Data-driven Power Loss Identification through Physics-Based Thermal Model Backpropagation
By: Mattia Scarpa , Francesco Pase , Ruggero Carli and more
Potential Business Impact:
Finds hidden energy waste using heat.
Digital twins for power electronics require accurate power losses whose direct measurements are often impractical or impossible in real-world applications. This paper presents a novel hybrid framework that combines physics-based thermal modeling with data-driven techniques to identify and correct power losses accurately using only temperature measurements. Our approach leverages a cascaded architecture where a neural network learns to correct the outputs of a nominal power loss model by backpropagating through a reduced-order thermal model. We explore two neural architectures, a bootstrapped feedforward network, and a recurrent neural network, demonstrating that the bootstrapped feedforward approach achieves superior performance while maintaining computational efficiency for real-time applications. Between the interconnection, we included normalization strategies and physics-guided training loss functions to preserve stability and ensure physical consistency. Experimental results show that our hybrid model reduces both temperature estimation errors (from 7.2+-6.8{\deg}C to 0.3+-0.3{\deg}C) and power loss prediction errors (from 5.4+-6.6W to 0.2+-0.3W) compared to traditional physics-based approaches, even in the presence of thermal model uncertainties. This methodology allows us to accurately estimate power losses without direct measurements, making it particularly helpful for real-time industrial applications where sensor placement is hindered by cost and physical limitations.
Similar Papers
Study Design and Demystification of Physics Informed Neural Networks for Power Flow Simulation
Machine Learning (CS)
Makes power grids safer and more reliable.
Predictive Digital Twins for Thermal Management Using Machine Learning and Reduced-Order Models
Machine Learning (CS)
Predicts car part heat to prevent problems.
Data-Driven Temperature Modelling of Machine Tools by Neural Networks: A Benchmark
Machine Learning (CS)
Fixes machines to make things more perfectly.