An energy-attributed graph approach for the purposes of FDI in a heated two-tank system

Uren, Kenneth R.; Schoor, George van; Auret, Lidia

doi:10.1016/j.ifacol.2019.09.181

Cited by 6 publications

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The approach, according to De Bruin et al and Uren et al, uses the eigenvalues from the cost matrices to generate a qualitative fault signature [23] , [44] . The first step of implementation involves a comparison of each of the 18 sets of operational node signature matrices

, with

.…”

Section: Fdi Techniquesmentioning

confidence: 99%

A comparative study of energy graph-based fault detection and isolation techniques applied to a lignite plant

Smith,

van Schoor,

Uren

et al. 2023

Heliyon

Self Cite

View full text Add to dashboard Cite

, with

.…”

Section: Fdi Techniquesmentioning

confidence: 99%

A comparative study of energy graph-based fault detection and isolation techniques applied to a lignite plant

Smith,

van Schoor,

Uren

et al. 2023

Heliyon

Self Cite

View full text Add to dashboard Cite

“…Therefore, attributed graph representations were introduced by [17] to combine energy and structural information in a holistic way. This energy graph-based approach was improved by extracting eigenvector information from a graph cost matrix [18]. The aim with of using eigenvector information is to obtain more descriptive residuals for fault detection and diagnosis.…”

mentioning

confidence: 99%

An Energy Graph-Based Approach to Fault Diagnosis of a Transcritical CO2 Heat Pump

et al. 2020

Self Cite

View full text Add to dashboard Cite

The objective of this paper is to describe an energy-based approach to visualize, identify, and monitor faults that may occur in a water-to-water transcritical CO 2 heat pump system. A representation using energy attributes allows the abstraction of all physical phenomena present during operation into a compact and easily interpretable form. The use of a linear graph representation, with heat pump components represented as nodes and energy interactions as links, is investigated. Node signature matrices are used to present the energy information in a compact mathematical form. The resulting node signature matrix is referred to as an attributed graph and is populated in such a way as to retain the structural information, i.e., where the attribute points to in the physical system. To generate the energy and exergy information for the compilation of the attributed graphs, a descriptive thermal-fluid model of the heat pump system is developed. The thermal-fluid model is based on the specifications of and validated to the actual behavioral characteristics of a physical transcritical CO 2 heat pump test facility. As a first step to graph-matching, cost matrices are generated to represent a characteristic residual between a normal system node signature matrix and a faulty system node signature matrix. The variation in the eigenvalues and eigenvectors of the characteristic cost matrices from normal conditions to a fault condition was used for fault characterization. Three faults, namely refrigerant leakage, compressor failure and gas cooler fouling, were considered. The paper only aims to introduce an approach, with the scope limited to illustration at one operating point and considers only three relatively large faults. The results of the proposed method show promise and warrant further work to evaluate its sensitivity and robustness for small faults.Energies 2020, 13, 1783 2 of 34 Zogg also refers to a field survey [2] that indicates that the average energetic efficiencies of the heat pumps installed in Swiss buildings were about 10% lower than the efficiencies measured under standard laboratory conditions. It was also found that for some installations the performance degradation was as high as 30%. A similar study in the United States of over 55,000 air conditioning units showed that more than 90% were operating with one or more kinds of faults [3]. Yet another study surveying roof-top units indicates an operating efficiency of 80% of designed performance with 63% of these units having performance degradation due to refrigerant leakage [4]. These observations motivate the need for diagnostic techniques which ideally can monitor these energy conversion systems during transient and steady-state operating conditions.In the review papers by Zhao et al., and Shi and O'Brien [5,6] the classification of fault detection methods for energy systems are presented. Fault detection methods can be classified into two subcategories: data-based and knowledge-based. The knowledge-based methods rely on domain and prior knowledge to deri...

show abstract

An energy graph eigendecomposition approach to fault detection and isolation applied to a gas-to-liquids process

Greyling¹,

Schoor²,

Uren³

et al. 2022

Computers & Chemical Engineering

View full text Add to dashboard Cite

An energy-attributed graph approach for the purposes of FDI in a heated two-tank system

Cited by 6 publications

References 8 publications

A comparative study of energy graph-based fault detection and isolation techniques applied to a lignite plant

A comparative study of energy graph-based fault detection and isolation techniques applied to a lignite plant

An Energy Graph-Based Approach to Fault Diagnosis of a Transcritical CO2 Heat Pump

An energy graph eigendecomposition approach to fault detection and isolation applied to a gas-to-liquids process

Contact Info

Product

Resources

About