Physics-guided neural networks with engineering domain knowledge for hybrid process modeling

As neural networks are more frequently used to solve problems in science and engineering, the methods used to incorporate scientific knowledge into these networks are becoming increasingly complex. This work breaks down these complicated techniques into a set of basic strategies which can easily be applied to diverse situations. Several novel neural networks are built using the categories laid out in this work. These networks are tested on simulated data from a continuous stirred tank reactor (CSTR) model to evaluate the advantages provided by each network. The three points demonstrated in this work are: (1) architectural hybrid models can speed up convergence and reduce the amount of data necessary to train a model; (2) adding a physics-guided loss function can improve model generalization and make models more physically consistent; (3) using physics-guided initialization and transfer learning improves accuracy and speeds up convergence, but can harm generalizability if used incorrectly.

Automated summary

As neural networks are increasingly used in science and engineering, incorporating scientific knowledge into these networks has become more complex. This work simplifies these techniques into basic strategies that can be easily applied in different situations. The study builds novel neural networks based on these strategies, and evaluates their advantages using simulated data from a CSTR model.