A method and device for detecting three-dimensional dose distribution of accelerator beam loss

Abstract

The present invention provides a method and device for detecting the three-dimensional dose distribution of accelerator beam loss, comprising a vacuum tube wall for running beam particles, at least four optical fibers evenly distributed on the outside of the vacuum tube wall, and a data acquisition component and a data processing component connected to the optical fibers in sequence; the data acquisition component is configured to sample and obtain a beam loss timing signal, and the data processing component is configured to process the beam loss timing signal to obtain a three-dimensional beam loss dose image of the optical fiber. The three-dimensional energy distribution of beam loss monitored by the detection method and device of the present invention monitors the beam loss dose distribution from a global perspective, laying the foundation for subsequent exploration of the beam loss generation mechanism. In addition, the present invention takes into account both the accuracy and efficiency of the calculation through two-step interpolation: circumferential interpolation and cubic spline interpolation.

Description

Technical Field

[0001] The present invention belongs to the fields of accelerator radiation protection and accelerator beam measurement technology, and particularly relates to a method and device for detecting three-dimensional dose distribution of accelerator beam loss. Background Art

[0002] Beam loss is an inevitable phenomenon in accelerator operation. It not only damages the accelerator's materials but also impacts the health of experimental personnel. Therefore, comprehensive monitoring of beam loss in particle accelerators is crucial.

[0003] Traditional beam loss monitoring focuses on key locations. For example, dual-pin or scintillator probes are placed at locations prone to beam loss, allowing the beam loss dose to be detected at known locations. Alternatively, Cerenkov fibers are placed against the vacuum tube wall, allowing the beam loss location and dose to be detected in a known direction. Both approaches provide only a localized view of beam loss and fail to provide a comprehensive view of beam loss, enabling analysis of its causes and better protection of equipment and personnel.

[0004] Therefore, it is urgent to design a detection method for the three-dimensional dose distribution of accelerator beam loss to monitor the beam loss dose distribution from a global perspective. Summary of the Invention

[0005] The object of the present invention is to provide a method and device for detecting three-dimensional dose distribution of accelerator beam loss, so as to monitor the beam loss dose distribution from a global perspective.

[0006] To achieve the above-mentioned objectives, the present invention provides a device for detecting the three-dimensional dose distribution of accelerator beam loss, comprising a vacuum tube wall for running beam particles, at least four optical fibers evenly distributed on the outside of the vacuum tube wall, and a data acquisition component and a data processing component connected to the optical fibers in sequence; the data acquisition component is configured to sample and obtain a beam loss timing signal, and the data processing component is configured to process the beam loss timing signal to obtain a three-dimensional beam loss dose image of the optical fiber.

[0007] There are 12 optical fibers evenly distributed on the outside of the vacuum tube wall, and the optical fibers are Cerenkov fibers. The data acquisition component is connected to the upstream ends of all the optical fibers in the beam running direction.

[0008] The data acquisition component includes a front-end signal conditioning unit and an ADC module connected in sequence. The front-end signal conditioning unit is used to convert the optical signal into a voltage signal, and the ADC module is used to sample and obtain a beam loss timing signal.

[0009] The ADC module uses a 4-channel ADC with a sampling rate of 5 GHz to cyclically collect voltage signals of all optical fibers through time division multiplexing technology.

[0010] The data processing component includes a preprocessing device and a reconstruction processing device connected in sequence; the preprocessing device is configured to preprocess the beam loss timing signal to obtain a digital signal of the beam loss position and loss dose of the optical fiber; the reconstruction processing device is configured to reconstruct the digital signals of the beam loss position and loss dose of multiple optical fibers from digital signals expanded along the Z-axis direction into a three-dimensional beam loss dose image through a three-dimensional reconstruction algorithm.

[0011] The pre-processing device is an FPGA, and the reconstruction processing device is an ARM processor.

[0012] In another aspect, the present invention provides a method for detecting three-dimensional dose distribution of accelerator beam loss, comprising:

[0013] S1: Build a device to detect the three-dimensional dose distribution based on the accelerator beam loss mentioned above;

[0014] S2: Use the data acquisition component to set sampling to obtain the beam loss timing signal;

[0015] S3: Using the data processing component, a three-dimensional beam loss dose image of the optical fiber is obtained according to the beam loss timing signal.

[0016] Step S3 specifically includes:

[0017] S31: using a preprocessing device, preprocessing the beam loss timing signal to obtain a digital signal of the beam loss position and loss dose of the optical fiber;

[0018] S32: using a reconstruction processing device and a three-dimensional reconstruction algorithm, reconstructing the digital signals of the beam loss positions and loss doses of the plurality of optical fibers from the digital signals expanded along the Z-axis direction into a three-dimensional beam loss dose image.

[0019] In step S31, according to the optical fiber bundle loss position and the dose detection algorithm, the time difference between the trigger signal time of the beam and the time when the optical fiber receives the signal and the optical fiber bundle loss position satisfy the following formula:

[0020] ,

[0021] in, is the time difference between the trigger signal time of the beam and the time when the optical fiber receives the signal, t1 is the trigger signal time, and t2 is the arrival time of the signal at the upstream end of the optical fiber; c is the speed of light, n is the refractive index of the optical fiber, and L is the relative distance from the optical fiber bundle loss position to the starting point of the upstream end of the optical fiber.

[0022] The step S32 specifically includes:

[0023] A1: The digital signals of the beam loss position and loss dose of different optical fibers are used as the different optical fiber angles digital signal , perform circumferential interpolation on the angle dimension of the signal on the XY plane to obtain the first interpolation angle The first interpolated signal ;

[0024] A2: For each XY plane, at the first interpolation angle The first interpolated signal Further interpolation is performed on the basis to obtain a continuous second interpolation signal;

[0025] A3: Based on the continuous second interpolation signal of different XY planes , signal interpolation is performed along the Z-axis direction to obtain a continuous three-dimensional signal along the Z-axis direction, and the three-dimensional beam damage dose image is obtained by integration.

[0026] In step A1, a circumferential difference is performed to obtain a first interpolation angle The first interpolated signal , specifically including:

[0027] A11: A Gaussian kernel function with good adaptability to angular periodicity is selected as the radial basis function;

[0028] A12: Divide the circumferential 360° into multiple first interpolation angles , and solve the weight coefficient of each fiber according to the radial basis function , i is the fiber number;

[0029] A13: According to the weight coefficient of each optical fiber and different fiber angles on the same XY plane digital signal , get the first interpolation angle The first interpolated signal ;

[0030] Among them, the first interpolation angle The first interpolated signal for:

[0031] ,

[0032] in, is the jth first interpolation angle, j is the ordinal number of the first interpolation angle, the range of j is 0-359 and it is a continuous natural number, the first interpolation angle The range is 0°-359°; is the fiber angle, i is the fiber number; when i is 1-12, the fiber angle Take 0°, 30°, 60°, 90°, 120°, 150°, 180°, 210°, 240°, 270°, 300°, and 330° respectively; is the weight coefficient of the i-th optical fiber, is the digital signal of the i-th optical fiber;

[0033] In steps A2 and A3, cubic spline interpolation is used for interpolation.

[0034] The three-dimensional dose distribution detection method and device for accelerator beam loss in this invention monitors the three-dimensional energy distribution of beam loss from a global perspective, laying the foundation for subsequent research into the mechanisms of beam loss generation. Furthermore, the invention achieves a balance between computational accuracy and efficiency through a two-step interpolation process: circumferential interpolation and cubic spline interpolation. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a schematic diagram of the Cherenkov fiber laying method.

[0036] Figure 2 It is an overall framework diagram of the detection equipment for three-dimensional dose distribution of accelerator beam loss of the present invention.

[0037] Figure 3 The present invention is a flowchart of a front-end signal conditioning unit of a detection device for three-dimensional dose distribution of accelerator beam loss.

[0038] Figure 4 This is a waveform diagram of the signal received by the ARM module.

[0039] Figure 5 It is a schematic diagram of a one-dimensional digital signal expanded along the Z axis, where Figure 5 The left part is the original image; the right part is the expanded image.

[0040] Figure 6 It is a timing diagram of the original signal collected by a single optical fiber through a photomultiplier tube.

[0041] Figure 7 It is a three-dimensional beam loss dose image. DETAILED DESCRIPTION

[0042] The present invention will be further described below with reference to specific examples. It should be understood that the following examples are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0043] The method and device for detecting three-dimensional dose distribution of accelerator beam loss of the present invention are mainly based on the following principles:

[0044] This method utilizes at least four optical fibers evenly distributed along the outer wall of the vacuum tube. Compared to existing techniques that rely on only one fiber, this method enables three-dimensional restoration of data from multiple fibers, thereby providing a global monitoring result of beam loss dose distribution. Specifically, the electrical signals extracted from the 12 optical fibers are collected using a time-division multiplexed, four-channel, 5GHz acquisition card. The collected signals are then restored and reconstructed in three dimensions using an ARM Cortex-A9 processor.

[0045] like Figure 1 and Figure 2 As shown, the apparatus for detecting the three-dimensional dose distribution of accelerator beam loss of the present invention includes a vacuum tube wall 10 for carrying beam particles, at least four optical fibers 20 evenly distributed outside the vacuum tube wall, and a data acquisition component 30 and a data processing component 40 connected in sequence to the optical fibers. The data acquisition component 30 is configured to sample and obtain a beam loss timing signal, and the data processing component 40 is configured to process the beam loss timing signal to obtain a three-dimensional beam loss dose image of the optical fiber.

[0046] The beam particles are running at the inner center of the vacuum tube wall 10 . The beam movement direction is the Z axis direction. The XY plane is a plane perpendicular to the beam movement direction, where X is the horizontal direction and Y is the vertical direction.

[0047] In this embodiment, the number of optical fibers 20 evenly distributed on the outside of the vacuum tube wall is 12, and the optical fibers are Cerenkov fibers. Therefore, the 12 50-meter-long Cerenkov optical fibers are closely attached to the outside of the vacuum tube wall and evenly spaced 30° apart. The optical fiber at the top in the vertical direction is optical fiber No. 1. Rotating 30° clockwise, the optical fibers No. 2 and No. 3 appear in sequence, and so on, rotating 360° back to optical fiber No. 1.

[0048] The data acquisition assembly 30 connects to all optical fibers 20 at their upstream ends in the direction of beam travel to extract the Cherenkov optical signal. Extracting the upstream signal and then subtracting it from the timing signal is a currently mainstream processing method that avoids optical signal stacking. The upstream end of the optical fiber is the upstream end of the beam travel direction. The two are synonymous: upstream refers to the direction of the incoming beam, and downstream refers to the direction of the outgoing beam.

[0049] like Figure 2As shown, the data acquisition component 30 includes a front-end signal conditioning unit 31 and an ADC module 32, which are connected in sequence. The front-end signal conditioning unit 31 is used to convert the optical signal into a voltage signal, and the ADC module 32 is used to sample and obtain a beam loss timing signal. The data processing component 40 includes an FPGA 41 and an ARM processor 42, which are connected in sequence. The FPGA is a preprocessing device configured to preprocess the beam loss timing signal to obtain a digital signal of the optical fiber's beam loss position and loss dose. The ARM processor 42 is a reconstruction processing device configured to reconstruct the digital signals of the beam loss position and loss dose of multiple optical fibers from the digital signals expanded along the Z-axis direction into a three-dimensional beam loss dose image using a three-dimensional reconstruction algorithm.

[0050] The optical signal output by each Cherenkov fiber is converted by the front-end signal conditioning unit 31 into a voltage signal that meets the amplitude requirements of the ADC module 32. The signal is then sampled by the ADC at a 5 GHz sampling rate to generate a beam-loss timing signal, which is then transmitted to the FPGA. The FPGA preprocesses the beam-loss timing signal, converting it into a digital signal representing the fiber's beam-loss position and dose loss using a fiber-loss position and dose detection algorithm. The beam-loss position and dose loss signals are then transmitted to the ARM processor 42 (i.e., an ARM Cortex-A9) via an internal bus protocol, where the ARM processor 42 performs three-dimensional recovery of the dose loss.

[0051] In this embodiment, if Figure 3 As shown, in order to convert the optical signal into a voltage signal that meets the requirements, the front-end signal conditioning unit 31 includes a photomultiplier tube 311, a transimpedance amplifier 312, and an adjustable attenuator 313 connected in sequence. Thus, the optical signal at the upstream end of the optical fiber 20 is converted by the photomultiplier tube 311 into a negative current pulse signal with a minimum rising edge of approximately 2-3ns. The signal is then converted into a positive voltage pulse signal without broadening by the 3.9GHz high-gain bandwidth transimpedance amplifier 312. The signal is then conditioned by the high-bandwidth adjustable attenuator 313 to a voltage range that meets the amplitude requirements of the ADC module 32.

[0052] The ADC module 32 uses a 4-channel 5GHz sampling rate ADC to cyclically collect the voltage signals of all 12 optical fibers through time division multiplexing technology. The continuous analog signal with a minimum rising edge of 2ns is converted into a discrete digital signal with an interval of 0.2ns and then transmitted to the FPGA.

[0053] The principle of the fiber beam loss location and dose detection algorithm is that when a beam is lost at a certain fiber coverage location, an optical signal is generated at that location and transmitted to the starting point at the upstream end of the fiber. This transmission process takes time, so the time difference between the trigger signal time of the beam and the time the fiber receives the signal corresponds to different fiber beam loss locations. According to the fiber beam loss location and dose detection algorithm, the time difference between the trigger signal time of the beam and the time the fiber receives the signal and the fiber beam loss location satisfy the following formula:

[0054]

[0055] in, is the time difference between the trigger signal time of the beam and the time when the optical fiber receives the signal, t1 is the trigger signal time (that is, when the particle bunch reaches the starting point of the upstream end of the optical fiber), t2 is the arrival time of the signal at the upstream end of the optical fiber (that is, the time when the Cherenkov light signal generated at the intermediate position is received at the upstream end of the optical fiber), t2 is collected by the optical fiber, and t1 is directly provided by the accelerator; c is the speed of light, n is the refractive index of the optical fiber, and L is the relative distance from the optical fiber beam loss position to the starting point of the upstream end of the optical fiber. It represents the time it takes for the beam to move from the upstream end of the optical fiber through the vacuum tube wall 10 to the position where the optical fiber bundle is lost.

[0056] It should be noted that the above formula is only applicable to the signal drawn out from the upstream end. Therefore, the present invention needs to draw out the signal of the Cherenkov fiber at the upstream end in the direction of beam flow, and then take the difference between the upstream end signal and the timing signal to obtain the fiber bundle loss position.

[0057] As described above, digital signals indicating the fiber's beam loss position and lost dose are transmitted via an internal bus protocol to an ARM processor 42 (i.e., an ARM Cortex-A9), where the lost dose is restored in three dimensions. Since there are 12 optical fibers evenly distributed along the outer wall of the vacuum tube 10, 12 one-dimensional digital signals along the Z-axis are transmitted to the ARM processor 42. The signal's horizontal coordinate, Z_n, represents the fiber's beam loss position along the Z-axis, converted from the fiber's original time-domain data. The vertical coordinate represents the signal intensity, or lost dose. The relationship between fiber signal intensity and lost dose requires actual calibration and the creation of a calibration curve. One known method involves inserting a fluorescent screen at a specific location to cause complete beam loss at that location, measuring the signal amplitude at that location, and then varying the beam energy multiple times to obtain multiple amplitudes and plot a calibration curve. In other words, the signal intensity of each time-domain signal can be considered a one-dimensional projection of the beam loss along the fiber it resides on, reflecting the beam loss dose distribution along the line where that fiber resides.

[0058] The ARM processor 42 is configured to reconstruct a three-dimensional beam loss dose image from the digital signals of the beam loss positions and loss doses of the 12 optical fibers, as measured along the Z axis, using a three-dimensional reconstruction algorithm. The beam movement direction is the Z axis, and the XY plane is a plane perpendicular to the beam movement direction, where X is the horizontal direction and Y is the vertical direction.

[0059] Since the method of the present invention is used to measure the distribution of secondary particles generated by beam loss particles impacting the vacuum tube wall 10 , only the radiation distribution on the wall surface of the vacuum tube wall 10 is considered, and the radiation situation inside the vacuum tube wall 10 is not considered.

[0060] Specifically, the ARM processor 42 is configured as follows:

[0061] Step A1: The digital signals of the beam loss position and loss dose of different optical fibers are used as the different optical fiber angles. digital signal , perform circumferential interpolation on the angle dimension of the signal on the XY plane to obtain the first interpolation angle The first interpolated signal ;

[0062] Among them, the digital signal The independent variable in is the z-axis coordinate, so it corresponds to the position of the optical fiber bundle loss, and the digital signal The intensity of itself is the signal intensity, which corresponds to the loss dose of the optical fiber. Therefore, the first interpolation signal The intensity also corresponds to the loss dose of the optical fiber.

[0063] Perform circumferential difference to obtain the first interpolation angle The first interpolated signal , specifically including:

[0064] Step A11: Selecting a Gaussian kernel function that has good adaptability to angular periodicity as a radial basis function (RBF);

[0065] Gaussian kernel function for:

[0066] ,

[0067] in, is the shape parameter, and the empirical value is =1 / (30°× ), For any angle, is the fiber angle, that is, the angles of the 12 known optical fibers; r is the minimum arc length between the two angles, and the min function selects the minimum value between the two angles.

[0068] Due to the angle The value of has a cyclic characteristic. To avoid Gibbs oscillation, the minimum arc length r between the two angles needs to be used to calculate and correct the result.

[0069] Step A12: Divide the circumferential 360° into multiple first interpolation angles , and solve the weight coefficient of each fiber according to the radial basis function , i is the fiber number;

[0070] Among them, the first interpolation angle The range is 0°-359°, j is the ordinal number of the first interpolation angle, the range of j is 0-359 and is a continuous natural number, that is, the first interpolation angle The fiber number i ranges from 1 to 12 and is a continuous natural number.

[0071] Among them, the weight coefficient Determined by solving the following system of linear equations:

[0072] ,

[0073] in, is the Gaussian kernel function, is the angle, 、…、 is the first interpolation angle.

[0074] Step A13: Based on the weight coefficient of each fiber and different fiber angles on the same XY plane digital signal , get the first interpolation angle The first interpolated signal ;

[0075] Among them, the first interpolation angle The first interpolated signal for:

[0076]

[0077] in, is the jth first interpolation angle, j is the ordinal number of the first interpolation angle, the range of j is 0-359 and it is a continuous natural number, the first interpolation angle The range is 0°-359°; is the fiber angle, i is the fiber number, i ranges from 1 to 12 and is a continuous natural number; when i is 1 to 12, the fiber angle Take 0°, 30°, 60°, 90°, 120°, 150°, 180°, 210°, 240°, 270°, 300°, and 330° respectively; is the weight coefficient of the i-th optical fiber, is the digital signal of the i-th optical fiber.

[0078] That is, the fiber angle With the first interpolation angle There is overlap, that is, the fiber angle when i is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 The first interpolation angles when j is 0, 30, 60, 90, 120, 150, 180, 210, 240, 270, 300, and 330 degrees respectively . The function is 12 products sum.

[0079] Step A2: For each XY plane, at the first interpolation angle The first interpolated signal Further interpolation is performed on the basis to obtain a continuous second interpolation signal;

[0080] Step A2 specifically includes:

[0081] Step A21: Obtain the first interpolation angle The first interpolated signal The polar coordinates of the sampling points and the Cartesian coordinates of the sampling points The conversion formula is used to obtain the first interpolation signal in the Cartesian coordinate system. .

[0082] The conversion from polar coordinates to Cartesian coordinates is done here because the subsequent Z-axis direction is also a cubic spline interpolation of the Cartesian coordinate system, which facilitates algorithm sharing and debugging.

[0083] First interpolation signal The polar coordinates of the sampling points and the Cartesian coordinates of the sampling points The conversion formula is:

[0084]

[0085] in, 、 are the X-axis coordinate and Y-axis coordinate of the sampling point of the j-th first interpolation angle in Cartesian coordinates, is the jth first interpolation angle, j is the ordinal number of the first interpolation angle, R is the radius of the vacuum tube wall, and z is the Z-axis coordinate.

[0086] Step A22: For each sampling point of the first interpolation signal in the XY plane, the first interpolation signal in the Cartesian coordinate system is Perform secondary interpolation between adjacent sampling points to obtain a continuous second interpolation signal .

[0087] Wherein, the quadratic interpolation is preferably cubic spline interpolation.

[0088] When performing cubic spline interpolation, the continuous second interpolation signal Satisfies the following formula:

[0089]

[0090] in, are the X-axis coordinates of the sampling points of the 0th, 1st, 2nd, ..., 359th first interpolation angle in Cartesian coordinates, is a function of the second interpolation signal with the X-axis coordinate as the independent variable, is the second interpolation signal, and its value at each sampling point of the first interpolation angle is equal to the first interpolation signal .

[0091] The sampling point of the first interpolation angle is the interpolation position of step A1 above. In other words, the present invention can As a continuous function that changes with the X-axis coordinate of the sampling point Cartesian coordinate, each continuous Cartesian coordinate is obtained The intensity of the second interpolation signal at , the curve of each two boundary points is a cubic polynomial curve, and these curves meet the conditions of continuity and smoothness. The intensity of corresponds to the loss dose of the optical fiber, and the intensity of the second interpolation signal also corresponds to the loss dose of the optical fiber.

[0092] Step A3: Based on the continuous second interpolation signal of different XY planes , signal interpolation is performed along the Z-axis direction to obtain a continuous three-dimensional signal along the Z-axis direction, and the three-dimensional beam damage dose image is obtained by integration.

[0093] The signal interpolation is performed along the Z-axis direction, preferably using cubic spline interpolation.

[0094] like Figure 5 As shown, in this embodiment, along the Z-axis direction (i.e., the beam movement direction), 12 discrete signals (only 7 are shown in the figure) are sampled at intervals of 0.2 ns along the Z-axis direction. By conversion, the interval between two sampling points is 0.024 m.

[0095] For different XY planes, at angles The optical fiber signal at is S(z), then at the angle Interpolated signal along the Z axis for:

[0096]

[0097] Where m is the number of sampling points along the Z axis of the optical fiber, is the Z-axis coordinate of the m-th sampling point, is the interpolation signal along the Z axis, and its value at the sampling point is equal to the second interpolation signal .

[0098] When the sampling rate is 5 GHz, the number of sampling points m along the Z axis of the optical fiber is:

[0099]

[0100] Where L is the length of the optical fiber, c is the speed of light, and n is the refractive index.

[0101] Assume that the length of the optical fiber is L = 50m, the refractive index n = 1.5, and the speed of light c = 3× m / s. For example, if the ADC sampling rate is 5 GHz, m is 2083.

[0102] Based on the above-mentioned detection device for three-dimensional dose distribution of accelerator beam loss, the method for detecting three-dimensional dose distribution of accelerator beam loss includes the following steps:

[0103] Step S1: Building the above-mentioned accelerator beam loss three-dimensional dose distribution detection equipment;

[0104] Step S2: using the data acquisition component 30 to sample and obtain a beam loss timing signal;

[0105] The data acquisition component 30 includes a front-end signal conditioning unit 31 and an ADC module 32 connected in sequence. The front-end signal conditioning unit 31 is used to convert the optical signal into a voltage signal, and the ADC module 32 is used to sample and obtain a beam loss timing signal.

[0106] Step S3: using the data processing component 40 to process the beam loss timing signal to obtain a three-dimensional beam loss dose image of the optical fiber.

[0107] Step S3 specifically includes:

[0108] Step S31: using a preprocessing device to preprocess the beam loss timing signal to obtain a digital signal of the beam loss position and loss dose of the optical fiber;

[0109] Among them, the FPGA pre-processes the beam loss timing signal to convert the beam loss timing signal into a digital signal of the optical fiber's beam loss position and loss dose through the optical fiber beam loss position and dose detection algorithm.

[0110] According to the fiber bundle loss position and dose detection algorithm, the time difference between the trigger signal time of the beam and the time when the fiber receives the signal and the fiber bundle loss position satisfy the following formula:

[0111]

[0112] in, is the time difference between the trigger signal time of the beam and the time when the optical fiber receives the signal, t1 is the trigger signal time (that is, when the particle bunch reaches the starting point of the upstream end of the optical fiber), t2 is the arrival time of the signal at the upstream end of the optical fiber (that is, the time when the Cherenkov light signal generated at the intermediate position is received at the upstream end of the optical fiber), t2 is collected by the optical fiber, and t1 is directly provided by the accelerator; c is the speed of light, n is the refractive index of the optical fiber, and L is the relative distance from the optical fiber beam loss position to the starting point of the upstream end of the optical fiber. It represents the time it takes for the beam to move from the upstream end of the optical fiber through the vacuum tube wall 10 to the position where the optical fiber bundle is lost.

[0113] Step S32: using a reconstruction processing device and a three-dimensional reconstruction algorithm, reconstructing the digital signals of the beam loss positions and loss doses of the multiple optical fibers from the digital signals expanded along the Z-axis direction into a three-dimensional beam loss dose image.

[0114] The step S32 specifically includes:

[0115] Step A1: The digital signals of the beam loss position and loss dose of different optical fibers are used as the different optical fiber angles. digital signal , perform circumferential interpolation on the angle dimension of the signal on the XY plane to obtain the first interpolation angle The first interpolated signal ;

[0116] Step A2: For each XY plane, at the first interpolation angle The first interpolated signal Further interpolation is performed on the basis to obtain a continuous second interpolation signal;

[0117] Step A3: Based on the continuous second interpolation signal of different XY planes , signal interpolation is performed along the Z-axis direction to obtain a continuous three-dimensional signal along the Z-axis direction, and the three-dimensional beam damage dose image is obtained by integration.

[0118] The specific implementation process of steps A1 to A3 is exactly the same as above.

[0119] Experimental results:

[0120] In the experiment, the Cherenkov fiber used was a Thorlabs FP600ERT, which transmits optical signals with a wavelength range of 400-2000 nm. The PMT used was a Hamamatsu H10721-110, with a typical rise time of 0.57 ns, a received wavelength range of 230-700 nm, a maximum ripple noise of 0.1 mV, and a typical cathode brightness sensitivity of 105 μA / lm. Twelve 50-meter-long FP600ERT Cherenkov fibers were evenly attached to the outer wall of the linear accelerator's vacuum tube. The upstream end of each fiber was connected to a PMT, which converted the optical signal into a negative current. This was then converted to a positive voltage by a transimpedance amplifier. After the signal was adjusted to an appropriate voltage by an adjustable attenuator, it was converted to a digital signal by an ADC with a 5 GHz sampling rate.

[0121] FPGA uses optical fiber bundle loss position and dose detection algorithm Convert the raw signal into a digital signal containing the true physical position and loss dose.

[0122] Under normal circumstances, within a short period of time, without changing external conditions, the position and dose of the beam loss caused by the beam hitting the vacuum tube wall remain almost unchanged. Based on this, the signals generated by 12 optical fibers are collected using a 4-channel acquisition device through time-division multiplexing technology. Therefore, three consecutive beam clusters are required to detect a complete three-dimensional beam loss dose.

[0123] The refractive index of a 50m long optical fiber is n=1.5, and the speed of light is c=3× m / s, the ADC sampling rate is 5 GHz, and the digital signal collected by each optical fiber along the Z-axis direction of the particle movement is 2083 effective points; along the XY plane perpendicular to the direction of particle movement, there are 12 discrete points because 12 optical fibers are evenly distributed.

[0124] The ADC module 32 first needs to interpolate 12 discrete points at intervals of 30° to complete the XY plane points: select the Gaussian kernel function As radial basis function and calculate the weight coefficients of different fibers , then calculate the first interpolation angle The first interpolated signal : Step A21: Get the first interpolation angle The first interpolated signal The polar coordinates of the sampling points and the Cartesian coordinates of the sampling points The conversion formula is used to obtain the first interpolation signal in the Cartesian coordinate system. Step A22: For each sampling point of the first interpolation signal in the XY plane, the first interpolation signal in the Cartesian coordinate system is Perform secondary interpolation between adjacent sampling points to obtain a continuous second interpolation signal Step A3: The second interpolation signal is obtained based on the continuous second interpolation signal of different XY planes. , perform signal interpolation along the Z-axis direction to obtain a continuous three-dimensional signal along the Z-axis direction.

[0125] The final three-dimensional signal, namely the three-dimensional beam damage dose image, is as follows Figure 7 shown.

[0126] The method and device for detecting three-dimensional dose distribution of accelerator beam loss of the present invention have the following advantages:

[0127] (1) Traditional beam loss monitoring can only focus on beam loss locally, but cannot monitor beam loss from a global perspective to analyze the causes of beam loss and better protect the safety of machines and personnel. The three-dimensional energy distribution of beam loss monitored by the present invention monitors the beam loss dose distribution from a global perspective, laying the foundation for subsequent research on the mechanism of beam loss generation.

[0128] (2) 12 optical fibers are evenly distributed on the outside of the vacuum tube wall. The original signal collected by each optical fiber through the photomultiplier tube is as follows: Figure 6 As shown by the yellow line in the figure, the outgoing electrical signal is captured by a time-division multiplexed 4-channel 5GHz acquisition card and then transmitted to the ARM Cortex-A9 via an internal bus protocol. The ARM Cortex-A9 first uses a Gaussian kernel as the radial basis function and calculates weight coefficients for different optical fibers for circumferential interpolation. The 12 discrete points are then converted to a Cartesian coordinate system and then cubic spline interpolation is performed in the XY plane to convert the discrete points into multiple piecewise continuous functions. Circumferential interpolation operates on a circle and is highly adaptable to factors such as angular distortion. Circumferential interpolation converts 12 discrete points into 360 discrete points, while cubic spline interpolation simply makes the range between the two endpoints continuous, resulting in a continuous function. This two-step interpolation balances computational accuracy and efficiency. In the Z-axis direction of the particle motion, cubic spline interpolation is performed after calculating the initial position, ensuring continuous and smooth operation along the Z-axis. Finally, the collected signal is reconstructed into three dimensions to obtain Figure 7 The pattern shown.

[0129] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Various modifications are possible. Any simple, equivalent changes and modifications made in accordance with the claims and description of the present invention are within the scope of protection of the patent claims. Anything not fully described in this invention is conventional technology.

Claims

1. A device for detecting three-dimensional dose distribution of accelerator beam loss, characterized in that: The device comprises a vacuum tube wall for running beam particles, at least four optical fibers evenly distributed on the outside of the vacuum tube wall, and a data acquisition component and a data processing component sequentially connected to the optical fibers; The data acquisition component is configured to sample and obtain a beam loss timing signal, and the data processing component is configured to process and obtain a three-dimensional beam loss dose image of the optical fiber according to the beam loss timing signal; The data processing component includes a preprocessing device and a reconstruction processing device connected in sequence; the preprocessing device is configured to preprocess the beam loss timing signal to obtain a digital signal of the beam loss position and loss dose of the optical fiber; the reconstruction processing device is configured to reconstruct the digital signals of the beam loss position and loss dose of multiple optical fibers from digital signals expanded along the Z-axis direction into a three-dimensional beam loss dose image through a three-dimensional reconstruction algorithm; the preprocessing device is an FPGA, the reconstruction processing device is an ARM processor, and the Z-axis direction is the beam movement direction.

2. The device for detecting three-dimensional dose distribution of accelerator beam loss according to claim 1, characterized in that: There are 12 optical fibers evenly distributed on the outside of the vacuum tube wall, and the optical fibers are Cerenkov fibers. The data acquisition component is connected to the upstream ends of all the optical fibers in the beam running direction.

3. The device for detecting three-dimensional dose distribution of accelerator beam loss according to claim 1, characterized in that: The data acquisition component includes a front-end signal conditioning unit and an ADC module connected in sequence. The front-end signal conditioning unit is used to convert the optical signal into a voltage signal, and the ADC module is used to sample and obtain a beam loss timing signal.

4. The device for detecting three-dimensional dose distribution of accelerator beam loss according to claim 3, characterized in that: The ADC module uses a 4-channel ADC with a sampling rate of 5 GHz to cyclically collect voltage signals of all optical fibers through time division multiplexing technology.

5. A method for detecting three-dimensional dose distribution of accelerator beam loss, characterized in that: include: Step S1: constructing a detection device for three-dimensional dose distribution of accelerator beam loss according to any one of claims 1 to 4; Step S2: using a data acquisition component to sample and obtain a beam loss timing signal; Step S3: using a data processing component to process the beam loss timing signal to obtain a three-dimensional beam loss dose image of the optical fiber.

6. The method for detecting three-dimensional dose distribution of accelerator beam loss according to claim 5, characterized in that: Step S3 specifically includes: Step S31: using a preprocessing device to preprocess the beam loss timing signal to obtain a digital signal of the beam loss position and loss dose of the optical fiber; Step S32: using a reconstruction processing device and a three-dimensional reconstruction algorithm, reconstructing the digital signals of the beam loss positions and loss doses of multiple optical fibers from the digital signals expanded along the Z-axis direction into a three-dimensional beam loss dose image, where the Z-axis direction is the beam movement direction.

7. The method for detecting three-dimensional dose distribution of accelerator beam loss according to claim 6, characterized in that: In step S31, according to the optical fiber bundle loss position and the dose detection algorithm, the time difference between the trigger signal time of the beam and the time when the optical fiber receives the signal and the optical fiber bundle loss position satisfy the following formula: ， in, is the time difference between the trigger signal time of the beam and the time when the optical fiber receives the signal, t1 is the trigger signal time, and t2 is the arrival time of the signal at the upstream end of the optical fiber; c is the speed of light, n is the refractive index of the optical fiber, and L is the relative distance from the optical fiber bundle loss position to the starting point of the upstream end of the optical fiber.

8. The method for detecting three-dimensional dose distribution of accelerator beam loss according to claim 6, characterized in that: The step S32 specifically includes: Step A1: The digital signals of the beam loss position and loss dose of different optical fibers are used as the different optical fiber angles. digital signal , perform circumferential interpolation on the angle dimension of the signal on the XY plane to obtain the first interpolation angle The first interpolated signal ; Step A2: For each XY plane, at the first interpolation angle The first interpolated signal Further interpolation is performed on the basis to obtain a continuous second interpolation signal; Step A3: Based on the continuous second interpolation signal of different XY planes , perform signal interpolation along the Z-axis direction to obtain continuous three-dimensional signals along the Z-axis direction, and integrate them to obtain a three-dimensional beam damage dose image; The Z-axis direction is the beam movement direction, and the XY plane is a plane perpendicular to the beam movement direction, where X is the horizontal direction and Y is the vertical direction.

9. The method for detecting three-dimensional dose distribution of accelerator beam loss according to claim 8, characterized in that: In step A1, a circumferential difference is performed to obtain a first interpolation angle The first interpolated signal , specifically including: Step A11: Select a Gaussian kernel function that has good adaptability to angular periodicity as the radial basis function; Step A12: Divide the circumferential 360° into multiple first interpolation angles , and solve the weight coefficient of each fiber according to the radial basis function , i is the fiber number; Step A13: Based on the weight coefficient of each fiber and different fiber angles on the same XY plane digital signal , get the first interpolation angle The first interpolated signal ; Among them, the first interpolation angle The first interpolated signal for: ， in, is the jth first interpolation angle, j is the ordinal number of the first interpolation angle, the range of j is 0-359 and it is a continuous natural number, the first interpolation angle The range is 0°-359°; is the fiber angle, i is the fiber number; when i is 1-12, the fiber angle Take 0°, 30°, 60°, 90°, 120°, 150°, 180°, 210°, 240°, 270°, 300°, and 330° respectively; is the weight coefficient of the i-th optical fiber, is the digital signal of the i-th optical fiber; In steps A2 and A3, cubic spline interpolation is used for interpolation.

Images