Machine Learning Model for Event-Based Prognostics in Gas Circulator Condition Monitoring

Abstract

The Strathprints institutional repository (https://strathprints.strath.ac.uk) is a digital archive of University of Strathclyde research outputs. It has been developed to disseminate open access research outputs, expose data about those outputs, and enable the management and persistent access to Strathclyde's intellectual output. Abstract-Gas circulator (GC) units are an important rotating asset used in the Advanced Gas-cooled Reactor (AGR) design, facilitating the flow of CO2 gas through the reactor core. The ongoing maintenance and examination of these machines is important for operators in order to maintain safe and economic generation. GCs experience a dynamic duty cycle with periods of non-steady state behavior at regular refuelling intervals, posing a unique analysis problem for reliability engineers.In line with the increased data volumes and sophistication of available technologies, the investigation of predictive and prognostic measurements has become a central interest in rotating asset condition monitoring. However, many of the state-of-theart approaches finding success deal with the extrapolation of stationary time series feeds, with little to no consideration of more-complex but expected events in the data.In this paper we demonstrate a novel modelling approach for examining refuelling behaviors in GCs, with a focus on estimating their health state from vibration data. A machine learning model was constructed using the operational history of a unit experiencing an eventual inspection-based failure. This new approach to examining GC condition is shown to correspond well with explicit remaining useful life (RUL) measurements of the case study, improving on the existing rudimentary extrapolation methods often employed in rotating machinery health monitoring.

Self‐learning and self‐adaptive resource allocation for cloud‐based software services

“…In this experiment, the value of QoS is mainly determined by response time RT, and it is mapped into the range of [0, 1], as shown in Figure 5. And, the weights of fitness function and cost model are defined in Formula (12).…”

“…Machine learning that gives computers the ability to learn without being explicitly programmed means it gives system the ability to learn from data, and it has made great achievements in many fields. (–) The self‐learning ability is needed in self‐adaptive resource allocation because self‐adaptive management is based on the knowledge, experience, and rules, which are hard to obtain. However, there are still two issues that have to be dealt with.…”

Predicting Remaining Useful Life Based on Hilbert–Huang Entropy with Degradation Model

“…Commonly used frequency-domain vibration signal detection methods include fast Fourier transforms, power spectra, and filtering. However, each of these methods has flaws and loses some characteristics of nonlinear vibrations, meaning that the monitoring frequency band has a decisive impact on the analysis results [38][39][40][41]. e timefrequency analysis method can accurately describe the time and frequency characteristics of fault signals and has more advantages.…”

AESGRU: An Attention-Based Temporal Correlation Approach for End-to-End Machine Health Perception

“…During the past decades, traditional machine learning (ML) models has received great success in various applications, including the health perception [6]. Some algorithms, such as support vector machine (SVM) [7], [8], random forests (RF) [9], and regression model [10], have achieved remarkably results. But notably, as shown in Figure 1, ML algorithms generally require feature engineering to extract important features, which may result in additional human labor and substantial expertise to complete efficiently.…”