Robust Fault Detection Using Subspace Aided Data Driven Design

Hussain, Abrar; Khan, Abdul Qayyum; Abid, Muhammad

doi:10.1002/asjc.1086

Cited by 12 publications

(10 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where • † denotes the Moore Penrose pseudo-inverse of the matrix •. ξ/ψis a shorthand for the projection of the row space of matrix ξ∈R p×σ on the row space of matrix ψ∈R q×σ [27] ξ/ψ = def ξ × Π ψ = ξψ T × (ψψ T ) † × ψ (27) Π ψ ⊥ is the geometric operator that projects the row space of a matrix onto the orthogonal complement of the row space of matrix ψ∈R q×σ :…”

Section: Data-driven Residual Generationmentioning

confidence: 99%

“…ξ / ψ is a shorthand for the projection of the row space of matrix ξ ∈ R p×N on the row space of matrix ψ ∈ R q×N [27]

bold-italicξ / bold-italicψ \overset{def}{=} bold-italicξ \times Π_{ψ} = ξ ψ^{T} \times false(ψ ψ^{T} false) † \times bold-italicψ

Π ψ ⊥ is the geometric operator that projects the row space of a matrix onto the orthogonal complement of the row space of matrix ψ ∈ R q×N :

bold-italicξ / ψ^{⊥} \overset{def}{=} bold-italicξ \times Π_{ψ^{⊥}}

where Π ψ ⊥ = I N − Π ψ .…”

Section: Robust Fault‐detection Design Of Wind Turbinesmentioning

confidence: 99%

See 1 more Smart Citation

Robust fault‐detection based on residual K–L divergence for wind turbines

Zhang

Wang

Xia

et al. 2019

IET Renewable Power Generation

View full text Add to dashboard Cite

A robust fault-detection design based on residual Kullback-Leibler (K-L) divergence, which is applied to a 5 MW offshore wind turbine (WT) benchmark, is presented. The main challenges of the wind turbine fault detection lie in its complex operation conditions and disturbances as well as measurement noise. For overcoming these difficulties, the measured data are divided on the basis of the operation conditions of WT. The robust residual generator based on parity vector is adopted to calculate the residual under different operation conditions. The K-L divergence based on probability density function is employed to measure the residual. Then, the threshold for the fault detection is determined in line with both false alarm rate and missed detection rate. The simulation results show that the performance and effectiveness of the proposed robust fault detection are better than compared with other data-driven fault-detection approaches.

show abstract

Section: Data-driven Residual Generationmentioning

confidence: 99%

“…ξ / ψ is a shorthand for the projection of the row space of matrix ξ ∈ R p×N on the row space of matrix ψ ∈ R q×N [27]

bold-italicξ / bold-italicψ \overset{def}{=} bold-italicξ \times Π_{ψ} = ξ ψ^{T} \times false(ψ ψ^{T} false) † \times bold-italicψ

Π ψ ⊥ is the geometric operator that projects the row space of a matrix onto the orthogonal complement of the row space of matrix ψ ∈ R q×N :

bold-italicξ / ψ^{⊥} \overset{def}{=} bold-italicξ \times Π_{ψ^{⊥}}

where Π ψ ⊥ = I N − Π ψ .…”

Section: Robust Fault‐detection Design Of Wind Turbinesmentioning

confidence: 99%

Robust fault‐detection based on residual K–L divergence for wind turbines

Zhang

Wang

Xia

et al. 2019

IET Renewable Power Generation

View full text Add to dashboard Cite

show abstract

“…in the process industry. The interested reader is referred to Ding et al (2009), Qin (2009), Wang et al (2011), Yin et al (2012a, 2012b), Hussain et al (2016), Jamil et al (2016), Yin et al (2017) and Tariq et al (2019) for some recent developments in this area.…”

Section: Introductionmentioning

confidence: 99%

Parity-based robust data-driven fault detection for nonlinear systems using just-in-time learning approach

Abbasi

Khan

Abid

et al. 2020

Transactions of the Institute of Measurement and Control

Self Cite

View full text Add to dashboard Cite

This paper discusses a framework for fault detection in nonlinear processes. A novel subspace aided parity-based data-driven technique is proposed using the so-called just-in-time learning (JITL) approach. The idea of the proposed technique is to choose a set of the most similar data samples from the database for each test data sample using the JITL approach to address the nonlinearity problem. The parity vector is constructed for each test sample using the selected data samples to generate the residual signal. The salient features of the proposed technique include easy and simple implementation together with effectiveness in fault detection in the presence of disturbances and measurement noise. The fault detection scheme is so designed that it exhibits robustness against sensor noise and disturbances and sensitivity to faults. A case study for the fault detection of a nuclear research reactor (NRR) is presented to demonstrate the efficacy of the technique. The NRR is a highly nonlinear and complex process. Two faults of NRR, namely external reactivity insertion and control rod withdrawal, are successfully detected using the proposed approach.

show abstract

“…Fault diagnosis techniques deal with the problem of detecting and isolating failures occurring in physical components and in the instrumentation of a system. There are essentially two approaches to fault diagnosis: the model-based [1,2] and the data-based approach [3,4]. Model-based approaches exploit measured inputoutput signals and a mathematical model of the system to derive diagnostic indicators, known as residuals, that are sensitive to faults.…”

Section: Introductionmentioning

confidence: 99%

“…Data-based approaches use residual generators that are derived directly from experimental data using system identification techniques [5]. Data-based approaches are preferable where a physical knowledge of the system is not available or when the system input-output relations are too complex [4,6]. Today, fault diagnosis is a mature research area, the main research directions of which can be categorized as parity-equations, state-observers (or state estimators), parameter estimation, and statistical process control [7].…”

Section: Introductionmentioning

confidence: 99%

Data‐based design of robust fault detection and isolation residuals via LASSO optimization and Bayesian filtering

Cascianelli

Crocetti

Ricci

et al. 2020

Asian Journal of Control

View full text Add to dashboard Cite

In this paper, a data‐based approach for the design of structured residual subsets for the robust isolation of sensor faults is proposed. Linear regression models are employed to estimate faulty signals and to build a set of primary residuals. L1‐regularized least squares estimation is used to identify model parameters and to enforce sparsity of the solutions by increasing the regularization weight. In this way, it is possible to generate a set of residuals generators with different fault sensitivity. Then, a residual selection procedure based on fault sensitivity maximization is proposed to extract a minimum size subset of structured residuals that allows for isolation of the faulty sensor. To overcome modelling uncertainty, a robust recursive Bayesian Filter has been employed to process, online, the distance of the residuals from nominal fault directions, providing a fault probability for each sensor. The proposed method has been validated by designing and testing a fault isolation scheme for six aircraft sensors using multi‐flight experimental data of a P92 Tecnam aircraft.

show abstract

Robust Fault Detection Using Subspace Aided Data Driven Design

Cited by 12 publications

References 28 publications

Robust fault‐detection based on residual K–L divergence for wind turbines

Robust fault‐detection based on residual K–L divergence for wind turbines

Parity-based robust data-driven fault detection for nonlinear systems using just-in-time learning approach

Data‐based design of robust fault detection and isolation residuals via LASSO optimization and Bayesian filtering

Contact Info

Product

Resources

About