Deep Learning-Based Meta-Modeling for Multi-Objective Technology Optimization of Electrical Machines

Optimization of rotating electrical machines is both time- and computationally expensive. Because the different parametrization, design optimization commonly executed separately for each machine technology. In this paper, we present application a variational auto-encoder (VAE) to optimize two technologies simultaneously, namely an asynchronous permanent magnet synchronous machine. After training, employ deep neural network decoder as meta-models predict global key performance indicators (KPIs) generate associated new designs, respectively, through unified latent space in loop. Numerical results demonstrate concurrent parametric multi-objective technology high-dimensional space. The VAE-based approach quantitatively compared classical learning-based direct KPIs prediction.