Online Program Home
My Program

Abstract Details

Activity Number: 390 - Advanced Fault Detection and Attribution in Large and Complex Data Streams
Type: Topic Contributed
Date/Time: Tuesday, July 30, 2019 : 2:00 PM to 3:50 PM
Sponsor: Quality and Productivity Section
Abstract #301758 Presentation
Title: Fault Attribution in a Complex, Nonstationary, and Temporally Dependent Wastewater Treatment System
Author(s): Molly Klanderman*
Companies: Baylor University
Keywords: Fault attribution; Fault detection; Fused lasso; Multivariate statistical process monitoring; Phase I monitoring

Decentralized wastewater treatment (WWT) facilities monitor many features that are complexly related. The ability to detect faults quickly and accurately identify the variables that are affected by a fault are vital to maintaining proper system operation. Various multivariate fault detection and attribution methods have been proposed, but the methods require the data to be independent and identically distributed when the process is in control (IC), and most require a distributional assumption. We propose a distribution-free fault attribution method for nonstationary and dependent processes. We detrend the data using IC observations to account for expected changes due external or user-controlled factors. Next, we perform fused lasso, which penalizes differences in consecutive observations, to detect faults and identify affected variables. To account for autocorrelation, the regularization parameter is chosen using an estimated effective sample size in the Extended Bayesian Information Criterion. We demonstrate the performance of our method compared to a state-of-the-art competitor in a simulation study. Finally, we apply our method to WWT facility data with a known fault.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program