Safety interlocking control method and system for particle accelerator operation maintenance

Abstract

The invention belongs to the field of particle accelerators, and provides a safety interlocking control method and system for particle accelerator operation maintenance, and the method comprises the steps: selecting a corresponding maintenance mode according to the operation state of a particle accelerator, determining maintenance personnel according to the selected maintenance mode, and triggering the interlocking state of a laboratory shielding door of the particle accelerator; monitoring the radiation level in the particle accelerator laboratory, and releasing the interlocking state of the shielding door of the particle accelerator laboratory when the radiation level is lower than a safety threshold value; performing face recognition on the maintainers based on a pre-trained face recognition model, and performing counting and identity information storage on the maintainers entering the particle accelerator laboratory; after the maintenance is finished, the identity information is compared with the identity information of the maintenance personnel entering the particle accelerator laboratory; and if the comparison succeeds, the anti-opening state of the shielding door of the particle accelerator laboratory is relieved until all the maintainers entering the particle accelerator laboratory leave, and safety maintenance is completed.

Description

Technical Field

[0001] The present invention belongs to the technical field of particle accelerators, and particularly relates to a safety interlock control method and system for the operation and maintenance of particle accelerators. Background Art

[0002] The statements in this section merely provide background technical information related to the present invention and do not necessarily constitute prior art.

[0003] Particle accelerators are widely used in the fields of medicine, scientific research, and industry, and high-energy radiation is generated during their operation. To ensure the safety of operators and maintenance personnel, special protection measures are required during the operation and maintenance of accelerators. Traditional safety management for accelerator maintenance mostly relies on manual management, which has certain risks of missed inspections and misoperations. Especially in emergency situations such as fault repairs and accident emergency repairs, there is often a lack of intelligent monitoring and personnel management, which may lead to the risk of personnel being detained and accidentally irradiated.

[0004] In addition, existing particle accelerator laboratories use face recognition to register the personnel entering and leaving the particle accelerator laboratory and set relevant permissions, and conduct intelligent management and control of the personnel entering and leaving the particle accelerator laboratory, so as to achieve the safety protection of the particle accelerator laboratory; when the particle accelerator laboratory is in emergency situations such as fault repairs and accident emergency repairs, some malicious attackers, in order to steal the confidential materials and technologies of the laboratory, will forge face masks or use network malicious attacks on the face recognition system, and try to mix into the maintenance personnel during the fault repair and emergency repair stages to enter the particle accelerator laboratory, resulting in the leakage of laboratory technical information. Summary of the Invention

[0005] To solve the above problems, the present invention proposes a safety interlock control method and system for the operation and maintenance of particle accelerators. The present invention ensures that the accelerator room is always in a safe state during personnel maintenance through means such as automatic detection, face recognition, and counting, avoids radiation hazards caused by personnel detention or misoperation, and improves the safety and efficiency of the accelerator maintenance process. At the same time, in order to prevent non-operation and maintenance personnel from maliciously fraudulently attacking the face recognition system and thus entering the production plant and affecting the production operation of equipment.

[0006] According to some embodiments, the first solution of the present invention provides a safety interlock control method for the operation and maintenance of particle accelerators, adopting the following technical solutions:

[0007] A safety interlock control method for the operation and maintenance of particle accelerators, comprising:

[0008] Obtaining the operating status of the particle accelerator, and selecting a corresponding maintenance mode according to the operating status of the particle accelerator, determining the maintenance personnel according to the selected maintenance mode, and triggering the interlocking state of the particle accelerator laboratory shielding door;

[0009] Monitor the radiation level in the particle accelerator laboratory and release the interlock state of the particle accelerator laboratory shielding door when the radiation level is below the safety threshold;

[0010] Perform facial recognition on maintenance personnel based on a pre-trained facial recognition model, count maintenance personnel entering the particle accelerator laboratory, and store their identity information;

[0011] Real-time monitoring of the maintenance status of the particle accelerator laboratory, the number of maintenance personnel, and the location information of the maintenance personnel, and triggering the anti-opening status of the particle accelerator laboratory shielding door;

[0012] When the maintenance is completed, the maintenance personnel passing through the entrance of the particle accelerator laboratory will be recognized and counted again, and the identity information will be compared with that of the maintenance personnel entering the particle accelerator laboratory;

[0013] If the comparison is successful, the anti-opening state of the particle accelerator laboratory shielding door will be released until all maintenance personnel who entered the particle accelerator laboratory have left and the safety maintenance is completed.

[0014] Furthermore, the obtaining of the operating state of the particle accelerator and selecting a corresponding maintenance mode according to the operating state of the particle accelerator are specifically:

[0015] When the particle accelerator operates normally, routine maintenance is selected and maintenance personnel are determined;

[0016] When the particle accelerator is in faulty operation, fault maintenance is selected and maintenance personnel are determined;

[0017] When an operational accident occurs in a particle accelerator, emergency repair is selected and maintenance personnel are determined.

[0018] Furthermore, the face recognition of the maintenance personnel is performed based on the pre-trained face recognition model, specifically:

[0019] Obtain the maintenance personnel's face RGB image and perform preprocessing;

[0020] Extract facial features and auxiliary features based on the preprocessed face RGB image;

[0021] The extracted facial features, facial feature ratio features and expression features are converted into feature vectors in the embedding space, and the converted feature vectors are fused to obtain a comprehensive face feature vector;

[0022] Classify according to the comprehensive face feature vector using a classification function to determine whether a mask is worn;

[0023] Match the comprehensive face feature vector with the face feature vectors in the pre-constructed operation and maintenance database to determine the maintenance personnel.

[0024] Further, the facial features of the five sense organs include eyes, nose, mouth, chin, and facial contour;

[0025] The eyes include the interpupillary distance, eye shape, and eyelid angle; the nose includes the bridge width, length, and tip shape; the mouth includes the lip shape and the degree of curvature of the corners of the mouth; the chin includes the chin contour, width, and prominence; the facial contour includes the face shape and hairline.

[0026] Further, the auxiliary features are skin texture, local features of the five sense organs, symmetry and ratio of the overall face, and facial expressions.

[0027] Further, when extracting features based on the preprocessed face RGB image, delete the background, clothing, hair, and environmental noise features, and extract the facial features of the five sense organs and auxiliary features.

[0028] Further, the matching of the comprehensive face feature vector with the face feature vectors in the pre-constructed operation and maintenance database to determine the maintenance personnel is specifically as follows:

[0029] Calculate the Euclidean distances between all the face feature vectors in the pre-constructed operation and maintenance database and the comprehensive face feature vector respectively;

[0030] Take the face feature vector corresponding to the minimum calculated Euclidean distance as the face feature vector of the maintenance personnel, and the recognition is successful.

[0031] According to some embodiments, the second solution of the present invention provides a safety interlock control system for the operation and maintenance of a particle accelerator, adopting the following technical solution:

[0032] A safety interlock control system for the operation and maintenance of a particle accelerator, comprising:

[0033] A status recognition module, configured to obtain the operation status of the particle accelerator, select a corresponding maintenance mode according to the operation status of the particle accelerator, determine the maintenance personnel according to the selected maintenance mode, and at the same time trigger the interlock status of the shielding door of the particle accelerator laboratory;

[0034] A shielding door control module, configured to monitor the radiation level in the particle accelerator laboratory, and release the interlock status of the shielding door of the particle accelerator laboratory when the radiation level is lower than the safety threshold;

[0035] The face recognition module is configured to perform face recognition on maintenance personnel based on a pre-trained face recognition model, and count and store the identity information of the maintenance personnel entering the particle accelerator laboratory;

[0036] The maintenance status monitoring module is configured to monitor the maintenance status of the particle accelerator laboratory, the number of maintenance personnel, and the location information of the maintenance personnel in real time, and trigger the anti-opening state of the shielding door of the particle accelerator laboratory;

[0037] The exit detection module is configured to, after the maintenance is completed, perform face recognition and counting on the maintenance personnel passing by the entrance of the particle accelerator laboratory again, and compare with the identity information of the maintenance personnel entering the particle accelerator laboratory;

[0038] The shielding door anti-opening release module is configured to release the anti-opening state of the shielding door of the particle accelerator laboratory if the comparison is successful, until all the maintenance personnel entering the particle accelerator laboratory have left, and complete the safe maintenance.

[0039] According to some embodiments, the third aspect of the present invention provides a computer-readable storage medium.

[0040] A computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the steps in a safety interlock control method for particle accelerator operation and maintenance as described in the first aspect above.

[0041] According to some embodiments, the fourth aspect of the present invention provides a computer device.

[0042] A computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, it implements the steps in a safety interlock control method for particle accelerator operation and maintenance as described in the first aspect above.

[0043] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0044] The present invention realizes the automatic unlocking of the shielding door through radiation detection, can quickly respond to different maintenance modes, and ensures the efficiency of the maintenance process. It integrates intelligent management functions such as face recognition, radiation monitoring, and personnel counting, monitors the entry and exit of personnel in real time, effectively avoids misoperations or personnel detention problems caused by negligence, and ensures personnel safety. It provides three maintenance modes: regular maintenance, fault maintenance, and emergency repair according to the actual situation, can flexibly adapt to different maintenance needs, and meet the diverse requirements of regular operations and emergency management. It can automatically generate maintenance records and save relevant data, which is convenient for future traceability and management, and reduces the errors of manual records and management.

[0045] The present invention controls the unlocking state of the shielding door through automatic radiation detection, ensuring that it is only unlocked when the radiation level in the room drops below the safe level, thus avoiding unnecessary radiation risks. Three maintenance modes (routine, fault, and emergency repair) are introduced, and the corresponding maintenance processes are automatically matched through face recognition, realizing the intelligence and customization of the maintenance process. Through the dual confirmation method combining face recognition and automatic counting functions, it automatically checks whether there are any remaining personnel in the room after maintenance is completed, ensuring the safe evacuation of personnel in high-radiation areas. Using face recognition technology for identity confirmation and automatic counting effectively improves the safety of the maintenance process and the accuracy of data management, especially significantly enhancing the operation efficiency and safety in the case of faults and emergency repairs. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings forming a part of this invention are used to provide a further understanding of the invention. The schematic embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0047] Figure 1 is a flowchart of a safety interlock control method for the operation and maintenance of a particle accelerator in an embodiment of the present invention;

[0048] Figure 2 is a flowchart for training a face recognition model in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0050] It should be noted that the following detailed description is illustrative and is intended to provide a further description of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0051] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments of the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0052] In the case of no conflict, the embodiments in the present invention and the features in the embodiments may be combined with each other.

[0053] Embodiment 1

[0054] The present embodiment provides a safety interlocking control method for particle accelerator operation and maintenance. The present embodiment uses the method applied to a server as an example for illustration. It is understandable that the method can also be applied to a terminal, and can also be applied to a terminal, a server, and a system, and is implemented through the interaction between the terminal and the server. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud operation and maintenance databases, cloud computing, cloud functions, cloud storage, network servers, cloud communications, middleware services, domain name services, security services CDN, and big data and artificial intelligence platforms. The terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The terminal and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in this application. In the present embodiment, the method includes the following steps:

[0055] Obtaining the operating status of the particle accelerator, and selecting a corresponding maintenance mode according to the operating status of the particle accelerator, determining the maintenance personnel according to the selected maintenance mode, and triggering the interlocking state of the particle accelerator laboratory shielding door;

[0056] Monitor the radiation level in the particle accelerator laboratory and release the interlock state of the particle accelerator laboratory shielding door when the radiation level is below the safety threshold;

[0057] Perform facial recognition on maintenance personnel based on a pre-trained facial recognition model, count maintenance personnel entering the particle accelerator laboratory, and store their identity information;

[0058] Real-time monitoring of the maintenance status of the particle accelerator laboratory, the number of maintenance personnel, and the location information of the maintenance personnel, and triggering the anti-opening status of the particle accelerator laboratory shielding door;

[0059] When the maintenance is completed, the maintenance personnel passing through the entrance of the particle accelerator laboratory will be recognized and counted again, and the identity information will be compared with that of the maintenance personnel entering the particle accelerator laboratory;

[0060] If the comparison is successful, the anti-opening state of the particle accelerator laboratory shielding door will be released until all maintenance personnel who entered the particle accelerator laboratory have left and the safety maintenance is completed.

[0061] like Figure 1 As shown, the method described in this embodiment specifically includes:

[0062] Step 1: Monitor the operating status of the particle accelerator in real time, match the maintenance mode according to the operating status of the particle accelerator, record the maintenance mode to determine the maintenance personnel, and trigger the interlocking status of the particle accelerator laboratory shielding door.

[0063] The system is set up with three maintenance modes: routine maintenance, fault maintenance and accident repair, specifically:

[0064] When the particle accelerator operates normally, routine maintenance is selected and maintenance personnel are determined;

[0065] When the particle accelerator is in faulty operation, fault maintenance is selected and maintenance personnel are determined;

[0066] When an operational accident occurs in a particle accelerator, emergency repair is selected and maintenance personnel are determined.

[0067] Added manual emergency unlocking function: In accident repair mode, a manual unlocking function is provided to allow quick access to the accelerator room in an emergency.

[0068] Step 2: Radiation detection and unlocking of shielding doors: After confirming the maintenance mode, the particle accelerator will be shut down. When the particle accelerator is shut down, radiation detection will be automatically started to monitor the radiation level inside the particle accelerator laboratory in real time. When the radiation level drops below the safe level, the interlocking state of the particle accelerator laboratory shielding door will be released, and the shielding door of the particle accelerator laboratory will be automatically unlocked to allow maintenance personnel to enter.

[0069] Step 3: Face recognition: When the maintenance personnel enter the shielding door, the maintenance personnel are recognized based on the pre-trained face recognition model, and the maintenance personnel entering the particle accelerator laboratory are counted and their identity information is stored. Specifically:

[0070] Obtain the maintenance personnel's face RGB image and perform preprocessing;

[0071] Based on the preprocessed face RGB image, extract facial features, expression features and auxiliary features;

[0072] The extracted facial features, facial feature ratio features and expression features are converted into feature vectors in the embedding space, and the converted feature vectors are fused to obtain a comprehensive face feature vector;

[0073] According to the comprehensive feature vector of the face, a classification function is used to classify the face to determine whether the face is wearing a mask;

[0074] The maintenance personnel are determined by matching the comprehensive facial feature vector with the facial feature vector in the pre-built operation and maintenance database, specifically:

[0075] respectively calculating the Euclidean distances between all face feature vectors and the face comprehensive feature vector in the pre-built operation and maintenance database;

[0076] The face feature vector corresponding to the minimum Euclidean distance calculated is used as the face feature vector of the maintenance personnel, and the recognition is successful.

[0077] For malicious attacks on the face recognition system, the common methods are to attack the system through the network, find system vulnerabilities or forge masks, etc. The method of forging masks has advantages such as low cost and easy acquisition. This method mainly aims at face fraud by forging masks. In this embodiment, a face recognition model is proposed based on the principle of the convolutional neural network algorithm.

[0078] As Figure 2 shown, the training process of the face recognition model is specifically as follows:

[0079] Construct a database; use a large number of known masked face RGB sample images and unknown masked face RGB sample images to construct a database, and divide the database into a training set, a validation set, and a test set.

[0080] Design a face recognition model - MSAKCNN, and construct a face recognition model based on the CNN as the basic model.

[0081] Use the constructed database to train the face recognition model, that is, use the training set for training, and use the validation set to verify the face recognition model after training with the training set.

[0082] This model has five layers: an input layer, a hidden layer, a feature extraction layer, an optimal layer, and an output layer.

[0083] Input layer: In the initialization process, a data object or variable is assigned a starting value. The initial value of the equation is provided by Equation (1):

[0084]

[0085] where, M * (h g ) is a function that aggregates all input data features, h g is the face RGB image data set, λ is the feature in the face RGB image data set, is the data element. In a specific example, the total number of components is 2, and the data set length is 7553 RGB.

[0086] Hidden layer: Preprocess the face RGB image output by the input layer. The main function is to eliminate errors, specifically:

[0087] R * =[h g -ε(n)] (2);

[0088] where, ε(n) is the error value, and R * is the preprocessing layer.

[0089] ε(n) is the error value: It represents the unnecessary information or data defects in the data that interfere with the analysis results. That is to say, the preprocessing layer clears the interference features in the face RGB image output by the input layer. For example, it mainly includes: Image quality defects: Blurred, uneven brightness, reflection, low resolution. Statistical outliers: Accidentally entered incorrect data, such as non-face images or partially damaged files.

[0090] Feature extraction layer: Feature extraction involves transforming rough information into processable mathematical elements while preserving the data of the first batch of information. The algorithm takes into account features such as masked and unmasked image datasets.

[0091] M * (h g+1 ) = p g -d g (3);

[0092] Among them, p g is the key feature, d g is the noise feature, h g+1 is the face RGB image dataset.

[0093] That is to say, the feature extraction layer uses convolutional layers and pooling layers to extract multi-level key features from the preprocessed face RGB image using different filters. The key features include facial feature features and auxiliary features.

[0094] Function of the pooling layer: Pooling operations (such as max pooling) retain the key information in the face RGB image through the dimensionality reduction process, further removing noise features, thereby strengthening the attention to facial features.

[0095] Facial feature features include eyes, nose, mouth, chin, and facial contour;

[0096] Eyes include eye distance, eye shape, and eyelid angle; Nose includes nose bridge width, length, and tip shape; Mouth includes lip shape and mouth corner curvature; Chin includes chin contour, width, and prominence; Facial contour includes face shape and hairline.

[0097] Auxiliary features are skin texture, local features of facial features, symmetry and ratio of the overall face, and facial expressions.

[0098] Skin texture: For example, facial smoothness and wrinkle distribution. Local features of facial features: For example, high-detail information such as nostril and eyebrow boundaries. Symmetry and ratio of the overall face (such as the three-part-five-eyes ratio). Dynamic features (only used to enhance the function of the data): Facial expressions (smiling, frowning, etc.), used as an auxiliary verification means to reduce the dimensionality of the feature map and reduce the amount of calculation.

[0099] The noise features include background, clothing, hair, and environmental noise.

[0100] Through the multi-layer structure of the convolutional neural network, this algorithm can adaptively adjust which data is "desired" during the training process. During the training process, the network will automatically learn how to distinguish different individuals through specific features (such as the relative position of eyes, the shape of the nose, etc.), so as to select "desired data" in practical applications.

[0101] After feature extraction, the extracted facial feature points and auxiliary features of facial features will be converted into feature vectors in the embedding space. Each image is mapped to a point in a high-dimensional space, and points with closer distances represent similar facial features. In this case, the Euclidean distance between features is used to determine whether two images belong to the same class. This embedding-based feature matching further ensures that only feature data related to identity recognition is selected.

[0102] Optimal layer: That is, the classification layer. The classification network selects the class with the highest output value. When the classification neural network and the prediction neural network are integrated into a hybrid system, the classification neural network becomes very powerful. The model proposed in the present invention can detect attack behaviors in the classification layer:

[0103]

[0104] Among them, M * is the classification function, CM * is the classification, detect attacks.

[0105] Output layer: After processing the classification result of the optimal layer through the output layer - fully connected layer, it is output.

[0106] Optimization algorithm: Dynamically adjust the parameters of the CNN to optimize feature extraction and classification performance. Through iterative search, find the optimal hyperparameter settings.

[0107] During the training process of this face recognition model, the weights are adjusted continuously through backpropagation to automatically select the most effective features. Through multiple iterations and learning of the training dataset, the network can finally identify which features are helpful for the recognition task.

[0108] Exception handling mechanism: For input errors or fraudulent behaviors, the model triggers a security alarm. Use threshold detection to identify malicious operations or identity theft.

[0109] Automatic counting and personnel management: When a person enters the accelerator room, the system will record the information of all entering personnel and automatically count. The system updates the number of personnel in real time to ensure the consistency of the information of entering and exiting personnel and prevent personnel from staying.

[0110] Step 4: Monitor the maintenance status of the particle accelerator laboratory, the number of maintenance personnel, and the location information of the maintenance personnel in real time, and trigger the anti-opening status of the particle accelerator laboratory shielding door.

[0111] Step 5: Personnel inventory and safety confirmation after maintenance: After the maintenance is completed, the system will use facial recognition and automatic counting functions to confirm that all entrants have exited the accelerator room to ensure that no one is left in the high radiation area.

[0112] If the confirmation fails, return to step 4 to continue monitoring the location information and number of maintenance personnel; after confirmation, the system re-locks the shielding door and ends this maintenance process.

[0113] Optimize alarm prompts: Introduce voice alarms or visual alarm prompts to ensure that stranded personnel receive reminders in the shortest time possible to enter the room, further improving safety. Integrated operation records: The system can integrate operation log recording functions to facilitate the traceability and management of the maintenance process and provide data support for subsequent safety reviews.

[0114] Embodiment 2

[0115] This embodiment provides a safety interlocking control system for particle accelerator operation and maintenance, including:

[0116] A state recognition module is configured to obtain the operating state of the particle accelerator, select a corresponding maintenance mode according to the operating state of the particle accelerator, determine a maintenance personnel according to the selected maintenance mode, and trigger an interlocking state of a particle accelerator laboratory shielding door;

[0117] A shielding door control module is configured to monitor the radiation level in the particle accelerator laboratory and release the interlocking state of the shielding door of the particle accelerator laboratory when the radiation level is lower than a safety threshold;

[0118] A face recognition module is configured to perform face recognition on maintenance personnel based on a pre-trained face recognition model, and count and store identity information of maintenance personnel entering the particle accelerator laboratory;

[0119] A maintenance status monitoring module is configured to monitor the maintenance status of the particle accelerator laboratory, the number of maintenance personnel and the location information of the maintenance personnel in real time, and trigger the anti-opening state of the shielding door of the particle accelerator laboratory;

[0120] The exit detection module is configured to perform facial recognition again and count the maintenance personnel passing through the entrance of the particle accelerator laboratory after the maintenance is completed, and compare the facial recognition information with the identity information of the maintenance personnel entering the particle accelerator laboratory;

[0121] The shield door anti-opening release module is configured to release the anti-opening state of the shield door of the particle accelerator laboratory when the comparison is successful, until all maintenance personnel who have entered the particle accelerator laboratory have left, completing the safety maintenance.

[0122] Among them, a manual emergency unlocking function is also set in the status recognition module. In the accident emergency repair mode, the manual emergency unlocking function is provided to quickly enter the particle accelerator laboratory in case of emergency.

[0123] An alarm unit is also set in the maintenance status monitoring module to ensure that the detained personnel receive reminders within the shortest time of entering the room, further improving safety.

[0124] The system is also provided with an operation record log module, which integrates operation log recording for the traceability and management of the maintenance process, providing data support for subsequent safety reviews.

[0125] The examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the content disclosed in the first embodiment above. It should be noted that the above modules, as part of the system, can be executed in a computer system such as a set of computer-executable instructions.

[0126] In the above embodiments, the descriptions of each embodiment have their own focuses. For parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0127] The proposed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the above module division is only a logical function division. In actual implementation, there can be other division methods. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0128] Embodiment Three

[0129] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the steps in a safety interlock control method for particle accelerator operation and maintenance as described in the first embodiment above.

[0130] Embodiment Four

[0131] This embodiment provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps in a safety interlock control system for particle accelerator operation and maintenance as described in the first embodiment above.

[0132] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories and optical memories, etc.) that contain computer-usable program code.

[0133] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0134] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0135] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are performed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0136] Those of ordinary skill in the art can understand that all or part of the processes in the above-described method embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-described method embodiments. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0137] Although the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, it is not a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications or deformations that can be made without creative efforts on the basis of the technical solution of the present invention are still within the protection scope of the present invention.

Claims

1. A safety interlocking control method for particle accelerator operation and maintenance, characterized in that: include: Obtaining the operating status of the particle accelerator, and selecting a corresponding maintenance mode according to the operating status of the particle accelerator, determining the maintenance personnel according to the selected maintenance mode, and triggering the interlocking state of the particle accelerator laboratory shielding door; Monitor the radiation level in the particle accelerator laboratory and release the interlock state of the particle accelerator laboratory shielding door when the radiation level is below the safety threshold; Perform facial recognition on maintenance personnel based on a pre-trained facial recognition model, count maintenance personnel entering the particle accelerator laboratory, and store their identity information; Real-time monitoring of the maintenance status of the particle accelerator laboratory, the number of maintenance personnel, and the location information of the maintenance personnel, and triggering the anti-opening status of the particle accelerator laboratory shielding door; When the maintenance is completed, the maintenance personnel passing through the entrance of the particle accelerator laboratory will be recognized and counted again, and the identity information will be compared with that of the maintenance personnel entering the particle accelerator laboratory; If the comparison is successful, the anti-opening state of the particle accelerator laboratory shielding door will be released until all maintenance personnel who entered the particle accelerator laboratory have left and the safety maintenance is completed.

2. A safety interlocking control method for particle accelerator operation and maintenance as claimed in claim 1, characterized in that: The step of obtaining the operating status of the particle accelerator and selecting a corresponding maintenance mode according to the operating status of the particle accelerator is specifically as follows: When the particle accelerator operates normally, routine maintenance is selected and maintenance personnel are determined; When the particle accelerator is in faulty operation, fault maintenance is selected and maintenance personnel are determined; When an operational accident occurs in a particle accelerator, emergency repair is selected and maintenance personnel are determined.

3. A safety interlocking control method for particle accelerator operation and maintenance as claimed in claim 1, characterized in that: The face recognition of the maintenance personnel based on the pre-trained face recognition model is specifically performed as follows: Obtain the maintenance personnel's face RGB image and perform preprocessing; Extract facial features and auxiliary features based on the preprocessed face RGB image; The extracted facial features, facial feature ratio features and expression features are converted into feature vectors in the embedding space, and the converted feature vectors are fused to obtain a comprehensive face feature vector; According to the comprehensive feature vector of the face, a classification function is used to classify the face to determine whether the face is wearing a mask; The maintenance personnel are determined by matching the comprehensive facial feature vector with the facial feature vector in the pre-built operation and maintenance database.

4. A safety interlocking control method for particle accelerator operation and maintenance as claimed in claim 3, characterized in that: The facial features include eyes, nose, mouth, chin, and facial contour; The eyes include the distance between the eyes, the shape of the eyes, and the angle of the eyelids; the nose includes the width, length, and shape of the nose bridge; the mouth includes the shape of the lips and the curvature of the mouth corners; the chin includes the chin outline, width, and protrusion; the facial contour includes the face shape and hairline.

5. A safety interlocking control method for particle accelerator operation and maintenance as claimed in claim 3, characterized in that: The auxiliary features are skin texture, local features of facial features, overall symmetry and ratio of the face, and facial expressions.

6. A safety interlocking control method for particle accelerator operation and maintenance as claimed in claim 3, characterized in that: When performing feature extraction based on the preprocessed face RGB image, the background, clothing, hair and environmental noise are deleted, and the facial features and auxiliary features are extracted.

7. A safety interlocking control method for particle accelerator operation and maintenance as claimed in claim 3, characterized in that: The maintenance personnel are determined by matching the comprehensive facial feature vector with the facial feature vector in the pre-built operation and maintenance database, specifically: respectively calculating the Euclidean distances between all face feature vectors and the face comprehensive feature vector in the pre-built operation and maintenance database; The facial feature vector corresponding to the minimum Euclidean distance is calculated and used as the facial feature vector of the maintenance personnel, and the recognition is successful.

8. A safety interlocking control system for particle accelerator operation and maintenance, characterized in that: include: A state recognition module is configured to obtain the operating state of the particle accelerator, select a corresponding maintenance mode according to the operating state of the particle accelerator, determine a maintenance personnel according to the selected maintenance mode, and trigger an interlocking state of a particle accelerator laboratory shielding door; A shielding door control module is configured to monitor the radiation level in the particle accelerator laboratory and release the interlocking state of the shielding door of the particle accelerator laboratory when the radiation level is lower than a safety threshold; A face recognition module is configured to perform face recognition on maintenance personnel based on a pre-trained face recognition model, and count and store identity information of maintenance personnel entering the particle accelerator laboratory; A maintenance status monitoring module is configured to monitor the maintenance status of the particle accelerator laboratory, the number of maintenance personnel and the location information of the maintenance personnel in real time, and trigger the anti-opening state of the shielding door of the particle accelerator laboratory; The exit detection module is configured to perform facial recognition again and count the maintenance personnel passing through the entrance of the particle accelerator laboratory after the maintenance is completed, and compare the facial recognition information with the identity information of the maintenance personnel entering the particle accelerator laboratory; The shielding door anti-opening release module is configured to release the anti-opening state of the particle accelerator laboratory shielding door if the comparison is successful, until all maintenance personnel entering the particle accelerator laboratory have left and the safety maintenance is completed.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of a safety interlocking control method for particle accelerator operation and maintenance as described in any one of claims 1 to 7 are implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the safety interlocking control method for particle accelerator operation and maintenance as described in any one of claims 1-7 are implemented.

Images