Physics-guided Neural Networks (PGNN): An Application in Lake Temperature Modeling

This paper introduces a novel framework for learning data science models by using the scientific knowledge encoded in physics-based models. This framework, termed as physicsguided neural network (PGNN), leverages the output of physics-based model simulations along with observational features to generate predictions using a neural network architecture. Further, we present a novel class of learning objective for training neural networks, which ensures that the model predictions not only show lower errors on the training data but are also consistent with the known physics. We illustrate the effectiveness of PGNN for the problem of lake temperature modeling, where physical relationships between the temperature, density, and depth of water are used in the learning of neural network model parameters. By using scientific knowledge to guide the construction and learning of neural networks, we are able to show that the proposed framework ensures better generalizability as well as physical consistency of results.