Constructing neural network based models for simulating dynamical systems

C Legaard, T Schranz, G Schweiger, J Drgoňa… - ACM Computing …, 2023 - dl.acm.org
Dynamical systems see widespread use in natural sciences like physics, biology, and
chemistry, as well as engineering disciplines such as circuit analysis, computational fluid …

[HTML][HTML] The Cyclostratigraphy Intercomparison Project (CIP): consistency, merits and pitfalls

M Sinnesael, D De Vleeschouwer, C Zeeden… - Earth-Science …, 2019 - Elsevier
Cyclostratigraphy is an important tool for understanding astronomical climate forcing and
reading geological time in sedimentary sequences, provided that an imprint of insolation …

Learning nonlinear state–space models using autoencoders

D Masti, A Bemporad - Automatica, 2021 - Elsevier
We propose a methodology for the identification of nonlinear state–space models from
input/output data using machine-learning techniques based on autoencoders and neural …

Model Predictive Control when utilizing LSTM as dynamic models

M Jung, PR da Costa Mendes, M Önnheim… - … Applications of Artificial …, 2023 - Elsevier
The prediction model is the most important part of an MPC strategy. The accuracy of such a
model influences the quality of predictions and control performance of the algorithm. In some …

[HTML][HTML] Continuous-time system identification with neural networks: Model structures and fitting criteria

M Forgione, D Piga - European Journal of Control, 2021 - Elsevier
This paper presents tailor-made neural model structures and two custom fitting criteria for
learning dynamical systems. The proposed framework is based on a representation of the …

Learning nonlinear state-space models using deep autoencoders

D Masti, A Bemporad - 2018 IEEE Conference on Decision and …, 2018 - ieeexplore.ieee.org
We introduce a new methodology for the identification of nonlinear state-space models
using machine-learning techniques based on deep autoencoders for dimensionality …

Comparing methods of DC motor control for UUVs

R Shah, T Sands - Applied Sciences, 2021 - mdpi.com
Featured Application Underwater vehicle control surfaces motor control. Abstract Adaptive
and learning methods are proposed and compared to control DC motors actuating control …

Kernel-based methods for Volterra series identification

A Dalla Libera, R Carli, G Pillonetto - Automatica, 2021 - Elsevier
Volterra series approximate a broad range of nonlinear systems. Their identification is
challenging due to the curse of dimensionality: the number of model parameters grows …

Multiple shooting for training neural differential equations on time series

EM Turan, J Jäschke - IEEE Control Systems Letters, 2021 - ieeexplore.ieee.org
Neural differential equations have recently emerged as a flexible data-driven/hybrid
approach to model time-series data. This letter experimentally demonstrates that if the data …

Improved initialization of state-space artificial neural networks

M Schoukens - 2021 European Control Conference (ECC), 2021 - ieeexplore.ieee.org
The identification of black-box nonlinear statespace models requires a flexible
representation of the state and output equation. Artificial neural networks have proven to …