Power Converter Loop Gain Identification Using Transient ML Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies for analyzing and compensating control loops in switched-mode power supply devices are cumbersome, time-consuming, and often inaccurate due to the need for intrusive measurements and limitations of conventional modeling approaches, which fail to account for nonlinearities and changing parameters over time and operating conditions.
Innovation Solution
A method using a machine-learned model trained on transient output voltage waveforms to predict the frequency-domain loop response of a control loop, allowing for non-invasive and efficient identification and adjustment of the control loop configuration, incorporating machine-learning techniques to improve accuracy and speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If conventional measurement methods are used to identify frequency response, then measurement accuracy can be maintained, but the process becomes time-consuming and intrusive
Solution Approach 1:
The patent replaces conventional mechanical measurement systems (network analyzers with physical connections and perturbation signal injection) with a machine-learning-based computational system. The ML model processes transient voltage waveforms to predict frequency response characteristics, eliminating the need for intrusive physical measurements while maintaining accuracy and significantly reducing measurement time.
2Device complexity
If analytical linearized models are used for control loop compensation, then the design process is simplified, but accuracy deteriorates due to nonlinearities and changing parameters
Solution Approach 1:
The patent transforms the approach by changing from fixed analytical parameters to dynamic, data-driven parameters. The machine-learning model is trained on transient voltage waveforms that capture the actual nonlinear behavior and parameter variations of the power supply device under different operating conditions. This allows the system to maintain simplicity while accurately representing nonlinearities and parameter changes through learned patterns rather than complex analytical models.
Data Source
AI summary
Technologies are provided for identification of closed-loop gain response and compensation of power supply devices. In an aspect, a computing device can receive data indicative of a transient output voltage of a power supply device. The computing device also can determine frequency-domain loop response of a control loop of the power supply device by applying a machine-learned model to the data indicative of the transient output voltage. In addition, or in other aspects, the computing device also can adjust one or multiple compensation component(s) of the power supply device in order to achieve a satisfactory performance during operation of the power supply device.