CN-121978897-A - Power semiconductor device control method, electronic device and storage medium

CN121978897ACN 121978897 ACN121978897 ACN 121978897ACN-121978897-A

Abstract

The application provides a power semiconductor device control method, electronic equipment and a storage medium, wherein the method comprises the steps of obtaining current working state parameters of a target power semiconductor device in real time, determining performance evaluation weight coefficients according to the current junction temperature, constructing a weighted objective function by combining the current working state parameters, carrying out iterative optimization calculation on the weighted objective function under the condition of meeting the constraint of a preset safe working area to obtain target control parameters, wherein the target control parameters comprise a grid driving voltage amplitude value and a grid resistance value, and generating a driving control signal according to the target control parameters to control a grid of the target power semiconductor device. The application can realize the collaborative optimization of the power semiconductor device in the full working condition range by sensing the running state of the device in real time and dynamically generating the driving parameters based on the multi-objective optimization algorithm.

Inventors

ZHANG BAO
DANG WEN
HUANG QISHENG

Assignees

西安电子科技大学

Dates

Publication Date: 20260505
Application Date: 20251212

Claims (10)

1. A method of controlling a power semiconductor device, comprising: Acquiring current working state parameters of a target power semiconductor device in real time, wherein the current working state parameters comprise current junction temperature; Determining a performance evaluation weight coefficient according to the current junction temperature, and constructing a weighted objective function by combining the current working state parameters; under the condition of meeting the constraint of a preset safe working area, carrying out iterative optimization calculation on the weighted objective function to obtain objective control parameters, wherein the objective control parameters comprise a grid driving voltage amplitude value and a grid resistance value; and generating a driving control signal to control the grid electrode of the target power semiconductor device according to the target control parameter.
2. The method according to claim 1, wherein the current operating state parameters further include an on-current, an off-voltage change rate, an on-voltage drop, and a switching loss, wherein the determining the performance evaluation weight coefficient according to the current junction temperature and constructing the weighted objective function in combination with the current operating state parameters includes: Calculating normalized total power loss, normalized switching time and normalized junction temperature fluctuation amplitude according to a preset normalized reference value and the current working state parameter respectively; Dynamically adjusting the total power loss, the switching time and the performance evaluation weight coefficient of the junction temperature fluctuation range based on the current junction temperature respectively; And summing the normalized total power loss, the normalized switching time and the normalized junction temperature fluctuation amplitude weighted according to the performance evaluation weight coefficient to obtain the weighted objective function.
3. The method for controlling a power semiconductor device according to claim 2, wherein the performing iterative optimization calculation on the weighted objective function under the condition that a preset safe operating area constraint is satisfied to obtain a target control parameter includes: encoding the gate drive voltage magnitude and the gate resistance value into a particle position vector; calculating inertia weight through nonlinear self-adaptive adjustment according to the current junction temperature, and updating the speed and the position of particles based on the inertia weight; Calculating to obtain a punishment item value according to the state parameters corresponding to the updated particle positions and the constraint of the safe working area; Summing the weighted objective function and the penalty term value to obtain a particle fitness value; And updating the individual history optimal solution and the global history optimal solution according to the fitness value until a preset iteration termination condition is met, and determining the final global history optimal solution as the target control parameter.
4. The method for controlling a power semiconductor device according to claim 3, wherein the safe operating region constraint includes a voltage safe constraint, a current safe constraint, a temperature safe constraint, and a change rate constraint, and the calculating the penalty term value according to the state parameter corresponding to the updated particle position and the safe operating region constraint includes: calculating the difference value between the turn-off voltage peak predicted value, the turn-on current peak predicted value, the real-time junction temperature predicted value, the voltage change rate predicted value and the current change rate predicted value corresponding to the particle position and the corresponding safety boundary value respectively; and calculating the punishment item value according to the difference value exceeding the safety boundary value.
5. A method of controlling a power semiconductor device according to claim 3, wherein said calculating an inertial weight by nonlinear adaptive adjustment based on said current junction temperature comprises: Obtaining maximum allowable junction temperature, attenuation coefficient, thermal coupling coefficient and maximum and minimum values of inertial weight; Calculating the temperature ratio of the current junction temperature to the maximum allowable junction temperature; calculating a first adjusting component which is exponentially attenuated along with the temperature rise according to the temperature ratio and the attenuation coefficient; calculating a second adjusting component which has a linear attenuation trend along with the temperature rise according to the temperature ratio; and carrying out superposition summation on the minimum value, the product of the first adjusting component and the inertia weight range and the product of the second adjusting component and the thermal coupling coefficient to obtain the inertia weight.
6. The method according to claim 1, wherein generating a drive control signal to control the gate of the target power semiconductor device according to the target control parameter comprises sequentially performing the following steps: A pre-charging stage, namely adopting a controllable current source mode to control the grid voltage to rise to a preset multiple of the threshold voltage of the target power semiconductor device at a preset rising rate; A main driving stage, namely outputting driving voltage according to the grid driving voltage amplitude in the target control parameter by adopting a constant voltage source mode so as to conduct the target power semiconductor device; And the voltage stabilization maintaining stage is to stabilize the grid voltage at a preset maintaining voltage by using an active clamp circuit after the main driving stage.
7. The method of controlling a power semiconductor device according to claim 6, wherein in the voltage stabilization holding stage, the method further comprises: monitoring the dropping amplitude of the grid voltage in real time through a voltage detection comparator; And if the drop amplitude exceeds the preset percentage of the nominal value, calculating the required compensation charge quantity according to the drop amplitude, and triggering a charge pump circuit to secondarily inject driving charge into the grid electrode until the grid electrode voltage is recovered to the target value.
8. The method of controlling a power semiconductor device according to claim 1, wherein when the target power semiconductor device is a plurality of devices connected in parallel, the method further comprises: during the conduction period of the target power semiconductor device, current values of all parallel branches are respectively collected through sampling resistors; calculating current unbalance according to the maximum value, the minimum value and the average value of the current values of each parallel branch circuit; And if the current imbalance exceeds a preset non-uniformity threshold, for the target power semiconductor device with a higher current value, performing an operation of increasing a gate driving resistance or reducing a voltage maintaining voltage in a voltage stabilizing and maintaining stage until the current imbalance is lower than the preset non-uniformity threshold.
9. An electronic device comprising a memory, a processor, the memory storing a computer program, the processor implementing the power semiconductor device control method according to any one of claims 1 to 8 when executing the computer program.
10. A computer-readable storage medium, characterized in that the storage medium stores a program that is executed by a processor to realize the power semiconductor device control method according to any one of claims 1 to 8.

Description

Power semiconductor device control method, electronic device and storage medium Technical Field The present application relates to the field of power semiconductor devices, and in particular, to a power semiconductor device control method, an electronic device, and a storage medium. Background Power semiconductor devices, such as insulated gate bipolar transistors and silicon carbide metal oxide semiconductor field effect transistors, are core components of power electronic conversion devices, and need to be turned on and off by a gate driving circuit to realize conversion and control of electric energy. In the related art, a gate driving circuit generally employs a fixed driving voltage and gate resistance, or a simple passive adjustment is made only for a single variable. However, the junction temperature, load current and other conditions in the actual running process are dynamically changed in real time. The driving mode of the fixed parameters cannot adapt to the dynamic change, for example, the switching speed is slow and the loss is large under the light-load low-temperature working condition, and the risk of overvoltage or overheat is also faced under the heavy-load high-temperature working condition. Although some simple self-adaptive driving schemes exist in the prior art, the method is often limited to passive adjustment of a single target, and is lack of collaborative optimization on multiple performance indexes such as loss, speed, thermal stability and the like, so that the performance of the device under all working conditions is difficult to be optimal. Disclosure of Invention The present application aims to solve at least one of the technical problems existing in the prior art. Therefore, the application provides a control method of a power semiconductor device, electronic equipment and a storage medium, which can realize the collaborative optimization of the power semiconductor device in a full working condition range by sensing the running state of the device in real time and dynamically generating driving parameters based on a multi-objective optimization algorithm. In a first aspect, an embodiment of the present application provides a method for controlling a power semiconductor device, including: Acquiring current working state parameters of a target power semiconductor device in real time, wherein the current working state parameters comprise current junction temperature; Determining a performance evaluation weight coefficient according to the current junction temperature, and constructing a weighted objective function by combining the current working state parameters; under the condition of meeting the constraint of a preset safe working area, carrying out iterative optimization calculation on the weighted objective function to obtain objective control parameters, wherein the objective control parameters comprise a grid driving voltage amplitude value and a grid resistance value; and generating a driving control signal to control the grid electrode of the target power semiconductor device according to the target control parameter. In some embodiments, the current operating state parameter further includes an on-current, an off-voltage change rate, an on-voltage drop, and a switching loss, and the determining the performance evaluation weight coefficient according to the current junction temperature and constructing the weighted objective function in combination with the current operating state parameter includes: Calculating normalized total power loss, normalized switching time and normalized junction temperature fluctuation amplitude according to a preset normalized reference value and the current working state parameter respectively; Dynamically adjusting the total power loss, the switching time and the performance evaluation weight coefficient of the junction temperature fluctuation range based on the current junction temperature respectively; And summing the normalized total power loss, the normalized switching time and the normalized junction temperature fluctuation amplitude weighted according to the performance evaluation weight coefficient to obtain the weighted objective function. In some embodiments, the performing iterative optimization calculation on the weighted objective function under the condition that the constraint of the preset safe working area is satisfied to obtain the objective control parameter includes: encoding the gate drive voltage magnitude and the gate resistance value into a particle position vector; calculating inertia weight through nonlinear self-adaptive adjustment according to the current junction temperature, and updating the speed and the position of particles based on the inertia weight; Calculating to obtain a punishment item value according to the state parameters corresponding to the updated particle positions and the constraint of the safe working area; Summing the weighted objective function and the penalty term value to obtain a particle