A physical experiment rack high-voltage pulse power protection system
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XIAN JIANNING ELECTRONIC TECH CO LTD
- Filing Date
- 2026-03-28
- Publication Date
- 2026-06-19
Smart Images
Figure CN122246647A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of high-voltage pulse power supply and pulse power control technology, and in particular to a high-voltage pulse power supply protection system for a physical experiment rack. Background Technology
[0002] In precision physics experiments such as the electron cyclotron resonance heating of nuclear fusion in fully superconducting tokamaks, the rack-mounted high-voltage pulse power supply needs to output extremely steep high-voltage pulses to physical loads such as microwave cyclotron tubes. During the transient process of the pulse rise edge, the inherent parasitic capacitance of the physical load will generate a huge normal capacitive displacement current.
[0003] Traditional rack-mounted high-voltage pulse power supply protection systems mostly employ hardware comparators based on one-dimensional static current extreme value comparison as their core architecture. Because traditional hardware comparators lack the boundary conditions for judging transient electrical change rates and actual physical arc losses, the system's underlying physical logic is completely unable to effectively distinguish between normal capacitive charging surges and very early, minute arcing currents under extremely steep pulse fast edges. This causes the protection device to easily misjudge high-frequency charging surges as insulation short circuits during the pulse voltage build-up phase, leading to frequent current-limiting malfunctions and forcibly interrupting irreversible physical experiments. Summary of the Invention
[0004] To overcome the above shortcomings, this invention provides a high-voltage pulse power supply protection system for physical experimental racks, aiming to improve the problem that most traditional protection systems use static threshold hard shutdown, which cannot distinguish between normal charging spikes and minor arcing under extremely steep pulse fast edges, thus causing frequent malfunctions.
[0005] In a first aspect, the present invention provides the following technical solution: a method for protecting a high-voltage pulse power supply for a physical experiment rack, comprising the following steps: S1, acquire the transient voltage sampling sequence and transient current sampling sequence at the output terminal of the rack high-voltage pulse power supply; S2, when the rising edge of the output pulse is detected, a preset leading edge blanking window is triggered. The current change rate and voltage drop rate are calculated synchronously within the preset leading edge blanking window to identify and filter out the fast-edge capacitive surge characteristics caused by the nonlinear parasitic capacitance of the physical load. S3, after the preset leading edge blanking window ends, the instantaneous energy integral of the transient current is calculated in real time, and when the instantaneous energy integral exceeds the ignition energy threshold set based on the dynamic envelope, it is determined that the physical load has an insulation breakdown physical singularity and a targeted intervention command is generated. S4, receive the targeted intervention command, output a microsecond-level blocking signal to the pulse width modulation controller to turn off the main power switch, and simultaneously trigger the active clamping discharge circuit to absorb the residual arc energy in the physical circuit, and then generate a quenching state signal. S5, receive the quenching state signal and count the historical fault frequency. When the historical fault frequency exceeds the preset target aging index, dynamically increase the ignition energy threshold according to the preset adaptive step size.
[0006] By adopting the above technical solution, the leading-edge capacitive surge is filtered out and active clamping arc extinguishing is performed by combining energy integration. This accurately decouples the normal displacement current from the insulation breakdown singularity, thereby improving the problem that most traditional protection systems use static threshold hard shutdown, which cannot distinguish between normal charging spikes and minor arcing under extremely steep pulse fast edges, thus causing frequent malfunctions.
[0007] Optionally, in S1, acquiring the transient voltage sampling sequence and transient current sampling sequence at the output terminal of the rack high-voltage pulse power supply includes: The continuous analog voltage and continuous analog current of the rack high-voltage pulse power supply RF output port are synchronously acquired through an RF analog-to-digital converter. The continuous analog voltage and the continuous analog current are discretized according to a preset high-frequency sampling rate to generate the transient voltage sampling sequence and the transient current sampling sequence.
[0008] Optionally, in S2, the current change rate and voltage drop rate are simultaneously calculated within the preset leading-edge blanking window to identify and filter out fast-edge capacitive surge characteristics caused by the nonlinear parasitic capacitance of the physical load, including: Within the preset leading-edge blanking window, the derivative of the transient current sampling sequence of the current sampling period is calculated as the current change rate, and the absolute value of the derivative of the transient voltage sampling sequence is calculated as the voltage drop rate. Calculate the ratio coefficient between the rate of change of current and the rate of decrease of voltage; When the ratio coefficient is within a preset safe range that characterizes the charging and discharging characteristics of the inherent parasitic capacitance, the current current spike is determined to be the fast-edge capacitive surge characteristic, and the underlying hardware overcurrent triggering logic is shielded within the preset leading-edge blanking window.
[0009] Optionally, in S3, the ignition energy threshold set based on the dynamic envelope includes: Extract the baseline operating data of the rack high-voltage pulse power supply under normal no-load and resistive rated load conditions; Based on the aforementioned benchmark operating data, a normal current fluctuation boundary is constructed in the time domain; The normal current fluctuation boundary is extended outward by a preset redundancy margin to generate the dynamic envelope that defines the safe operating area, and the energy integral safety extreme value corresponding to the current moment on the dynamic envelope is extracted as the ignition energy threshold.
[0010] Optionally, in S3, the instantaneous energy integral of the transient current is calculated in real time, and when the instantaneous energy integral exceeds the arcing energy threshold set based on the dynamic envelope, a physical singularity of insulation breakdown of the physical load is determined and a targeted intervention command is generated, including: Extract the discrete current values from the transient current sampling sequence; The instantaneous energy integral is obtained by rolling summation of the product of the square of the discrete current value and the discrete sampling time interval using a sliding time window; The instantaneous energy integral is compared with the ignition energy threshold in real time. When it is determined that the instantaneous energy integral is greater than the ignition energy threshold, the targeted intervention command is output.
[0011] Optionally, in S4, the output of the microsecond-level blocking signal to the pulse width modulation controller to turn off the main power switch and simultaneously trigger the active clamping discharge circuit to absorb the residual arc energy in the physical circuit includes: By forcibly pulling the drive pin level of the pulse width modulation controller low through the underlying hardware interrupt channel, the main power switch of the rack high voltage pulse power supply is hard-shut down, blocking the continued feeding of bus energy to the physical circuit. A high-level trigger signal is output to turn on the clamping switch device connected in parallel with the physical load, and the inductive residual arc energy stored in the physical circuit is introduced into the discharge resistor network connected in series with the clamping switch device for dissipation, thereby forcibly extinguishing the arc.
[0012] Optionally, in S5, the step of statistically analyzing historical fault frequencies, and dynamically adjusting the ignition energy threshold by a preset adaptive step size when the historical fault frequency exceeds a preset target aging index, includes: Extract the total number of running pulses and the number of ignition actions that trigger the quenching state signal within the preset statistical period; Calculate the ratio of the number of ignition actions to the total number of operating pulses to obtain the actual failure probability; The actual failure probability is compared with the preset probability upper limit constant that characterizes the preset target material aging index; When the actual failure probability is determined to be greater than the preset probability upper limit constant, it is determined that the physical load target material has suffered severe insulation aging, and the ignition energy threshold is controlled to increase by one preset adaptive step size.
[0013] Secondly, the present invention provides the following technical solution: a high-voltage pulse power supply protection system for a physical experiment rack, comprising the following modules: The high-frequency transient waveform sampling module is used to acquire the transient voltage sampling sequence and transient current sampling sequence at the output terminal of the rack high-voltage pulse power supply. The fast-edge capacitive surge feature shielding module is used to trigger a preset leading edge blanking window when the rising edge of the output pulse is detected. Within the preset leading edge blanking window, the current change rate and voltage drop rate are calculated simultaneously to identify and filter out the fast-edge capacitive surge feature caused by the nonlinear parasitic capacitance of the physical load. The ignition energy integration and diagnosis module is used to calculate the instantaneous energy integration of the transient current in real time after the preset leading edge blanking window ends, and when the instantaneous energy integration exceeds the ignition energy threshold set based on the dynamic envelope, it determines that the physical load has an insulation breakdown physical singularity and generates a targeted intervention command. The hardware-level blocking and clamping execution module is used to receive the targeted intervention command, output a microsecond-level blocking signal to the pulse width modulation controller to turn off the main power switch, and simultaneously trigger the active clamping discharge circuit to absorb the residual arc energy in the physical circuit, and then generate a quenching state signal. The historical fault statistics and threshold compensation module is used to receive the quenching state signal and count the historical fault frequency. When the historical fault frequency exceeds the preset target aging index, the ignition energy threshold is dynamically increased by a preset adaptive step size.
[0014] Thirdly, the invention provides the following technical solution: a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned method for protecting a high-voltage pulse power supply for a physical experiment rack.
[0015] Fourthly, the present invention provides the following technical solution: a readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the above-mentioned method for protecting a high-voltage pulse power supply for a physical experimental rack.
[0016] The present invention has the following beneficial effects: 1. In this invention, by filtering out the leading-edge capacitive surge and combining it with energy integration for active clamping arc extinguishing, the normal displacement current and insulation breakdown singularity are precisely decoupled. This improves the problem that traditional protection systems mostly use static threshold hard shutdown, which cannot distinguish between normal charging spikes and minor arcing under extremely steep pulse fast edges, thus causing frequent malfunctions.
[0017] 2. In this invention, the charging and discharging characteristics of the parasitic capacitance of the physical load are accurately identified by calculating the ratio coefficient of the current change rate to the voltage drop rate. This improves the problem that traditional waveform identification mostly uses simple amplitude comparison, which lacks a judgment boundary for transient change rate, thus causing high-frequency charging surges to be misjudged as insulation short circuits.
[0018] 3. In this invention, the instantaneous energy integral of the transient current is calculated in real time and compared with the threshold set by the dynamic envelope, thereby quantifying the real physical arc loss. This improves the problem that traditional fault diagnosis mostly uses instantaneous current extreme value comparison, which is difficult to eliminate high-frequency wandering glitches, resulting in serious missed detection of breakdown singularity identification.
[0019] 4. In this invention, by statistically analyzing the frequency of historical faults and dynamically increasing the ignition energy threshold, the diagnostic benchmark is calibrated according to the actual aging and loss of the physical target material. This improves the problem that traditional control logic mostly uses fixed static parameters, which cannot detect the pressure drift caused by the aging of the target material, thus causing the judgment sensitivity to gradually fail. Attached Figure Description
[0020] Figure 1 This is a flowchart of a high-voltage pulse power supply protection method for a physical experiment rack proposed in this invention; Figure 2 This is an architectural diagram of a high-voltage pulse power supply protection system for a physical experiment rack proposed in this invention; Figure 3 This is a block diagram of the computer equipment and readable storage medium for a high-voltage pulse power supply protection method for a physical experiment rack proposed in this invention. Detailed Implementation
[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0022] Example 1: In a first embodiment of the present invention, the present invention provides a method for protecting a high-voltage pulse power supply for a physical experiment rack, such as... Figure 1 As shown, it includes the following steps: S1, acquire the transient voltage sampling sequence and transient current sampling sequence at the output terminal of the rack high-voltage pulse power supply; Furthermore, in S1, acquiring the transient voltage sampling sequence and transient current sampling sequence at the output terminal of the rack high-voltage pulse power supply includes: The continuous analog voltage and continuous analog current of the rack high-voltage pulse power supply RF output port are synchronously acquired through an RF analog-to-digital converter. The continuous analog voltage and continuous analog current are discretized according to the preset high-frequency sampling rate to generate transient voltage sampling sequences and transient current sampling sequences.
[0023] Specifically, step S1 in the entire rack-mounted high-voltage pulse power supply protection system undertakes the data pre-acquisition task of converting the underlying physical signals to the digital domain. The output pulse rise edge of the rack-mounted high-voltage pulse power supply is extremely steep, and nanosecond-level high-frequency transient fluctuations are easily generated at the load end. Conventional sampling methods cannot accurately capture the true electrical characteristics of such fast-edge signals. This step, based on the high-frequency cycle set at the hardware layer, converts the analog electrical quantities in the high-voltage main circuit into the basic digital source required for subsequent protection logic judgment.
[0024] The data input terminal connects to the RF output port of the rack-mounted high-voltage pulse power supply. The RF analog-to-digital converter, acting as a hardware actuator, directly probes this physical node, extracting continuously occurring analog voltage and current. Employing an RF-level synchronous acquisition mode ensures strict phase alignment of the acquired voltage and current waveforms on the absolute time axis, avoiding transient distortion caused by the time difference between dual-channel sampling.
[0025] After the physical quantities are extracted, the data stream proceeds to the discretization process. The control layer, based on the crystal oscillator clock configuration and a preset high-frequency sampling rate, performs time-interval truncation and quantization conversion on continuous analog voltages and currents. The mathematical expression for this discretization sampling process is defined as follows: ; ; ; in This represents the preset high-frequency sampling rate value of the system's underlying configuration. This represents the discrete sampling time interval corresponding to the sampling rate mentioned above. This represents the continuous analog voltage variables acquired by the front-end data acquisition system. This represents the continuous analog current variable acquired by the pre-acquisition system. This represents the sequence index value in the discrete sampling process, and is limited to positive integers. This represents the discrete voltage values within the transient voltage sampling sequence output after discretization. This represents the discrete current values within the transient current sampling sequence output after discretization.
[0026] After undergoing the aforementioned hardware acquisition and numerical discretization process, the original continuous analog electrical quantities are reconstructed into an arrayed data stream that can be directly processed by the controller. The final output transient voltage and transient current sampling sequences completely retain the nonlinear evolution characteristics of the original high-frequency pulse fast edges, and are directly supplied to subsequent modules for characteristic shielding of capacitive surges and instantaneous integral comparison of ignition energy.
[0027] S2, when the rising edge of the output pulse is detected, a preset leading edge blanking window is triggered. Within the preset leading edge blanking window, the current change rate and voltage drop rate are calculated simultaneously to identify and filter out the fast-edge capacitive surge characteristics caused by the nonlinear parasitic capacitance of the physical load. Furthermore, in S2, the rate of change of current and the rate of drop of voltage are calculated simultaneously within a preset leading-edge blanking window to identify and filter out fast-edge capacitive surge characteristics caused by the nonlinear parasitic capacitance of the physical load, including: Within the preset leading-edge blanking window, the derivative of the transient current sampling sequence of the current sampling period is calculated as the current change rate, and the absolute value of the derivative of the transient voltage sampling sequence is calculated as the voltage drop rate. Calculate the ratio coefficient of the rate of change of current to the rate of decrease of voltage; When the ratio coefficient is within the preset safe range that characterizes the charging and discharging characteristics of the inherent parasitic capacitance, the current current spike is determined to be a fast-edge capacitive surge, and the underlying hardware overcurrent triggering logic is shielded within the preset leading-edge blanking window.
[0028] Specifically, step S2 executes the interception and judgment from signal feature extraction to hardware anti-malfunction logic. At the instant the rack-mounted high-voltage pulse power supply outputs a steep pulse fast edge, the nonlinear parasitic capacitance at the load end will inevitably trigger a charging / discharging peak current with a very large amplitude. Conventional fixed protection logic is highly prone to false alarms during this extremely brief state. The core function of this step is to utilize the dynamic electrical change rate characteristics within the leading-edge blanking time window to purely isolate and allow normal capacitive charging surges at the physical level.
[0029] The input terminal of the data stream receives the transient voltage sampling sequence and transient current sampling sequence acquired previously. When the underlying hardware detects that the output pulse is in the rising edge ramp-up state, the system immediately activates and enters the preset leading edge blanking window.
[0030] During the blanking window, the computational unit performs differentiation operations on the discrete voltage and discrete current sequences, respectively. The mathematical formulas involved in the feature solution are defined as follows: ; ; ; in This represents the scalar quantity representing the rate of change of current obtained from the solution. This represents the transient current value in the input sequence. This represents the time variable corresponding to discrete sampling. This indicates that the time derivative operation is performed on the transient current. This represents the scalar value representing the voltage drop rate obtained from the solution. This represents the transient voltage value in the input sequence. This indicates that the time derivative of the transient voltage is performed, and the outermost vertical line symbol indicates that the absolute value of the derivative is extracted. This represents the ratio coefficient obtained using the results of the first two operations.
[0031] After parameter calculation is completed, the system proceeds to the safety limit verification node. A preset safety range characterizing the inherent parasitic capacitance charging and discharging characteristics is pre-programmed into the system. The threshold of this range strictly anchors the charge transfer limit of the real physical medium during normal voltage build-up. The system compares the ratio coefficient obtained in real-time with the preset safety range. When the conditional judgment ratio coefficient falls exactly within the preset safety range, the system confirms, from a physical mechanism perspective, that the current surge is purely a fast-edge capacitive surge.
[0032] Once the determination result is established, the system immediately performs a shielding action on the underlying electrical structure. Within the entire preset leading-edge blanking window, the trigger signal output channel of the underlying hardware overcurrent determination logic is forcibly cut off. This execution flow completely eliminates hardware shutdown faults caused by normal parasitic displacement current during the initial voltage build-up phase, ensuring the complete output of the high-frequency, high-voltage pulse fast edge and the continuity of subsequent experiments.
[0033] S3, after the preset leading edge blanking window ends, calculates the instantaneous energy integral of the transient current in real time, and when the instantaneous energy integral exceeds the ignition energy threshold set based on the dynamic envelope, determines that the physical load has an insulation breakdown physical singularity and generates a targeted intervention command. Furthermore, in S3, the ignition energy threshold set based on the dynamic envelope includes: Extract the baseline operating data of the rack high-voltage pulse power supply under normal no-load and resistive rated load conditions; A normal current fluctuation boundary is constructed in the time domain based on the baseline operating data; The normal current fluctuation boundary is extended outward by a preset redundancy margin to generate a dynamic envelope that defines the safe operating area, and the energy integral safety extreme value corresponding to the current moment on the dynamic envelope is extracted as the ignition energy threshold.
[0034] In S3, the instantaneous energy integral of the transient current is calculated in real time. When the instantaneous energy integral exceeds the arcing energy threshold set based on the dynamic envelope, a physical singularity of insulation breakdown in the physical load is determined, and targeted intervention commands are generated, including: Extract discrete current values from transient current sampling sequences; The instantaneous energy integral is obtained by rolling summation of the product of the square of the discrete current value and the discrete sampling time interval using a sliding time window; The instantaneous energy integral is compared with the ignition energy threshold in real time. When the instantaneous energy integral is determined to be greater than the ignition energy threshold, a targeted intervention command is output.
[0035] Specifically, step S3 in the rack-mounted high-voltage pulse power supply protection system undertakes the core tasks of fault singularity identification and command issuance. This step abandons the hardware comparison method of using a single static current extreme value in traditional protection schemes, and instead uses the cumulative effect of instantaneous Joule energy to quantify the real physical loss of the arc, accurately eliminate high-frequency wandering glitches, and lock the real physical singularity of insulation breakdown.
[0036] The system control unit first establishes the judgment benchmark. The data extraction end reads the benchmark operating data obtained from the rack high-voltage pulse power supply under normal no-load and resistive rated load conditions. The calculation core plots the normal current fluctuation boundary in the time dimension based on the above benchmark operating data. To prevent conventional electrical fluctuations from touching the boundary, the system extends the normal current fluctuation boundary outward by a preset redundancy margin, thereby generating a dynamic envelope that defines the safe operating area. The system extracts the energy integral safety extreme value corresponding to the current calculation moment on this dynamic envelope in real time along the time axis and directly assigns it as the ignition energy threshold. .
[0037] After the timing period of the preset leading-edge blanking window ends, the system officially enters the physical singularity diagnosis process. The input end extracts the discrete current values within the transient current sampling sequence. The computation module calls the sliding time window logic embedded in the underlying layer to perform mathematical calculations on the feature quantities within the window coverage area, calculating the instantaneous energy integral. The specific mathematical formula for integral summation is defined as follows: ; in This represents the instantaneous energy integral scalar generated by the solution. This represents the sorting index value of each discrete sampling point within the current sliding time window. This represents the total number of sampling points contained within the sliding time window set at the underlying level. Represents the corresponding number The specific value of the discrete current at each discrete sampling point. This represents the discrete sampling time interval between two adjacent discrete sampling points.
[0038] The comparison execution module receives the calculated instantaneous energy integral in real time and compares it with the previously determined ignition energy threshold. Within each computation cycle, the system performs a hard comparison between the instantaneous energy integral and the ignition energy threshold. When the underlying logic determines that the instantaneous energy integral exceeds the ignition energy threshold, the system physically confirms that a substantial insulation breakdown has occurred at the physical load end. After confirmation, the system immediately generates a targeted intervention command and dispatches this command as the final output data stream of this step to the subsequent hardware execution stage.
[0039] By setting the energy boundary based on the dynamic envelope and accumulating the work done by the real electric arc using a sliding time window, the fundamental defects of traditional one-dimensional extreme value comparators, which are prone to missing reports when faced with small sparks and to false alarms when faced with capacitive spikes, are completely solved, ensuring the accuracy of targeted intervention commands.
[0040] S4 receives the targeted intervention command, outputs a microsecond-level blocking signal to the pulse width modulation controller to turn off the main power switch, and simultaneously triggers the active clamping discharge circuit to absorb the residual arc energy in the physical circuit, and then generates a quenching state signal. Furthermore, in S4, a microsecond-level blocking signal is output to the pulse width modulation controller to turn off the main power switch, and simultaneously triggers the active clamping discharge circuit to absorb residual arc energy in the physical circuit, including: By forcibly pulling the drive pin level of the pulse width modulation controller low through the underlying hardware interrupt channel, the main power switch of the rack high voltage pulse power supply is hard-shut down, blocking the continued feeding of bus energy to the physical circuit. A high-level trigger signal is output to turn on the clamping switch device connected in parallel with the physical load, which dissipates the inductive residual arc energy stored in the physical circuit into the bleeder resistor network connected in series with the clamping switch device, thus forcibly extinguishing the arc.
[0041] Specifically, step S4 acts as the underlying physical blocking and active energy absorption execution node in the entire rack high-voltage pulse power supply protection system. Its core function is to completely cut off the continuous energy injection path of the bus and, by constructing a low-impedance bypass dissipation channel, forcibly convert the ionized destructive arc energy into Joule heat for consumption, thereby avoiding the risk of breakdown caused by the huge back electromotive force induced by the traditional protection device relying solely on the unidirectional hard-cut-off switch.
[0042] The flow of control commands and hardware actions unfolds sequentially along a defined physical path. The input terminal receives targeted intervention commands from the preceding diagnostic logic as the trigger source for the action. After the hardware execution layer intercepts this command, it directly invokes the underlying hardware interrupt channel, bypassing the scanning cycle of the regular software program, and forcibly pulls the drive pin level of the pulse width modulation controller low. This microsecond-level blocking signal directly cuts off the drive source of the main power switch of the rack high-voltage pulse power supply, physically blocking the continued feeding of bus energy to the load end where the breakdown has occurred.
[0043] At the same timing node where the main power switch is turned off, the system synchronously outputs a high-level trigger signal. This trigger signal turns on the clamping switch directly connected in parallel with the physical load, instantly opening the active clamping discharge circuit. The inductive residual arc energy that was originally accumulated inside the physical circuit loses its original flow path and is entirely diverted into the discharge resistor network connected in series with the clamping switch for dissipation.
[0044] The physical process of dissipating residual arc energy can be defined by the following Joule heat integral formula: ; in This represents the actual residual arc energy consumed by the bleeder resistor network. This represents the quenching time span required from the start of conduction of the self-clamping switch device to the complete extinguishing of the fault arc. This represents the transient discharge current flowing through the active clamp discharge circuit in real time. It represents the equivalent physical resistance value of the bleeder resistor network in the underlying hardware architecture. This represents the time variable used in continuous integral calculations.
[0045] After undergoing high-intensity dissipation work by the resistor network, the fault arc at the physical load end is forcibly extinguished due to rapid energy depletion. After the arc is completely extinguished, the underlying monitoring circuit extracts the action termination node information and generates a quenching state signal, which is then reported as the final output data of step S4 to the subsequent historical fault statistics and threshold compensation module.
[0046] S5 receives the quenching status signal and counts the historical fault frequency. When the historical fault frequency exceeds the preset target aging index, the ignition energy threshold is dynamically increased according to the preset adaptive step size.
[0047] Furthermore, in S5, historical fault frequencies are statistically analyzed. When the historical fault frequency exceeds the preset target aging index, the ignition energy threshold is dynamically adjusted upwards by a preset adaptive step size, including: Extract the total number of running pulses and the number of ignition actions that trigger the quenching state signal within the preset statistical period; Calculate the ratio of the number of ignition actions to the total number of operating pulses to obtain the actual failure probability; The actual failure probability is compared with the preset probability upper limit constant that characterizes the preset target material aging index. When the actual failure probability is determined to be greater than the preset upper limit constant, it is determined that the physical load target material has suffered severe insulation aging, and the ignition energy threshold is increased by a preset adaptive step size.
[0048] Specifically, step S5 establishes a data-driven threshold adaptive compensation mechanism within the rack-mounted high-voltage pulse power supply protection system. Long-cycle physical experiments inevitably lead to physical wear and tear on the physical load target. This step transforms the objective statistical values of the underlying hardware's actions into the basis for adjusting the judgment parameters.
[0049] Data flow begins with receiving the quenching status signal generated in the preceding steps. The system control unit extracts the total number of running pulses and the number of ignition actions that trigger the quenching status signal within a preset statistical period.
[0050] The computation module then proceeds to the numerical calculation stage, determining the ratio of the number of ignition actions to the total number of operating pulses to obtain the actual failure probability. The specific probability extraction formula is defined as follows: ; in This represents the actual failure probability scalar obtained from the solution. This represents the cumulative number of ignition actions that trigger the quenching state signal within a preset statistical period. This represents the total number of operating pulses actually sent by the system to the physical load within the same preset statistical period.
[0051] After the calculation is completed, the system proceeds to the status determination node. The system compares the actual failure probability with the preset probability upper limit constant that characterizes the preset target material aging index.
[0052] When the actual failure probability exceeds the preset probability upper limit constant, the system establishes severe insulation aging of the physical load target material at the physical level. The system then increases the ignition energy threshold, which is dependent on the preceding comparison logic, by a preset adaptive step size. The specific dynamic threshold update formula is as follows: ; in This represents the new ignition energy threshold generated after the compensation correction is completed. This represents the old ignition energy threshold currently mounted and executed by the system. This represents the preset adaptive step size value set internally by the system. The newly generated ignition energy threshold will replace the old ignition energy threshold as the updated judgment criterion in the subsequent protection logic's cyclic detection.
[0053] By receiving quenching state signals, statistically analyzing historical fault frequencies, and dynamically adjusting the ignition energy threshold according to a preset adaptive step size, the system is endowed with the ability to track the dynamic evolution of the physical wear state of the target material. This improves upon the traditional protection methods, which mostly use fixed static protection thresholds and cannot detect the non-stationary drift of the pressure resistance boundary caused by the aging of the experimental target material, resulting in the judgment boundary deviating from the actual working conditions of the physical entity and the frequent false triggering of system shutdowns.
[0054] Example 2: This technology is applied in the electron cyclotron resonance heating system of a fully superconducting tokamak nuclear fusion reactor, specifically as a rack-mounted high-voltage pulse power supply device to drive a high-power cyclotron tube. Under the operating conditions of this high-energy physics experiment, the rack-mounted high-voltage pulse power supply needs to continuously output high-frequency high-voltage pulses with extremely steep rise edges to loads such as microwave cyclotron tubes. The rise time of the fast edge of the pulse is typically less than 20 nanoseconds, and the operating cycle of a single physics experiment is extremely long.
[0055] In this application scenario, due to the inherent nonlinear parasitic capacitance characteristics inside the gyrotron, the extremely steep high-voltage pulse fast edge can instantly generate a huge capacitive surge current. Traditional protection systems mostly use one-dimensional static current comparators, which cannot distinguish between normal surges and early arcing within nanosecond-level transients. This can easily lead to false shutdowns, thus forcibly interrupting irreversible fusion experiments. At the same time, once insulation breakdown actually occurs inside the gyrotron vacuum cavity, traditional protection mechanisms, relying solely on disconnecting the main power switch, are completely unable to consume the inductive residual energy accumulated in the physical circuit. This not only makes it difficult to extinguish the arc quickly, but the huge back electromotive force generated can also easily backfire and break down the expensive power devices at the bottom of the power supply. In addition, under long-period high-frequency subpulse bombardment, the gyrotron cathode target will inevitably undergo physical aging, causing its actual insulation breakdown energy boundary to drift non-stationarily. Traditional protection logic based on fixed static parameters cannot dynamically sense and adapt to this hardware loss, and is prone to severe threshold mismatch in the middle and later stages of the experiment, ultimately causing frequent false triggering or complete failure of protection actions. To solve the above problems, this invention provides a high-voltage pulse power supply protection system for a physical experiment rack, the structure of which is as follows: Figure 2 As shown. The specific implementation process of this system is as follows: The high-frequency transient waveform sampling module extracts the transient electrical sequence from the output of the high-voltage pulse power supply in the rack at the bottom layer, providing a basic digital source for the system. When the output pulse is in the rising edge state, the fast-edge capacitive surge characteristic shielding module activates the preset leading-edge blanking window. Combined with the current change rate and voltage drop rate calculated within this time span, it directly strips and filters out the normal fast-edge capacitive surge caused by the nonlinear parasitic capacitance of the physical load, eliminating the potential hardware mis-triggered risks in the initial stage of voltage build-up.
[0056] After the preset leading-edge blanking window expires, the ignition energy integration and diagnostic module takes over the monitoring process. It abandons the single-point extreme value comparison method and instead accumulates the instantaneous energy integration of the transient current in real time, comparing it against the ignition energy threshold set based on the dynamic envelope. This module accurately locates the physical singularity of insulation breakdown based on the truly accumulated Joule energy and issues targeted intervention commands, effectively filtering out high-frequency wandering glitches and interference.
[0057] After the hardware-level blocking and clamping execution module intercepts the intervention command, it implements physical blocking. By sending a microsecond-level blocking signal, it forces the pulse width modulation controller to turn off the main power switch, cuts off the continuous energy injection path of the bus, and triggers the active clamping discharge circuit to force the residual arc energy accumulated in the physical circuit to be converted into heat dissipation, extinguishing the fault arc to avoid the risk of back electromotive force breakdown caused by unidirectional hard turn-off. The action ends and a quenching state signal is generated.
[0058] The historical fault statistics and threshold compensation module accumulates the historical fault frequency based on the quenching state signal. When the frequency data exceeds the preset target aging index, it establishes that the load target has undergone insulation degradation. Accordingly, it dynamically adjusts the ignition energy threshold on which the previous diagnostic steps depend by a preset adaptive step size, giving the protection system the parameter calibration capability to track the actual aging state of the physical equipment and avoid threshold failure in the later stages of long-cycle experiments.
[0059] Example 3 In a third embodiment of the present invention, based on the same inventive concept, the present invention proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the image recognition method described in the above embodiments.
[0060] Example 4 According to the fourth embodiment of the present invention, based on the same inventive concept, the present invention proposes a computer device comprising: a processor and a memory; the processor and the memory communicate with each other; the memory is used to store instructions; the processor is used to execute the instructions in the memory to perform the image recognition method of the above embodiment.
[0061] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0062] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for protecting a high-voltage pulse power supply for a physical experiment rack, characterized in that, Includes the following steps: S1, acquire the transient voltage sampling sequence and transient current sampling sequence at the output terminal of the rack high-voltage pulse power supply; S2, when the rising edge of the output pulse is detected, a preset leading edge blanking window is triggered. The current change rate and voltage drop rate are calculated synchronously within the preset leading edge blanking window to identify and filter out the fast-edge capacitive surge characteristics caused by the nonlinear parasitic capacitance of the physical load. S3, after the preset leading edge blanking window ends, the instantaneous energy integral of the transient current is calculated in real time, and when the instantaneous energy integral exceeds the ignition energy threshold set based on the dynamic envelope, it is determined that the physical load has an insulation breakdown physical singularity and a targeted intervention command is generated. S4, receive the targeted intervention command, output a microsecond-level blocking signal to the pulse width modulation controller to turn off the main power switch, and simultaneously trigger the active clamping discharge circuit to absorb the residual arc energy in the physical circuit, and then generate a quenching state signal. S5, receive the quenching state signal and count the historical fault frequency. When the historical fault frequency exceeds the preset target aging index, dynamically increase the ignition energy threshold according to the preset adaptive step size.
2. The method for protecting a high-voltage pulse power supply for a physical experiment rack according to claim 1, characterized in that, In S1, acquiring the transient voltage sampling sequence and transient current sampling sequence at the output terminal of the rack high-voltage pulse power supply includes: The continuous analog voltage and continuous analog current of the rack high-voltage pulse power supply RF output port are synchronously acquired through an RF analog-to-digital converter. The continuous analog voltage and the continuous analog current are discretized according to a preset high-frequency sampling rate to generate the transient voltage sampling sequence and the transient current sampling sequence.
3. The method for protecting a high-voltage pulse power supply for a physical experiment rack according to claim 1, characterized in that, In S2, the current change rate and voltage drop rate are simultaneously calculated within the preset leading-edge blanking window to identify and filter out fast-edge capacitive surge characteristics caused by the nonlinear parasitic capacitance of the physical load, including: Within the preset leading-edge blanking window, the derivative of the transient current sampling sequence of the current sampling period is calculated as the current change rate, and the absolute value of the derivative of the transient voltage sampling sequence is calculated as the voltage drop rate. Calculate the ratio coefficient between the rate of change of current and the rate of decrease of voltage; When the ratio coefficient is within a preset safe range that characterizes the charging and discharging characteristics of the inherent parasitic capacitance, the current current spike is determined to be the fast-edge capacitive surge characteristic, and the underlying hardware overcurrent triggering logic is shielded within the preset leading-edge blanking window.
4. The method for protecting a high-voltage pulse power supply for a physical experiment rack according to claim 1, characterized in that, In S3, the ignition energy threshold set based on the dynamic envelope includes: Extract the baseline operating data of the rack high-voltage pulse power supply under normal no-load and resistive rated load conditions; Based on the aforementioned benchmark operating data, a normal current fluctuation boundary is constructed in the time domain; The normal current fluctuation boundary is extended outward by a preset redundancy margin to generate the dynamic envelope that defines the safe operating area, and the energy integral safety extreme value corresponding to the current moment on the dynamic envelope is extracted as the ignition energy threshold.
5. The method for protecting a high-voltage pulse power supply for a physical experiment rack according to claim 4, characterized in that, In S3, the instantaneous energy integral of the transient current is calculated in real time, and when the instantaneous energy integral exceeds the ignition energy threshold set based on the dynamic envelope, a physical load insulation breakdown physical singularity is determined and a targeted intervention command is generated, including: Extract the discrete current values from the transient current sampling sequence; The instantaneous energy integral is obtained by rolling summation of the product of the square of the discrete current value and the discrete sampling time interval using a sliding time window; The instantaneous energy integral is compared with the ignition energy threshold in real time. When it is determined that the instantaneous energy integral is greater than the ignition energy threshold, the targeted intervention command is output.
6. The method for protecting a high-voltage pulse power supply for a physical experiment rack according to claim 1, characterized in that, In S4, the step of outputting a microsecond-level blocking signal to the pulse width modulation controller to turn off the main power switch and simultaneously triggering the active clamping discharge circuit to absorb residual arc energy in the physical circuit includes: By forcibly pulling the drive pin level of the pulse width modulation controller low through the underlying hardware interrupt channel, the main power switch of the rack high voltage pulse power supply is hard-shut down, blocking the continued feeding of bus energy to the physical circuit. A high-level trigger signal is output to turn on the clamping switch device connected in parallel with the physical load, and the inductive residual arc energy stored in the physical circuit is introduced into the discharge resistor network connected in series with the clamping switch device for dissipation, thereby forcibly extinguishing the arc.
7. The method for protecting a high-voltage pulse power supply for a physical experiment rack according to claim 1, characterized in that, In S5, the statistical analysis of historical fault frequencies, when the historical fault frequency exceeds a preset target aging index, dynamically adjusting the ignition energy threshold by a preset adaptive step size includes: Extract the total number of running pulses and the number of ignition actions that trigger the quenching state signal within the preset statistical period; Calculate the ratio of the number of ignition actions to the total number of operating pulses to obtain the actual failure probability; The actual failure probability is compared with the preset probability upper limit constant that characterizes the preset target material aging index; When the actual failure probability is determined to be greater than the preset probability upper limit constant, it is determined that the physical load target material has suffered severe insulation aging, and the ignition energy threshold is controlled to increase by one preset adaptive step size.
8. A high-voltage pulse power supply protection system for a physical experiment rack, characterized in that, A method for protecting a high-voltage pulse power supply for a physical experimental rack as described in any one of claims 1-7 includes the following modules: The high-frequency transient waveform sampling module is used to acquire the transient voltage sampling sequence and transient current sampling sequence at the output terminal of the rack high-voltage pulse power supply. The fast-edge capacitive surge feature shielding module is used to trigger a preset leading edge blanking window when the rising edge of the output pulse is detected. Within the preset leading edge blanking window, the current change rate and voltage drop rate are calculated simultaneously to identify and filter out the fast-edge capacitive surge feature caused by the nonlinear parasitic capacitance of the physical load. The ignition energy integration and diagnosis module is used to calculate the instantaneous energy integration of the transient current in real time after the preset leading edge blanking window ends, and when the instantaneous energy integration exceeds the ignition energy threshold set based on the dynamic envelope, it determines that the physical load has an insulation breakdown physical singularity and generates a targeted intervention command. The hardware-level blocking and clamping execution module is used to receive the targeted intervention command, output a microsecond-level blocking signal to the pulse width modulation controller to turn off the main power switch, and simultaneously trigger the active clamping discharge circuit to absorb the residual arc energy in the physical circuit, and then generate a quenching state signal. The historical fault statistics and threshold compensation module is used to receive the quenching state signal and count the historical fault frequency. When the historical fault frequency exceeds the preset target aging index, the ignition energy threshold is dynamically increased by a preset adaptive step size.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements a high-voltage pulse power supply protection method for a physical experiment rack as described in any one of claims 1-7.
10. A readable storage medium, characterized in that, The readable storage medium stores a computer program, which, when executed by a processor, implements a high-voltage pulse power supply protection method for a physical experiment rack as described in any one of claims 1-7.