PCA Methods and Evidence Based Filtering for Robust Aircraft Sensor Fault Diagnosis

Cartocci, Nicholas; Napolitano, Marcello R.; Valigi, Paolo; Crocetti, Francesco; Fravolini, Mario Luca

doi:10.1109/med48518.2020.9182973

Cited by 9 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since real sensor flight data with sensor faults are not easily available, additive faults are artificially injected on the fault-free data to simulate the occurrence of a sensor fault (this approach of generating artificial faulty data is widely used in the FDi community, see for instance [ 11 , 12 , 13 , 14 , 15 , 16 ]).…”

Section: Machine Learning-based Fault Isolation and Estimationmentioning

confidence: 99%

“…Today, multivariate statistical process monitoring (SPM) methods, such as principal component analysis (PCA) [ 11 ] (and its variants) and parity space approaches [ 12 , 13 , 14 , 15 ], are widespread techniques used for system monitoring and fault diagnosis purposes. The widespread use of PCA-based monitoring techniques is due to its simplicity and its capability to efficiently manage large quantities of multivariate data.…”

Section: Introductionmentioning

confidence: 99%

“…The widespread use of PCA-based monitoring techniques is due to its simplicity and its capability to efficiently manage large quantities of multivariate data. In this context, widespread approaches used for fault isolation (FI) are the so-called contribution plots methods, such as the reconstruction-based contributions method [ 11 , 13 ]. Although powerful and effective, this method might produce incorrect fault isolations due to the so-called smearing effect, i.e., the influence of a faulty sensor measurement on non-faulty sensors contributions.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Data-Driven Fault Diagnosis Techniques: Non-Linear Directional Residual vs. Machine-Learning-Based Methods

Cartocci

Napolitano

Crocetti

et al. 2022

Sensors

View full text Add to dashboard Cite

Linear dependence of variables is a commonly used assumption in most diagnostic systems for which many robust methodologies have been developed over the years. In case the system nonlinearities are relevant, fault diagnosis methods, relying on the assumption of linearity, might potentially provide unsatisfactory results in terms of false alarms and missed detections. In recent years, many authors have proposed machine learning (ML) techniques to improve fault diagnosis performance to mitigate this problem. Although very powerful, these techniques require faulty data samples that are representative of any fault scenario. Additionally, ML techniques suffer from issues related to overfitting and unpredictable performance in regions which are not fully explored in the training phase. This paper proposes a non-linear additive model to characterize the non-linear redundancy relationships among the system signals. Using the multivariate adaptive regression splines (MARS) algorithm, these relationships are identified directly from the data. Next, the non-linear redundancy relationships are linearized to derive a local time-dependent fault signature matrix. The faulty sensor can then be isolated by measuring the angular distance between the column vectors of the fault signature matrix and the primary residual vector. A quantitative analysis of fault isolation and fault estimation performance is performed by exploiting real data from multiple flights of a semi-autonomous aircraft, thus allowing a detailed quantitative comparison with state-of-the-art machine-learning-based fault diagnosis algorithms.

show abstract

Section: Machine Learning-based Fault Isolation and Estimationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Data-Driven Fault Diagnosis Techniques: Non-Linear Directional Residual vs. Machine-Learning-Based Methods

Cartocci

Napolitano

Crocetti

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Another approach for FI and FE is the well-known reconstruction-based method [ 38 , 46 , 47 ]. In this approach, for each fault direction

, the fault amplitude, at time instant k , is estimated by computing the fault amplitude

that minimizes the norm-two of the reconstruction errors defined as:

…”

Section: Sensor Fi and Fe Based On Primary Residualsmentioning

confidence: 99%

A Comprehensive Case Study of Data-Driven Methods for Robust Aircraft Sensor Fault Isolation

Cartocci

Napolitano

Fravolini

2021

Sensors

View full text Add to dashboard Cite

Recent catastrophic events in aviation have shown that current fault diagnosis schemes may not be enough to ensure a reliable and prompt sensor fault diagnosis. This paper describes a comparative analysis of consolidated data-driven sensor Fault Isolation (FI) and Fault Estimation (FE) techniques using flight data. Linear regression models, identified from data, are derived to build primary and transformed residuals. These residuals are then implemented to develop fault isolation schemes for 14 sensors of a semi-autonomous aircraft. Specifically, directional Mahalanobis distance-based and fault reconstruction-based techniques are compared in terms of their FI and FE performance. Then, a bank of Bayesian filters is proposed to compute, in flight, the fault belief for each sensor. Both the training and the validation of the schemes are performed using data from multiple flights. Artificial faults are injected into the fault-free sensor measurements to reproduce the occurrence of failures. A detailed evaluation of the techniques in terms of FI and FE performance is presented for failures on the air-data sensors, with special emphasis on the True Air Speed (TAS), Angle of Attack (AoA), and Angle of Sideslip (AoS) sensors.

show abstract

“…The main contributions of this work are given as follows: An improved PCA‐based detection method is developed to solve the influence of attack detector on defense methods. Compared with the work in [25], the proposed method can enhance the detection rate and reduce the false‐positive rate by introducing the mathematical transformation principle method. A genetic optimization algorithm‐based defense method is proposed to address the limitation of the precomputed design parameter matrix. Compared with the existing defense works in [26] and [27], the developed method can quickly eliminate the influence of FDIA on voltage. …”

Section: Introductionmentioning

confidence: 99%

A novel detection and defense mechanism against false data injection attack in smart grids

Cui

Gao

Guo

2023

IET Generation Trans & Dist

View full text Add to dashboard Cite

As the next generation of green power system, smart grids have gradually enhanced the operation efficiency of power system. Meanwhile, the application of communication and intelligent technologies make the power grid more vulnerable to the emerging cyber‐physical attacks, such as the false data injection attack (FDIA). Particularly, the deception property of the FDIA on the output measurement estimation can fool the current security mechanism without triggering an alarm. Motivated by this problem, this paper aims at developing a novel detection and recovery mechanism against FDIA in smart grid. Based on the established state space grid model derived from the three‐phase sinusoidal voltage equations, an improved principal component analysis (PCA)‐based detection method is proposed. By introducing the mathematical transformation principle method, the detection performance such as detection rate and false positive rate can be improved. To keep the stable running of power system, a genetic optimization algorithm‐based linear quadratic regulator (LQR) defense method is developed. In addition, to improve the response performance to external attacks, an artificial intelligence method named genetic optimization algorithm is introduced to optimize the robust performance of the proposed defense method. Finally, the simulation results on the IEEE 6‐bus and 118‐bus grid system demonstrate the superiority of the proposed genetic algorithm optimization‐based LQR defense method.

show abstract

PCA Methods and Evidence Based Filtering for Robust Aircraft Sensor Fault Diagnosis

Cited by 9 publications

References 24 publications

Data-Driven Fault Diagnosis Techniques: Non-Linear Directional Residual vs. Machine-Learning-Based Methods

Data-Driven Fault Diagnosis Techniques: Non-Linear Directional Residual vs. Machine-Learning-Based Methods

A Comprehensive Case Study of Data-Driven Methods for Robust Aircraft Sensor Fault Isolation

A novel detection and defense mechanism against false data injection attack in smart grids

Contact Info

Product

Resources

About