title: Physics-informed neural networks for data-free surrogate modelling and engineering optimization – An example from composite manufacturing authors: Tobias Würth, Constantin Krauß, Clemens Zimmerling, Luise Kärger year: 2023 URL: https://doi.org/10.5445/IR/1000159290 Drive: https://1drv.ms/b/s!Ar4x-UlrYAiZla5XEQtEiYnlp5ggtA?e=6TW8va –-

• PINNs are alternative to SBO methods

• Reference 6: one NN per output

• Used same weight for all losses (domain, B.C., I.C.)

• NNs provide automatic jacobian and hessian evaluations, what can be used for great benefit of optimization algorithms

• Implemented in SciANN [[@Haghighat2021sciann]] wrapper to [[@Tensorflow2015Whitepaper]]

• Solves the diffusion equation

• Kamal-Malkin model can be seem as a more general parametrization as JMKA $\rightarrow$ maybe useful in cement hydration modeling?

• Interesting to write a tutorial about!