Adaptive Online Multivariate Signal Extraction With Locally Weighted Robust Polynomial Regression

Journal Article · 09 May 2023 · Data Science in Science

DOI: https://doi.org/10.1080/26941899.2023.2200856 · OSTI ID:1973155

Klanderman, Molly C.  ^[1]; Lee, Junho  ^[1]; Villez, Kris  ^[2]; Cath, Tzahi Y.  ^[3]; Hering, Amanda S.  ^[1]

---

1. Baylor Univ., Waco, TX (United States)
2. Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
3. Colorado School of Mines, Golden, CO (United States)

High-frequency, multivariate data collected in real-time and used to control or make decisions regarding a process’ operation often contain some noise and outliers. Thus, a method to extract the signal is needed in order to reduce the number and magnitude of control-based adjustments that are implemented. Such a method must be (i) online, depending only on past and current observations; (ii) fast, producing a smooth value more quickly than the measurement frequency; (iii) robust, ignoring brief bursts of erroneously measured values; (iv) multivariate, ignoring observations that are jointly unusual; (v) adaptive, adjusting to periods of rapid fluctuation in the signal versus periods of stability; and (vi) purely data-driven, not incorporating any information about the process from which the data are collected. Most existing methods are only able to address a subset of these six features. Furthermore, we also require the method to be nonlinear, providing a local nonlinear estimate of the signal. In this work, we propose a novel, real-time signal extraction method based on a local, robust polynomial fit. We demonstrate the performance of our method compared to a state-of-the-art competitor through simulation. For illustration, the methodology is applied to data collected from a reverse osmosis water treatment process.

• Research Organization: National Alliance for Water Innovation (NAWI), Berkeley, CA (United States); Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
• Sponsoring Organization: National Science Foundation (NSF); USDOE Office of Energy Efficiency and Renewable Energy (EERE); USDOE Office of Energy Efficiency and Renewable Energy (EERE), Energy Efficiency Office. Advanced Materials & Manufacturing Office (AMMTO); USDOE Office of Energy Efficiency and Renewable Energy (EERE), Energy Efficiency Office. Advanced Materials & Manufacturing Technologies Office (AMMTO)
• Grant/Contract Number: AC05-00OR22725
• OSTI ID: 1973155
• Journal Information: Data Science in Science, Journal Name: Data Science in Science Journal Issue: 1 Vol. 2; ISSN 2694-1899
• Publisher: Taylor & Francis OnlineCopyright Statement
• Country of Publication: United States
• Language: English