Radio frequency cavity ignition position diagnosis method based on differential flight time

Abstract

The invention relates to the technical field of particle accelerators, and provides a radio frequency cavity ignition position diagnosis method based on differential flight time. According to the method, a dual-channel differential flight time measurement framework is adopted, and fast-response arc light detectors are symmetrically installed at the two ends of a radio frequency cavity, so that light signals of a sparking event are captured in real time, and dual-channel original waveform data are recorded; performing dynamic amplitude normalization processing on the signal to eliminate the gain difference of the sensor; step response parameters are analyzed to calculate time difference, a slope-shared least square fitting model is adopted to reversely deduce a theoretical time point, the time difference is converted into a spatial distance difference in combination with the electromagnetic wave propagation speed, and finally the ignition point position is determined. The centimeter-level positioning is realized, the anti-interference capability and the positioning stability are greatly improved, the real-time online monitoring and quick response capability is realized, the hardware cost is reduced by more than 90%, seamless integration with the existing system can be realized, and the convenience and the economical efficiency of field application are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of particle accelerators, and in particular to a radio frequency cavity spark position diagnosis method based on differential time of flight. Background Art

[0002] As a core component of a particle accelerator, the operational stability of the RF resonant cavity directly determines the accelerator's performance. Under high electric field gradient conditions, RF sparking frequently occurs, leading to a series of risks such as reduced acceleration efficiency, equipment damage, and operational interruption. Therefore, accurate and rapid diagnosis of sparking locations within the RF cavity is crucial to ensuring stable particle accelerator operation.

[0003] Currently, existing spark location technologies fall into three main categories: The first is detection methods based on physical responses, including low-level signal analysis, pickup signal positioning, and vacuum gauge response-assisted judgment. These methods essentially use coupler phase differences or vacuum timing sequence differences to perform cavity-level positioning. However, due to limitations in signal bandwidth and spatial resolution, their positioning accuracy cannot exceed the accuracy of the cavity length. For example, the low-level method in CNSS can only identify specific cavities, while the pickup method has positioning errors exceeding meters due to probe spacing limitations. The second is acoustic detection based on energy propagation characteristics: piezoelectric sensors are used to capture spark shock waves. However, this method is affected by acoustic wave attenuation, multipath effects, and the mechanical coupling characteristics of the cavity, resulting in actual positioning errors often reaching tens of centimeters. The third is arc location based on optical signal detection: While capable of quickly identifying spark events, traditional solutions only use single-ended detectors, lack the time difference measurement dimension, and cannot achieve spatial coordinate resolution.

[0004] The above three methods generally have problems such as insufficient positioning accuracy (>1m), weak anti-electromagnetic interference ability, and inability to respond in real time in complex cavity structures, making it difficult for operation and maintenance personnel to quickly locate centimeter-level defects or contamination points in the cavity. This is mainly because the existing technology fails to effectively utilize the propagation delay characteristics of the electromagnetic waves / optical signals generated by the spark event in the cavity medium, and simply relies on signal amplitude or single propagation path analysis, resulting in spatial resolution being 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 achieve high precision, anti-interference, and real-time response. Summary of the Invention

[0006] In response to the deficiencies in the prior art, the present invention aims to provide a radio frequency cavity spark position diagnosis method based on differential time of flight, which is suitable for high-precision, real-time diagnosis of the spark position in the radio frequency resonant cavity, and is particularly suitable for locating radio frequency cavity spark faults in particle accelerators under high electric field gradient conditions.

[0007] The technical solution adopted by the present invention to solve the technical problem is: a radio frequency cavity spark position diagnosis method based on differential time of flight, using a dual-channel differential time of flight measurement architecture, including the following steps:

[0008] S1: Fast-response arc detectors are symmetrically installed at both ends of the RF cavity and connected to the detectors via optical fiber links to capture the optical signal of the ignition event in real time and record the dual-channel raw waveform data;

[0009] S2: Perform dynamic amplitude normalization on the two-channel signals, linearly scale the channel 2 signal based on the steady-state value and the initial value to eliminate sensor gain differences;

[0010] S3: Analyze the step response parameters of the normalized signal, calculate the time difference between the 10% and 90% steady-state threshold points, and intercept the rising segment data of the two channels;

[0011] S4: The shared slope least squares method is used to fit the model and reverse the theoretical time point to calculate the high-precision time difference (Δt);

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

[0013] S6: Calculate the specific location of the ignition point based on the total length of the cavity through geometric relationships.

[0014] The dynamic amplitude normalization processing in S2 includes: linearly scaling the channel 2 signal based on the steady-state value and the initial value, forcibly aligning the amplitude ranges of the dual-channel signals, and eliminating the influence of sensor gain differences on time extraction.

[0015] The least squares fitting model of the shared slope in S4 includes:

[0016] The least squares method is used to constrain the two-channel signals to share the same slope, and the time difference measurement error is reduced to ≤1ns by fitting the theoretical time point instead of direct threshold triggering.

[0017] The radio frequency cavity spark position 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 spark events.

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

[0019] The radio frequency cavity ignition position diagnosis method also includes: seamless integration with the existing accelerator control system through software upgrade, without the need for large-scale modification, thereby improving the convenience and economy of on-site application.

[0020] The dual-channel differential time-of-flight measurement architecture includes: deploying fast-response arc light detectors at both ends of the RF cavity, synchronously collecting optical signals and calculating nanosecond-level time differences (Δt), breaking through the limitations of cavity length on positioning accuracy and achieving sub-centimeter coordinate solution.

[0021] The radio frequency cavity spark position diagnosis method also includes: only detectors need to be installed at both ends of the cavity and signals need to be transmitted through optical fiber links, which is compatible with existing accelerator structures and reduces hardware costs by more than 90%.

[0022] The radio frequency cavity spark position diagnosis method also includes: using dynamic amplitude normalization and shared slope least squares fitting algorithm to effectively eliminate sensor gain differences and random noise interference, thereby significantly improving anti-electromagnetic interference capability and positioning stability.

[0023] The radio frequency cavity spark position diagnosis method also includes: achieving sub-centimeter spatial resolution positioning through nanosecond time difference inversion, which is significantly better than the meter-level accuracy of traditional low-level methods, pickup methods and acoustic methods.

[0024] The beneficial effects of the present invention are: the present invention uses a dual-channel differential time-of-flight measurement architecture to accurately analyze the nanosecond time difference of the arc light signal reaching the detectors at both ends of the cavity, and combines it with the electromagnetic wave propagation velocity model to achieve sub-centimeter positioning, which is significantly better than the meter-level accuracy of traditional low-level methods, Pickup methods and acoustic methods; in particular, the present invention adopts a dynamic amplitude normalization algorithm and a shared slope least squares fitting algorithm to effectively eliminate sensor gain differences and random noise interference, thereby greatly improving the anti-electromagnetic interference capability and positioning stability.

[0025] The present invention combines dual-end synchronous acquisition with high-precision clock synchronization technology to achieve real-time online monitoring and rapid response to ignition events, effectively meeting the operation and maintenance requirements under high electric field gradient conditions; at the same time, the system structure of the present invention is compact, and only detectors need to be installed at both ends of the cavity and signals are transmitted through optical fiber links. It is compatible with existing accelerator structures, and the hardware cost is reduced by more than 90%. The optical fiber links and high-speed data acquisition cards can be flexibly deployed in radio frequency cavities of different sizes and shapes; the present invention can be seamlessly integrated with the existing accelerator control system through software upgrades, without the need for large-scale modifications, thereby improving the convenience and economy of on-site applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a schematic diagram of the differential technology principle used in the present invention based on dual-channel time of flight;

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

[0028] Figure 3 It is the original waveform result diagram of the present invention;

[0029] Figure 4 It is the fitting result diagram of the present invention;

[0030] Figure 5 This is a waveform alignment result diagram in the present invention. DETAILED DESCRIPTION

[0031] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments;

[0032] like Figure 1-5 As shown, a method for diagnosing the position of a radio frequency cavity spark based on differential time of flight adopts a dual-channel differential time of flight measurement architecture and includes the following steps:

[0033] S1: Fast-response arc detectors are symmetrically installed at both ends of the RF cavity and connected to the detectors via optical fiber links to capture the optical signal of the ignition event in real time and record the dual-channel raw waveform data;

[0034] S2: Performing dynamic amplitude normalization processing on the two-channel signals, linearly scaling the channel 2 signal based on the steady-state value and the initial value to eliminate sensor gain differences; the dynamic amplitude normalization processing includes: linearly scaling the channel 2 signal based on the steady-state value and the initial value to forcibly align the amplitude ranges of the two-channel signals to eliminate the impact of sensor gain differences on time extraction;

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

[0036] S4: A shared-slope least-squares fitting model is used to reverse-calculate the theoretical time point to calculate a high-precision time difference (Δt). This shared-slope least-squares fitting model involves 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 to reduce the time difference measurement error to ≤1ns.

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

[0038] S6: Calculate the specific location of the ignition point based on the total length of the cavity through geometric relationships.

[0039] The radio frequency cavity spark position 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 spark events.

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

[0041] The dual-channel differential time-of-flight measurement architecture includes: deploying fast-response arc light detectors at both ends of the RF cavity, synchronously collecting optical signals and calculating nanosecond-level time differences (Δt), breaking through the limitations of cavity length on positioning accuracy and achieving sub-centimeter coordinate solution.

[0042] The radio frequency cavity spark position diagnosis method is seamlessly integrated with the existing accelerator control system through software upgrades, eliminating the need for large-scale modifications and improving the convenience and economy of on-site applications. It only requires detectors to be installed at both ends of the cavity and signals to be transmitted via optical fiber links, making it compatible with existing accelerator structures and reducing hardware costs by more than 90%. Dynamic amplitude normalization and shared slope least squares fitting algorithms are used to effectively eliminate sensor gain differences and random noise interference, significantly improving electromagnetic interference resistance and positioning stability. Sub-centimeter spatial resolution positioning is achieved through nanosecond time difference inversion, significantly surpassing the meter-level accuracy of traditional low-level methods, pickup methods, and acoustic methods.

[0043] like Figure 1 and 2 As shown, the present invention embodies the core idea of ​​differential measurement, that is, by measuring the time difference between the two channel signals, combined with parameters such as the propagation speed of electromagnetic waves in the cavity medium, the ignition position can be accurately determined, and by accurately analyzing the nanosecond time difference between the arc signal reaching the detectors at both ends of the cavity, combined with the electromagnetic wave propagation speed model, sub-centimeter positioning can be achieved. More specifically, the present invention has a detector on each side, which is a setting in which fast-response arc detectors are symmetrically installed at both ends of the RF cavity. Through this dual-channel architecture, the relevant signals of the ignition event at both ends of the RF cavity can be captured separately, providing basic data for the subsequent differential measurement. There is a time difference between the signals captured by the two detectors. In the actual RF cavity ignition position diagnosis process, the light signal generated by the ignition event will arrive at the detectors on both sides at different times. This time difference (Δt = |t1-t2|) is the key parameter for calculating the ignition position in the subsequent implementation process, that is, recording the dual-channel original waveform data.

[0044] Figure 3 The waveforms of channel 1 and channel 2 are presented in Figure 3It can be seen that the waveforms of the two channels differ in features such as the rising edge, reflecting the signal inconsistency caused by factors such as detector gain difference and signal transmission. It is necessary to perform dynamic amplitude normalization on the signal to eliminate the impact of these differences on subsequent time difference measurements. Figure 3 The difference in the waveform, especially the time difference of the rising edge, is the basis for calculating the time difference (Δt) between the two channel signals. By accurately measuring these time differences and combining them with the propagation speed of electromagnetic waves in the cavity medium, accurate diagnosis of the spark position can be achieved. When calculating the time difference, specific threshold points are usually selected, such as the 10% and 90% steady-state threshold points. Figure 3 As a result of multiple experiments, characteristics such as the steady-state value and initial value of the waveform can provide experimental results to justify the selection of the threshold point.

[0045] Figure 4 The figure shows the raw data (solid line) and the fitted curve (dashed line) for Channel 1 and Channel 2 during the experiment. This demonstrates the technical solution's process of analyzing the step response parameters of the normalized signal, intercepting the rising segment data of both channels, and fitting the model using the least squares method with a shared slope. This curve fitting allows for more accurate determination of the signal's characteristic time points, enabling high-precision time difference calculation.

[0046] Figure 5 The figure illustrates the process of aligning the two-channel waveforms during the experiment, which is relevant to the technical solution's approach of eliminating the impact of sensor gain differences on time extraction. Since different detectors may have different gain, resulting in inconsistencies in the amplitude and timing of the two-channel signals, waveform alignment allows for more accurate calculation of the signal time difference. The annotation "90% point time alignment after fitting" in the figure indicates that a specific alignment standard was used in multiple experiments, using the 90% point of the fitted waveform as the reference point for time alignment. This alignment method helps improve the accuracy of time difference measurement and reduce errors caused by inaccurate selection of signal feature points. The figure shows the waveforms of channel 1 after fitting (solid line) and channel 2 after fitting and alignment (dashed line). A comparison shows that the aligned waveforms of the two channels are more consistent in time, providing a reliable basis for accurate calculation of the time difference (Δt). The aligned waveforms more accurately determine characteristic time points such as the rising edge of the signal. Combined with the least squares fitting model of the shared slope (Claim 3), theoretical time points can be inferred to calculate the time difference with high precision. Combined with the propagation speed of electromagnetic waves in the cavity medium (v=c / n, c is the speed of light, n is the refractive index), the time difference is converted into a spatial distance difference (ΔL=v×Δt), and the specific position of the ignition point is then calculated.

[0047] The following is further described through specific examples:

[0048] Example 1:

[0049] This embodiment is based on a dual-channel differential time-of-flight measurement architecture. It uses fast-response arc detectors installed at both ends of the RF cavity to capture the optical signal of the ignition event in real time. By accurately calculating the time difference between the light signal reaching the two detectors and combining it with the propagation speed of electromagnetic waves in the cavity medium, accurate diagnosis of the ignition location can be achieved.

[0050] Implementation Process

[0051] S1: Signal acquisition and preprocessing

[0052] Fast-response arc flash detectors are symmetrically mounted at both ends of the RF cavity. Their nanosecond response time ensures accurate capture of the optical signal generated by the arc flash event. The detectors are connected to a data acquisition system via an optical fiber link, capturing the arc flash event's optical signal in real time and recording dual-channel raw waveform data. The optical fiber link offers advantages such as immunity to electromagnetic interference and low transmission loss, ensuring stable signal transmission.

[0053] S2: Dynamic amplitude normalization

[0054] Because different detectors may have different gain, the amplitudes of the two-channel signals may be inconsistent, affecting the accuracy of time difference measurement. Therefore, dynamic amplitude normalization is performed on the two-channel signals. The specific method is to linearly scale the channel 2 signal based on the steady-state value and the initial value, forcing the amplitude range of the two-channel signals to align, eliminating the influence of sensor gain differences on time extraction.

[0055] S3: Time difference calculation

[0056] The step response parameters of the normalized signal are analyzed, and the time difference between the 10% and 90% steady-state threshold points is calculated. The rising segment of the data for both channels is then intercepted. A shared slope least squares fitting model is used to constrain the two channel signals to share the same slope. By fitting theoretical time points instead of direct threshold triggering, the time difference measurement error is reduced to ≤1ns, resulting in a highly accurate time difference (Δt).

[0057] S4: Position calculation

[0058] By 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). Given the total length of the cavity, the specific location of the ignition point can be calculated through geometric relationships.

[0059] That is, in a radio frequency cavity of a particle accelerator, the time difference between the two-channel optical 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 =1m. Combined with geometric relationships such as the total length of the cavity, the position of the ignition point in the cavity can be accurately calculated.

[0060] Example 2:

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

[0062] Implementation Process

[0063] S1: Optimizing signal acquisition

[0064] The fast-response arc detectors installed at both ends of the RF cavity utilize advanced sensor technology, offering higher sensitivity and lower noise levels. Furthermore, the data acquisition system utilizes a high-speed, high-precision data acquisition card with a sampling rate reaching the GHz level, ensuring accurate recording of subtle changes in the optical signal. The detectors are connected to the data acquisition system via an optical fiber link, capturing the optical signal of the spark event in real time and recording dual-channel raw waveform data.

[0065] S2: Improved dynamic amplitude normalization

[0066] In the dynamic amplitude normalization process, an adaptive algorithm is used to automatically adjust the linear scaling factor 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] When calculating the time difference, in addition to using a least-squares fit model with a shared slope, a machine learning algorithm is introduced to analyze and predict signal characteristics, further optimizing the fitting model and improving the accuracy of the time difference calculation. Through multiple measurements and data analysis, the time difference measurement error is controlled within 0.5ns.

[0069] S4: Accurate position calculation

[0070] By combining a more accurate electromagnetic wave propagation velocity model and considering the impact of factors such as cavity medium temperature and pressure on propagation velocity, the time difference is converted into a spatial distance difference. Using high-precision geometric measurement data and the total cavity length, a more complex geometric algorithm is used to calculate the specific location of the ignition point.

[0071] That is, in the radio frequency cavity of another particle accelerator, the time difference between the two-channel optical signals measured by the optimized system is 3ns, considering that the actual propagation speed of the cavity medium is 2.2×10 8 m / s, then the spatial distance difference ΔL=2.2×10 8×3×10 -9 =0.66m. Through precise geometric calculation, the position of the ignition point in the cavity can be determined more accurately, providing more reliable fault diagnosis information for operation and maintenance personnel.

[0072] In summary, it can be seen from the above two embodiments that the radio frequency cavity spark position diagnosis method based on differential time of flight of the present invention 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 position of a radio frequency cavity spark based on differential time of flight, characterized by: A dual-channel differential time-of-flight measurement architecture is used, which includes the following steps: S1: Fast-response arc detectors are symmetrically installed at both ends of the RF cavity and connected to the detectors via optical fiber links to capture the optical signal of the ignition event in real time and record the dual-channel raw waveform data; S2: Perform dynamic amplitude normalization on the two-channel signals, linearly scale the channel 2 signal 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 intercept the rising segment data of the two channels; S4: The shared slope least squares method is used to fit the model and reverse the theoretical time point to calculate the high-precision time difference (Δt); S5: Combined with the propagation speed of electromagnetic waves in the cavity medium (v = c / n, c is the speed of light, 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 based on the total length of the cavity through geometric relationships.

2. The method for diagnosing radio frequency cavity spark position based on differential time of flight according to claim 1, characterized in that: The dynamic amplitude normalization processing in S2 includes: linearly scaling the channel 2 signal based on the steady-state value and the initial value, aligning the amplitude ranges of the dual-channel signals, and eliminating the influence of sensor gain differences on time extraction.

3. The method for diagnosing radio frequency cavity spark position based on differential time of flight according to claim 1, characterized in that: The least squares fitting model for the shared slope in S4 includes: using the least squares method to constrain the two channel signals to share the same slope, replacing direct threshold triggering with fitting theoretical time points, and reducing the time difference measurement error to ≤1ns.

4. The method for diagnosing radio frequency cavity spark position based on differential time of flight according to claim 1, characterized in that: The radio frequency cavity spark position 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 spark events.

5. The method for diagnosing radio frequency cavity spark position based on differential time of flight according to claim 1, characterized in that: The radio frequency cavity spark position diagnosis method also includes: the system structure is compact, and the optical fiber link and the high-speed data acquisition card can be flexibly arranged in radio frequency cavities of different sizes and shapes.

6. The method for diagnosing radio frequency cavity spark position based on differential time of flight according to claim 1, characterized in that: The radio frequency cavity ignition position diagnosis method further includes: seamlessly integrating with an existing accelerator control system through software upgrade.

7. The method for diagnosing radio frequency cavity spark position based on differential time of flight according to claim 1, characterized in that: The dual-channel differential time-of-flight measurement architecture includes: deploying fast-response arc light detectors at both ends of the RF cavity, synchronously collecting optical signals and calculating nanosecond-level time differences (Δt), breaking through the limitations of cavity length on positioning accuracy and achieving centimeter-level coordinate solution.

8. The method for diagnosing radio frequency cavity spark position based on differential time of flight according to claim 1, characterized in that: The radio frequency cavity spark position diagnosis method also includes: only detectors need to be installed at both ends of the cavity and signals need to be transmitted through optical fiber links, which is compatible with existing accelerator structures and reduces hardware costs by more than 90%.

9. The method for diagnosing radio frequency cavity spark position based on differential time of flight according to claim 1, characterized in that: The radio frequency cavity spark position diagnosis method also includes: using dynamic amplitude normalization and shared slope least squares fitting algorithm to effectively eliminate sensor gain differences and random noise interference, thereby significantly improving anti-electromagnetic interference capability and positioning stability.

10. The method for diagnosing radio frequency cavity spark position based on differential time of flight according to claim 1, characterized in that: The radio frequency cavity spark position diagnosis method also includes: achieving centimeter-level 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.

Images