Improving Root Cause Analysis by Detecting and Removing Transient Changes in Oscillatory Time Series with Application to a 1,3-Butadiene Process

Abstract

Oscillations occurring in industrial process plants often reflect the presence of severe disturbances affecting process operations. Accurate detection and root-cause analysis of oscillations is of great interest for the economic viability of the process operation. Standard oscillation detection and root cause analysis methods require a large enough number of data samples. Unrelated transient changes superimposed on the oscillation pattern reduce the number of useful data samples. The present paper proposes simple heuristic methods to effectively detect and remove two types of transient changes from oscillatory signals, namely step changes and spikes. The proposed methods are used to pre-process oscillatory time series. The accuracy gained when using auto-correlation function method for oscillation detection 1 and transfer entropy method for oscillation propagation 2 is experimentally evaluated. The methods are carried out on a 1.3-Butadiene production process where several measurements showed an established oscillation occurring after a production level change.

Towards Ontology Reshaping for KG Generation with User-in-the-Loop: Applied to Bosch Welding

“…KGs provide semantically structured information that can be interpreted by computing machines [27,39], and an efficient foundation for standardised ways of data retrieval and analytics to support data driven methods. Data driven methods have been widely used in industries [17,18,30,31], especially machine learning [26,29,[35][36][37]. The problem of transforming a bigger ontology to a smaller ontology of the same domain is often referred to as ontology modularisation [1][2][3][4]14] and ontology summarisation [19].…”

SemML: Facilitating Development of ML Models for Condition Monitoring with Semantics

“…In the Oil and Gas industry [36], examples includes equipment and process monitoring in off-shore platforms and oil reservoirs at Equinor [14]. Another example is the detection and processing of disturbances in chemical or process industry for root-cause-analysis at ABB and INEOS [8].…”

“…Indeed, modern machines and production systems are equipped with sensors that constantly collect and send data and with control units that monitor and process these data, coordinate machines and manufacturing environment and send messages, notifications, requests. Availability of these voluminous data has led to a large growth of interest in data analysis for a wide range of industrial applications [5][6][7][8], especially the use of Machine Learning (ML) approaches for condition monitoring for manufacturing processes, machines, oil, gas and chemical systems, and products by predicting system disturbance, machines' down-times or the quality of manufactured products [9]. Such approaches allow to analyse large amount of data and gain fruitful insights for condition monitoring.…”

Machine learning with domain knowledge for predictive quality monitoring in resistance spot welding

“…The common quality monitoring practice is to estimate or assess the product quality after a manufacturing operation happens, so as to provide results for future manufacturing operations. While predictive quality monitoring, is to predict the quality of a manufacturing operation before the actual operation happens, so that necessary measures can be undertaken even before the quality deficiency happens (Zhou et al 2020a(Zhou et al ). et al 2019.…”