Showcasing My Work
This project aims to develop a surrogate model that predicts thermal behaviour in electronic circuits more quicl;y than traditional simulations.
The project uses machine learning techniques to construct a model that can efficiently approximatee the outcomes of complex thermal simulations. This surrogate model allows forn rapid testing and iteraion in electronic design processes, crucial for thermal management in high-performance electronics.
The model is trained using a dataset generated from an automated data collection script. Training involves adjusting the model weights to minimize the error between the predicted temperatures and actual simulated temperatures.