Sparse neural networks with skip-connections for nonlinear system identification

01/02/2023
by   Erlend Torje Berg Lundby, et al.
0

Data-driven models such as neural networks are being applied more and more to safety-critical applications, such as the modeling and control of cyber-physical systems. Despite the flexibility of the approach, there are still concerns about the safety of these models in this context, as well as the need for large amounts of potentially expensive data. In particular, when long-term predictions are needed or frequent measurements are not available, the open-loop stability of the model becomes important. However, it is difficult to make such guarantees for complex black-box models such as neural networks, and prior work has shown that model stability is indeed an issue. In this work, we consider an aluminum extraction process where measurements of the internal state of the reactor are time-consuming and expensive. We model the process using neural networks and investigate the role of including skip connections in the network architecture as well as using l1 regularization to induce sparse connection weights. We demonstrate that these measures can greatly improve both the accuracy and the stability of the models for datasets of varying sizes.

READ FULL TEXT
research
09/13/2022

Sparse deep neural networks for modeling aluminum electrolysis dynamics

Artificial neural networks have a broad array of applications today due ...
research
09/22/2022

A novel corrective-source term approach to modeling unknown physics in aluminum extraction process

With the ever-increasing availability of data, there has been an explosi...
research
12/21/2022

Modelling Controllers for Cyber Physical Systems Using Neural Networks

Model Predictive Controllers (MPC) are widely used for controlling cyber...
research
09/11/2021

Structure-preserving Sparse Identification of Nonlinear Dynamics for Data-driven Modeling

Discovery of dynamical systems from data forms the foundation for data-d...
research
05/31/2018

Long-time predictive modeling of nonlinear dynamical systems using neural networks

We study the use of feedforward neural networks (FNN) to develop models ...
research
08/01/2019

Neural Simplex Architecture

We present the Neural Simplex Architecture (NSA), a new approach to runt...
research
12/02/2022

Guaranteed Conformance of Neurosymbolic Models to Natural Constraints

Deep neural networks have emerged as the workhorse for a large section o...

Please sign up or login with your details

Forgot password? Click here to reset