Operable adaptive sparse identification of systems: Application to chemical processes

Abstract

Over the past few decades, several data-driven methods have been developed for identifying a model that accurately describes the process dynamics. Lately, sparse identification of nonlinear dynamics (SINDy) has delivered promising results for various nonlinear processes. However, at any instance of plant-model mismatch or process upset, retraining the model using SINDy is computationally expensive and cannot guarantee to catch up with rapidly changing dynamics. Hence, we propose operable adaptive sparse identification of systems (OASIS) framework that extends the capabilities of SINDy for accurate, automatic, and adaptive approximation of process models. First, we use SINDy to obtain multiple models from historical data for varying input settings. Next, using these models and their training data, we build a deep neural network that is incorporated in a model predictive control framework for closed-loop operation. We demonstrate the OASIS methodology on the identification and control of a continuous stirred tank reactor.

A hybrid science‐guided machine learning approach for modeling chemical processes: A review

“…They used sparse regression in combination with feature selection to identify accurate models in an adaptive model identification methodology which requires much less than data that current methods. In a similar study Bhadiraju et al 121 have developed a modified SINDy approach that is helpful in cases of plant model mismatch and does not require retraining and hence computationally less expensive.…”

Deterministic Chaos Detection and Simplicial Local Predictions Applied to Strawberry Production Time Series

“…In this article, we discuss an alternative non-linear method for the cases in which the resulting series is not stationary. Although we find many emerging non-linear techniques that can be used to make both short-term and long-term predictions on non-stationary chaotic data, such as the sparse identification of nonlinear dynamics (SINDy) algorithm [11] widely used to model non-linear dynamic systems and make predictions on them [12][13][14][15], or non-linear systems reconstruction techniques that allow the regeneration of time series subjected to white noise, which would allow a new study of stationarity and eliminate the disturbances associated with the observed variable [16,17], in this work, we focus on maximal Lyapunov exponents.…”

On the Physical Consistency of Evolution Laws Obtained with Sparse Regression

“…Alternatively, regression techniques are more proper for providing the interpretability that NN does not have, which makes them an attractive tool to extract dynamic evolution laws from datasets [13]. In this scenario, stands out the technique known as sparse identification of nonlinear dynamical systems (SINDy) [13][14][15][16], a sparse regression method to identify evolution equations from data that proven to be efficient in different areas and problems, and stands out for three aspects: (i) interpretability of the obtained equation; (ii) excellent generalization (extrapolation) capability; (iii) computational efficiency. As several dynamic systems have evolution laws with only a few terms, SINDy looks for a sparse and parsimonious differential equation that best fits the known data.…”