loading page

Machine Learning Modeling and Predictive Control of Nonlinear Processes Using Noisy Data
  • +1
  • Zhe Wu,
  • David Rincon,
  • Junwei Luo,
  • Panagiotis Christofides
Zhe Wu
University of California, Los Angeles
Author Profile
David Rincon
University of California, Los Angeles
Author Profile
Junwei Luo
University of California, Los Angeles
Author Profile
Panagiotis Christofides
University of California, Los Angeles
Author Profile

Abstract

This work focuses on machine learning modeling and predictive control of nonlinear processes using noisy data. We use long short-term memory (LSTM) networks with training data corrupted by two types of noise: Gaussian and non-Gaussian noise, to train the process model that will be used in a model predictive controller (MPC). We first discuss the LSTM training with noisy data following a Gaussian distribution, and demonstrate that the standard LSTM network is capable of capturing the underlying process dynamic behavior. Subsequently, given that the standard LSTM performs poorly on a noisy dataset from industrial operation (i.e., non-Gaussian noisy data), we propose an LSTM network using Monte Carlo dropout method to reduce the over-fitting to noisy data. Furthermore, an LSTM network using co-teaching training method is proposed to further improve its approximation performance when clean data from a nonlinear model capturing the nominal process state evolution is available.

Peer review status:UNDER REVIEW

20 Aug 2020Submitted to AIChE Journal
26 Aug 2020Assigned to Editor
26 Aug 2020Submission Checks Completed
28 Aug 2020Reviewer(s) Assigned