can both precisely adjust in the current environment and quickly adapt to the new environment; : environmental difference, obtained by comparing the data acquired by the sensor network in the new environment and the old environment, representing the difference between the current environment and the original environment; : deep neural network model, which, based on the input of data such as air flow speed, vibration range, temperature and humidity changes, and the deviation from the reference value, outputs a control strategy. It can be understood as the immediate optimization of the device by the deep learning model; ; Where, is the weight, is the bias term; : represents the transfer learning function, used to enhance the adaptability of the system in the new environment. By fine - tuning the parameters of the deep neural network model, the control strategy can be adapted to the new environment; ; Where, and are the fine - tuning parameters in the transfer learning process; Model usage: Obtain the initial control strategy: Calculate based on the deviation between the collected data and the benchmark value, input the data into the deep neural network model, and the model outputs an initial control strategy; Optimization of the deep learning model: The deep neural network learns how to adjust device parameters according to the input data through the weight matrix and bias to achieve optimized control. The model will continuously be trained with historical data to gradually improve the accuracy of the control strategy; Adjustment of transfer learning: When the device is transferred to a new experimental environment, transfer learning fine-tunes the parameters of the deep neural network based on the differences between the original environment and the new environment. This process helps the system quickly adapt to the new environment without the need for a large amount of new data; Adaptive and feedback: The outputs of the deep learning and transfer learning models will jointly determine the control decisions of the device. Whenever new data is input into the system, the models will continuously update the output and optimize the parameters to ensure that the device always operates in the best safety and energy efficiency state.

[0044] Embodiment 2: This embodiment also provides a computer device applicable to the situation of a behavior analysis control method for the safe operation of a fume hood, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement a behavior analysis control method for the safe operation of a fume hood as proposed in the above embodiment.

[0045] This embodiment also provides a storage medium with a computer program stored thereon, and when the program is executed by the processor, it implements a behavior analysis control method for the safe operation of a fume hood as proposed in the above embodiment.

[0046] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0047] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, etc., all kinds of media that can store program codes.

[0048] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or in combination with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in combination with an instruction execution system, apparatus, or device.

[0049] More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection parts with one or more wirings (electronic devices), portable computer disk cartridges (magnetic devices), random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories (EPROMs or flash memories), fiber optic devices, and portable compact disc read-only memories (CDROMs). Additionally, a computer-readable medium can even be paper or other suitable media on which a program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, then editing, interpreting, or otherwise processing it as appropriate, and then storing it in a computer memory.

[0050] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following techniques known in the art or a combination thereof can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application specific integrated circuits with suitable combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0051] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A method for analyzing and controlling the behavior of safe operation of a fume hood, characterized in that, It includes the following steps: S1. Define the safety operation criteria and real-time monitoring system for the fume hood to obtain the operation data of the fume hood; S2. Use machine learning algorithms to analyze the behavior patterns based on the operation data of the fume hood to obtain the behavior analysis results; S3. Conduct safety warnings based on the behavior analysis results to obtain warning signals; S4. Take automatic control measures according to the warning signals to obtain control results; S5. Optimize the energy efficiency according to the control results, adopt optimization algorithms to optimize the control parameters, and obtain the optimization plan; S6. Conduct long-term operation data feedback according to the optimization plan to obtain the optimized operation data; S7. Further optimize the behavior patterns and energy efficiency based on the operation data to obtain the optimization results; S8. Conduct automatic upgrade and self-adjustment of the system according to the optimization results.

2. The behavioral analysis and control method for the safe operation of a fume hood according to claim 1, wherein, In S1, the safety operation criteria and real-time monitoring system for the fume hood include: Define the safety operation criteria for the fume hood to obtain the reference parameters; Real-time monitor the status of the fume hood through the sensor network to obtain the operation data of the fume hood; Use Internet of Things technology to transmit the real-time monitoring data to the data processing center; Judge whether the fume hood is within the safety reference range; If it exceeds the safety reference range, record the abnormal event and generate an alarm signal.

3. The behavioral analysis and control method for the safe operation of a fume hood according to claim 1, characterized in that, The steps for conducting safety warnings in S3 include: Based on the behavior analysis results, the system judges whether there are abnormal behavior patterns; If there are abnormal behavior patterns, generate warning signals based on the pattern recognition algorithm; Compare the behavior patterns with the historical data to judge whether they belong to known dangerous patterns; If they belong to known dangerous patterns, issue an emergency alarm and give suggestions on the adoption of corresponding safety measures.

4. The behavioral analysis and control method for the safe operation of a fume hood according to claim 3, characterized in that, The steps for comparing the behavior patterns with the historical data include: Data collection and preprocessing: Obtain real-time monitoring data through the sensor network, preprocess the data, and obtain the preprocessed data; Behavior pattern recognition: The system analyzes the preprocessed data through machine learning algorithms and identifies the behavior patterns; Storage and management of historical data: Store the historical operation data and the labeled behavior patterns in the historical database of the system, and calibrate the historical data as the comparison standard; Pattern comparison and matching: After identifying the behavior pattern at the current moment, the system compares it with the historical data, calculates the similarity between the current pattern and the historical pattern, and matches it with the historical pattern according to the similarity; Decision-making and feedback: If the similarity between the new behavior pattern and the abnormal pattern is high, trigger safety warnings or automatic adjustment measures; if the similarity with the normal pattern is high, maintain the current operation status and continue to monitor the data changes.

5. The behavioral analysis and control method for the safe operation of a fume hood according to claim 1, characterized in that, The steps for taking automatic control measures in S4 include: Real-time detect the warning signals; After detecting the warning signals, the system automatically starts the preset safety control measures; Use adaptive control technology to adjust according to the feedback results of the safety control measures to improve the system's response ability to abnormal situations.

6. The behavioral analysis and control method for the safe operation of a fume hood according to claim 1, characterized in that, The steps for conducting long-term operation data feedback in S6 include: The system continuously monitors the optimized operation data; The system continuously evaluates the durability and reliability of the optimization effect according to the changes in the real-time data; The system makes adjustments according to the evaluation results to ensure that the equipment operates in a safe and efficient state.

7. The behavioral analysis and control method for the safe operation of a fume hood according to claim 1, characterized in that, The method for further optimizing the behavior pattern and energy efficiency in S7 is as follows: Based on historical data and real-time feedback, the system uses deep reinforcement learning technology with multi-factor dynamic adjustment, combines the results of behavior analysis and energy efficiency optimization, continuously learns, and improves the control algorithm by modifying the algorithm weight coefficients to further optimize the system performance.

8. The behavioral analysis and control method for the safe operation of a fume hood according to claim 7, characterized in that, The deep reinforcement learning technology with multi-factor dynamic adjustment includes: Model construction: Construct the following reward function for the training and optimization of reinforcement learning: ； ; Where: : Airflow velocity, : Vibration frequency change difference, : Temperature and humidity change, obtained through the sensor network, where is calculated by the weighted average method: is the temperature, is the humidity, is the temperature weight coefficient, is the humidity weight coefficient, ; , , : Reference parameter, defined in step S1; : Deviation of air flow velocity, the smaller the deviation, the greater the reward; : The deviation of the vibration frequency, the smaller the deviation, the greater the reward; : The deviation of temperature and humidity, the smaller the deviation, the greater the reward; : Energy efficiency score, the larger the more energy-efficient; : Ideal power consumption; :Actual power consumption; , , , : Weight coefficient.

9. The behavioral analysis and control method for the safe operation of a fume hood according to claim 1, wherein, The steps for the automatic upgrade and self-adjustment of the system in S8 include: Based on the optimization results, the system continuously updates its control strategy through deep learning and transfer learning technologies, thereby performing self-adjustment and upgrade.

10. The behavioral analysis and control method for the safe operation of a fume hood according to claim 9, wherein, The deep learning and transfer learning technologies include: Model construction: Combining deep learning and transfer learning, construct the following model: ； Where: : The control output at time ; : It is the weight coefficient between deep learning and transfer learning, controlling the relative influence in the final outputs of both, and is adjusted according to the actual effect; : Environmental differences, obtained by comparing data acquired by the sensor network in the new environment and the old environment; : A deep learning function that outputs a control strategy based on the input data and continuously optimizes it; : Transfer learning function that adjusts the control strategy to adapt to the new environment.

Images