Method and device for measuring the transverse working point of a beam
By adaptive smoothing and normalization of the Schottky signal, combined with exponential averaging and Kalman filtering, the problems of low signal-to-noise ratio and rapid frequency changes in transverse operating point measurement in proton therapy synchrotrons were solved, achieving real-time and stable transverse operating point measurement.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANGHAI INSTITUTE OF APPLIED PHYSICS CHINESE ACADEMY OF SCIENCES
- Filing Date
- 2026-01-14
- Publication Date
- 2026-05-29
AI Technical Summary
In existing proton therapy synchrotrons, the low beam intensity leads to a poor signal-to-noise ratio, the transverse operating point measurement is easily misled by noise, and it is difficult to measure accurately when the cyclotron frequency changes rapidly during the acceleration phase. The longitudinal and transverse components in the spectrum are mixed, resulting in severe interference.
By sampling, adaptively smoothing, normalizing, and resampling the Schottky signal, combined with exponential averaging and Kalman filtering, an optimal estimate of the reference transverse operating point is formed, suppressing noise interference and tracking the transverse operating point in real time.
It achieves real-time and stable transverse operating point measurement under low signal-to-noise ratio conditions, has strong noise resistance, adapts to rapid changes in cyclotron frequency, is not affected by occasional noise, has low computational load, and is suitable for compact devices.
Smart Images
Figure CN122110199A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of particle accelerator beam diagnostics technology, and more specifically to a method and apparatus for measuring the transverse operating point of a beam. Background Technology
[0002] Synchrotrons require real-time monitoring of the beam's transverse operating point (also known as the transverse oscillation operating point) during operation to ensure the beam operates within a permissible stable operating region and avoid transverse resonance. Traditionally, this involves applying external excitation to the beam (e.g., using a slow-extraction kicker to scan the beam), then using a beam position detector (BPM) to acquire the response signal and perform spectral analysis to extract the peak position and obtain the transverse operating point.
[0003] The above method has three typical limitations. First, in compact devices such as proton therapy synchrotrons, the beam intensity is low, and the signal-to-noise ratio (SNR) of the transverse Schottky signal is often at the level of approximately -20 dB to -10 dB, making the simple "finding the highest peak" method very susceptible to noise misleading. Second, during the acceleration (energy boost) phase, the cyclotron frequency of the beam changes rapidly over a very short time, causing the transverse sidebands to shift or even stretch along the absolute frequency axis, making it impossible to rely on long-term averaging to improve the SNR, and the traditional multi-frame averaging method is no longer applicable. Third, the spectrum often contains both longitudinal and transverse components, coherence lines, stray electromagnetic interference, etc., causing the location of a single peak to not always correspond to the true transverse operating point.
[0004] Therefore, there is an urgent need in this field for a lateral operating point measurement method that can output in real time during the power-up process at a low signal-to-noise ratio of -20dB and is robust to occasional noise and transient distortion. Summary of the Invention
[0005] The purpose of this invention is to provide a method and apparatus for measuring the transverse working point of a beam, so as to solve the problems existing in the prior art.
[0006] To achieve the above objectives, the present invention provides a method for measuring the transverse operating point of a beam, comprising:
[0007] The Schottky signal is sampled in chronological order. A preset time period is set, and the Schottky signal collected in each preset time period is taken as a frame of data. The current preset time period is taken as the current frame, the previous preset time period is taken as the previous frame, and the next preset time period is taken as the next frame.
[0008] The power spectrum of the current frame is obtained based on the current frame data;
[0009] Based on the power spectrum of the current frame, the power spectrum of the target frequency segment of the current frame under the unified physical frequency axis is obtained, and used as the target power spectrum of the current frame.
[0010] An adaptive smoothing filter is applied to the target power spectrum of the current frame to obtain the smoothed target power spectrum of the current frame.
[0011] The frequency coordinates of the smoothed target power spectrum of the current frame are normalized to dimensionless operating point coordinates to obtain the normalized power spectrum of the current frame.
[0012] The normalized power spectrum of the current frame is resampled to obtain the resampled power spectrum of the current frame;
[0013] Based on the resampled power spectrum of the current frame, a reference power spectrum of the current frame is formed using an exponential averaging mechanism; based on the reference power spectrum of the current frame, a reference lateral operating point of the current frame is determined.
[0014] Based on the resampled power spectrum of the current frame, determine the candidate lateral operating point of the current frame;
[0015] The current frame and several preceding frames are combined into a frame sequence. Based on the reference lateral working point and candidate lateral working point of each frame in the frame sequence, the optimal estimate of the lateral working point of the current frame is determined.
[0016] Optionally, adaptive smoothing filtering is performed on the target power spectrum of the current frame to obtain a smoothed target power spectrum of the current frame, specifically including:
[0017] Predetermine the frequency range corresponding to the region that does not belong to the transverse beam signal, and use it as the suppression frequency range;
[0018] The power spectral density values within the suppression frequency range of the target power spectrum of the current frame are suppressed to obtain the suppressed power spectrum of the current frame;
[0019] A smoothing window is pre-determined based on the widening of the lateral sidebands of the Schottky signal;
[0020] Based on the smoothing window, the suppressed power spectrum of the current frame is smoothed and filtered to obtain the smoothed target power spectrum of the current frame.
[0021] Optionally, the frequency coordinates of the smoothed target power spectrum of the current frame are normalized to dimensionless operating point coordinates to obtain the normalized power spectrum of the current frame, specifically including:
[0022] Divide each frequency coordinate of the target power spectrum after smoothing and filtering in the current frame by the cyclotron frequency of the current frame to obtain the dimensionless value corresponding to that frequency coordinate, and use the decimal part of the dimensionless value as the operating point coordinate corresponding to that frequency coordinate.
[0023] The coordinates of each working point are folded into the interval [0, 0.5) to obtain the normalized power spectrum of the current frame.
[0024] Optionally, the normalized power spectrum of the current frame is resampled to obtain the resampled power spectrum of the current frame, specifically including:
[0025] The interval [0, 0.5) is divided into equal intervals with a preset interval to obtain the coordinates of multiple discrete points;
[0026] The power spectral density values corresponding to the coordinates of each working point of the normalized power spectrum of the current frame are projected onto the coordinates of each discrete point in a soft-assignment manner to obtain the resampled power spectral density values corresponding to the coordinates of each discrete point. The coordinates of each discrete point and its corresponding resampled power spectral density values form the resampled power spectrum of the current frame.
[0027] Optionally, the power spectral density values corresponding to the working point coordinates of the normalized power spectrum of the current frame are projected onto the discrete point coordinates using a soft-assignment method to obtain the resampled power spectral density values corresponding to the discrete point coordinates. Specifically, this includes:
[0028] For each working point coordinate of the normalized power spectrum of the current frame, determine the coordinates of two discrete points adjacent to the working point coordinate; based on the distance between the working point coordinate and the two adjacent discrete point coordinates, determine the allocation weight of the working point coordinate to the two adjacent discrete point coordinates; based on the allocation weight of the working point coordinate to the two adjacent discrete point coordinates, allocate the power spectral density value corresponding to the working point coordinate to the two adjacent discrete point coordinates.
[0029] The weights assigned to each discrete point coordinate in the normalized power spectrum of the current frame are summed to obtain the cumulative weight of that discrete point coordinate.
[0030] The power spectral density values assigned to each working point coordinate of the normalized power spectrum of the current frame and the power spectral density values of each discrete point coordinate are summed to obtain the cumulative power spectral density value of that discrete point coordinate.
[0031] The resampled power spectral density value of each discrete point coordinate is obtained by dividing the cumulative power spectral density value of that discrete point coordinate by the cumulative weight of that discrete point coordinate.
[0032] Optionally, based on the reference power spectrum of the current frame, the reference lateral operating point of the current frame is determined, specifically including:
[0033] Extract the coordinates of the discrete point corresponding to the maximum power spectral density value of the reference power spectrum of the current frame, and use them as the coordinates of the peak discrete point of the current frame;
[0034] Median filtering is performed based on the peak discrete point coordinates of the current frame and the peak discrete point coordinates of several previous frames to obtain the median filtered peak discrete point coordinates of the current frame, which serves as the reference horizontal working point for the current frame.
[0035] Based on the resampled power spectrum of the current frame, candidate lateral operating points for the current frame are determined, specifically including:
[0036] The overall score of a discrete point coordinate is determined based on the power spectral density value corresponding to the coordinate of each discrete point in the resampled power spectrum of the current frame and the distance between the coordinate of that discrete point and the reference lateral working point of the current frame.
[0037] The coordinates of the discrete point with the highest overall score are selected as the candidate discrete point coordinates for the current frame.
[0038] Median filtering is performed on the candidate discrete point coordinates of the current frame and the candidate discrete point coordinates of several previous frames to obtain the median-filtered candidate discrete point coordinates of the current frame, which are then used as candidate lateral working points for the current frame.
[0039] Optionally, the optimal estimate of the lateral working point of the current frame is determined based on the reference lateral working point and candidate lateral working points of each frame in the frame sequence, specifically including:
[0040] Determine the predicted value and uncertainty of the lateral working point of the current frame;
[0041] The fused observations and observation variance of the lateral working point of the current frame are determined based on the reference lateral working point and candidate lateral working point of each frame in the frame sequence.
[0042] The predicted value and uncertainty of the horizontal working point of the current frame are corrected by using the fused observation value of the horizontal working point of the current frame, so as to obtain the optimal estimate and uncertainty of the horizontal working point of the current frame.
[0043] Optionally, the fused observations and observation variance of the lateral working point of the current frame are determined based on the reference lateral working point and candidate lateral working points of each frame in the frame sequence, specifically including:
[0044] Calculate the variance, mean square error, or median absolute deviation of the reference lateral working point and candidate lateral working point for each frame in the frame sequence, and use them as the noise variance of the reference lateral working point and candidate lateral working point for the current frame, respectively.
[0045] The weights of the reference lateral operating point and the candidate lateral operating point of the current frame and the observation variance of the lateral operating point of the current frame are determined based on the noise variance of the reference lateral operating point and the candidate lateral operating point of the current frame.
[0046] The reference lateral working point and the candidate lateral working point of the current frame are fused based on their weights to obtain the fused observation value of the lateral working point of the current frame.
[0047] Optionally, the predicted value and uncertainty of the lateral operating point in the current frame are corrected using the fused observations of the lateral operating point in the current frame, so as to obtain the optimal estimate and uncertainty of the lateral operating point in the current frame, specifically including:
[0048] The Kalman filter gain for the current frame is determined based on the predicted value of the lateral working point and the fused observation value.
[0049] The optimal estimate of the lateral operating point for the current frame is determined based on the predicted value of the lateral operating point, the fused observation value, and the Kalman gain.
[0050] The uncertainty of the optimal estimate of the lateral working point for the current frame is determined based on the uncertainty of the predicted value of the lateral working point for the current frame and the Kalman gain.
[0051] Another aspect of the present invention provides a measuring device for the transverse operating point of a beam, comprising:
[0052] The sampling module is used to sample the Schottky signal in chronological order, set a preset time period, and take the Schottky signal collected in each preset time period as a frame of data; wherein, the current preset time period is taken as the current frame, the previous preset time period is taken as the previous frame, and the next preset time period is taken as the next frame.
[0053] The first acquisition module is used to acquire the power spectrum of the current frame based on the current frame data;
[0054] The second acquisition module is used to acquire the power spectrum of the target frequency segment of the current frame under the unified physical frequency axis based on the power spectrum of the current frame, and use it as the target power spectrum of the current frame.
[0055] A smoothing filter module is used to perform adaptive smoothing filtering on the target power spectrum of the current frame to obtain the smoothed target power spectrum of the current frame.
[0056] The normalization module is used to normalize the frequency coordinates of the smoothed and filtered target power spectrum of the current frame into dimensionless operating point coordinates to obtain the normalized power spectrum of the current frame.
[0057] The resampling module is used to resample the normalized power spectrum of the current frame to obtain the resampled power spectrum of the current frame.
[0058] The reference point determination module is used to form a reference power spectrum of the current frame based on the resampled power spectrum of the current frame using an exponential averaging mechanism; and to determine the reference lateral operating point of the current frame based on the reference power spectrum of the current frame.
[0059] The candidate point determination module is used to determine candidate lateral working points of the current frame based on the resampled power spectrum of the current frame.
[0060] The estimation module is used to form a frame sequence by combining the current frame and several preceding frames, and to determine the optimal estimate of the lateral working point of the current frame based on the reference lateral working point and candidate lateral working point of each frame in the frame sequence. Attached Figure Description
[0061] Figure 1 A flowchart illustrating a method for measuring the transverse operating point of a beam according to an embodiment of the present invention;
[0062] Figure 2 This is a structural block diagram of a beam transverse operating point measuring device according to an embodiment of the present invention. Detailed Implementation
[0063] The preferred embodiments of the present invention are given below with reference to the accompanying drawings and described in detail.
[0064] like Figure 1 As shown, this embodiment of the invention provides a method for measuring the transverse operating point of a beam, which includes the following steps:
[0065] S100: Sample the Schottky signal in chronological order, set a preset time period, and take the Schottky signal collected in each preset time period as a frame of data; wherein, the current preset time period is taken as the current frame, the previous preset time period is taken as the previous frame, and the next preset time period is taken as the next frame.
[0066] The Schottky signal is a radio frequency signal generated by the spontaneous current fluctuations of the beam in an accelerator. It consists of the incoherent superposition of the transverse and longitudinal motions of a large number of particles. Its power spectrum contains not only transverse operating point information, but also beam parameters such as emittance, chromaticity, and synchronous oscillation frequency. Therefore, the transverse operating point of the beam can be measured non-destructively through the Schottky signal.
[0067] The Schottky signal is a weak radio frequency signal (typically in the microvolt to millivolt range) induced by the beam current on the pickup electrode. It cannot be directly sampled and requires preprocessing before sampling. Specifically, a wideband pickup electrode can be used to cover a broad frequency band. The radio frequency signal on the pickup electrode is sequentially passed through a bandpass filter and low-noise amplification before entering an ADC (analog-to-digital converter). The ADC converts the analog signal to a digital signal, thus completing the sampling of the Schottky signal. The resulting Schottky signal is a digital signal in the time domain, which is a digital signal-time series. The digital signal represents the signal value of the Schottky signal, and the time represents the sampling time corresponding to that signal value. Since each frame of data is a Schottky signal acquired within a preset time period, each frame of data is a signal-time series, and each frame has the same length.
[0068] The preset time period is set to no more than 1ms to ensure that the cyclotron frequency drift during the energy-boosting phase does not cause significant broadening within the same time period.
[0069] S200: Obtain the power spectrum of the current frame based on the current frame data.
[0070] Step S200 specifically includes:
[0071] S210: Perform a Fast Fourier Transform (FFT) on the current frame data to obtain the spectrum of the current frame;
[0072] S220: Calculate the power spectrum of the current frame based on the spectrum of the current frame.
[0073] The current frame data is in the time domain. By performing an FFT on it, it can be transformed from the time domain to the frequency domain, yielding the spectrum of the current frame. By squaring and normalizing the spectrum, the power spectrum can be obtained. The horizontal axis of the power spectrum is frequency, and the vertical axis is the power spectral density value. That is, the power spectrum is a power density-frequency sequence. The power spectrum can amplify the power characteristics of the signal while weakening the interference of isolated noise spikes (the noise power distribution is more uniform). Therefore, it is more suitable for identifying the location of lateral sidebands in low signal-to-noise ratio signals.
[0074] The methods for calculating FFT and power spectrum are well known in the field, and their specific steps and principles will not be elaborated here.
[0075] S300: Based on the power spectrum of the current frame, obtain the power spectrum of the target frequency segment of the current frame under the unified physical frequency axis, and use it as the target power spectrum of the current frame.
[0076] Step S300 specifically includes:
[0077] S310: Extract the power spectrum of the target frequency segment from the power spectrum of the current frame to obtain the extracted power spectrum of the current frame; where the target frequency segment is the frequency range covered by the center frequency and effective bandwidth of the transverse detector; in this way, only the power spectrum of the frequency segment corresponding to the bandwidth of the transverse detector can be retained, and frequency regions unrelated to the target frequency segment can be filtered out to reduce invalid noise sources.
[0078] S320: Reconstruct the frequency axis of the power spectrum after the current frame is truncated to obtain the target power spectrum of the current frame under a unified physical frequency axis.
[0079] In step S320, the mirror position of the actual power spectrum in the baseband can be calculated based on the sampling conditions of the Schottky signal (including sampling method and sampling rate, etc.), and the frequency coordinates are restored to the true physical frequency, so that the power spectrum of all frames is under a unified physical frequency axis. Step S320 specifically includes:
[0080] Predetermine the approximate physical frequency band (center frequency and bandwidth) and sampling rate of the Schottky signal;
[0081] Based on the relationship between sampling rate and bandwidth, the baseband mirror interval corresponding to the physical frequency band is determined according to the spectrum mirroring pattern (i.e., which Nyquist interval it is in and the start and end frequency index in the FFT result).
[0082] The spectrum data within the baseband mirror interval is shifted and flipped as necessary, and each FFT frequency point is mapped back to its corresponding physical frequency, forming a lookup table from the baseband frequency point index to the physical frequency.
[0083] By rearranging and labeling the frequency coordinates using a lookup table, the frequency coordinates of the power spectrum of the current frame can be mapped to a unified physical frequency axis, thus enabling the reconstruction and unification of the frequency axis under the condition of undersampling of bandpass signals.
[0084] S400: Perform adaptive smoothing filtering on the target power spectrum of the current frame to obtain the smoothed target power spectrum of the current frame.
[0085] Step S400 specifically includes:
[0086] S410: Predetermine the frequency range corresponding to the region that does not belong to the transverse beam signal, and use it as the suppression frequency range;
[0087] S420: Suppress the power spectral density values within the suppression frequency range of the target power spectrum of the current frame (e.g., set to 0 or a very low value) to obtain the suppressed power spectrum of the current frame.
[0088] S430: Predetermines the smoothing window based on the widening of the lateral sidebands of the Schottky signal;
[0089] S440: Smooth the suppressed power spectrum of the current frame based on the smoothing window to obtain the smoothed target power spectrum of the current frame.
[0090] In step S430, the widening of the lateral sideband can be calculated using the following formula:
[0091]
[0092] in, To broaden, Harmonic number, The cyclotron frequency, In order to disperse, The sliding phase factor, For horizontal working points The decimal part, For color products, For synchronous oscillation amplitude, This is the synchronous oscillation frequency.
[0093] The formula for calculating a smooth window is:
[0094]
[0095] in, To smooth the window, This represents the frequency resolution of the power spectrum. The smoothing window is positively correlated with the broadening; the larger the broadening, the larger the smoothing window, and vice versa, thus maintaining resolution.
[0096] In step S440, smoothing filtering refers to, for each frequency point of the power spectrum, taking that frequency point as the center, taking (N) values to the left and right of that frequency point. f The power spectral density values of -1) / 2 adjacent frequency points are weighted and averaged (e.g., equal-weighted average or Gaussian-weighted average). The weighted average power spectral density value is used as the power spectral density value of that frequency point in the smoothed power spectrum. Through adaptive smoothing filtering, the random noise floor can be reduced without significantly sacrificing frequency resolution.
[0097] S500: Normalize the frequency coordinates of the smoothed target power spectrum of the current frame to dimensionless operating point coordinates to obtain the normalized power spectrum of the current frame.
[0098] The transverse operating point of the beam is physically represented by two sidebands surrounding a harmonic cyclotron frequency. Their position on the absolute frequency axis shifts as the cyclotron frequency changes. To identify the operating point using the same coordinate system even with constantly changing cyclotron frequencies, the power spectrum can be converted from absolute frequency coordinates to operating point coordinates. Step S500 specifically includes:
[0099] S510: Divide each frequency coordinate of the smoothed target power spectrum of the current frame by the cyclotron frequency of the current frame to obtain the dimensionless value corresponding to the frequency coordinate, and use the decimal part of the dimensionless value as the operating point coordinate corresponding to the frequency coordinate; the cyclotron frequency of the current frame can be directly obtained through the existing accelerator control system.
[0100] S520: Fold the coordinates of each working point into the interval [0, 0.5) to obtain the normalized power spectrum of the current frame.
[0101] Step S520 specifically includes:
[0102] For each working point coordinate, if the working point coordinate is less than 0.5, the working point coordinate remains unchanged; if the working point coordinate is greater than or equal to 0.5, the working point coordinate is replaced by 1 minus the working point coordinate, thereby mirroring the working point coordinate to the interval [0, 0.5).
[0103] From a physical perspective, the upper and lower transverse sidebands are symmetrically distributed with respect to their respective harmonics. For example, if the cyclotron frequency is 1MHz and the fractional part of the transverse operating point is approximately 0.3, then two sidebands will appear near a certain harmonic: the lower sideband... This corresponds to q=0.7; (The text above is incomplete and likely refers to a different topic.) , corresponding to q=0.3, where, Let q be the cyclotron frequency and q be the fractional part of the transverse operating point. Following the folding rules described above, 0.3 remains unchanged, and 0.7 is mapped to 1 − 0.7 = 0.3. Thus, the upper and lower sidebands are effectively merged into the same operating point coordinate q = 0.3. Through this operation, all transverse sidebands can be processed within a unified interval [0, 0.5), eliminating their dependence on the current cyclotron frequency and the selection of upper and lower sidebands.
[0104] S600: Resample the normalized power spectrum of the current frame to obtain the resampled power spectrum of the current frame.
[0105] Step S600 specifically includes:
[0106] S610: Divide the interval [0, 0.5) into equal intervals with a preset interval to obtain the coordinates of multiple discrete points;
[0107] S620: The power spectral density values corresponding to the coordinates of each working point of the normalized power spectrum of the current frame are projected onto the coordinates of each discrete point in a soft-assignment manner to obtain the resampled power spectral density values corresponding to the coordinates of each discrete point. The coordinates of each discrete point and its corresponding resampled power spectral density values form the resampled power spectrum of the current frame.
[0108] In step S610, the preset interval can be selected as needed, for example, it can be set to 1 / 2048, so that 1024 discrete point coordinates can be obtained in the interval [0, 0.5).
[0109] In step S620, soft allocation means that the power spectral density value corresponding to each operating point coordinate is distributed to the coordinates of two discrete points adjacent to that operating point coordinate according to distance. The advantage of doing so is that it can avoid the spectral peak jumping back and forth between discrete points due to slight changes in the cyclotron frequency. For example, suppose a certain operating point coordinate is q orig Its corresponding power spectral density value is E orig q orig The coordinates q of the adjacent discrete point k and q k+1 Between, q orig With q k The distance between them is d left =q orig -q k q orig With q k+1 The distance between them is d right =q k+1 -q orig Then the working point coordinates q orig Given the coordinates q of the discrete point k The allocation weight is w left =d left / Δq, coordinates of the working point q orig The weights assigned to the coordinates of discrete points are w. right =d right / Δq, Δq=q k+1 -q k Working point coordinates q orig Assigned to discrete point coordinates q k The power spectral density is E k =w left ×E orig Working point coordinates q orig Assigned to discrete point coordinates q k+1 The power spectral density is E k+1 =w right ×E origThe cumulative weight of each discrete point coordinate is obtained by summing the weights assigned to each working point coordinate of the normalized power spectrum of the current frame. The cumulative power spectral density value of each discrete point coordinate is obtained by summing the power spectral density values assigned to each working point coordinate of the normalized power spectrum of the current frame. The resampled power spectral density value of each discrete point coordinate is obtained by dividing the cumulative power spectral density value by the cumulative weight assigned to that discrete point coordinate. By dividing by the cumulative weight, the average power density level can be obtained, avoiding the overall increase in background noise due to the superposition of multiple segments.
[0110] S700: Based on the resampled power spectrum of the current frame, a reference power spectrum of the current frame is formed using an exponential averaging mechanism; based on the reference power spectrum of the current frame, the reference lateral operating point of the current frame is determined.
[0111] Assume the resampled power spectrum of the current frame is S t The reference power spectrum of the current frame is R. t The reference power spectrum of the previous frame is R. t-1 The formula for calculating the reference power spectrum of the current frame is: R t =αS t +(1-α)R t-1 Where α is the weight of the current frame, t is the frame number of the current frame, t-1 is the frame number of the previous frame, and R0=0, meaning the reference power spectrum of the initial frame is αS1. Thus, the weight of the current frame is α (e.g., 0.2), and the weight of the previous frame is determined by R... t-1 Indirect contribution weights, with earlier frames weighted according to (1-α). 2 (1-α) 3 …gradual decay means that the most recent frames contribute the most to the reference power spectrum of the current frame, while the influence of earlier frames becomes smaller and smaller. Therefore, the reference power spectrum will quickly absorb the latest lateral operating point information without being completely deflected by a single noise spike, thus forming a smooth, memory-based historical reference power spectrum.
[0112] Based on the reference power spectrum of the current frame, determine the reference lateral operating point of the current frame, specifically including:
[0113] Extract the coordinates of the discrete point corresponding to the maximum power spectral density value of the reference power spectrum of the current frame, and use them as the coordinates of the peak discrete point of the current frame;
[0114] Median filtering is performed based on the peak discrete point coordinates of the current frame and the peak discrete point coordinates of several previous frames to obtain the median-filtered peak discrete point coordinates of the current frame, which serves as the reference horizontal working point for the current frame.
[0115] Median filtering involves sorting the peak discrete point coordinates of the current frame and several preceding frames from smallest to largest, then taking the middle peak discrete point coordinate as the median-filtered discrete point coordinate, which is used as the reference horizontal working point for the current frame. This way, if the peak discrete point coordinates of a frame experience a significant jump due to transient interference, while the frames before and after remain stable, outliers are ignored, and the filtered output is closer to the normal level of neighboring frames. Furthermore, because the number of frames used for median filtering is relatively small, the ability to track slow changes in the true working point is largely unaffected. Therefore, the median-filtered peak discrete point coordinates retain the temporal smoothness of the peak discrete point coordinates while further suppressing occasional jumps, ensuring temporal stability and noise resistance.
[0116] S800: Determine candidate lateral working points for the current frame based on the resampled power spectrum of the current frame.
[0117] Step S800 specifically includes:
[0118] S810: Determine the overall score of the discrete point coordinates based on the power spectral density value corresponding to each discrete point coordinate of the resampled power spectrum of the current frame and the distance between the discrete point coordinates and the reference lateral working point of the current frame.
[0119] S820: Select the coordinates of the discrete point with the highest comprehensive score as the candidate discrete point coordinates for the current frame;
[0120] S830: Median filtering is performed based on the candidate discrete point coordinates of the current frame and the candidate discrete point coordinates of several previous frames to obtain the median-filtered candidate discrete point coordinates of the current frame, which are used as candidate horizontal working points of the current frame.
[0121] In step S810, the larger the power spectral density value, the more likely it is to be a true transverse operating point signal, and the higher its score will be. The closer the discrete point is to the reference transverse operating point, the less likely it is to be a noise isolated point, and the higher its score will be. The contribution of the power spectral density value and the contribution of the distance of each discrete point are weighted according to a certain weight to obtain the comprehensive score of the discrete point.
[0122] S900: Combine the current frame and several preceding frames into a frame sequence, and determine the optimal estimate of the horizontal working point of the current frame based on the reference horizontal working point and candidate horizontal working point of each frame in the frame sequence.
[0123] In some embodiments, the optimal estimate of the horizontal working point of the current frame can be determined based on the Kalman filter algorithm, and step S900 specifically includes:
[0124] S910: Determine the predicted value and uncertainty of the horizontal working point of the current frame.
[0125] Between adjacent frames, the change in the lateral operating point is usually slow. Therefore, a simple constant-value model can be used, assuming that there is no significant jump in the lateral operating point between two adjacent frames. Thus, the predicted value of the lateral operating point in the current frame can be taken as the optimal estimate of the lateral operating point in the previous frame. The uncertainty of the predicted value of the lateral operating point in the current frame can be taken as the sum of the uncertainty of the optimal estimate of the lateral operating point in the previous frame and the variance of the process noise. The calculation formula is as follows: x pred =x prev P pred =P prev +Q, where x prev x is the predicted value of the horizontal working point in the current frame. prev P is the optimal estimate of the horizontal working point of the previous frame. pred P represents the uncertainty of the predicted value of the lateral working point in the current frame. prev Q represents the uncertainty of the optimal estimate of the horizontal working point in the previous frame, and Q represents the noise variance.
[0126] S920: Determine the fused observations and observation variance of the lateral working point of the current frame based on the reference lateral working point and candidate lateral working point of each frame in the frame sequence.
[0127] Step S920 specifically includes:
[0128] The variance, mean square error, or median absolute deviation of the reference lateral working point and candidate lateral working point of each frame in the frame sequence are calculated and used as the noise variance of the reference lateral working point and candidate lateral working point of the current frame, respectively. The noise variance reflects the degree of fluctuation.
[0129] The weights of the reference lateral working point and the candidate lateral working point of the current frame and the observation variance of the lateral working point of the current frame are determined based on the noise variance of the reference lateral working point and the candidate lateral working point of the current frame.
[0130] The reference lateral working point and the candidate lateral working point of the current frame are fused based on their weights to obtain the fused observation value of the lateral working point of the current frame.
[0131] Assuming the noise variance of the reference horizontal operating point in the current frame is R1, the noise variance of the candidate horizontal operating point in the current frame is R2, the weight of the reference horizontal operating point in the current frame is w1, and the weight of the candidate horizontal operating point in the current frame is w2, then we have: w1 = (1 / R1) / ((1 / R1) + (1 / R2)), w2 = (1 / R2) / ((1 / R1) + (1 / R2)). By fusing the two horizontal operating points in the current frame using w1 and w2 (i.e., multiplying each horizontal operating point by its respective weight and then adding the results), we can obtain the fused observation z. fuseBy fusing the noise variances of the two transverse operating points, the observation variance R can be obtained. fuse R fuse =1 / ((1 / R1)+(1 / R2)).
[0132] The actual meaning of the fusion process is: to treat the reference lateral working point of the current frame as a "stable but potentially lagging" measurement, and the candidate lateral working point of the current frame as a "timely but potentially fluctuating" measurement, and to decide which one to trust based on the degree of fluctuation of the two, and output a fused prediction value.
[0133] S930: Use the fused observations of the horizontal working point of the current frame to correct the predicted value and uncertainty of the horizontal working point of the current frame, so as to obtain the optimal estimate and uncertainty of the horizontal working point of the current frame.
[0134] Step S930 specifically includes:
[0135] Calculate the Kalman gain: K=P pred / (P) pred +R fuse );
[0136] Update state estimate: x new =x pred +K(z) fuse -x pred ), where x new This is the optimal estimate of the horizontal working point for the current frame;
[0137] Update uncertainty: P new =(1-K)P pred ;where P new The uncertainty is the optimal estimate of the horizontal working point for the current frame.
[0138] After obtaining the current frame, the method for measuring the lateral operating point of the beam in this embodiment of the invention will take the next frame as the current frame again and repeat steps S200-S900, thereby outputting the optimal estimate of the lateral operating point of the beam in each frame. As can be seen from the above description, the optimal estimate of the lateral operating point output in step S900 is the decimal part of the lateral operating point.
[0139] The method for measuring the transverse operating point of the beam according to embodiments of the present invention has the following beneficial effects:
[0140] 1) Real-time performance: Outputs horizontal working points on a frame time scale of less than 1ms, with low computational load;
[0141] 2) Noise immunity: It can maintain stable estimation even when the signal-to-noise ratio is as low as -20dB;
[0142] 3) Robustness: Not easily affected by occasional noise spikes, electronic interference, or instantaneous spectral distortion;
[0143] 4) Sustainability: It can process beam data in real time during the acceleration phase (the phase of rapid change in cyclotron frequency) without interruption.
[0144] like Figure 2 As shown, this embodiment of the invention provides a measuring device for the transverse working point of a beam, which includes a sampling module 10, a first acquisition module 20, a second acquisition module 30, a smoothing filter module 40, a normalization module 50, a resampling module 60, a reference point determination module 70, a candidate point determination module 80, and an estimation module 90.
[0145] The sampling module 10 is used to sample the Schottky signal in chronological order, set a preset time period, and take the Schottky signal collected in each preset time period as a frame of data; wherein, the current preset time period is taken as the current frame, the previous preset time period is taken as the previous frame, and the next preset time period is taken as the next frame.
[0146] The first acquisition module 20 is used to acquire the power spectrum of the current frame based on the current frame data;
[0147] The second acquisition module 30 is used to acquire the power spectrum of the target frequency segment of the current frame under the unified physical frequency axis based on the power spectrum of the current frame, and use it as the target power spectrum of the current frame.
[0148] The smoothing filter module 40 is used to perform adaptive smoothing filtering on the target power spectrum of the current frame to obtain the smoothed target power spectrum of the current frame.
[0149] The normalization module 50 is used to normalize the frequency coordinates of the smoothed target power spectrum of the current frame to dimensionless operating point coordinates, so as to obtain the normalized power spectrum of the current frame.
[0150] The resampling module 60 is used to resample the normalized power spectrum of the current frame to obtain the resampled power spectrum of the current frame.
[0151] The reference point determination module 70 is used to form a reference power spectrum of the current frame based on the resampled power spectrum of the current frame using an exponential averaging mechanism; and to determine the reference lateral operating point of the current frame based on the reference power spectrum of the current frame.
[0152] The candidate point determination module 80 is used to determine the candidate lateral working points of the current frame based on the resampled power spectrum of the current frame.
[0153] The estimation module 90 is used to form a frame sequence by combining the current frame and several preceding frames of the current frame, and to determine the optimal estimate of the horizontal working point of the current frame based on the reference horizontal working point and candidate horizontal working point of each frame in the frame sequence.
[0154] The sampling module 10, the first acquisition module 20, the second acquisition module 30, the smoothing filter module 40, the normalization module 50, the resampling module 60, the reference point determination module 70, the candidate point determination module 80, and the estimation module 90 are the functional modules corresponding to steps S100-S900 in the method embodiment. Their specific implementation methods can be found in the description in the method embodiment, and will not be repeated here.
[0155] Another embodiment of the present invention provides a readable storage medium having a computer program stored thereon. When the computer program is executed in a computer, it causes the computer to perform the steps of the method for measuring the transverse operating point of the beam in the above embodiments of the present invention.
[0156] Another embodiment of the present invention provides an electronic device, which includes a memory and a processor. The memory stores executable code. When the processor executes the executable code, it performs the steps of the method for measuring the transverse operating point of a beam in the above embodiments of the present invention.
[0157] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.
[0158] For ease of description, the above apparatus is described by dividing it into various functional units. Of course, in implementing this invention, the functions of each unit can be implemented in one or more software and / or hardware components.
[0159] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0160] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0161] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0162] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0163] In a typical configuration, an electronic device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0164] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.
[0165] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by electronic devices. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.
[0166] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0167] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0168] This invention can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.
[0169] The various embodiments in this invention are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.
[0170] The foregoing has described specific embodiments of the invention. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0171] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of the invention. Various variations can be made to the above embodiments of the present invention. That is, all simple and equivalent changes and modifications made based on the claims and description of this invention fall within the protection scope of the claims of this patent. All aspects not described in detail in this invention are conventional technical content.
Claims
1. A method for measuring the transverse operating point of a beam, characterized in that, include: The Schottky signal is sampled in chronological order. A preset time period is set, and the Schottky signal collected in each preset time period is taken as a frame of data. The current preset time period is taken as the current frame, the previous preset time period is taken as the previous frame, and the next preset time period is taken as the next frame. The power spectrum of the current frame is obtained based on the current frame data; Based on the power spectrum of the current frame, the power spectrum of the target frequency segment of the current frame under the unified physical frequency axis is obtained, and used as the target power spectrum of the current frame. An adaptive smoothing filter is applied to the target power spectrum of the current frame to obtain the smoothed target power spectrum of the current frame. The frequency coordinates of the smoothed target power spectrum of the current frame are normalized to dimensionless operating point coordinates to obtain the normalized power spectrum of the current frame. The normalized power spectrum of the current frame is resampled to obtain the resampled power spectrum of the current frame; Based on the resampled power spectrum of the current frame, a reference power spectrum of the current frame is formed using an exponential averaging mechanism; based on the reference power spectrum of the current frame, a reference lateral operating point of the current frame is determined. Based on the resampled power spectrum of the current frame, determine the candidate lateral operating point of the current frame; The current frame and several preceding frames are combined into a frame sequence. Based on the reference lateral working point and candidate lateral working point of each frame in the frame sequence, the optimal estimate of the lateral working point of the current frame is determined.
2. The method for measuring the transverse operating point of a beam according to claim 1, characterized in that, Adaptive smoothing filtering is performed on the target power spectrum of the current frame to obtain the smoothed target power spectrum of the current frame, specifically including: Predetermine the frequency range corresponding to the region that does not belong to the transverse beam signal, and use it as the suppression frequency range; The power spectral density values within the suppression frequency range of the target power spectrum of the current frame are suppressed to obtain the suppressed power spectrum of the current frame; A smoothing window is pre-determined based on the widening of the lateral sidebands of the Schottky signal; Based on the smoothing window, the suppressed power spectrum of the current frame is smoothed and filtered to obtain the smoothed target power spectrum of the current frame.
3. The method for measuring the transverse operating point of a beam according to claim 1, characterized in that, The frequency coordinates of the smoothed target power spectrum of the current frame are normalized to dimensionless operating point coordinates to obtain the normalized power spectrum of the current frame, specifically including: Divide each frequency coordinate of the target power spectrum after smoothing and filtering in the current frame by the cyclotron frequency of the current frame to obtain the dimensionless value corresponding to that frequency coordinate, and use the decimal part of the dimensionless value as the operating point coordinate corresponding to that frequency coordinate. The coordinates of each working point are folded into the interval [0, 0.5) to obtain the normalized power spectrum of the current frame.
4. The method for measuring the transverse operating point of a beam according to claim 3, characterized in that, The normalized power spectrum of the current frame is resampled to obtain the resampled power spectrum of the current frame, specifically including: The interval [0, 0.5) is divided into equal intervals with a preset interval to obtain the coordinates of multiple discrete points; The power spectral density values corresponding to the coordinates of each working point of the normalized power spectrum of the current frame are projected onto the coordinates of each discrete point in a soft-assignment manner to obtain the resampled power spectral density values corresponding to the coordinates of each discrete point. The coordinates of each discrete point and its corresponding resampled power spectral density values form the resampled power spectrum of the current frame.
5. The method for measuring the transverse operating point of a beam according to claim 4, characterized in that, The power spectral density values corresponding to the working point coordinates of the normalized power spectrum of the current frame are projected onto the discrete point coordinates using a soft-assignment method to obtain the resampled power spectral density values corresponding to the discrete point coordinates. Specifically, this includes: For each working point coordinate of the normalized power spectrum of the current frame, determine the coordinates of two discrete points adjacent to the working point coordinate; based on the distance between the working point coordinate and the two adjacent discrete point coordinates, determine the allocation weight of the working point coordinate to the two adjacent discrete point coordinates; based on the allocation weight of the working point coordinate to the two adjacent discrete point coordinates, allocate the power spectral density value corresponding to the working point coordinate to the two adjacent discrete point coordinates. The weights assigned to each discrete point coordinate in the normalized power spectrum of the current frame are summed to obtain the cumulative weight of that discrete point coordinate. The power spectral density values assigned to each working point coordinate of the normalized power spectrum of the current frame and the power spectral density values of each discrete point coordinate are added together to obtain the cumulative power spectral density value of that discrete point coordinate. The resampled power spectral density value of each discrete point coordinate is obtained by dividing the cumulative power spectral density value of that discrete point coordinate by the cumulative weight of that discrete point coordinate.
6. The method for measuring the transverse operating point of a beam according to claim 1, characterized in that, Based on the reference power spectrum of the current frame, determine the reference lateral operating point of the current frame, specifically including: Extract the coordinates of the discrete point corresponding to the maximum power spectral density value of the reference power spectrum of the current frame, and use them as the coordinates of the peak discrete point of the current frame; Median filtering is performed based on the peak discrete point coordinates of the current frame and the peak discrete point coordinates of several previous frames to obtain the median filtered peak discrete point coordinates of the current frame, which serves as the reference horizontal working point for the current frame. Based on the resampled power spectrum of the current frame, candidate lateral operating points for the current frame are determined, specifically including: The overall score of a discrete point coordinate is determined based on the power spectral density value corresponding to the coordinate of each discrete point in the resampled power spectrum of the current frame and the distance between the coordinate of that discrete point and the reference lateral working point of the current frame. The coordinates of the discrete point with the highest overall score are selected as the candidate discrete point coordinates for the current frame. Median filtering is performed on the candidate discrete point coordinates of the current frame and the candidate discrete point coordinates of several previous frames to obtain the median-filtered candidate discrete point coordinates of the current frame, which are then used as candidate lateral working points for the current frame.
7. The method for measuring the transverse operating point of a beam according to claim 1, characterized in that, The optimal estimate of the lateral working point of the current frame is determined based on the reference lateral working point and candidate lateral working points of each frame in the frame sequence, specifically including: Determine the predicted value and uncertainty of the lateral working point of the current frame; The fused observations and observation variance of the lateral working point of the current frame are determined based on the reference lateral working point and candidate lateral working point of each frame in the frame sequence. The predicted value and uncertainty of the horizontal working point of the current frame are corrected by using the fused observation value of the horizontal working point of the current frame, so as to obtain the optimal estimate and uncertainty of the horizontal working point of the current frame.
8. The method for measuring the transverse operating point of a beam according to claim 7, characterized in that, The fused observations and observation variance of the lateral working point of the current frame are determined based on the reference lateral working point and candidate lateral working point of each frame in the frame sequence, specifically including: Calculate the variance, mean square error, or median absolute deviation of the reference lateral working point and candidate lateral working point for each frame in the frame sequence, and use them as the noise variance of the reference lateral working point and candidate lateral working point for the current frame, respectively. The weights of the reference lateral operating point and the candidate lateral operating point of the current frame and the observation variance of the lateral operating point of the current frame are determined based on the noise variance of the reference lateral operating point and the candidate lateral operating point of the current frame. The reference lateral working point and the candidate lateral working point of the current frame are fused based on their weights to obtain the fused observation value of the lateral working point of the current frame.
9. The method for measuring the transverse operating point of a beam according to claim 7, characterized in that, The predicted value and uncertainty of the lateral operating point in the current frame are corrected using the fused observations of the current frame's lateral operating point to obtain the optimal estimate and uncertainty of the current frame's lateral operating point. Specifically, this includes: The Kalman filter gain for the current frame is determined based on the predicted value of the lateral working point and the fused observation value. The optimal estimate of the lateral operating point for the current frame is determined based on the predicted value of the lateral operating point, the fused observation value, and the Kalman gain. The uncertainty of the optimal estimate of the lateral working point in the current frame is determined based on the uncertainty of the predicted value of the lateral working point in the current frame and the Kalman gain.
10. A device for measuring the transverse operating point of a beam, characterized in that, include: The sampling module is used to sample the Schottky signal in chronological order, set a preset time period, and take the Schottky signal collected in each preset time period as a frame of data; wherein, the current preset time period is taken as the current frame, the previous preset time period is taken as the previous frame, and the next preset time period is taken as the next frame. The first acquisition module is used to acquire the power spectrum of the current frame based on the current frame data; The second acquisition module is used to acquire the power spectrum of the target frequency segment of the current frame under the unified physical frequency axis based on the power spectrum of the current frame, and use it as the target power spectrum of the current frame. A smoothing filter module is used to perform adaptive smoothing filtering on the target power spectrum of the current frame to obtain the smoothed target power spectrum of the current frame. The normalization module is used to normalize the frequency coordinates of the smoothed and filtered target power spectrum of the current frame into dimensionless operating point coordinates to obtain the normalized power spectrum of the current frame. The resampling module is used to resample the normalized power spectrum of the current frame to obtain the resampled power spectrum of the current frame. The reference point determination module is used to form a reference power spectrum of the current frame based on the resampled power spectrum of the current frame using an exponential averaging mechanism; and to determine the reference lateral operating point of the current frame based on the reference power spectrum of the current frame. The candidate point determination module is used to determine candidate lateral working points of the current frame based on the resampled power spectrum of the current frame. The estimation module is used to form a frame sequence by combining the current frame and several preceding frames, and to determine the optimal estimate of the lateral working point of the current frame based on the reference lateral working point and candidate lateral working point of each frame in the frame sequence.