Accelerator and wall surface temperature detection method thereof, storage medium and product

Abstract

The invention provides an accelerator and a wall surface temperature detection method thereof, a storage medium and a product, the accelerator is provided with an infrared thermal imaging device, the method comprises the following steps: in a temperature detection stage, acquiring external measurable data of the accelerator, the external measurable data at least comprising first acquisition data acquired by the infrared thermal imaging device; preprocessing the external measurable data to obtain external feature data; and inputting the external characteristic data into the trained temperature detection model, and processing the external characteristic data by the temperature detection model to obtain a detection result representing the wall surface temperature of the accelerator. The infrared thermal image device is configured outside the accelerator to collect the external infrared thermal image data of the accelerator, and the collected external measurable data is analyzed and processed in combination with the artificial intelligence model, so that the high-precision detection of the wall surface temperature of the accelerating tube is realized, and the operation stability of the accelerator is guaranteed.

Description

Technical Field

[0001] This application relates to the field of accelerators, and more particularly to an accelerator and a method for detecting the wall temperature thereon, a storage medium, and a product thereof. Background Technology

[0002] Electron beam irradiation technology has been widely used in fields such as irradiation sterilization, food preservation, material modification, and industrial flaw detection. As electron beam irradiation technology develops towards higher precision and higher power, the microwave loss and beam load effect of accelerators using S, C, X, and Ku bands (corresponding to frequencies of 2-18 GHz) have increased significantly, and the accelerator structure has become more precise. This places higher demands on the thermal management of accelerators: in order to avoid thermal deformation and field distribution distortion of the accelerator, and to ensure beam quality and equipment life, it is necessary to perform high-precision online monitoring of the accelerator tube wall temperature.

[0003] However, due to the metal shielding and strong electromagnetic interference that typically exist in the operating environment of accelerators, conventional non-contact temperature measurement methods are difficult to use for high-precision detection of the accelerator tube wall temperature. Summary of the Invention

[0004] This application provides an accelerator and its wall temperature detection method, storage medium, and product, aiming to perform high-precision detection of the wall temperature of the accelerator tube.

[0005] The technical solution of this application embodiment is implemented as follows: In a first aspect, embodiments of this application provide a method for detecting the wall temperature of an accelerator, wherein the accelerator is equipped with an infrared thermal imaging device, and the method includes: During the temperature detection phase, external measurable data of the accelerator are acquired, including at least the first acquisition data collected by the infrared thermal imaging device. The externally measurable data is preprocessed to obtain external feature data; The external feature data is input into the trained temperature detection model, and after processing by the temperature detection model, the detection result characterizing the wall temperature of the accelerator is obtained.

[0006] In some implementations, the method further includes: During the model training phase, external measurable data and wall temperature detection data of the accelerator are acquired; The externally measurable data is preprocessed to obtain external feature data; Using the external feature data as input data and the wall temperature detection data as label data, a sample training set is constructed; Based on the sample training set, train the temperature detection model to be trained to obtain the trained temperature detection model; Among them, the accelerator tube used for model training has a fiber optic grating temperature acquisition device pre-embedded inside, which is used to collect wall surface temperature detection data.

[0007] In some implementations, based on the sample training set, a temperature detection model to be trained is trained to obtain a trained temperature detection model, including: The external feature data is input into the temperature detection model to be trained to obtain the temperature prediction result; The deviation between the temperature prediction result and the corresponding wall temperature detection data is used as the loss value of the loss function. Based on the loss value, the parameters of the temperature detection model to be trained are optimized until the loss value converges to the target loss threshold, thus obtaining the trained temperature detection model.

[0008] In some implementations, the preprocessing of the externally measurable data to obtain external feature data includes: Feature extraction is performed on the first collected data to obtain infrared feature data; The infrared feature data and the second acquired data in the external measurable data are fused to obtain the external feature data.

[0009] In some implementations, the second acquired data includes one or more of the following: cooling system data, beam load data, and vacuum level data; wherein the cooling system data includes one or more of the following: coolant inlet temperature, coolant outlet temperature, and coolant flow rate; and the beam load data includes one or more of the following: beam power and pulse repetition frequency.

[0010] In some implementations, the first acquired data includes a temperature distribution image of the outer surface of the accelerator tube; the step of extracting features from the first acquired data to obtain infrared feature data includes: Feature extraction is performed on the temperature distribution image to obtain infrared feature data; The infrared feature data includes one or more of the following: maximum temperature, average temperature, temperature gradient, and hotspot area coordinates.

[0011] In some implementations, the method further includes: comparing the detection results with a set temperature threshold; If the detection result exceeds the set temperature threshold, an early warning message will be sent.

[0012] Secondly, embodiments of this application provide an accelerator configured with an infrared thermal imaging device; the accelerator further includes: a processor and a memory for storing a computer program capable of running on the processor; Wherein, when the processor is used to run a computer program, it performs the steps of the method as described in the first aspect.

[0013] Thirdly, embodiments of this application provide a storage medium storing a computer program, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0014] Fourthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0015] The technical solution provided in this application embodiment involves equipping the accelerator with an infrared thermal imaging device. During the temperature detection phase, external measurable data of the accelerator is acquired, including at least the first acquisition data collected by the infrared thermal imaging device. The external measurable data is preprocessed to obtain external feature data. This external feature data is then input into a trained temperature detection model, and after processing by the model, a detection result characterizing the accelerator wall temperature is obtained. Thus, this application embodiment, by equipping the accelerator with an infrared thermal imaging device to acquire external infrared thermal image data, and combining this with an artificial intelligence model to analyze and process the acquired external measurable data, achieves high-precision detection of the accelerator tube wall temperature, ensuring the accelerator's operational stability. Attached Figure Description

[0016] Figure 1 This is a schematic flowchart of the wall temperature detection method for an accelerator according to an embodiment of this application; Figure 2 This is a first structural schematic diagram of the acceleration structure according to an embodiment of this application; Figure 3 This is a schematic diagram of the second structure of the acceleration structure according to an embodiment of this application; Figure 4 This is a schematic diagram of the data acquisition and processing process of the accelerator in an embodiment of this application; Figure 5 This is a schematic diagram of the wall temperature detection device of the accelerator according to an embodiment of this application; Figure 6 This is a schematic diagram of the accelerator structure according to an embodiment of this application.

[0017] It should be noted that the terms "first" and "second" mentioned above are only used to distinguish between different options and do not represent the degree of superiority or inferiority of the options or their priority in the implementation process. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] Before providing a detailed description of the technical solution of this application, in order to enable those skilled in the art to fully understand the technical solution of this application, the relevant technical terms involved in the technical solution will be explained first.

[0020] An accelerator, also known as a particle accelerator, is a device that accelerates charged particles (such as electrons, protons, and ions) using electric fields, magnetic fields, or other forces, enabling the charged particles to acquire high energy. It typically consists of a particle source, an accelerating structure, a focusing system, a cooling system, a vacuum system, and a control system. It is widely used in fields such as irradiation sterilization, food preservation, material modification, and industrial flaw detection. Among these, the accelerating structure is the core component of the accelerator that enables the particle energy to be increased. It is used to establish an accelerating field so that charged particles continuously gain energy as they pass through the accelerating structure. Types of accelerating structures include, but are not limited to, electrostatic accelerating structures, high-frequency accelerating structures, and inductive accelerating structures.

[0021] Here, the accelerator in this application includes a high-frequency acceleration structure, that is, a high-frequency accelerating electric field is used to enable charged particles to gain energy.

[0022] Wall surface temperature: refers to the surface temperature of the wall of components with wall structures such as pipes, containers, and equipment cavities. It is a basic indicator reflecting the thermal state of such components.

[0023] Here, the wall temperature of the accelerator in this application specifically refers to the inner wall surface temperature of the accelerating tube in the accelerating structure. This inner wall surface temperature is directly affected by the heat generated by microwave loss and beam load effect during the operation of the accelerating tube.

[0024] Furthermore, embodiments of this application provide a method for detecting the wall temperature of an accelerator, such as... Figure 1 As shown, the method includes: Step 101: In the temperature detection stage, acquire external measurable data of the accelerator. The external measurable data includes at least the first acquisition data collected by the infrared thermal imaging device.

[0025] Step 102: Preprocess the externally measurable data to obtain external feature data.

[0026] Step 103: Input the external feature data into the trained temperature detection model. After processing by the temperature detection model, the detection results characterizing the wall temperature of the accelerator are obtained.

[0027] Here, the accelerator in this embodiment includes an acceleration structure, specifically a high-frequency acceleration structure. In some embodiments, the high-frequency acceleration structure is suitable for high-frequency electromagnetic fields in the S-band (corresponding to a frequency of 2-4 GHz), C-band (corresponding to a frequency of 4-8 GHz), X-band (corresponding to a frequency of 8-12 GHz), and Ku-band (corresponding to a frequency of 12-18 GHz), and particle acceleration can be achieved by matching microwave signals of the corresponding bands.

[0028] Here, the acceleration structure of the accelerator in this embodiment is as follows: Figure 2 The acceleration structure includes an acceleration tube 1. The acceleration tube 1 is a sealed structure used to establish a high-frequency accelerating electric field within a sealed environment, allowing charged particles to pass through and acquire energy, while isolating the external environment to maintain the vacuum conditions required for acceleration.

[0029] It should be noted that accelerators using S, C, X, and Ku band electron beams typically have thin-walled accelerating tubes with strict dimensional precision control to adapt to the characteristics of high-frequency electric fields. Therefore, the accelerating tube structure is sensitive to temperature changes; even a small temperature rise can trigger thermal deformation of the tube wall, thereby disrupting the uniformly distributed high-frequency accelerating electric field within the tube, leading to field distortion and ultimately affecting beam quality such as focusing and acceleration stability. Furthermore, long-term thermal stress can shorten the equipment's lifespan. Simultaneously, during accelerator operation, the high-frequency microwaves from the high-frequency accelerating electric field generate microwave losses as they propagate within the accelerating tube. The interaction between the electron beam and the inner wall of the accelerating tube also induces beam current loading effects. If the accelerator uses high-frequency electron beams such as S, C, X, and Ku, microwave losses and beam current loading effects will significantly increase, resulting in concentrated heating and a rapid temperature rise rate in the accelerating tube. Therefore, for accelerators using S, C, X, and Ku band electron beams, high-precision online monitoring of the accelerating tube wall temperature is necessary during operation to strictly manage thermal risks.

[0030] However, since the accelerator tube is a sealed structure, it is not possible to directly detect the temperature of the inner wall of the accelerator tube using contact temperature measurement methods. Furthermore, the operating environment of accelerators usually involves metal shielding and strong electromagnetic interference, and conventional non-contact temperature measurement methods such as microwave temperature measurement cannot achieve high-precision detection of the inner wall temperature of the accelerator tube due to the superposition of strong electromagnetic signals from the accelerator.

[0031] Based on this, the present application provides a temperature measurement method that combines externally measurable data with an artificial intelligence model, enabling non-contact, high-precision measurement of the wall temperature of the accelerator tube in a high-frequency accelerator structure. That is, the temperature of the accelerator tube wall is not directly detected on the inner wall of the accelerator tube, but is indirectly derived and predicted by collecting easily detectable data from the outside of the accelerator tube.

[0032] Specifically, such as Figure 2As shown in the embodiment of this application, the acceleration structure is configured with a sensor system. The sensor system includes an infrared thermal imaging device 207. The infrared thermal imaging device 207 is installed in an unobstructed position on the outer shell of the observable acceleration tube 1, and the lens of the infrared thermal imaging device 207 faces the acceleration tube 1 to collect infrared thermal image data on the surface of the outer shell of the acceleration tube, i.e., the first acquisition data.

[0033] In some embodiments, the infrared thermal imaging device 207 is an infrared thermal imager; correspondingly, the first acquisition data collected by the infrared thermal imaging device 207 includes a temperature distribution image of the outer surface of the accelerator tube 1. It should be noted that the embodiments of this application do not specifically limit the form of the infrared thermal imaging device 207. The infrared thermal imaging device 207 can be the infrared thermal imager described above, or it can be an infrared thermal imaging module or a composite sensing device with infrared thermal imaging function.

[0034] It should be noted that infrared thermal imaging devices receive infrared radiation signals emitted by target objects and convert these signals into quantifiable infrared thermal image data, enabling non-contact detection of the surface temperature distribution of the target object. Although infrared thermal imaging devices cannot directly detect the wall temperature of the accelerator tube, since the accelerator tube is a one-piece sealed structure, the heat generated inside the tube due to microwave loss, beam load, and other factors during operation will be transferred to the outer surface through heat conduction through the tube wall, forming a correlation between wall temperature, tube wall heat conduction, and outer surface temperature. That is, when the wall temperature of the accelerator tube changes, it will also indirectly cause a change in the outer surface temperature of the accelerator tube. Therefore, by using an infrared thermal imaging device to collect the directly measurable outer surface temperature distribution of the accelerator tube, the first collected data can serve as key basic data for predicting the wall temperature of the accelerator tube.

[0035] Understandably, when infrared thermal imaging devices acquire thermal images, they do not need to contact the accelerator tube, thus not affecting the normal operation of the accelerator. Furthermore, they can obtain three-dimensional temperature distribution data of the outer surface of the accelerator tube, rather than single-point temperature detection data, thereby improving the accuracy of wall temperature prediction.

[0036] It should be noted that although changes in wall temperature will indirectly cause changes in the first data collected by the infrared thermal imaging device, the first data alone cannot accurately reflect the pattern and magnitude of changes in wall temperature. Therefore, this application also introduces an artificial intelligence model. By constructing a temperature detection model with artificial intelligence algorithms as its core, and using external measurable data containing infrared thermal imaging data as model input, high-precision prediction of the wall temperature of the accelerating tube can be achieved.

[0037] In some embodiments, the temperature detection model is a deep learning model. The deep learning model is trained using a large number of training samples containing external measurable data and wall temperature data, so that the output of the trained temperature detection model converges to the actual wall temperature.

[0038] Accordingly, during the manufacturing stage of the accelerated structure, the temperature detection model needs to be pre-trained. For example, the method further includes: during the model training stage, acquiring external measurable data and wall temperature detection data of the accelerator; preprocessing the external measurable data to obtain external feature data; using the external feature data as input data and the wall temperature detection data as label data to construct a sample training set; and training the temperature detection model to be trained based on the sample training set to obtain the trained temperature detection model.

[0039] Here, the acceleration structure used for model training is as follows: Figure 3 As shown, with Figure 2 Compared to the acceleration structure shown, the sensor system also includes a fiber optic grating temperature acquisition device 205 (also known as a fiber optic grating temperature sensor). During the manufacturing of the acceleration tube 1 used for model training, the fiber optic grating temperature acquisition device 205 is pre-embedded inside the acceleration tube 1 to collect wall temperature detection data.

[0040] Here, a fiber Bragg grating temperature acquisition device 205 is pre-embedded at the temperature-sensitive point or heat concentration point inside the accelerating tube 1. The fiber Bragg grating temperature acquisition device 205 is attached to the inner wall of the accelerating tube. The fiber Bragg grating temperature acquisition device 205 has excellent radiation resistance and can work stably in the strong electromagnetic interference environment of the accelerator. When the wall temperature changes, it will affect the center wavelength of the internal grating and cause a shift. By detecting the wavelength shift, the detection result characterizing the wall temperature of the accelerating tube 1 can be obtained. Here, the accelerator tube 1 is equipped with a vacuum penetration component based on the pre-embedded position of the fiber Bragg grating temperature acquisition device 205. The signal line of the fiber Bragg grating temperature acquisition device 205 passes through the vacuum penetration component and exits from the vacuum tube. The accelerator uses this signal line to obtain the acquisition results of the fiber Bragg grating temperature acquisition device 205 and ensures the vacuum condition inside the accelerator tube 1.

[0041] It should be noted that the acceleration tube 1 used only for model training has a fiber optic temperature acquisition device 205 pre-embedded inside and a vacuum penetration component is provided; in order to ensure the vacuum requirement of the acceleration tube 1, the actual application acceleration structure does not have a fiber optic temperature acquisition device 205 pre-embedded and does not have a vacuum penetration component.

[0042] Understandably, during the model training phase, a sensor system including a fiber optic temperature acquisition device 205 and an infrared thermal imaging device 207 collects external measurable data and wall temperature detection data of the accelerator. Based on the acquisition time, the external feature data obtained from the processing of the external measurable data is paired with the wall temperature detection data to construct labeled sample data for model training. The label of the sample data is the wall temperature detection data collected by the fiber optic temperature acquisition device 205. After constructing a sample dataset including multiple sample data, the temperature detection model is trained using the sample dataset, so that the model output converges to the detection results of the fiber optic temperature acquisition device 205. Then, the trained temperature detection model is used to predict the wall temperature of accelerator tubes of the same specifications.

[0043] For example, training a temperature detection model to be trained based on a sample training set to obtain a trained temperature detection model includes: inputting external feature data into the temperature detection model to be trained to obtain a temperature prediction result; using the deviation between the temperature prediction result and the corresponding wall temperature detection data as the loss value of the loss function, and optimizing the parameters of the temperature detection model to be trained based on the loss value until the loss value converges to the target loss threshold to obtain a trained temperature detection model.

[0044] Among them, the temperature prediction result is the output result of the temperature detection model, which represents the prediction result of the wall temperature of the key position or region inside the accelerating tube. The key position or region is associated with the pre-embedded position of the fiber optic grating temperature acquisition device 205.

[0045] Understandably, the loss function is used to quantify the degree of deviation between the temperature prediction result and the actual wall temperature. During the training process, the temperature detection model learns under supervision and continuously iterates and optimizes the model parameters, so that the loss value of the loss function continuously decreases, and the model output converges. The loss function includes, but is not limited to, the mean squared error loss function and the mean absolute error loss function. The model parameter optimization algorithm includes, but is not limited to, the stochastic gradient descent method, the Adam optimization algorithm, and the adaptive momentum estimation algorithm. This application does not specifically limit the specific algorithms used in this embodiment.

[0046] It should be noted that, in order to improve the accuracy and robustness of the temperature detection model output under different operating conditions, during the model training phase, the external measurable data and wall temperature detection data of the accelerator were collected under different power and cooling conditions, thereby constructing a sample dataset covering different power and cooling conditions.

[0047] It should be noted that, in order to further improve the detection accuracy and precision of the wall temperature, the external measurable data of the accelerator obtained in this application embodiment includes not only the first acquisition data collected by the infrared thermal imaging device 207, but also the second acquisition data collected by other sensing devices of the sensor system. That is, this application embodiment specifically provides a temperature measurement method that combines external measurable multi-physics field coupling data with artificial intelligence model, so as to realize non-contact high-precision measurement of the wall temperature of the accelerator tube of the high-frequency acceleration structure.

[0048] In some embodiments, the second acquired data includes one or more of the following: cooling system data, beam load data, and vacuum level data.

[0049] In some embodiments, the cooling system data includes one or more of the following: coolant inlet temperature, coolant outlet temperature, and coolant flow rate.

[0050] In some embodiments, the beam load data includes one or more of the following: beam power and pulse repetition frequency.

[0051] Here, the accelerator also includes a cooling system, which relies on the circulation of coolant to remove the heat generated during the operation of the accelerator tube and maintain a stable wall temperature. Figure 2 and Figure 3 In the acceleration structure shown, the sensor system also includes an inlet flow sensor 201, an outlet flow sensor 202, an inlet temperature sensor 203, an outlet temperature sensor 204, and a vacuum sensor 206. The inlet flow sensor 201 and inlet temperature sensor 203 are located at the inlet of the cooling system, while the outlet flow sensor 202 and outlet temperature sensor 204 are located at the outlet of the cooling system. The inlet flow sensor 201 and outlet flow sensor 202 are used together to collect the coolant flow rate, the inlet temperature sensor 203 is used to collect the coolant inlet temperature, and the outlet temperature sensor 204 is used to collect the coolant outlet temperature. The vacuum sensor 206 is located on the vacuum tube 1 and is used to collect vacuum data characterizing the vacuum level inside the vacuum tube 1. Here, beam power characterizes beam energy intensity, which directly affects the amount of beam load generated inside the accelerating tube, and pulse repetition frequency characterizes the frequency of beam output, which directly affects the rate of heat accumulation inside the accelerating tube; in some embodiments, the sensor system also includes a beam power acquisition device (not shown in the figure) and a beam pulse counting device (not shown in the figure), the beam power acquisition device is used to acquire beam power, and the beam pulse counting device is used to acquire beam pulse frequency.

[0052] It is understood that, in addition to the first acquired data which has the highest correlation with the wall temperature, the aforementioned second acquired data also have varying degrees of correlation with the wall temperature, and each of the second acquired data is easily directly detectable external accelerator data. This embodiment of the application constructs multiphysics acquisition data by acquiring the first acquired data and one or more of the aforementioned second acquired data, and inputs the corresponding external feature data obtained from processing the multiphysics acquisition data into the temperature detection model, thereby achieving high-precision detection of the accelerator tube wall temperature. It is also understood that this embodiment of the application acquires external infrared thermal image data of the accelerator by configuring an infrared thermal imaging device outside the accelerator, and combines this with an artificial intelligence model to analyze and process the acquired external measurable data, thereby achieving high-precision detection of the accelerator tube wall temperature and ensuring the operational stability of the accelerator.

[0053] For example, preprocessing externally measurable data to obtain external feature data includes: extracting features from the first acquired data to obtain infrared feature data; and fusing the infrared feature data with the second acquired data from the externally measurable data to obtain external feature data.

[0054] Here, the above-mentioned preprocessing method for externally measurable data is used for preprocessing externally measurable data in the temperature detection stage and for preprocessing externally measurable data in the model training stage.

[0055] In some embodiments, the first acquisition data includes a temperature distribution image of the outer surface of the accelerator tube 1 acquired by the infrared thermal imaging device 207. Accordingly, feature extraction is performed on the first acquisition data to obtain infrared feature data, including: feature extraction of the temperature distribution image to obtain infrared feature data.

[0056] The infrared feature data includes one or more of the following: the highest temperature, average temperature, temperature gradient, and hot spot coordinates in the temperature distribution image.

[0057] Here, the external feature data needs to be converted into a data format that the temperature detection model can support for analysis and processing; in some embodiments, the infrared feature data is in vector form, which is generated by vector fusion of the infrared thermal images extracted from the first acquisition data; correspondingly, the external feature data is a comprehensive feature vector, which is generated by vector fusion of the infrared feature vector and the aforementioned second acquisition features.

[0058] In some embodiments, in order to eliminate the dimensional differences between infrared feature data and second acquired data and avoid imbalance in feature weight allocation, before performing feature fusion on the infrared feature data and the second acquired data in external measurable data to obtain external feature data, the method further includes: normalizing the infrared feature data and the second acquired data.

[0059] It should be noted that, in addition to making high-precision predictions of the wall temperature of the accelerating tube during the temperature detection stage, the accelerator in this application embodiment also constructs online learning sample data based on the collected data and detection results, and continuously optimizes the model to adapt to the aging degree of the accelerator based on the online learning sample data.

[0060] In some embodiments, after obtaining the detection result characterizing the wall temperature of the accelerator, the method further includes: comparing the detection result with a set temperature threshold; if the detection result exceeds the set temperature threshold, sending an early warning message.

[0061] Here, the temperature threshold is set to the safety threshold preset by the accelerator. If the wall temperature of the accelerator tube exceeds this safety threshold, the accelerator tube is currently at risk of thermal deformation of the tube wall.

[0062] Here, the warning message indicates that the accelerator tube is currently at risk of thermal deformation, reminding relevant personnel to handle the abnormal state of the accelerator in a timely manner. The forms of warning messages include, but are not limited to: audible and visual alarm signals, pop-up warning displays, and remote push notifications.

[0063] It is understood that during accelerator operation, in addition to detecting the wall temperature of the accelerator tube, this embodiment of the application also performs real-time thermal management based on the comparison results of the detection results and the set temperature threshold, thereby eliminating potential thermal anomalies during accelerator operation and forming a closed-loop management of detection-early warning-optimization.

[0064] exist Figure 2 and Figure 3The acceleration structure shown is also equipped with an intelligent operation and maintenance system, which includes a data processing device 301, a model processing device 302, and an online monitoring device 303. The data processing device 301 is connected to the inlet flow sensor 201, the outlet flow sensor 202, the inlet temperature sensor 203, the outlet temperature sensor 204, the vacuum sensor 206, and the infrared thermal imaging device 207, respectively. It is used to synchronously acquire data from each of the connected sensor systems during accelerator operation and to preprocess the acquired external measurable data. During the model training phase, the data processing device 301 is also connected to the fiber optic grating temperature acquisition device 205. The model processing device 302 carries a temperature detection model, which receives external feature data output by the data processing device 301 and analyzes and processes the external feature data based on the temperature detection model, outputting the detection results of the wall temperature at key locations or areas of the accelerator tube. The online monitoring device 303 acquires the wall temperature at key locations or areas calculated by the model processing device 302 in real time and compares it with a pre-set temperature threshold. When the calculated wall temperature exceeds the set temperature threshold, or the accelerator's operating time exceeds a set duration threshold, the online monitoring device 303 issues an early warning to avoid unplanned operational interruptions, thus transforming passive maintenance into intelligent operation and maintenance.

[0065] In some embodiments, the flowchart of the accelerator's data acquisition and processing is as follows: Figure 4 As shown. The original physical parameters of the accelerator's high-frequency acceleration structure are collected by a pre-embedded radiation-resistant fiber optic grating sensor, a cooling system temperature and flow sensor, a vacuum and beam load parameter sensor, and an infrared thermal imager. The original physical parameters collected by each sensor are input to the data acquisition and signal processing unit of the model processing device. The data acquisition and signal processing unit performs data fusion and feature vector extraction. The resulting external feature data is analyzed and processed by an artificial intelligence model, outputting the detection result characterizing the wall temperature of the accelerating tube. The final output detection result and all original physical parameters are input to the online monitoring device of the intelligent operation and maintenance system to achieve functions such as temperature monitoring, field distribution monitoring, threshold monitoring, safety interlocking, predictive maintenance, and life assessment. To implement the method of this application embodiment, this application embodiment also provides an accelerator wall temperature detection device, which corresponds to the above-described accelerator wall temperature detection method. The steps in the above-described accelerator wall temperature detection method embodiment are also fully applicable to this device embodiment.

[0066] like Figure 5As shown, the accelerator wall temperature detection device according to an embodiment of this application includes an acquisition module 501, a processing module 502, and a detection module 503. The acquisition module 501 is used to acquire external measurable data of the accelerator during the temperature detection stage. The external measurable data includes at least first acquisition data collected by an infrared thermal imaging device. The processing module 502 is used to preprocess the external measurable data to obtain external feature data. The detection module 503 is used to input the external feature data into a trained temperature detection model, and after processing by the temperature detection model, obtain the detection result characterizing the accelerator wall temperature. The accelerator is equipped with an infrared thermal imaging device.

[0067] In some embodiments, the acquisition module 501 is further configured to: acquire external measurable data and wall temperature detection data of the accelerator during the model training phase; correspondingly, during the model training phase, the processing module 502 is further configured to: preprocess the external measurable data to obtain external feature data; and construct a sample training set by using the external feature data as input data and the wall temperature detection data as label data. The accelerator used for model training has a fiber Bragg grating temperature acquisition device pre-embedded inside the accelerator tube, which is used to acquire the wall temperature detection data.

[0068] In some embodiments, the wall temperature detection device of the accelerator further includes a training module 504, which is used to: train the temperature detection model to be trained based on the sample training set, and obtain the trained temperature detection model.

[0069] In some embodiments, the training module 504 is specifically used to: input external feature data into the temperature detection model to be trained to obtain temperature prediction results; use the deviation between the temperature prediction results and the corresponding wall temperature detection data as the loss value of the loss function, and optimize the parameters of the temperature detection model to be trained based on the loss value until the loss value converges to the target loss threshold to obtain the trained temperature detection model.

[0070] In some embodiments, the processing module 502 is specifically used to: extract features from the first acquired data to obtain infrared feature data; and fuse the infrared feature data with the second acquired data in the external measurable data to obtain external feature data.

[0071] In some embodiments, the second acquired data includes one or more of the following: cooling system data, beam load data, and vacuum level data; wherein, the cooling system data includes one or more of the following: coolant inlet temperature, coolant outlet temperature, and coolant flow rate; and the beam load data includes one or more of the following: beam power and pulse repetition frequency.

[0072] In some embodiments, the first acquired data includes a temperature distribution image of the outer surface of the accelerator tube; correspondingly, the processing module 502 is specifically used to: extract features from the temperature distribution image to obtain infrared feature data; wherein, the infrared feature data includes one or more of the following data: maximum temperature, average temperature, temperature gradient, and hot spot area coordinates.

[0073] In some embodiments, the detection module 503 is further configured to: compare the detection result with a set temperature threshold; and send an early warning message if the detection result exceeds the set temperature threshold.

[0074] It should be noted that the accelerator wall temperature detection device provided in the above embodiments is only illustrated by the division of the above-described program modules when detecting the accelerator wall temperature. In practical applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device can be divided into different program modules to complete all or part of the processing described above. In addition, the accelerator wall temperature detection device and the accelerator wall temperature detection method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0075] Based on the hardware implementation of the above program modules, and in order to implement the wall temperature detection method of the accelerator in this application embodiment, this application embodiment also provides an accelerator, such as... Figure 6 As shown, accelerator 600 includes at least one processor 601, memory 602, user interface 603, and at least one network interface 604. The various components in accelerator 600 are coupled together via bus system 605. It can be understood that bus system 605 is used to implement communication between these components. In addition to a data bus, bus system 605 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 6 The general designated all buses as Bus System 605.

[0076] The user interface 603 may include a monitor, keyboard, mouse, trackball, click wheel, buttons, touchpad, or touch screen.

[0077] The memory 602 in this embodiment is used to store various types of data to support the operation of the accelerator 600. Examples of such data include any computer program used to operate on the accelerator 600.

[0078] The accelerator wall temperature detection method disclosed in this application can be applied to or implemented by the processor 601. The processor 601 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the accelerator wall temperature detection method can be completed by the integrated logic circuitry in the hardware of the processor 601 or by instructions in software form. The processor 601 can be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 601 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software modules can be located in a storage medium, specifically memory 602. The processor 601 reads information from memory 602 and, in conjunction with its hardware, completes the steps of the accelerator wall temperature detection method provided in the embodiments of this application.

[0079] In an exemplary embodiment, the accelerator 600 may be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), FPGAs, general-purpose processors, controllers, microcontrollers (MCUs), microprocessors, or other electronic components to perform the aforementioned accelerator wall temperature detection method.

[0080] It is understood that memory 602 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), EEPROM, ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Sync Link Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM). The memory 602 described in this application embodiment is intended to include, but is not limited to, these and any other suitable types of memory.

[0081] It should be noted that the acceleration structure of Accelerator 600 also includesFigure 2 The acceleration structure shown includes various systems and components; the acceleration structure of the accelerator 600 used for model training also includes... Figure 3 The fiber optic grating temperature acquisition device 205 shown is shown.

[0082] In some embodiments, the processor 601 and memory 602 are configured in the intelligent operation and maintenance system of the accelerator 600.

[0083] In an exemplary embodiment, this application also provides a storage medium, namely a computer storage medium, specifically a computer-readable storage medium, such as a memory 602 storing a computer program. This computer program can be executed by the processor 601 of the accelerator 600 to complete the steps described in the accelerator wall temperature detection method of this application embodiment. The computer-readable storage medium can be a ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disc, or CD-ROM, etc.

[0084] In an exemplary embodiment, this application also provides a computer program product, including a computer program that can be executed by the processor 601 of the accelerator 600 to perform the steps described in the method of this application embodiment.

[0085] It should be noted that terms such as "first" and "second" are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0086] Furthermore, the technical solutions described in the embodiments of this application can be combined arbitrarily without conflict.

[0087] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for detecting the wall temperature of an accelerator, characterized in that, The accelerator is equipped with an infrared thermal imaging device, and the method includes: During the temperature detection phase, external measurable data of the accelerator are acquired, including at least the first acquisition data collected by the infrared thermal imaging device. The externally measurable data is preprocessed to obtain external feature data; The external feature data is input into the trained temperature detection model, and after processing by the temperature detection model, the detection result characterizing the wall temperature of the accelerator is obtained.

2. The method according to claim 1, characterized in that, The method further includes: During the model training phase, external measurable data and wall temperature detection data of the accelerator are acquired; The externally measurable data is preprocessed to obtain external feature data; Using the external feature data as input data and the wall temperature detection data as label data, a sample training set is constructed; Based on the sample training set, train the temperature detection model to be trained to obtain the trained temperature detection model; Among them, the accelerator tube used for model training has a fiber optic grating temperature acquisition device pre-embedded inside, which is used to collect wall surface temperature detection data.

3. The method according to claim 2, characterized in that, Based on the aforementioned sample training set, a temperature detection model is trained to obtain a trained temperature detection model, including: The external feature data is input into the temperature detection model to be trained to obtain the temperature prediction result; The deviation between the temperature prediction result and the corresponding wall temperature detection data is used as the loss value of the loss function. Based on the loss value, the parameters of the temperature detection model to be trained are optimized until the loss value converges to the target loss threshold, thus obtaining the trained temperature detection model.

4. The method according to claim 1 or 2, characterized in that, The preprocessing of the externally measurable data to obtain external feature data includes: Feature extraction is performed on the first collected data to obtain infrared feature data; The infrared feature data and the second acquired data in the external measurable data are fused to obtain the external feature data.

5. The method according to claim 4, characterized in that, The second acquired data includes one or more of the following: cooling system data, beam load data, and vacuum level data; wherein, the cooling system data includes one or more of the following: coolant inlet temperature, coolant outlet temperature, and coolant flow rate; and the beam load data includes one or more of the following: beam power and pulse repetition frequency.

6. The method according to claim 4, characterized in that, The first acquired data includes a temperature distribution image of the outer surface of the accelerator tube; the step of extracting features from the first acquired data to obtain infrared feature data includes: Feature extraction is performed on the temperature distribution image to obtain infrared feature data; The infrared feature data includes one or more of the following: maximum temperature, average temperature, temperature gradient, and hotspot area coordinates.

7. The method according to claim 1, characterized in that, The method further includes: comparing the detection results with a set temperature threshold; If the detection result exceeds the set temperature threshold, an early warning message will be sent.

8. An accelerator, characterized in that, The accelerator is equipped with an infrared thermal imaging device; the accelerator also includes a processor and a memory for storing computer programs that can run on the processor; When the processor is used to run a computer program, it executes the steps of the method according to any one of claims 1 to 7.

9. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images