A radio frequency cavity spark position diagnosis method based on differential time of flight

Abstract

The present application relates to the technical field of particle accelerator, and provides a radio frequency cavity spark position diagnosis method based on differential time of flight; the method adopts a double-channel differential time of flight measurement architecture, installs a fast-response arc light detector symmetrically at two ends of a radio frequency cavity, captures a spark event light signal in real time, and records double-channel original waveform data; then, dynamic amplitude normalization processing is performed on the signal to eliminate sensor gain difference; a time difference is calculated by analyzing a step response parameter, a least square method with a shared slope is used to fit a model to reverse a theoretical time point, the time difference is converted into a spatial distance difference by combining an electromagnetic wave propagation speed, and finally, a spark point position is determined. The present application realizes centimeter-level positioning, greatly improves anti-interference capability and positioning stability, has real-time online monitoring and fast response capability, reduces hardware cost by more than 90%, can be seamlessly integrated with an existing system, and improves convenience and economy of field application.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of particle accelerator technology, in particular to a radio frequency cavity spark position diagnosis method based on differential time of flight. BACKGROUND

[0002] As the core component of the particle accelerator, the stability of the radio frequency resonant cavity directly determines the performance of the accelerator. Under high electric field gradient conditions, radio frequency sparking phenomenon occurs frequently, which can cause a series of risks such as acceleration efficiency decline, equipment damage and operation interruption. Therefore, accurately and quickly diagnosing the spark position in the radio frequency cavity is crucial to ensure the stable operation of the particle accelerator.

[0003] At present, the existing spark positioning technology mainly includes the following three types: first, the detection method based on physical response: including low-level signal analysis, pickup signal positioning, vacuum gauge response auxiliary judgment, etc. The essence of these methods is to position the cavity by the phase difference of the coupler or the time difference of the vacuum gauge. However, due to the limitation of signal bandwidth and spatial resolution, the positioning accuracy cannot break through the cavity length order of magnitude. For example, the low-level method in CNSS can only identify the specific cavity, and the pickup method leads to a positioning error of more than one meter due to the limitation of probe spacing; the second is the acoustic detection based on energy propagation characteristics: using piezoelectric sensors to capture the shock wave of sparking, but this method is affected by sound wave attenuation, multipath effect and cavity mechanical coupling characteristics, and the actual positioning error is often tens of centimeters; the third is the arc light positioning based on optical signal detection: although it can quickly identify the sparking event, the traditional scheme only installs a single-end detector, lacks time difference measurement dimension, and cannot realize spatial coordinate solution.

[0004] The above three methods generally have the problems of insufficient positioning accuracy (>1m level), weak anti-electromagnetic interference ability and inability to respond in real time in complex cavity structures, which makes it difficult for maintenance personnel to quickly lock the centimeter-level defects or contamination points in the cavity. The main reason is that the existing technology cannot effectively utilize the propagation time delay characteristics of the electromagnetic wave / optical signal generated by the sparking event in the cavity medium, and simply relies on signal amplitude or single propagation path analysis, resulting in that the spatial resolution is limited by the hardware layout density.

[0005] Therefore, there is an urgent need for a radio frequency cavity spark position diagnosis method that can break through the spatial resolution limit of traditional methods and realize high-precision, anti-interference and real-time response. SUMMARY

[0006] In view of the deficiencies in the prior art, the present application aims to provide a radio frequency cavity spark position diagnosis method based on differential time of flight, which is suitable for high-precision and real-time diagnosis of the spark position in the radio frequency resonant cavity, and is especially suitable for positioning the spark fault in the particle accelerator radio frequency cavity under high electric field gradient conditions.

[0007] The technical scheme adopted by the present application to solve the technical problem is: a radio frequency cavity spark position diagnosis method based on differential time of flight, adopting a double-channel differential time of flight measurement architecture, comprising the following steps:

[0008] S1: symmetrically installing fast-response arc light detectors at both ends of the radio frequency cavity, connecting to the detectors through optical fiber links, capturing light signals of spark events in real time and recording double-channel original waveform data;

[0009] S2: performing dynamic amplitude normalization processing on the two-channel signals, linearly scaling the channel two signals based on the steady-state value and the initial value, and eliminating the gain difference of the sensor;

[0010] S3: analyzing the step response parameters of the normalized signals, calculating the time difference of the 10% and 90% steady-state threshold points, and intercepting the rising section data of the two channels;

[0011] S4: adopting a least squares fitting model with a shared slope to back-propagate the theoretical time point to calculate a high-precision time difference (Δt);

[0012] S5: combining the propagation speed of electromagnetic waves in the cavity medium (v=c / n, c is the speed of light, and n is the refractive index) to convert the time difference into a spatial distance difference (ΔL=v×Δt);

[0013] S6: combining the total length of the cavity to solve the specific position of the spark point through geometric relations.

[0014] The dynamic amplitude normalization processing in S2 includes: linearly scaling the channel two signals based on the steady-state value and the initial value, forcibly aligning the amplitude value range of the double-channel signals, and eliminating the influence of the gain difference of the sensor on time extraction.

[0015] The least squares fitting model with a shared slope in S4 includes:

[0016] Adopting a least squares method to constrain the two-channel signals to share the same slope, replacing direct threshold triggering with fitting a theoretical time point, and reducing the time difference measurement error to ≤1 ns.

[0017] The radio frequency cavity spark position diagnosis method further includes: combining double-end synchronous acquisition with high-precision clock synchronization technology to realize real-time online monitoring and rapid response to spark events.

[0018] The radio frequency cavity spark position diagnosis method further includes: the system structure is compact, the optical fiber links and high-speed data acquisition cards can be flexibly arranged in radio frequency cavities of different sizes and shapes, reducing the investment in additional hardware and reducing installation and maintenance costs.

[0019] The radio frequency cavity sparking position diagnosis method further comprises: being seamlessly integrated with an existing accelerator control system through software upgrading, without large-scale modification, to improve the convenience and economy of field application.

[0020] The double-channel differential time-of-flight measurement architecture comprises: arranging fast-response arc light detectors at both ends of the radio frequency cavity, calculating the nanosecond-level time difference (Delta t) by synchronously collecting light signals, breaking through the limitation of cavity length on positioning accuracy, and realizing sub-centimeter-level coordinate calculation.

[0021] The radio frequency cavity sparking position diagnosis method further comprises: only needing to install detectors at both ends of the cavity and transmitting signals through an optical fiber link, being compatible with an existing accelerator structure, and reducing hardware cost by more than 90%.

[0022] The radio frequency cavity sparking position diagnosis method further comprises: using dynamic amplitude normalization and shared slope least square fitting algorithm to effectively eliminate sensor gain difference and random noise interference, and greatly improving the anti-electromagnetic interference ability and positioning stability.

[0023] The radio frequency cavity sparking position diagnosis method further comprises: realizing sub-centimeter-level spatial resolution positioning through nanosecond-level time difference inversion, which is significantly better than the meter-level accuracy of traditional low-level method, Pickup method and acoustic method.

[0024] The beneficial effects of the present application are: through the double-channel differential time-of-flight measurement architecture, the nanosecond-level time difference of the arc light signal reaching the detectors at both ends of the cavity is accurately analyzed, and sub-centimeter-level positioning is realized in combination with the electromagnetic wave propagation speed model, which is significantly better than the meter-level accuracy of traditional low-level method, Pickup method and acoustic method; especially, the dynamic amplitude normalization algorithm and the shared slope least square fitting algorithm are used to effectively eliminate sensor gain difference and random noise interference, and greatly improve the anti-electromagnetic interference ability and positioning stability.

[0025] The present application combines double-end synchronous acquisition with high-precision clock synchronization technology to realize real-time online monitoring and rapid response to sparking events, effectively meeting the operation and maintenance requirements under high electric field gradient conditions; at the same time, the system structure of the present application is compact, only needs to install detectors at both ends of the cavity and transmit signals through an optical fiber link, is compatible with an existing accelerator structure, reduces hardware cost by more than 90%, and the optical fiber link and high-speed data acquisition card can be flexibly arranged in radio frequency cavities of different sizes and shapes; the present application can be seamlessly integrated with an existing accelerator control system through software upgrading, without large-scale modification, to improve the convenience and economy of field application. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 It is a differential technology principle diagram based on double-channel time-of-flight in the present application;

[0027] Figure 2is a flow chart of a differential method based on a dual-channel time of flight in the present application;

[0028] Figure 3 is a raw waveform result chart in the present application;

[0029] Figure 4 is a fitting result chart in the present application;

[0030] Figure 5 is a waveform alignment result chart in the present application. DETAILED DESCRIPTION

[0031] The present application is further described in detail below in conjunction with the accompanying drawings and specific embodiments;

[0032] As shown in the drawings, Figures 1-5 A radio frequency cavity spark position diagnosis method based on differential time of flight, which adopts a dual-channel differential time of flight measurement architecture, includes the following steps:

[0033] S1: symmetrically install fast-response arc light detectors at both ends of the radio frequency cavity, connect to the detectors through optical fiber links, capture light signals of spark events in real time and record dual-channel raw waveform data;

[0034] S2: perform dynamic amplitude normalization processing on the two-channel signals, linearly scale channel two signals based on the steady-state value and the initial value, and eliminate sensor gain differences; the dynamic amplitude normalization processing includes: linearly scaling channel two signals based on the steady-state value and the initial value, forcibly aligning the amplitude range of the dual-channel signals, and eliminating the influence of sensor gain differences on time extraction;

[0035] S3: analyze the step response parameters of the normalized signals, calculate the time difference of the 10% and 90% steady-state threshold points, and intercept the rising segment data of the two channels;

[0036] S4: adopt a least squares fitting model with a shared slope to back-propagate the theoretical time point to calculate a high-precision time difference (Δt); the least squares fitting model with a shared slope includes: using the least squares method to constrain the two-channel signals to share the same slope, replacing direct threshold triggering with fitting the theoretical time point, and reducing the time difference measurement error to ≤1 ns.

[0037] S5: combined with the propagation speed of electromagnetic waves in the cavity medium (v=c / n, c is the speed of light, and n is the refractive index), convert the time difference into a spatial distance difference (ΔL=v×Δt);

[0038] S6: combined with the total length of the cavity, solve the specific position of the spark point through geometric relationship.

[0039] The radio frequency cavity arcing location diagnosis method further includes: combining dual-end synchronous acquisition with high-precision clock synchronization technology to achieve real-time online monitoring and rapid response to arcing events.

[0040] The radio frequency cavity arcing location diagnosis method also includes: a compact system structure, with fiber optic links and high-speed data acquisition cards that can be flexibly deployed in radio frequency cavities of different sizes and shapes, reducing investment in new hardware and lowering installation and maintenance costs.

[0041] The dual-channel differential time-of-flight measurement architecture includes: fast-response arc detectors deployed at both ends of the radio frequency cavity; synchronous acquisition of optical signals and calculation of nanosecond-level time difference (Δt) to overcome the limitation of cavity length on positioning accuracy and achieve sub-centimeter-level coordinate calculation.

[0042] The radio frequency cavity arcing position diagnosis method is seamlessly integrated with the existing accelerator control system through software upgrades, requiring no large-scale modifications and improving the convenience and economy of field applications. It only requires the installation of detectors at both ends of the cavity and the transmission of signals through fiber optic links, making it compatible with existing accelerator structures and reducing hardware costs by more than 90%. It adopts dynamic amplitude normalization and shared slope least squares fitting algorithms to effectively eliminate sensor gain differences and random noise interference, significantly improving anti-electromagnetic interference capabilities and positioning stability. It achieves sub-centimeter spatial resolution positioning through nanosecond-level time difference inversion, which is significantly better than the meter-level accuracy of traditional low-level methods, pickup methods, and acoustic methods.

[0043] like Figure 1 and 2 As shown, this invention embodies the core idea of ​​differential measurement, which is to accurately determine the arcing location by measuring the time difference between the signals of the two channels and combining parameters such as the propagation speed of electromagnetic waves in the cavity medium. It achieves sub-centimeter-level positioning by accurately analyzing the nanosecond-level time difference between the arrival of the arc signal at the detectors at both ends of the cavity and combining this with an electromagnetic wave propagation speed model. More specifically, this invention has one detector on each side, a fast-response arc detector symmetrically installed at both ends of the RF cavity. This dual-channel architecture can capture the relevant signals of the arcing event at both ends of the RF cavity, providing basic data for subsequent differential measurements. There is a time difference between the signals captured by the two detectors. In the actual RF cavity arcing location diagnosis process, the light signal generated by the arcing event will arrive at the detectors on both sides at different times. This time difference (Δt=|t1-t2|) is a key parameter for calculating the arcing location in subsequent implementation, i.e., recording the original waveform data of the two channels.

[0044] Figure 3 The waveforms of channel one and channel two are displayed in the image. Figure 3As can be seen from the figures, the waveforms of the two channels differ in characteristics such as the rising edge, reflecting the inconsistency of the signals due to factors such as detector gain differences and signal transmission, and thus require dynamic amplitude normalization of the signals to eliminate the impact of these differences on subsequent time difference measurements. Figure 3 The differences in the waveforms, especially the time differences in the rising edges, are the basis for calculating the time difference (Δt) of the signals of the two channels. By accurately measuring these time differences and combining the propagation speed of electromagnetic waves in the cavity medium (v = c / n, c is the speed of light, and n is the refractive index), the precise diagnosis of the sparking position can be achieved. When calculating the time difference, specific threshold points such as the 10% and 90% steady-state threshold points are usually selected, Figure 3 As a result of multiple experiments, the steady-state values and initial values of the waveforms can provide experimental results to support the selection of threshold points.

[0045] Figure 4 The figures show the raw data (solid line) and the fitted curve (dashed line) of channel one and channel two during the experiment, which reflects the process of analyzing the normalized signal step response parameters, intercepting the rising segment data of the two channels, and fitting the model using the least squares method with shared slope in the technical solution. Through the fitted curve, the characteristic time points of the signal can be more accurately determined, and thus the high-precision time difference can be calculated.

[0046] Figure 5 The figures reflect the process of aligning the dual-channel waveforms during the experiment, which is related to eliminating the impact of sensor gain differences on time extraction in the technical solution. Due to the gain differences of different detectors, the amplitude and time of the two-channel signals are inconsistent, and through waveform alignment, the time difference of the signal can be more accurately calculated. The figure marks "time alignment after 90% point fitting", indicating that a specific alignment standard is used in multiple experiments, i.e. using the 90% point of the fitted waveform as the reference point for time alignment. This alignment method helps to improve the accuracy of time difference measurement and reduce errors caused by inaccurate selection of signal feature points. The figure shows the waveform of channel one after fitting (solid line) and the waveform of channel two after fitting and alignment (dashed line). As can be seen from the comparison, the aligned waveforms of the two channels are more consistent in time, which provides a reliable basis for accurately calculating the time difference (Δt). The aligned waveforms can more accurately determine the characteristic time points of the signal rising edge, and combined with the least squares fitting model with shared slope (claim 3), the theoretical time points can be deduced to calculate the high-precision time difference. Combined with the propagation speed of electromagnetic waves in the cavity medium (v = c / n, c is the speed of light, and n is the refractive index), the time difference is converted into a spatial distance difference (ΔL = v × Δt), and then the specific position of the sparking point is calculated.

[0047] The following will be further illustrated through specific embodiments:

[0048] Example One:

[0049] The embodiment is based on a double-channel differential time-of-flight measurement architecture, uses fast-response arc light detectors installed at both ends of a radio frequency cavity to capture light signals of sparking events in real time, and realizes accurate diagnosis of sparking positions by accurately calculating the time difference of light signals reaching the two detectors and combining the propagation speed of electromagnetic waves in the cavity medium.

[0050] Implementation process

[0051] S1: signal acquisition and preprocessing

[0052] Fast-response arc light detectors are symmetrically installed at both ends of the radio frequency cavity, and the response time of the detector reaches nanoseconds, ensuring that the light signals generated by the sparking event can be accurately captured. The detector is connected to the data acquisition system through an optical fiber link to capture the light signals of the sparking event in real time and record the double-channel raw waveform data. The optical fiber link has the advantages of anti-electromagnetic interference and low transmission loss, ensuring the stability of signal transmission.

[0053] S2: dynamic amplitude normalization

[0054] Because there may be gain differences between different detectors, the amplitudes of the two-channel signals are inconsistent, affecting the measurement accuracy of the time difference. Therefore, dynamic amplitude normalization processing is performed on the two-channel signals. The specific method is to perform linear scaling on the channel two signals based on the steady-state value and the initial value, forcibly align the amplitude range of the double-channel signals, and eliminate the influence of sensor gain difference on time extraction.

[0055] S3: time difference calculation

[0056] The step response parameters of the normalized signals are analyzed, the time difference of the 10% and 90% steady-state threshold points is calculated, and the rising section data of the two channels is intercepted. The least squares fitting model with shared slope is used to constrain the two-channel signals to share the same slope, and the time difference measurement error is reduced to ≤1 ns by replacing the direct threshold trigger with the fitting theoretical time point, so as to obtain a high-precision time difference (Δt).

[0057] S4: position solution

[0058] Combined with the propagation speed of electromagnetic waves in the cavity medium (v=c / n, c is the speed of light, and n is the refractive index), the time difference is converted into a spatial distance difference (ΔL=v×Δt). Given the total length of the cavity, the specific position of the sparking point can be solved through geometric relationship.

[0059] That is, in the radio frequency cavity of a certain particle accelerator, the time difference of the two-channel light signals measured by the above steps is 5 ns, and the propagation speed of electromagnetic waves in the cavity medium is 2×10 8 m / s, then the spatial distance difference ΔL=2×10 8 ×5×10-9 = 1 m. Combined with the total length of the cavity and other geometric relationships, the location of the sparking point in the cavity can be accurately calculated.

[0060] Example Two:

[0061] This embodiment is also based on a dual-channel differential time-of-flight measurement architecture, focusing on further improving the accuracy and reliability of sparking location diagnosis by optimizing signal processing algorithms and hardware configurations.

[0062] Implementation Process

[0063] S1: Optimize signal acquisition

[0064] The fast-response arc light detectors installed at both ends of the radio frequency cavity use more advanced sensor technology, with higher sensitivity and lower noise levels. At the same time, the data acquisition system uses high-speed, high-precision data acquisition cards with a sampling rate of GHz level, ensuring accurate recording of subtle changes in light signals. The detector is connected to the data acquisition system through an optical fiber link, capturing real-time light signals of sparking events and recording dual-channel raw waveform data.

[0065] S2: Improve dynamic amplitude normalization

[0066] In the dynamic amplitude normalization process, an adaptive algorithm is used to automatically adjust the linear scaling coefficient based on the real-time characteristics of the signal, further improving the accuracy of amplitude alignment. At the same time, the signal is filtered to remove high-frequency noise interference and improve signal quality.

[0067] S3: High-precision time difference calculation

[0068] In calculating the time difference, in addition to using the least squares fitting model with shared slope, a machine learning algorithm is introduced to analyze and predict signal characteristics, further optimizing the fitting model and improving the accuracy of time difference calculation. Through multiple measurements and data analysis, the time difference measurement error is controlled within 0.5 ns.

[0069] S4: Accurate position calculation

[0070] Combined with a more accurate electromagnetic wave propagation speed model, considering the influence of factors such as temperature and pressure of the cavity medium on the propagation speed, the time difference is converted into a spatial distance difference. Using high-precision geometric measurement data, combined with the total length of the cavity, the specific location of the sparking point is calculated through more complex geometric algorithms.

[0071] That is, in another particle accelerator radio frequency cavity, the optimized system measures the time difference of the two-channel light signals to be 3 ns, considering the actual propagation speed of the cavity medium to be 2.2 × 10 8 m / s, then the spatial distance difference ΔL = 2.2 × 10 8X3X10 -9 = 0.66m. Through accurate geometric solution, the position of the sparking point in the cavity can be determined more accurately, and more reliable fault diagnosis information can be provided for the operation and maintenance personnel.

[0072] In conclusion, through the above two embodiments, it can be seen that the radio frequency cavity sparking position diagnosis method based on differential time of flight has the advantages of high precision, anti-interference, real-time response, etc., and can effectively solve the problems existing in the prior art.

Claims

1. A method for diagnosing the arcing position of a radio frequency cavity based on differential time-of-flight, characterized in that: The dual-channel differential time-of-flight measurement architecture includes the following steps: S1: Fast-response arc detectors are symmetrically installed at both ends of the radio frequency cavity and connected to the detectors via fiber optic links to capture the optical signals of the arcing event in real time and record the raw waveform data of the dual channels. S2: Perform dynamic amplitude normalization on the two-channel signals, and linearly scale the channel two signals based on the steady-state value and the initial value to eliminate sensor gain differences; S3: Analyze the step response parameters of the normalized signal, calculate the time difference between the 10% and 90% steady-state threshold points, and extract the rising segment data of the two channels. S4: The least squares method with shared slope is used to fit the model and back-calculate the theoretical time point to calculate the high-precision time difference Δt; S5: Combining the propagation speed of electromagnetic waves in the cavity medium v ​​= c / n, where c is the speed of light and n is the refractive index, the time difference is converted into a spatial distance difference ΔL = v×Δt. S6: Determine the specific location of the ignition point by combining the total length of the cavity with geometric relationships.

2. The radio frequency cavity arcing position diagnosis method based on differential time-of-flight as described in claim 1, characterized in that: The dynamic amplitude normalization process in S2 includes: linearly scaling the channel 2 signal based on the steady-state value and the initial value to align the amplitude range of the dual-channel signal.

3. The radio frequency cavity arcing position diagnosis method based on differential time-of-flight as described in claim 1, characterized in that: The least squares fitting model for shared slope in S4 includes: using the least squares method to constrain the two channel signals to share the same slope, and replacing the direct threshold trigger by fitting the theoretical time point to reduce the time difference measurement error to ≤1ns.

4. The radio frequency cavity arcing position diagnosis method based on differential time-of-flight as described in claim 1, characterized in that: The radio frequency cavity arcing location diagnosis method further includes: combining dual-end synchronous acquisition with high-precision clock synchronization technology to achieve real-time online monitoring and rapid response to arcing events.

5. The radio frequency cavity arcing position diagnosis method based on differential time-of-flight as described in claim 1, characterized in that: The radio frequency cavity arcing location diagnosis method also includes: fiber optic links and high-speed data acquisition cards can be flexibly deployed in radio frequency cavities of different sizes and shapes.

6. The radio frequency cavity arcing position diagnosis method based on differential time-of-flight as described in claim 1, characterized in that: The radio frequency cavity arcing position diagnosis method also includes seamless integration with the existing accelerator control system through software upgrades.

7. The radio frequency cavity arcing position diagnosis method based on differential time-of-flight as described in claim 1, characterized in that: The dual-channel differential time-of-flight measurement architecture includes: fast-response arc detectors deployed at both ends of the radio frequency cavity, and centimeter-level coordinate calculation achieved by synchronously acquiring optical signals and calculating nanosecond-level time difference Δt.

8. The radio frequency cavity arcing position diagnosis method based on differential time-of-flight as described in claim 1, characterized in that: The radio frequency cavity arcing location diagnosis method also includes: only requiring the installation of detectors at both ends of the cavity and transmitting signals through an optical fiber link, which is compatible with existing accelerator structures.

9. The radio frequency cavity arcing position diagnosis method based on differential time-of-flight according to claim 1, characterized in that: The radio frequency cavity arcing location diagnosis method also includes: using dynamic amplitude normalization and shared slope least squares fitting algorithm.

10. The radio frequency cavity arcing position diagnosis method based on differential time-of-flight according to claim 1, characterized in that: The radio frequency cavity arcing location diagnosis method also includes: achieving centimeter-level spatial resolution positioning through nanosecond-level time difference inversion.

Images