Optimization method and system for electron beam high-voltage power supply based on optical fiber sensing

Abstract

This invention relates to the field of high-voltage power supply control and fiber optic sensing technology, and discloses a method and system for optimizing an electron beam high-voltage power supply based on fiber optic sensing. The method includes: establishing a thermally induced parameter drift correlation model based on the correlation between temperature and voltage ripple in the high-voltage power supply; generating a voltage compensation reference value for the high-voltage power supply based on the thermally induced parameter drift correlation model; feeding the voltage compensation reference value forward to the pulse width modulation control loop of the high-voltage power supply to generate a dynamic load adjustment reference value; and generating a composite control reference signal for the high-voltage power supply based on the voltage compensation reference value, the dynamic load adjustment reference value, and the insulation state. This invention can improve the efficiency of electron beam high-voltage power supply optimization based on fiber optic sensing.

Description

Technical Field

[0001] This invention relates to the fields of high-voltage power supply control and fiber optic sensing technology, and in particular to an electron beam high-voltage power supply optimization method and system based on fiber optic sensing. Background Technology

[0002] As a core component of precision machining and testing equipment, the stability of the electron beam high-voltage power supply directly determines the electron beam quality and process repeatability. Under continuous high-power operation, the thermal accumulation effect generated by the power devices and magnetic components inside the power supply can cause irreversible drift of key component parameters, leading to deterioration of output voltage ripple and deviation from the set value. Existing high-voltage power supply control architectures mostly rely on feedback closed-loop regulation of the output electrical signal, which has an inherent lag in response to thermally induced parameter drift and is difficult to effectively compensate for in the early stages of dynamic temperature field evolution. At the same time, the internal insulation material of the high-voltage power supply will gradually develop microscopic defects and induce partial discharge under the combined effect of strong electric field and thermal stress. Traditional electrical monitoring methods are limited by the high-potential isolation problem and cannot accurately extract weak insulation degradation characteristic signals in high common-mode interference environments, resulting in a lack of insulation condition assessment and further exacerbating the risk of performance degradation during long-term service of the power supply.

[0003] While fiber optic sensing technology possesses physical advantages such as intrinsic insulation and electromagnetic interference resistance, a systematic multi-parameter collaborative sensing and closed-loop control method has yet to be established in current high-voltage power supply applications. Current solutions primarily rely on single-parameter temperature monitoring, and the acquired distributed temperature data lacks a direct mapping to the voltage ripple control loop. This results in temperature compensation and dynamic load adjustment still depending on empirically preset fixed parameter tables. Furthermore, the coupling effect of ultrasonic signals generated by partial discharge in the insulation state with output voltage fluctuations is not incorporated into a unified control framework. Consequently, when faced with multiple disturbances such as sudden temperature changes, load jumps, and gradual insulation degradation, the steady-state accuracy and transient recovery capability of the output voltage are significantly reduced, making it difficult to meet the stringent output quality requirements of high-precision electron beam applications. Therefore, improving the optimization efficiency of electron beam high-voltage power supplies has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides an electron beam high-voltage power supply optimization method and system based on fiber optic sensing to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an electron beam high-voltage power supply optimization method based on fiber optic sensing, comprising:

[0006] Based on the correlation between temperature and voltage ripple in a high-voltage power supply, a thermally induced parameter drift correlation model is established, and a voltage compensation reference value for the high-voltage power supply is generated based on the thermally induced parameter drift correlation model.

[0007] The voltage compensation reference value is fed forward to the pulse width modulation control loop of the high voltage power supply to generate a dynamic load adjustment reference value.

[0008] A composite control reference signal for the high-voltage power supply is generated based on the voltage compensation reference value, the dynamic load adjustment reference value, and the insulation status.

[0009] In a preferred embodiment, before establishing the thermally induced parameter drift correlation model based on the correlation between temperature and voltage ripple in the high-voltage power supply, the method further includes:

[0010] Sensors are deployed in the high-potential nodes and insulating support structures of the high-voltage power supply.

[0011] Optical signals are introduced by combining insulated optical fibers to construct an optical fiber sensing and monitoring network covering the electromagnetic core area of ​​high-voltage power supplies.

[0012] In a preferred embodiment, after introducing optical signals through the composite of insulated optical fibers to construct an optical fiber sensing and monitoring network covering the electromagnetic core area of ​​the high-voltage power supply, the method further includes:

[0013] The center wavelength shift in the optical fiber sensing and monitoring network is used as distributed temperature field data.

[0014] The changes in the interference spectrum in the fiber optic sensing network are used as the ultrasonic characteristic signal of partial discharge in the insulation state.

[0015] The voltage ripple data and load current data at the power output terminal of the high-voltage power supply are synchronized through a high-voltage isolation sampling circuit.

[0016] In a preferred embodiment, establishing a thermally induced parameter drift correlation model based on the correlation between temperature and voltage ripple in the high-voltage power supply includes:

[0017] The distributed temperature field data and the output voltage ripple data are time-domain aligned to obtain a temperature-ripple correlation dataset.

[0018] Based on the temperature-ripple correlation dataset, the mapping relationship between temperature gradient change features and ripple amplitude fluctuation features is extracted to establish a thermal parameter drift correlation model.

[0019] In a preferred embodiment, the calculation formula for the thermally induced parameter drift correlation model is as follows:

[0020] ;

[0021] In the formula, The temperature is Predicted peak output voltage ripple at time The temperature is The initial output voltage ripple peak value at that time. It is the first-order temperature drift coefficient. The real-time average operating temperature, The preset reference temperature, It is the second-order temperature drift coefficient.

[0022] In a preferred embodiment, generating the voltage compensation reference value of the high-voltage power supply based on the thermally induced parameter drift correlation model includes:

[0023] The real-time temperature data of the high-voltage power supply at the current moment is input into the thermally induced parameter drift correlation model to obtain the predicted output voltage ripple peak value at the current moment.

[0024] Based on the difference between the predicted output voltage ripple peak value and the preset standard ripple reference value, the voltage compensation direction and compensation magnitude for offsetting the temperature drift effect are determined.

[0025] Based on the voltage compensation direction and the compensation amplitude, a voltage compensation reference quantity is generated that is superimposed on the basic control quantity of the high-voltage power supply.

[0026] In a preferred embodiment, feeding the voltage compensation reference value forward to the pulse width modulation control loop of the high-voltage power supply to generate a dynamic load adjustment reference value includes:

[0027] The voltage compensation reference value is input to the given signal input terminal of the pulse width modulation control loop in the high voltage power supply, and is superimposed with the original voltage feedback error signal of the pulse width modulation control loop in the time domain.

[0028] Based on the superimposed signal, the reference threshold of the comparator in the pulse width modulation control loop is adjusted so that the pulse width modulation control loop changes the modulation strategy in advance before the output fluctuation is caused by temperature change.

[0029] The pulse width modulation control loop generates a temperature-compensated feedforward control signal based on the adjusted reference threshold.

[0030] In a preferred embodiment, after the pulse width modulation control loop generates a temperature-compensated feedforward control signal based on an adjusted reference threshold, the method further includes:

[0031] The load dynamic response requirement is determined based on the rate of change of the load current data.

[0032] The temperature-compensated feedforward control signal is correlated with the load dynamic response requirements to determine the load adjustment margin under the premise of satisfying temperature compensation.

[0033] Based on the load adjustment margin, the dynamic load adjustment reference value is generated while ensuring the temperature compensation effect and taking into account the impact of load fluctuations on the output voltage.

[0034] In a preferred embodiment, generating the composite control reference signal for the high-voltage power supply based on the voltage compensation reference, the dynamic load adjustment reference, and the insulation state includes:

[0035] Based on the amplitude and frequency of occurrence of the partial discharge ultrasonic characteristic signal, the confidence descriptor of the insulation state is determined;

[0036] Using the confidence level description as a weighting benchmark, the voltage compensation reference and the dynamic load adjustment reference are weighted and synthesized to obtain a composite adjustment value;

[0037] The composite adjustment amount is converted into a composite control reference signal that can be recognized by the pulse width modulation control loop of the high-voltage power supply, wherein the composite control reference signal is used to simultaneously respond to the combined effects of temperature drift, load fluctuation and insulation degradation on the output voltage.

[0038] To address the aforementioned problems, the present invention also provides an electron beam high-voltage power supply optimization system based on fiber optic sensing, the system comprising:

[0039] The thermally induced ripple modeling and compensation generation module is used to establish a thermally induced parameter drift correlation model based on the correlation between temperature and voltage ripple in the high-voltage power supply, and generate the voltage compensation reference value of the high-voltage power supply based on the thermally induced parameter drift correlation model.

[0040] The feedforward dynamic adjustment module is used to feed the voltage compensation reference value to the pulse width modulation control loop of the high voltage power supply, thereby generating a dynamic load adjustment reference value.

[0041] The composite control generation module is used to generate a composite control reference signal for the high-voltage power supply based on the voltage compensation reference value, the dynamic load adjustment reference value, and the insulation state.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] 1. This invention constructs an optical fiber sensing monitoring network covering the electromagnetic core area of ​​a high-voltage power supply to simultaneously acquire distributed temperature field data and ultrasonic characteristic signals of partial discharge in insulation conditions. Based on this, a thermally induced parameter drift correlation model between temperature and voltage ripple is established, generating a voltage compensation reference quantity with temperature adaptive capability. This compensation reference quantity is directly applied to the pulse width modulation control loop in a feedforward manner, pre-adjusting the modulation strategy before output fluctuations are caused by temperature changes, eliminating output voltage deviations caused by thermal hysteresis, and significantly improving the output steady-state accuracy and ripple suppression level of the high-voltage power supply under continuous temperature variation conditions.

[0044] 2. Further, the voltage compensation reference quantity is combined with the dynamic load adjustment reference quantity obtained from the correlation analysis of the load current change rate, and an insulation state confidence description quantity is introduced for weighted synthesis to generate a composite control reference signal that can simultaneously respond to the effects of temperature drift, load fluctuation, and insulation degradation. This method integrates temperature compensation, dynamic load adjustment, and insulation safety constraints into a pulse width modulation control loop, effectively enhancing the comprehensive anti-interference capability and long-term operational consistency of the high-voltage power supply under multi-source disturbance coupling conditions, and improving the optimization efficiency and output quality of the electron beam high-voltage power supply from the control architecture level. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating an electron beam high-voltage power supply optimization method based on fiber optic sensing, provided in an embodiment of the present invention.

[0046] Figure 2 A functional block diagram of an electron beam high-voltage power supply optimization system based on fiber optic sensing, provided in an embodiment of the present invention;

[0047] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0048] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0049] This application provides a method for optimizing an electron beam high-voltage power supply based on fiber optic sensing. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cluster of cloud servers. The server can be an independent server or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0050] Reference Figure 1 The diagram shown is a flowchart illustrating an electron beam high-voltage power supply optimization method based on fiber optic sensing, according to an embodiment of the present invention. In this embodiment, the electron beam high-voltage power supply optimization method based on fiber optic sensing includes:

[0051] Based on the correlation between temperature and voltage ripple in the high-voltage power supply, a thermally induced parameter drift correlation model is established, and a voltage compensation reference value for the high-voltage power supply is generated based on the thermally induced parameter drift correlation model.

[0052] In this embodiment of the invention, before establishing the thermally induced parameter drift correlation model based on the correlation between temperature and voltage ripple in the high-voltage power supply, the method further includes:

[0053] Sensors are deployed in the high-potential nodes and insulating support structures of the high-voltage power supply.

[0054] Optical signals are introduced by combining insulated optical fibers to construct an optical fiber sensing and monitoring network covering the electromagnetic core area of ​​high-voltage power supplies.

[0055] Following the step of introducing optical signals through insulated optical fiber composite to construct an optical fiber sensing and monitoring network covering the electromagnetic core area of ​​the high-voltage power supply, the following is also included:

[0056] The center wavelength shift in the optical fiber sensing and monitoring network is used as distributed temperature field data.

[0057] The changes in the interference spectrum in the fiber optic sensing network are used as the ultrasonic characteristic signal of partial discharge in the insulation state.

[0058] The voltage ripple data and load current data at the power output terminal of the high-voltage power supply are synchronized through a high-voltage isolation sampling circuit.

[0059] The establishment of a thermally induced parameter drift correlation model based on the correlation between temperature and voltage ripple in a high-voltage power supply includes:

[0060] The distributed temperature field data and the output voltage ripple data are time-domain aligned to obtain a temperature-ripple correlation dataset.

[0061] Based on the temperature-ripple correlation dataset, the mapping relationship between temperature gradient change features and ripple amplitude fluctuation features is extracted to establish a thermal parameter drift correlation model.

[0062] The calculation formula for the thermal parameter drift correlation model is as follows:

[0063] ;

[0064] In the formula, The temperature is Predicted peak output voltage ripple at time The temperature is The initial output voltage ripple peak value at that time. It is the first-order temperature drift coefficient. The real-time average operating temperature, The preset reference temperature, It is the second-order temperature drift coefficient.

[0065] The step of generating the voltage compensation reference value of the high-voltage power supply based on the thermally induced parameter drift correlation model includes:

[0066] The real-time temperature data of the high-voltage power supply at the current moment is input into the thermally induced parameter drift correlation model to obtain the predicted output voltage ripple peak value at the current moment.

[0067] Based on the difference between the predicted output voltage ripple peak value and the preset standard ripple reference value, the voltage compensation direction and compensation magnitude for offsetting the temperature drift effect are determined.

[0068] Based on the voltage compensation direction and the compensation amplitude, a voltage compensation reference quantity is generated that is superimposed on the basic control quantity of the high-voltage power supply.

[0069] A fiber optic temperature sensor and an interferometric fiber optic ultrasonic sensor are installed in the high-potential node and insulation support structure of the high-voltage power supply. The high-potential node includes a heat dissipation substrate for power switching devices and the lead-out end of a high-frequency transformer winding. The insulation support structure includes an epoxy resin insulating pillar and an output ceramic insulating sleeve. The sensor uses polyimide-coated optical fiber directly attached to the surface being measured and is encapsulated and fixed with high-voltage resistant insulating colloid.

[0070] The sensing optical fiber is led from the high-voltage side to the ground potential side through an insulated optical fiber composite sleeve. The optical signal emitted by the broadband light source is transmitted inside the insulated optical fiber, and the outside is wrapped with a multi-layer silicone rubber insulating sheath to withstand the full rated high voltage potential difference. During the transmission of the optical fiber, the optical signal passes through the locations of each sensor and carries the corresponding wavelength and phase modulation information back to the demodulation equipment. This forms an optical fiber sensing and monitoring network covering the electromagnetic core area where the power conversion unit, high-voltage transformer, rectifier filter components and insulating support components are located.

[0071] In the fiber optic sensing and monitoring network, the broadband light source spectrum is split by a wavelength division multiplexer and then illuminates each fiber optic grating sensor. Each sensor reflects a narrowband light signal with a specific center wavelength and returns it to the wavelength demodulation module. The wavelength demodulation module tracks the offset value of the center wavelength of the reflected spectrum of each channel in real time and maps the offset value to the distributed temperature field data distributed along the fiber optic path inside the high-voltage power supply according to the spatial coordinates of the sensor's location.

[0072] The interferometric fiber optic ultrasonic sensor in the fiber optic sensing and monitoring network has an internal structure of an unbalanced Mach-Zehnder interferometer. When the ultrasonic pressure wave generated by partial discharge acts on the sensing fiber, it causes axial strain of the fiber and changes the optical path difference between the interferometer arms. The waveform of the intensity change of the interference spectrum output by the demodulator is used to extract the envelope peak and repetition rate in a specific frequency band of the waveform as the ultrasonic characteristic signal of partial discharge in the insulation state.

[0073] A high-voltage isolation sampling circuit is connected to the secondary side of the high-voltage divider and current transformer at the power output end. The high-voltage isolation sampling circuit adopts a cascaded structure of linear optocoupler and isolation amplifier to block the common-mode interference path between the high-voltage side and the low-voltage side. The sampling circuit synchronously acquires the instantaneous value of the output voltage ripple after voltage division and the instantaneous value of the load current after current-voltage conversion, and sends the sampling results to the signal acquisition unit in the form of low-voltage analog signal to complete the synchronous acquisition of voltage ripple data and load current data.

[0074] The temperature spatial distribution sequence at each sampling time in the distributed temperature field data is paired with the corresponding output voltage ripple data collected at the same time according to the timestamp. A hardware-triggered synchronization mechanism is used to ensure that the sampling clock edges of the temperature data and ripple data are aligned. After pairing, a temperature ripple association dataset is constructed with the temperature spatial distribution as the input parameter and the peak value of the output voltage ripple as the output label.

[0075] Based on data samples from multiple consecutive sampling periods in the temperature ripple correlation dataset, the temperature difference between adjacent sensor temperature measurement points is calculated along the fiber optic deployment path to obtain the temperature gradient change characteristics. At the same time, the fluctuation amplitude change of the output voltage ripple peak value within the corresponding time period is extracted to obtain the ripple amplitude fluctuation characteristics. The mapping relationship between the temperature gradient characteristics and the ripple amplitude fluctuation characteristics under different operating temperature ranges is established by piecewise linear regression. This mapping relationship is the thermally induced parameter drift correlation model.

[0076] The thermally induced parameter drift correlation model reads the average temperature of several temperature measurement points closest to the core power heating region in the current real-time temperature field data of the high-voltage power supply as the model input variables. The model internally stores the output voltage ripple reference value, first-order temperature drift contribution factor and second-order temperature drift contribution factor under the pre-calibrated reference temperature condition. The model calculates the first-order temperature drift contribution and second-order temperature drift contribution by the difference between the independent variables and the reference temperature, and then weights and superimposes the two with the reference value to output the predicted peak value of the output voltage ripple under the corresponding real-time temperature condition.

[0077] The real-time temperature field data obtained by the fiber optic sensing monitoring network at the current moment of the high-voltage power supply is input into the thermally induced parameter drift correlation model. The model outputs the predicted peak value of the output voltage ripple at the current moment. At the same time, the standard ripple reference value preset in the high-voltage power supply control parameter storage unit is retrieved. The ripple deviation is obtained by subtracting the predicted peak value of the output voltage ripple from the standard ripple reference value.

[0078] The voltage compensation direction is determined based on the sign of the ripple deviation. When the predicted ripple peak value is greater than the standard ripple reference value, the voltage compensation direction is reverse compensation to reduce the output voltage. When the predicted ripple peak value is less than the standard ripple reference value, the voltage compensation direction is positive compensation to increase the output voltage. The compensation amplitude is determined by scaling the absolute value of the ripple deviation. The scaling factor is preset in the compensation generation logic according to the gain characteristics of the high-voltage power supply output stage.

[0079] A simulated voltage compensation signal is generated according to the determined voltage compensation direction and compensation amplitude. This simulated voltage compensation signal is superimposed on the original basic control given voltage of the high voltage power supply in the simulated adder circuit. The voltage compensation reference quantity formed after superposition is sent as an independent control input signal to the feedforward input terminal of the pulse width modulation control loop.

[0080] The method for obtaining the initial output voltage ripple peak value under the preset reference temperature condition is as follows: when the high-voltage power supply is initially powered on and in a stable no-load condition, the output voltage ripple waveform is continuously acquired through a high-voltage isolation sampling circuit. The arithmetic mean of the ripple peak values ​​within multiple consecutive power frequency cycles is taken as the initial output voltage ripple peak value corresponding to the reference temperature and stored in the parameter storage unit of the thermally induced parameter drift correlation model. The real-time average operating temperature is obtained by taking the arithmetic mean of the temperature readings of the heat dissipation substrate of the power switching device and the lead-out terminal of the high-frequency transformer winding from the distributed temperature field data output by the fiber optic sensing monitoring network. This average value characterizes the overall temperature level of the thermally induced parameter drift sensitive area of ​​the high-voltage power supply. The method for determining the first-order and second-order temperature drift coefficients is to conduct a controlled temperature rise experiment on the high-voltage power supply. The ambient temperature is gradually increased at fixed temperature intervals above the preset reference temperature, and the steady-state measurement results of the output voltage ripple peak value at each temperature step are recorded. The correspondence between the temperature deviation value and the change in ripple peak value is fitted into a quadratic polynomial using the least squares method. The first-order term weighting factor obtained from the fitting is the first-order temperature drift coefficient, and the quadratic term weighting factor is the second-order temperature drift coefficient.

[0081] The computational logic expresses the quantitative correspondence between the peak output voltage ripple of the high-voltage power supply and the degree of deviation of the operating temperature from the reference temperature. The thermally induced parameter drift correlation model uses the deviation between the real-time average operating temperature and the preset reference temperature as the driving variable. It generates a linear drift contribution component by weighting the first-order temperature drift coefficient with a linear term of the deviation, and a nonlinear drift contribution component by weighting the second-order temperature drift coefficient with a quadratic term of the deviation. The linear and nonlinear drift contribution components are summed to form a comprehensive temperature drift influence factor. Finally, the comprehensive temperature drift influence factor is multiplied by the initial peak output voltage ripple under the reference temperature condition to obtain the predicted peak output voltage ripple at the current temperature. This model transforms the nonlinear cumulative effect of electrical parameter drift exhibited by the power devices and magnetic components inside the high-voltage power supply during temperature rise into a predicted value of the peak output voltage ripple.

[0082] When the real-time average operating temperature equals the preset reference temperature, the temperature deviation is zero, both the first-order and second-order temperature drift contribution components are zero, the comprehensive temperature drift influence factor equals value one, and the predicted output voltage ripple peak value equals the initial output voltage ripple peak value. When the real-time average operating temperature is higher than the preset reference temperature, the temperature deviation is positive, both the first-order and second-order temperature drift contribution components are positive, the comprehensive temperature drift influence factor is greater than value one, the predicted output voltage ripple peak value is higher than the initial output voltage ripple peak value, and the predicted output voltage ripple peak value shows an accelerating upward trend as the temperature deviation increases. When the real-time average operating temperature is lower than the preset reference temperature, the temperature deviation is negative, the first-order temperature drift contribution component is negative, the second-order temperature drift contribution component is positive and its amplitude is less than the absolute value of the first-order contribution component, the comprehensive temperature drift influence factor is less than value one, the predicted output voltage ripple peak value is lower than the initial output voltage ripple peak value, and the predicted output voltage ripple peak value shows a decelerating downward trend as the temperature deviation increases negatively.

[0083] The beneficial effects are as follows: By directly deploying fiber optic temperature sensors and interferometric fiber optic ultrasonic sensors within the high-potential nodes and insulating support structures of the high-voltage power supply, and encapsulating them in high-voltage insulating colloid, and using insulated fiber optic composite sleeves to safely lead the sensing optical signals out of the high-potential region, a fiber optic sensing and monitoring network covering the electromagnetic core area is constructed. This achieves in-situ, lossless acquisition and high-potential isolated transmission of multiple physical field parameters on the high-voltage side. The network simultaneously acquires distributed temperature field data, ultrasonic characteristic signals of partial discharge in insulation, and voltage ripple and load current data. After time-domain alignment processing, a temperature ripple correlation dataset is established, and the mapping relationship between temperature gradient change characteristics and ripple amplitude fluctuation characteristics is extracted to form a thermally induced parameter drift correlation model. Based on real-time temperature field data, the model predicts the peak value of the output voltage ripple and compares it with a standard ripple reference value to generate a voltage compensation reference value. This reference value is applied to the pulse width modulation control loop in a feedforward manner, thereby pre-adjusting the output voltage before the thermal accumulation effect manifests, effectively suppressing the output ripple degradation caused by temperature drift.

[0084] The voltage compensation reference value is fed forward to the pulse width modulation control loop of the high-voltage power supply to generate a dynamic load adjustment reference value.

[0085] In this embodiment of the invention, feeding the voltage compensation reference value forward to the pulse width modulation control loop of the high-voltage power supply to generate a dynamic load adjustment reference value includes:

[0086] The voltage compensation reference value is input to the given signal input terminal of the pulse width modulation control loop in the high voltage power supply, and is superimposed with the original voltage feedback error signal of the pulse width modulation control loop in the time domain.

[0087] Based on the superimposed signal, the reference threshold of the comparator in the pulse width modulation control loop is adjusted so that the pulse width modulation control loop changes the modulation strategy in advance before the output fluctuation is caused by temperature change.

[0088] The pulse width modulation control loop generates a temperature-compensated feedforward control signal based on the adjusted reference threshold.

[0089] After the pulse width modulation control loop generates a temperature-compensated feedforward control signal based on the adjusted reference threshold, the following is also included:

[0090] The load dynamic response requirement is determined based on the rate of change of the load current data.

[0091] The temperature-compensated feedforward control signal is correlated with the load dynamic response requirements to determine the load adjustment margin under the premise of satisfying temperature compensation.

[0092] Based on the load adjustment margin, the dynamic load adjustment reference value is generated while ensuring the temperature compensation effect and taking into account the impact of load fluctuations on the output voltage.

[0093] The voltage compensation reference quantity is connected to the given signal input terminal of the high-voltage power supply pulse width modulation control loop in the form of an analog voltage signal. This given signal input terminal and the voltage feedback error signal output terminal are connected to the non-inverting input node of the analog adder circuit. The analog adder circuit performs point-by-point summation of the instantaneous amplitude of the voltage compensation reference quantity and the instantaneous amplitude of the voltage feedback error signal in the time domain. The superimposed signal generated by the summation operation is sent as the adjusted error quantity to the subsequent signal conditioning link of the pulse width modulation control loop.

[0094] The superimposed signal is sent to the non-inverting input of the high-speed voltage comparator in the pulse width modulation control loop. The inverting input of the high-speed voltage comparator is connected to a sawtooth wave carrier signal generator with a fixed frequency. The change in the amplitude of the superimposed signal directly changes the phase position of the comparator output flip time relative to the start point of the sawtooth wave period. This flip time corresponds to the duty cycle boundary of the pulse width modulation signal. Thus, the modulation strategy is pre-adjusted by changing the equivalent threshold of the comparator input reference voltage. This adjustment is completed before the temperature change is actually transmitted to the power supply output.

[0095] The square wave pulse sequence generated at the output of the high-speed voltage comparator in the pulse width modulation control loop is amplified by the isolation drive circuit and sent to the gate drive terminal of the power switching device. The duty cycle of the square wave pulse sequence is determined by the adjusted reference threshold and includes the duty cycle correction introduced by temperature compensation. The power switching device performs turn-on and turn-off actions according to the square wave pulse sequence, thereby forming the switching voltage waveform corresponding to the temperature-compensated feedforward control signal at the front end of the output filter circuit.

[0096] The load current data synchronously output by the high-voltage isolation sampling circuit is processed by the differential operation circuit to obtain the instantaneous rate of change of the load current. The differential operation circuit, consisting of an operational amplifier and a resistor-capacitor network forming a high-pass filter, realizes the derivative function of the current signal with respect to time. The absolute value of the load current rate of change is compared with the load transient response trigger threshold preset in the threshold register. When the absolute value of the rate of change exceeds the trigger threshold, a load dynamic response demand flag signal is generated. This flag signal represents the urgency of the current load state for the output voltage regulation rate.

[0097] The duty cycle value corresponding to the temperature-compensated feedforward control signal and the load dynamic response demand flag signal are simultaneously input to the load adjustment margin determination logic unit. The load adjustment margin determination logic unit calculates the adjustment amount occupied by the current temperature compensation based on the available adjustment range of the current duty cycle value from the maximum duty cycle limit and the minimum duty cycle limit. Then, based on the trigger state of the load dynamic response demand flag signal, an additional duty cycle adjustment space is reserved to cope with the load step change. This reserved space is the load adjustment margin under the premise of satisfying temperature compensation.

[0098] After the load adjustment margin is determined, the duty cycle value corresponding to the temperature-compensated feedforward control signal and the duty cycle increment corresponding to the load adjustment margin are weighted and combined in the pulse width synthesis logic unit. The weighting weight is pre-fixed in the configuration register of the pulse width synthesis logic unit according to the priority of the load dynamic response requirement flag signal. After weighting and combination, a dynamic load adjustment reference value is generated. This dynamic load adjustment reference value is output in the form of pulse width modulation duty cycle correction value and can simultaneously maintain the ripple suppression effect achieved by temperature compensation and the transient response capability required by load fluctuation.

[0099] The beneficial effect is that the voltage compensation reference value is superimposed point-by-point in the time domain with the voltage feedback error signal through an analog adder to generate a superimposed signal. This directly adjusts the input reference threshold of the high-speed voltage comparator, allowing the pulse width modulation duty cycle to be corrected before temperature changes are transmitted to the output. The output is the switching voltage waveform corresponding to the temperature-compensated feedforward control signal. Simultaneously, the differential circuit extracts the instantaneous rate of change from the load current data and compares it with the trigger threshold to generate a load dynamic response demand flag signal. The load adjustment margin determination logic unit calculates the load adjustment margin under the premise of satisfying temperature compensation based on the current duty cycle limit. The pulse width synthesis logic unit weights and synthesizes the temperature compensation duty cycle and the load adjustment margin into a dynamic load adjustment reference value according to the priority of the load dynamic response demand. This process provides transient adjustment space for load step changes while maintaining the thermal ripple suppression effect, realizing the coordinated response of temperature feedforward compensation and load dynamic adjustment within the same control cycle.

[0100] A composite control reference signal for the high-voltage power supply is generated based on the voltage compensation reference value, the dynamic load adjustment reference value, and the insulation status.

[0101] In this embodiment of the invention, generating a composite control reference signal for the high-voltage power supply based on the voltage compensation reference value, the dynamic load adjustment reference value, and the insulation state includes:

[0102] Based on the amplitude and frequency of occurrence of the partial discharge ultrasonic characteristic signal, the confidence descriptor of the insulation state is determined;

[0103] Using the confidence level description as a weighting benchmark, the voltage compensation reference and the dynamic load adjustment reference are weighted and synthesized to obtain a composite adjustment value;

[0104] The composite adjustment amount is converted into a composite control reference signal that can be recognized by the pulse width modulation control loop of the high-voltage power supply, wherein the composite control reference signal is used to simultaneously respond to the combined effects of temperature drift, load fluctuation and insulation degradation on the output voltage.

[0105] The partial discharge ultrasonic characteristic signal is obtained by processing the interference spectrum intensity change waveform output by the interferometric fiber ultrasonic sensor through envelope detection and threshold discrimination circuit. The envelope detection circuit extracts the amplitude envelope curve in a specific frequency band of the interference spectrum intensity change waveform. The threshold discrimination circuit compares the amplitude of the envelope curve with the preset partial discharge event judgment level. Each time the envelope amplitude exceeds the judgment level, it is counted as a partial discharge event. The cumulative number of partial discharge events within a unit time window is the occurrence frequency of the partial discharge ultrasonic characteristic signal. At the same time, the maximum value of the envelope amplitude in a single partial discharge event is recorded as the amplitude of that event. The arithmetic mean of the amplitudes of all partial discharge events within a unit time window is taken as the amplitude of the partial discharge ultrasonic characteristic signal. The amplitude and the occurrence frequency are input together to the insulation state confidence evaluation logic unit.

[0106] The insulation condition confidence assessment logic unit internally stores a mapping table corresponding to the degree of insulation degradation and partial discharge characteristics. This mapping table is generated by applying step-increasing voltage and thermal stress to the same type of insulation material in an offline accelerated aging test and simultaneously recording the evolution of the partial discharge ultrasonic characteristic signal. The insulation condition confidence assessment logic unit performs interval matching between the amplitude of the currently acquired partial discharge ultrasonic characteristic signal and the amplitude range corresponding to each degradation level in the mapping table, and performs interval matching between the occurrence frequency and the frequency range corresponding to each degradation level in the mapping table. Combining the amplitude matching result and the frequency matching result, the logic unit queries the mapping table to output the corresponding insulation condition confidence descriptor. This confidence descriptor is a dimensionless value, ranging from zero to one. The lower the value, the more severe the insulation degradation and the stronger the constraint requirement for voltage compensation and load adjustment functions.

[0107] Using the insulation state confidence descriptor as a weighting benchmark, the voltage compensation reference and the dynamic load adjustment reference are weighted and synthesized. The weighting synthesis is performed by a cascaded circuit of a multi-input analog multiplier and adder. The voltage compensation reference is multiplied by the confidence descriptor through the first analog multiplier to generate a weighted voltage compensation component. The dynamic load adjustment reference is multiplied by the confidence descriptor through the second analog multiplier to generate a weighted load adjustment component. The weighted voltage compensation component and the weighted load adjustment component are summed in the analog adder to obtain the composite regulation. When the confidence descriptor decreases, the weighting coefficient decreases, which limits the contribution ratio of temperature compensation and load adjustment in the composite regulation. When the confidence descriptor is close to one, the weighting coefficient is close to one, which ensures that the contribution ratio of temperature compensation and load adjustment in the composite regulation is fully preserved.

[0108] The conversion of the composite regulation value into a composite control reference signal recognizable by the high-voltage power supply pulse width modulation control loop is accomplished by the voltage offset conversion circuit. The composite regulation value is an analog voltage signal. The voltage offset conversion circuit has a built-in adder and level shifting network to adjust the DC bias level of the analog voltage signal to the input common-mode voltage range allowed by the pulse width modulation control loop error amplifier. The adjusted signal is the composite control reference signal. The composite control reference signal is connected to the non-inverting input of the pulse width modulation control loop error amplifier to replace the original single voltage feedback signal. This allows the switching duty cycle generated by the pulse width modulation control loop to simultaneously respond to the ripple compensation requirements introduced by temperature drift, the dynamic adjustment requirements introduced by load fluctuations, and the adjustment limit constraints introduced by insulation degradation. The coupled influence of these three factors acts on the closed-loop regulation process of the output voltage through the same control channel.

[0109] The beneficial effect is that the amplitude and frequency of the partial discharge ultrasonic characteristic signal are used to determine the insulation state confidence descriptor by querying a mapping table through the insulation state confidence assessment logic unit. Using this confidence descriptor as a weighted benchmark, the voltage compensation reference and dynamic load adjustment reference are simulated and multiplied to obtain a composite adjustment value. This composite control reference signal is then formed by adjusting the bias level through a voltage offset conversion circuit and input to the in-phase input of the pulse width modulation control loop error amplifier. This process transforms the degree of insulation degradation into a confidence descriptor that constrains the adjustment authority of temperature compensation and load adjustment functions. The composite control reference signal synchronously responds to temperature drift ripple compensation requirements, load fluctuation dynamic adjustment requirements, and insulation degradation adjustment constraints within the same control channel, achieving comprehensive optimized control of the high-voltage power supply output voltage under multi-source disturbance coupling conditions.

[0110] like Figure 2 The diagram shown is a functional block diagram of an electron beam high-voltage power supply optimization system based on fiber optic sensing, provided in an embodiment of the present invention.

[0111] The electron beam high-voltage power supply optimization system 100 based on fiber optic sensing described in this invention can be installed in an electronic device. Depending on the functions implemented, the electron beam high-voltage power supply optimization system 100 may include a thermally induced ripple modeling and compensation generation module 101, a feedforward dynamic adjustment module 102, and a composite control generation module 103. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0112] In this embodiment, the functions of each module / unit are as follows:

[0113] The thermally induced ripple modeling and compensation generation module 101 is used to establish a thermally induced parameter drift correlation model based on the correlation between temperature and voltage ripple in the high-voltage power supply, and generate a voltage compensation reference value for the high-voltage power supply based on the thermally induced parameter drift correlation model.

[0114] The feedforward dynamic adjustment module 102 is used to feed forward the voltage compensation reference value to the pulse width modulation control loop of the high voltage power supply, thereby generating a dynamic load adjustment reference value.

[0115] The composite control generation module 103 is used to generate a composite control reference signal for the high-voltage power supply based on the voltage compensation reference value, the dynamic load adjustment reference value, and the insulation state.

[0116] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0117] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0118] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0119] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0120] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for optimizing an electron beam high-voltage power supply based on fiber optic sensing, characterized in that, The method includes: Based on the correlation between temperature and voltage ripple in a high-voltage power supply, a thermally induced parameter drift correlation model is established. Based on this model, a voltage compensation reference value for the high-voltage power supply is generated, including: The center wavelength shift in the fiber optic sensing network is used as distributed temperature field data. The changes in the interference spectrum in the fiber optic sensing monitoring network are used as ultrasonic characteristic signals of partial discharge in the insulation state. The voltage ripple data and load current data at the power output terminal of the high-voltage power supply are synchronized through a high-voltage isolation sampling circuit. The distributed temperature field data and the output voltage ripple data are time-domain aligned to obtain a temperature-ripple correlation dataset. Based on the temperature-ripple correlation dataset, the mapping relationship between temperature gradient change features and ripple amplitude fluctuation features is extracted to establish a thermal parameter drift correlation model. The calculation formula of the thermal parameter drift correlation model is as follows: ； In the formula, The temperature is Predicted peak output voltage ripple at time The temperature is The initial output voltage ripple peak value at that time. It is the first-order temperature drift coefficient. The real-time average operating temperature, The preset reference temperature, It is the second-order temperature drift coefficient; The voltage compensation reference value is fed forward to the pulse width modulation control loop of the high voltage power supply to generate a dynamic load adjustment reference value. Based on the voltage compensation reference value, the dynamic load adjustment reference value, and the insulation state, a composite control reference signal for the high-voltage power supply is generated, including: Based on the amplitude and frequency of occurrence of the partial discharge ultrasonic characteristic signal, the confidence descriptor of the insulation state is determined; Using the confidence level description as a weighting benchmark, the voltage compensation reference and the dynamic load adjustment reference are weighted and synthesized to obtain a composite adjustment value; The composite adjustment amount is converted into a composite control reference signal that can be recognized by the pulse width modulation control loop of the high-voltage power supply, wherein the composite control reference signal is used to simultaneously respond to the combined effects of temperature drift, load fluctuation and insulation degradation on the output voltage.

2. The method for optimizing an electron beam high-voltage power supply based on fiber optic sensing as described in claim 1, characterized in that, Before establishing the thermally induced parameter drift correlation model based on the correlation between temperature and voltage ripple in the high-voltage power supply, the following steps are also included: Sensors are deployed in the high-potential nodes and insulating support structures of the high-voltage power supply. Optical signals are introduced by combining insulated optical fibers to construct an optical fiber sensing and monitoring network covering the electromagnetic core area of ​​high-voltage power supplies.

3. The method for optimizing an electron beam high-voltage power supply based on fiber optic sensing as described in claim 1, characterized in that, The step of generating the voltage compensation reference value of the high-voltage power supply based on the thermally induced parameter drift correlation model includes: The real-time temperature data of the high-voltage power supply at the current moment is input into the thermally induced parameter drift correlation model to obtain the predicted output voltage ripple peak value at the current moment. Based on the difference between the predicted output voltage ripple peak value and the preset standard ripple reference value, the voltage compensation direction and compensation magnitude for offsetting the temperature drift effect are determined. Based on the voltage compensation direction and the compensation amplitude, a voltage compensation reference quantity is generated that is superimposed on the basic control quantity of the high-voltage power supply.

4. The method for optimizing an electron beam high-voltage power supply based on fiber optic sensing as described in claim 3, characterized in that, The step of feeding the voltage compensation reference value forward to the pulse width modulation control loop of the high-voltage power supply to generate a dynamic load adjustment reference value includes: The voltage compensation reference value is input to the given signal input terminal of the pulse width modulation control loop in the high voltage power supply, and is superimposed with the original voltage feedback error signal of the pulse width modulation control loop in the time domain. Based on the superimposed signal, the reference threshold of the comparator in the pulse width modulation control loop is adjusted so that the pulse width modulation control loop changes the modulation strategy in advance before the output fluctuation is caused by temperature change. The pulse width modulation control loop generates a temperature-compensated feedforward control signal based on the adjusted reference threshold.

5. The method for optimizing an electron beam high-voltage power supply based on fiber optic sensing as described in claim 4, characterized in that, After the pulse width modulation control loop generates a temperature-compensated feedforward control signal based on the adjusted reference threshold, the following is also included: The load dynamic response requirement is determined based on the rate of change of the load current data. The temperature-compensated feedforward control signal is correlated with the load dynamic response requirements to determine the load adjustment margin under the premise of satisfying temperature compensation. Based on the load adjustment margin, the dynamic load adjustment reference value is generated while ensuring the temperature compensation effect and taking into account the impact of load fluctuations on the output voltage.

6. An electron beam high-voltage power supply optimization system based on fiber optic sensing, used to implement the electron beam high-voltage power supply optimization method based on fiber optic sensing as described in claim 1, the system comprising: The thermally induced ripple modeling and compensation generation module is used to establish a thermally induced parameter drift correlation model based on the correlation between temperature and voltage ripple in the high-voltage power supply, and generate the voltage compensation reference value of the high-voltage power supply based on the thermally induced parameter drift correlation model. The feedforward dynamic adjustment module is used to feed the voltage compensation reference value to the pulse width modulation control loop of the high voltage power supply, thereby generating a dynamic load adjustment reference value. The composite control generation module is used to generate a composite control reference signal for the high-voltage power supply based on the voltage compensation reference value, the dynamic load adjustment reference value, and the insulation state.

Images