Approximating the full-field temperature evolution in 3D electronic systems from randomized "Minecraft" systems

Neural Networks as fast physics simulators have a large potential for many engineering design tasks. Prerequisites for a wide-spread application are an easy-to-use workflow for generating training datasets in a reasonable time, and the capability of the network to generalize to unseen systems. In contrast to most previous works where training systems are similar to the evaluation dataset, we propose to adapt the type of training system to the network architecture. Specifically, we apply a fully convolutional network and, thus, design 3D systems of randomly located voxels with randomly assigned physical properties. The idea is tested for the transient heat diffusion in electronic systems. Training only on random "Minecraft" systems, we obtain good generalization to electronic systems four times as large as the training systems (one-step prediction error of 0.07 % vs 0.8 %). 1 INTRODUCTION The idea of using Neural Networks (NNs) as trainable physics simulators has seen much attention sparked by recent impressive results [1, 2, 3]. Once trained such a NN could predict physical properties much faster than any standard simulation method. Potential benefits can be seen in many fields of application, e.g.