CN-121808355-B - Intelligent control system for intermediate frequency power supply

CN121808355BCN 121808355 BCN121808355 BCN 121808355BCN-121808355-B

Abstract

The invention relates to the technical field of power electronic converters and induction heating intelligent control, in particular to an intermediate frequency power supply intelligent control system, which comprises a data acquisition module, a characteristic extraction module, a state evaluation module and an adaptive control module, wherein the data acquisition module is configured to acquire end-side voltage data and end-side current data of power conversion hardware in real time, the characteristic extraction module is configured to receive the end-side voltage data and the end-side current data and extract broadband waveform distortion characteristics, the state evaluation module is configured to receive the broadband waveform distortion characteristics and calculate a model trust entropy value based on the broadband waveform distortion characteristics, the adaptive control module is configured to receive the model trust entropy value, trigger a second control mode to generate a second switch instruction and send the second switch instruction to the power conversion hardware, the model trust entropy value represents the predictive deviation degree of a preset robust control model on the current real physical state of the power conversion hardware, and the problem of insufficient fine granularity service parameter perception in the background technology is solved.

Inventors

KANG YUANZHENG
KANG HAOPENG
ZHANG HONGTAO

Assignees

西安蓝辉科技股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260312

Claims (9)

1. An intelligent control system for an intermediate frequency power supply, the system comprising: the data acquisition module is configured to acquire end-side voltage data and end-side current data of the power conversion hardware in real time; The characteristic extraction module is configured to receive the terminal side voltage data and the terminal side current data and extract broadband waveform distortion characteristics; The state evaluation module is configured to receive the broadband waveform distortion characteristics and calculate a model trust entropy value based on the broadband waveform distortion characteristics; The self-adaptive control module is configured to receive the model trust entropy value and compare the model trust entropy value with a preset safety threshold, and when the model trust entropy value is smaller than or equal to the preset safety threshold, the self-adaptive control module triggers a first control mode to generate a first switching instruction and send the first switching instruction to the power conversion hardware; the model trust entropy value characterizes the prediction deviation degree of a preset robust control model on the current real physical state of the power conversion hardware; the control target of the first control mode is to maximize active power output and dynamic response speed, and the control target of the second control mode is to inhibit microscopic thermal cycle and thermoelectric fatigue accumulation of the power conversion hardware; The state evaluation module includes: A sensorless measuring and calculating unit configured to input the broadband waveform distortion characteristic into a preset thermoelectric mapping matrix reflecting the association relation between waveform distortion and parasitic parameters under the condition that an external temperature sensor is not introduced, and calculate the internal parasitic parameter drift amount of the power conversion hardware; and an entropy value quantization unit configured to generate the model trust entropy value by calculating information entropy quantization of probability distribution of the difference value based on the difference value between the internal parasitic parameter drift amount and a reference parameter.
2. The intelligent control system of an intermediate frequency power supply according to claim 1, wherein the feature extraction module comprises: The interference separation unit is configured to analyze the terminal side voltage data and the terminal side current data by utilizing a frequency domain analysis algorithm and separate fundamental wave characteristics and high-frequency harmonic noise caused by coupling of a network side and a load side; and the distortion reconstruction unit is configured to reconstruct the broadband waveform distortion characteristic based on the high-frequency harmonic noise, filter the pseudo characteristic data with the amplitude lower than a preset noise threshold value, and further send the cleaned broadband waveform distortion characteristic to the state evaluation module.
3. The intelligent control system of an intermediate frequency power supply according to claim 2, wherein in the first control mode, the adaptive control module comprises: a reinforcement learning unit configured to run a deep reinforcement learning algorithm, performing dynamic resonance frequency tracking based on the terminal side voltage data and the terminal side current data; And the extremum optimizing unit is configured to calculate a duty ratio parameter meeting a zero voltage opening condition or a zero current opening condition according to the result of the dynamic resonance frequency tracking, and generate the first switching instruction according to the duty ratio parameter.
4. The intelligent control system of an intermediate frequency power supply according to claim 3, wherein in the second control mode, the adaptive control module comprises: A degraded operation unit configured to forgo performing the dynamic resonance frequency tracking and actively reduce a switching frequency of the power conversion hardware; and the heat distribution balancing unit is configured to calculate a phase parameter for forcing the system to enter a derating non-resonant operation mode by adjusting a phase shift angle of the driving signal, introduce reactive circulation into the power conversion hardware according to the phase parameter so as to balance heat distribution, and further generate the second switch instruction.
5. The intelligent control system of an intermediate frequency power supply of claim 1, wherein the adaptive control module is further configured to execute multi-stage boundary defense logic: setting a preset critical failure threshold value, wherein the preset critical failure threshold value is larger than the preset safety threshold value; and triggering a second control mode when the trust entropy value of the model is larger than the preset safety threshold value, wherein the second control mode specifically comprises the following steps: Executing the second control mode when the model trust entropy value is greater than the preset safety threshold and less than or equal to the preset critical failure threshold; When the model trust entropy value is larger than the preset critical failure threshold value, triggering a third control mode to generate a blocking pulse instruction and sending the blocking pulse instruction to the power conversion hardware; The control target of the third control mode is to cut off energy transmission within a preset hardware damage time boundary, and the hardware damage time boundary is set to be two milliseconds.
6. The intelligent control system of an intermediate frequency power supply according to any one of claims 1 to 5, further comprising: the trust reconstruction module is configured to collect feedback voltage data and feedback current data of the power conversion hardware under the drive of the second switch instruction during the operation of the second control mode; the model updating module is configured to calculate a prediction error by using the feedback voltage data and the feedback current data, adaptively correct bottom parameters of the preset robust control model based on the prediction error, and send a recovery instruction to the adaptive control module after the correction is completed; The adaptive control module is further configured to re-compare the model trust entropy value to the preset security threshold in response to the recovery instruction.
7. The intelligent control system of an intermediate frequency power supply according to claim 6, wherein the power conversion hardware comprises: The inverter bridge circuit is composed of a plurality of insulated gate bipolar transistor modules or silicon carbide power modules; and the driving protection circuit is configured to receive the first switch instruction or the second switch instruction and convert the first switch instruction or the second switch instruction into a physical level signal for driving the inverter bridge circuit to be turned on and off.
8. The intelligent control system for an intermediate frequency power supply according to claim 3, wherein, The load side is connected with a special metallurgical smelting furnace or a superconducting material heating device; The terminal side voltage data and the terminal side current data comprise nonlinear impedance mutation information caused by the special metallurgical smelting furnace or the superconducting material heating device in the phase change process.
9. The intelligent control system of an intermediate frequency power supply according to claim 1, wherein, The data acquisition module and the self-adaptive control module are deployed in the field edge controller to meet microsecond real-time control requirements; The feature extraction module and the state evaluation module are deployed in a cloud server and interact data with the field edge controller through an industrial network.

Description

Intelligent control system for intermediate frequency power supply Technical Field The invention relates to the technical field of intelligent control of a power electronic converter and induction heating, in particular to an intelligent control system of an intermediate frequency power supply. Background In the industrial scene of megawatt special metallurgy and superconducting material heating, the operation stability of the intermediate frequency power supply system not only depends on the real-time response of electrical parameters, but also depends on the deep perception of the physical state and service logic data of the underlying hardware; The conventional filter circuit and sampling mechanism often filter weak transient components in a high-frequency switching environment, so that fine-granularity service data capable of reflecting specific waveform distortion such as hardware aging, thermoelectric fatigue and the like cannot be obtained; the existing control logic pursues dynamic response speed when facing nonlinear impedance mutation caused by material phase change, lacks an adaptive adjustment mechanism based on the matching degree of a software model and the real state of hardware, and is easy to cause thermal collapse of the power module under complex interference; Therefore, how to realize the noninductive and fine-grained data acquisition of the bottom parasitic parameters and the macroscopic model trust degree in the high-frequency high-power operation environment of the medium-frequency power supply, and construct a multi-stage self-adaptive control system from performance optimization to active self-healing based on quantized model trust indexes becomes a key problem for improving the survival probability of power conversion hardware under severe working conditions. Disclosure of Invention The invention aims to provide an intelligent control system for an intermediate frequency power supply, which solves the following technical problems: the system is prevented from accelerating thermal collapse under complex phase change interference caused by pursuing dynamic response speed, and the noninductive zero-delay measurement and calculation of the internal microscopic physical state of the inverter can be realized without depending on an external physical sensor, so that microscopic thermal cycle and thermoelectric fatigue accumulation are actively restrained, and the survival probability and long-term process stability of the intermediate frequency power supply under severe working conditions are improved. The aim of the invention can be achieved by the following technical scheme: An intelligent control system for an intermediate frequency power supply, the system comprising: the data acquisition module is configured to acquire end-side voltage data and end-side current data of the power conversion hardware in real time; the characteristic extraction module is configured to receive the terminal side voltage data and the terminal side current data and extract broadband waveform distortion characteristics; the state evaluation module is configured to receive the broadband waveform distortion characteristics and calculate a model trust entropy value based on the broadband waveform distortion characteristics; The self-adaptive control module is configured to receive the model trust entropy value and compare the model trust entropy value with a preset safety threshold, and when the model trust entropy value is smaller than or equal to the preset safety threshold, the self-adaptive control module triggers the first control mode to generate a first switching instruction and send the first switching instruction to the power conversion hardware; the model trust entropy value characterizes the prediction deviation degree of a preset robust control model on the current real physical state of the power conversion hardware; the control objective of the first control mode is to maximize active power output and dynamic response speed, and the control objective of the second control mode is to suppress microscopic thermal cycling and thermal-electrical fatigue accumulation of the power conversion hardware. Optionally, the feature extraction module includes: the interference separation unit is configured to analyze the terminal side voltage data and the terminal side current data by utilizing a frequency domain analysis algorithm and separate fundamental wave characteristics and high-frequency harmonic noise caused by coupling of a network side and a load side; And the distortion reconstruction unit is configured to reconstruct broadband waveform distortion characteristics based on the high-frequency harmonic noise, filter pseudo characteristic data with amplitude lower than a preset noise threshold value, and send the cleaned broadband waveform distortion characteristics to the state evaluation module. Optionally, the state evaluation module includes: the sensorless measuring and calculating unit is configur