A method for predicting plasma disruption based on alfen eigenmodes

Abstract

This application relates to the field of plasma diagnostic technology and discloses a plasma breakup prediction method based on Alfvén eigenmodes. The method includes: acquiring a diagnostic signal of the plasma; performing a frequency domain transformation on the diagnostic signal to obtain a corresponding amplitude spectrum; detecting the presence of a pressure Alfvén eigenmode (BAE) signal based on the amplitude spectrum; and if the detected pressure Alfvén eigenmode (BAE) is excited by a tearing mode (TM), then the pressure Alfvén eigenmode (BAE) is considered a valid precursor to plasma breakup. This application enables accurate prediction of plasma breakup.

Description

Technical Field

[0001] This application relates to the field of plasma diagnostic technology, and in particular to a plasma breakup prediction method based on Alfvén eigenmodes. Background Technology

[0002] The Big Ripple is one of the most severe and rapid plasma instabilities in Tokamak plasmas. It causes the plasma to completely collapse from a high-temperature, high-confinement state within milliseconds, with destructive effects. The electromagnetic forces resulting from the ripple process can cause unacceptable damage to Tokamak devices. Therefore, the prediction and avoidance of Big Ripple is a crucial consideration for the operation of large Tokamak devices. Tear modes, a type of resistive magnetohydrodynamic instability, are one of the most significant and common causes of Big Ripple. Tear modes lead to the reconnection of magnetic field lines and the formation of magnetic island structures. The presence of these islands disrupts the integrity of the original magnetic surface, allowing particles and heat that were previously confined within the magnetic surface to rapidly traverse the plasma along the helical magnetic field lines within the island, resulting in increased local transport. When the island grows to a certain width, it interacts with the plasma rotation. Due to effects such as electromagnetic braking, the rotation of the island slows down until it stops, essentially "locked" in a fixed position on the device wall. This ultimately leads to the Big Ripple of the plasma.

[0003] Mode-locking in tearing mode is typically detected using a saddle coil. This is because the frequency of the mode-locking signal is usually below 1kHz, falling within the low-frequency magnetic disturbance range, and the saddle coil responds well to this range. Signal processing involves envelope analysis of the saddle coil signal to obtain the amplitude growth rate. Furthermore, by summarizing experimental patterns, the threshold for the amplitude growth rate was obtained. When the signal growth rate exceeds this threshold, it can be identified as a precursor to mode-locking. However, the drawback of this method is the growth rate threshold. The threshold for the rate of increase of the precursor to rupture needs to be set manually, and often changes with experimental conditions, meaning there is a lack of a universal and unified judgment standard. Furthermore, the time interval between mode-locking exceeding the threshold and rupture typically occurs on the order of milliseconds, providing limited time for rupture mitigation. In addition, not all plasma ruptures caused by tearing modes result in mode-locking; for example, when the tearing mode has developed to the point where the magnetic island width is a fraction of the plasma's small radius, nearing rupture, but due to external momentum injection, such as in the case of neutral beam injection, the rotational frequency of the tearing mode remains above 1 kHz until rupture occurs, and the rate of increase of the tearing mode before rupture occurs is significantly higher. The tearing mode width has approached zero, meaning it has reached saturation. At this point, it is impossible to predict plasma breakup in advance based on the tearing mode growth rate. Summary of the Invention

[0004] To address the limitations of existing tear mode locking detection methods, this application provides a plasma breakage prediction method based on Alfvén eigenmodes. This method can predict plasma breakage caused by tear modes a few milliseconds in advance, providing sufficient preparation time for breakage mitigation and thus enabling more effective prediction and mitigation of plasma breakage.

[0005] This application discloses a plasma fracture prediction method based on Alfvén eigenmodes, which includes: Step 1: Obtain the diagnostic signal of the plasma; the diagnostic signal is a discrete first time-series digital signal or a discrete second time-series digital signal; Step 2: Perform frequency domain transformation on the diagnostic signal to obtain the corresponding amplitude spectrum; Step 3: Based on the amplitude spectrum, detect whether there is a specific pressure Alfvén eigenmode BAE signal; Step 4: If the detected specific pressure Alfvén intrinsic mode BAE is excited by the tearing mode TM, then the specific pressure Alfvén intrinsic mode BAE is taken as an effective plasma disruption precursor.

[0006] Further, step 1 includes: During the tokamak discharge process, the plasma magnetic disturbance voltage signal induced by the magnetic diagnostic coil is acquired in real time. The original analog signal, i.e., the plasma magnetic disturbance voltage signal, is subjected to anti-aliasing filtering and analog-to-digital conversion to obtain a discrete first time-series digital signal. The sampling frequency setting of the analog-to-digital conversion needs to meet the Nyquist sampling theorem requirements for the acquisition of high-frequency BAE signals.

[0007] Furthermore, step 1 also includes: During the tokamak discharge process, microwaves of a predetermined frequency are emitted into the plasma via a microwave reflectometer, and reflected signals reflected from a predetermined cutoff layer of the plasma are received. The phase of the reflected signal... The change is proportional to the density perturbation of the cutoff layer; a sampling frequency capable of capturing the BAE band is used to measure the phase. Real-time acquisition and digital processing are performed to obtain discrete second time-series digital signals.

[0008] Further, step 2 includes: The diagnostic signal is segmented, with overlapping adjacent segments. A Fourier transform is performed on each segment to obtain the corresponding amplitude spectrum. , For time, For frequency.

[0009] Further, step 3 includes: Peak signals with amplitudes greater than a first preset amplitude threshold and durations greater than a first preset duration threshold are detected within a first frequency range to identify tearing mode TM signals; Detect newly generated peak signals with a duration longer than a second preset duration threshold within the second frequency range to identify the specific voltage Alfvén intrinsic mode (BAE) signal; Wherein, the lower limit of the second frequency range is greater than the upper limit of the first frequency range.

[0010] Furthermore, step 3 also includes: In the obtained amplitude spectrum, i.e., the density perturbation spectrum: Identify density perturbation features with a first frequency characteristic, excited by the tearing mode TM; Identify density perturbation features with second frequency characteristics excited by the specific pressure Alfvén eigenmode BAE; Among them, the microwave reflectometer is used to acquire local plasma measurement information. The frequency point corresponding to the specific pressure Alfvén eigenmode (BAE) may be different from the frequency point measured by magnetic diagnostics, and the second frequency feature is different from the first frequency feature.

[0011] Further, step 4 includes: When a specific pressure Alfvén intrinsic mode (BAE) signal exists and it has a temporal or nonlinear coupling relationship with the tearing mode (TM) signal, the plasma rupture precursor caused by the tearing mode (TM) signal is identified.

[0012] Furthermore, if the occurrence time of the specific pressure Alfvén intrinsic mode BAE signal is later than that of the identified tearing mode TM signal, then there is a timing correlation between the specific pressure Alfvén intrinsic mode BAE signal and the tearing mode TM signal. Bispectral analysis is performed on the signal period containing the tearing mode TM and the specific pressure Alfvén intrinsic mode BAE. Based on the bispectral analysis results, the peak values ​​corresponding to the tearing mode TM frequency, the specific pressure Alfvén intrinsic mode BAE frequency, and the sum or difference frequency components of the two are identified. According to the peak values, it is determined that there is nonlinear coupling between the tearing mode TM and the specific pressure Alfvén intrinsic mode BAE.

[0013] Furthermore, after step 4, the method further includes: Trigger warning and / or mitigation actions to prevent plasma disruption.

[0014] Furthermore, the triggering of the warning and / or mitigation actions includes: Trigger warning: Send a warning signal to the control system and human-machine interface of the plasma device. The warning signal is used to indicate that the specific pressure Alfvén eigenmode (BAE) excited by the tearing mode has been detected and the plasma is predicted to be at risk of rupture. Triggering mitigation action: The warning signal is output to the plasma rupture mitigation system. After receiving the warning signal, the plasma rupture mitigation system executes a preset rupture mitigation strategy within a preset delay range. The preset rupture mitigation strategy includes: controlling the high-pressure gas injection valve to inject preset gas into the plasma edge, and dissipating plasma energy through radiation cooling to avoid or reduce the damage caused by plasma rupture.

[0015] Due to the adoption of the above technical solution, this application has the following advantages: Predicting and mitigating fractures using BAE caused by mold-locking can achieve the following: 1. No need to manually set the mold-locking growth rate threshold, reducing the impact of human factors on prediction. 2. Uniform prediction standards that do not change with experimental parameters. 3. Prediction time can reach approximately 50ms, far exceeding the mold-locking development process, providing sufficient reaction time for fracture mitigation. 4. It can also predict fractures caused by unlocked mold-locking. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a schematic flowchart of a plasma fracture prediction method based on Alfvén eigenmodes according to an embodiment of this application. Figure 2 This is a schematic diagram of the spectrum and plasma current Ip timing before a plasma discharge breaks down according to an embodiment of this application; Figure 3 This is a schematic diagram of bispectral analysis of plasma diagnostic signals according to an embodiment of this application. Detailed Implementation

[0018] The present application will be further described in conjunction with the accompanying drawings and embodiments. The described embodiments are only some, not all, of the embodiments of the present application. All other embodiments obtained by those skilled in the art should fall within the protection scope of the embodiments of the present application.

[0019] See Figure 1 This application provides an embodiment of a plasma fracture prediction method based on Alfvén eigenmodes, which includes: Step 1: Obtain the diagnostic signal of the plasma; the diagnostic signal is a discrete first time-series digital signal or a discrete second time-series digital signal; Step 2: Perform frequency domain transformation on the diagnostic signal to obtain the corresponding amplitude spectrum; Step 3: Based on the amplitude spectrum, detect whether there is a specific pressure Alfvén eigenmode BAE signal; Step 4: If the detected specific pressure Alfvén intrinsic mode BAE is excited by the tearing mode TM, then the specific pressure Alfvén intrinsic mode BAE is taken as an effective plasma disruption precursor.

[0020] This application enables earlier, more general, and more reliable prediction of plasma rupture caused by tearing modes.

[0021] Optionally, step 1 includes: During the tokamak discharge process, the plasma magnetic disturbance voltage signal induced by the magnetic diagnostic coil is acquired in real time. The original analog signal, i.e., the plasma magnetic disturbance voltage signal, is subjected to anti-aliasing filtering and analog-to-digital conversion to obtain a discrete first time-series digital signal. The sampling frequency setting of the analog-to-digital conversion needs to meet the Nyquist sampling theorem requirements for the acquisition of high-frequency BAE signals.

[0022] Optionally, step 1 further includes: During the tokamak discharge process, microwaves of a predetermined frequency are emitted into the plasma via a microwave reflectometer, and reflected signals reflected from a predetermined cutoff layer of the plasma are received. The phase of the reflected signal... The change is proportional to the density perturbation of the cutoff layer; a sampling frequency capable of capturing the BAE band is used to measure the phase. Real-time acquisition and digital processing are performed to obtain discrete second time-series digital signals.

[0023] Optionally, step 2 includes: The diagnostic signal is segmented, with overlapping adjacent segments. A Fourier transform is performed on each segment to obtain the corresponding amplitude spectrum. , For time, For frequency.

[0024] Optionally, step 3 includes: Peak signals with amplitudes greater than a first preset amplitude threshold and durations greater than a first preset duration threshold are detected within a first frequency range to identify tearing mode TM signals; Detect newly generated peak signals with a duration longer than a second preset duration threshold within the second frequency range to identify the specific voltage Alfvén intrinsic mode (BAE) signal; Wherein, the lower limit of the second frequency range is greater than the upper limit of the first frequency range.

[0025] Optionally, step 3 further includes: In the obtained amplitude spectrum, i.e., the density perturbation spectrum: Identify density perturbation features with a first frequency characteristic, excited by the tearing mode TM; Identify density perturbation features with second frequency characteristics excited by the specific pressure Alfvén eigenmode BAE; Among them, the microwave reflectometer is used to acquire local plasma measurement information. The frequency point corresponding to the specific pressure Alfvén eigenmode (BAE) may be different from the frequency point measured by magnetic diagnostics, and the second frequency feature is different from the first frequency feature.

[0026] Optionally, step 4 includes: When a specific pressure Alfvén intrinsic mode (BAE) signal exists and it has a temporal or nonlinear coupling relationship with the tearing mode (TM) signal, the plasma rupture precursor caused by the tearing mode (TM) signal is identified.

[0027] Optionally, if the occurrence time of the pressure Alfvén intrinsic mode (BAE) signal is later than that of the identified tearing mode (TM) signal, then there is a time correlation between the pressure Alfvén intrinsic mode (BAE) signal and the tearing mode (TM) signal. Bispectral analysis is performed on the signal period containing the tearing mode TM and the specific pressure Alfvén intrinsic mode BAE. Based on the bispectral analysis results, the peak values ​​corresponding to the tearing mode TM frequency, the specific pressure Alfvén intrinsic mode BAE frequency, and the sum or difference frequency components of the two are identified. According to the peak values, it is determined that there is nonlinear coupling between the tearing mode TM and the specific pressure Alfvén intrinsic mode BAE.

[0028] Optionally, after step 4, the method further includes: Trigger warning and / or mitigation actions to prevent plasma disruption.

[0029] Optionally, the triggering of the warning and / or mitigation action includes: Trigger warning: Send a warning signal to the control system and human-machine interface of the plasma device. The warning signal is used to indicate that the specific pressure Alfvén eigenmode (BAE) excited by the tearing mode has been detected and the plasma is predicted to be at risk of rupture. Triggering mitigation action: The warning signal is output to the plasma rupture mitigation system. After receiving the warning signal, the plasma rupture mitigation system executes a preset rupture mitigation strategy within a preset delay range. The preset rupture mitigation strategy includes: controlling the high-pressure gas injection valve to inject preset gas into the plasma edge, and dissipating plasma energy through radiation cooling to avoid or reduce the damage caused by plasma rupture.

[0030] For ease of understanding, this application provides more specific embodiments: Example 1: Prediction of tearing mode fracture based on magnetic diagnostic signals This embodiment uses the signal from a magnetic diagnostic coil (such as a saddle coil or magnetic probe) conventionally installed on a tokamak device as the data source to implement the prediction method described in this application. Figure 2 and Figure 3 The experimental results shown are based on this type of signal.

[0031] Combination Figure 2 and Figure 3 The specific implementation steps of this embodiment are as follows: Step 1: Real-time data acquisition and preprocessing.

[0032] During the tokamak discharge process, the plasma magnetic disturbance voltage signal induced by the magnetic diagnostic coil is acquired in real time. This raw analog signal is then subjected to anti-aliasing filtering (e.g., a low-pass filter cutoff frequency of 100 kHz) and analog-to-digital conversion (ADC sampling frequency must be ≥200 kHz to meet the Nyquist sampling theorem requirements for high-frequency BAE signal acquisition) to obtain a discrete time-series digital signal. .

[0033] Step 2: Time-frequency analysis and spectrum acquisition.

[0034] For the preprocessed signal Continuous time-frequency analysis is performed to monitor the changes in the spectrum over time. Specifically, the Short Time Fourier Transform (STFT) method can be used: the signal is segmented (e.g., each segment corresponds to a 5-10 ms time window, with overlapping segments), and a Fourier transform is performed on each segment to obtain the amplitude spectrum within that time period. . Figure 2 The two-dimensional color spectrum in the upper middle part is generated by this method. Its horizontal axis is time, its vertical axis is frequency, and the color intensity represents the signal power.

[0035] Step 3: Feature pattern recognition.

[0036] In the real-time updated spectrum In the process, two types of feature patterns are identified in parallel: Low-frequency tearing mode (TM) identification: within the frequency range (For example, 1 kHz < Within the range of <10 kHz, look for significant and sustained peak signals. For example... Figure 2 As shown, a clear TM signal appears in the region below 10 kHz, starting from approximately 2050 ms.

[0037] High-frequency BAE identification: in a higher frequency range (For example, 20 kHz) Within the range of <40 kHz, look for newly emerging, sustained peak signals. For example... Figure 2As shown, a significant BAE signal peak appeared in the 20-30 kHz frequency band at approximately 2200 ms.

[0038] Step 4: Physical correlation confirmation and rupture precursor judgment.

[0039] This step is crucial to this application, as it confirms that the detected BAE is TM-triggered, thus identifying it as a valid precursor to rupture.

[0040] Timing correlation: Confirm that the BAE signal appears after the identified TM signal. For example... Figure 2 The TM already existed at 2050 ms, while the BAE only appeared at 2200 ms, which conforms to the causal logic of "TM first, then BAE is triggered".

[0041] Verification of Nonlinear Interactions: To further confirm the physical connection between BAE and TM, bispectral analysis can be performed on data including both TM and BAE periods. Bispectral analysis can detect nonlinear couplings between different frequency components of a signal. For example... Figure 3 As shown, there is a distinct peak in the bispectral diagram, the coordinates of which correspond to the TM frequency ( ), BAE frequency ( ) and their sum or difference frequencies (e.g. ). Figure 3 The black area indicated by the black arrow represents a strong nonlinear interaction between TM and BAE, providing direct evidence that "BAE is excited by TM". Figure 3 middle, For TM frequency, This refers to the BAE frequency.

[0042] Precursor determination: When both of the following conditions are met simultaneously, it is determined to be a precursor to plasma rupture caused by tearing mode: Condition 1: A continuous BAE signal is clearly identified in the spectrum. (frequency band).

[0043] Condition 2: The BAE signal has a time-series correlation with the existing TM signal, and / or nonlinear coupling is confirmed through bispectral analysis.

[0044] Step 5: Warning triggering and mitigation actions.

[0045] Once the signs of impending rupture are identified, the following actions will be triggered immediately: Issue an early warning: Send a high-level early warning signal to the device control system and operator interface, indicating that "BAE triggered by tearing mode has been detected, and rupture is predicted to occur in tens of milliseconds."

[0046] Automatic Triggering Mitigation System: Simultaneously, the warning signal is interlocked to the plasma fracture mitigation system. Upon receiving the signal, this system automatically executes the preset fracture mitigation plan within a very short delay (e.g., within 1-2 ms). Figure 2 As shown, there is a time window of approximately 50 ms from the appearance of BAE (~2200 ms) to the occurrence of a small fracture (~2250 ms). The mitigation system can perform actions within this window, such as triggering the high-pressure gas injection valve to inject a small amount of neon or argon gas into the plasma edge, thereby uniformly dissipating plasma energy through radiative cooling, thus avoiding or reducing the destructive force of the large fracture.

[0047] Technical effect: This embodiment successfully predicted a non-molding fracture ( Figure 2 (The TM frequency is always >1 kHz). Traditional methods based on TM growth rate or mode-locking fail in this case. However, this application utilizes BAE precursors to provide an effective early warning approximately 50 ms in advance, gaining valuable time for proactive mitigation.

[0048] Example 2: Tear Mode Fragmentation Prediction Based on Microwave Reflectometer Signals This embodiment demonstrates the universality of the method of this application for different diagnostic tools. A microwave reflectometer is used to measure the density perturbation signal in a localized area of ​​the plasma.

[0049] Step 1: Data collection.

[0050] In a tokamak discharge, a microwave reflectometer emits microwaves of a specific frequency into the plasma and receives the signal reflected back from a specific cutoff layer of the plasma. The phase change of the reflected signal is proportional to the density perturbation of that cutoff layer. The phase signal of this reflected signal is acquired and digitized in real time. The sampling frequency needs to be high enough to capture the BAE band (e.g., ≥500kHz).

[0051] Step 2: Spectrum analysis and pattern recognition.

[0052] right Perform a time-frequency analysis similar to step 2 in Example 1. In the obtained density perturbation spectrum: Identify density perturbation characteristics caused by TM at low frequencies (<10 kHz).

[0053] Identify density perturbation characteristics caused by BAE at high frequencies (e.g., 15-60 kHz). Since the reflectometer measures local information, the BAE signal may appear at a different specific frequency than in magnetic diagnostics, but its characteristics as a high-frequency mode are still evident.

[0054] Step 3: Association determination and early warning.

[0055] The same logic as step 4 in Example 1 is applied to determine whether a high-frequency BAE mode is present and whether it has a time-sequential or non-linear coupling relationship with a low-frequency TM mode. Once determined, the warning and mitigation process described in step 5 of Example 1 is triggered.

[0056] The beneficial effects of this embodiment are: it proves that this application does not rely on specific magnetic diagnostics; as long as the diagnostic means can sense the perturbation of plasma parameters (density, temperature, magnetic field) caused by BAE excited by TM, this method can be applied. This greatly broadens the applicable scenarios and device scope of this application.

[0057] In summary, the above embodiments specifically illustrate how to achieve early, general, and reliable fracture prediction by monitoring BAE excited by the tearing mode. Embodiment 1 details the process, key judgment steps (especially dual-spectrum verification), and significant technical effects with reference to the accompanying drawings; Embodiment 2 demonstrates the scalability of the technical solution of this application. These specific embodiments provide sufficient support for the scope of protection outlined in the claims.

[0058] This application obtains spectral information by performing Fourier transform on plasma diagnostic signals. Unlike previous methods that directly analyze low-frequency tearing modes, this method predicts plasma disruption by analyzing the Alfvén eigenmodes (BAEs) induced by the specific pressure of the tearing modes. In tokamak plasmas, strong tearing modes significantly alter the local magnetic field structure and plasma profile, thereby exciting BAEs. Tearing modes create the necessary conditions for BAE excitation by modifying the background profile and providing free energy. When a BAE induced by a tearing mode occurs, it indicates that the tearing mode has grown sufficiently to cause plasma disruption. Therefore, a BAE induced by a tearing mode can be considered a precursor to disruption caused by the tearing mode. For example... Figure 2 As shown, this is the spectrum before a plasma discharge breaks down. A distinct tearing mode appears around 2050 ms, and its intensity and frequency are relatively stable. The sign of plasma breakdown is a rapid rise in the plasma current Ip; a small breakdown occurs at 2250 ms, and a large breakdown at 2290 ms. The BAE (Blasing Electrode Effect) induced by the tearing mode appears around 2200 ms and persists for about 40 ms, with a small plasma breakdown occurring around 2250 ms. Therefore, using the BAE signal induced by the tearing mode as a prediction signal, the breakdown caused by the tearing mode can be predicted approximately 50 ms in advance. Furthermore, since the frequency of the tearing mode is greater than 1 kHz before the small breakdown occurs at 2250 ms, the breakdown caused by this tearing mode differs from conventional mode-locked breakdown and cannot be predicted using methods such as saddle coils. Figure 3The image shows the bispectral diagnostic signal when both tearing patterns and BAEs are present. The black signals in the image indicate the interaction between the tearing patterns and BAEs. Figure 2 The fact that the tear modulus persists longer than the base-effect (BAE) duration, where the BAE arises due to the tear modulus modifying the background profile and providing free energy, means that the BAE can serve as an indicator of tear modulus strength, thereby predicting fracture caused by the tear modulus.

[0059] The core of this application lies in the fundamental shift in prediction logic and detection object, namely: from "directly monitoring the low-frequency growth and mode-locking behavior of the tearing mode (TM) itself" to "monitoring the emergence of the high-frequency specific pressure Alfvén intrinsic mode (BAE) excited by the strong tearing mode, and using it as a precursor to rupture."

[0060] Figure 2 In the diagram, the low-frequency range (<10 kHz) represents the tearing mode (TM), and the high-frequency range (20~30 kHz) represents the Alfvén eigenmode (BAE) caused by specific pressure. The solid red line represents the plasma current. The sign of plasma disruption is a rapid rise in the plasma current Ip. In the diagram, a small disruption occurs at 2250 ms, and a large disruption occurs at 2290 ms.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application and not to limit them. Although this application has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of this application. Any modifications or equivalent substitutions that do not depart from the spirit and scope of this application should be covered within the protection scope of the claims of this application.

Claims

1. A plasma disruption prediction method based on Alfvén eigenmodes, characterized in that, include: Step 1: Obtain the diagnostic signal of the plasma; the diagnostic signal is a discrete first time-series digital signal or a discrete second time-series digital signal; Step 2: Perform frequency domain transformation on the diagnostic signal to obtain the corresponding amplitude spectrum; Step 3: Based on the amplitude spectrum, detect whether there is a specific pressure Alfvén eigenmode BAE signal; Step 4: If the detected specific pressure Alfvén intrinsic mode BAE is excited by the tearing mode TM, then the specific pressure Alfvén intrinsic mode BAE is taken as an effective plasma disruption precursor.

2. The method according to claim 1, characterized in that, Step 1 includes: During the tokamak discharge process, the plasma magnetic disturbance voltage signal induced by the magnetic diagnostic coil is acquired in real time. The original analog signal, i.e., the plasma magnetic disturbance voltage signal, is subjected to anti-aliasing filtering and analog-to-digital conversion to obtain a discrete first time-series digital signal. The sampling frequency setting of the analog-to-digital conversion needs to meet the Nyquist sampling theorem requirements for the acquisition of high-frequency BAE signals.

3. The method according to claim 1, characterized in that, Step 1 further includes: During the tokamak discharge process, microwaves of a predetermined frequency are emitted into the plasma via a microwave reflectometer, and reflected signals reflected from a predetermined cutoff layer of the plasma are received. The phase of the reflected signal... The change is proportional to the density perturbation of the cutoff layer; a sampling frequency capable of capturing the BAE band is used to measure the phase. Real-time acquisition and digital processing are performed to obtain discrete second time-series digital signals.

4. The method according to claim 2 or 3, characterized in that, Step 2 includes: The diagnostic signal is segmented, with overlapping adjacent segments. A Fourier transform is performed on each segment to obtain the corresponding amplitude spectrum. , For time, For frequency.

5. The method according to claim 2, characterized in that, Step 3 includes: Peak signals with amplitudes greater than a first preset amplitude threshold and durations greater than a first preset duration threshold are detected within a first frequency range to identify tearing mode TM signals; Detect newly generated peak signals with a duration longer than a second preset duration threshold within the second frequency range to identify the specific voltage Alfvén intrinsic mode (BAE) signal; Wherein, the lower limit of the second frequency range is greater than the upper limit of the first frequency range.

6. The method according to claim 3, characterized in that, Step 3 also includes: In the obtained amplitude spectrum, i.e., the density perturbation spectrum: Identify density perturbation features with a first frequency characteristic, excited by the tearing mode TM; Identify density perturbation features with second frequency characteristics excited by the specific pressure Alfvén eigenmode BAE; Among them, the microwave reflectometer is used to acquire local plasma measurement information. The frequency point corresponding to the specific pressure Alfvén eigenmode (BAE) may be different from the frequency point measured by magnetic diagnostics, and the second frequency feature is different from the first frequency feature.

7. The method according to claim 1, characterized in that, Step 4 includes: When a specific pressure Alfvén intrinsic mode (BAE) signal exists and it has a temporal or nonlinear coupling relationship with the tearing mode (TM) signal, the plasma rupture precursor caused by the tearing mode (TM) signal is identified.

8. The method according to claim 7, characterized in that, If the occurrence time of the specific pressure Alfvén intrinsic mode (BAE) signal is later than that of the identified tearing mode (TM) signal, then there is a timing correlation between the specific pressure Alfvén intrinsic mode (BAE) signal and the tearing mode (TM) signal. Bispectral analysis is performed on the signal period containing the tearing mode TM and the specific pressure Alfvén intrinsic mode BAE. Based on the bispectral analysis results, the peak values ​​corresponding to the tearing mode TM frequency, the specific pressure Alfvén intrinsic mode BAE frequency, and the sum or difference frequency components of the two are identified. According to the peak values, it is determined that there is nonlinear coupling between the tearing mode TM and the specific pressure Alfvén intrinsic mode BAE.

9. The method according to claim 1, characterized in that, After step 4, the following is also included: Trigger warning and / or mitigation actions to prevent plasma disruption.

10. The method according to claim 9, characterized in that, The triggering of warnings and / or mitigation actions includes: Trigger warning: Send a warning signal to the control system and human-machine interface of the plasma device. The warning signal is used to indicate that the specific pressure Alfvén eigenmode (BAE) excited by the tearing mode has been detected and the plasma is predicted to be at risk of rupture. Triggering mitigation action: The warning signal is output to the plasma rupture mitigation system. After receiving the warning signal, the plasma rupture mitigation system executes a preset rupture mitigation strategy within a preset delay range. The preset rupture mitigation strategy includes: controlling the high-pressure gas injection valve to inject preset gas into the plasma edge, and dissipating plasma energy through radiation cooling to avoid or reduce the damage caused by plasma rupture.

Images