Fig. 6: Comparison of multi-level physics-informed neural network prediction using a different number of collocation points. | Communications Engineering

Fig. 6: Comparison of multi-level physics-informed neural network prediction using a different number of collocation points.

From: Multi-level physics informed deep learning for solving partial differential equations in computational structural mechanics

Fig. 6

a Beam structure case with a different number of collocation points. The number of data points np at the interior zone is selected as 10, 25, 50, and 100. b Shell structure case with a different number of collocation points. np represents the number of data points at the interior zone, which is given by 100, 500, 1000, and 2000. nb denotes the number of data points at the boundaries.

Back to article page