Deep auto-encoder observer multiple-model fast aircraft actuator fault diagnosis algorithm

Ma, Jun; Shi-hong, NI; Xie, Wei; Dong, Wenhan

doi:10.1007/s12555-016-0160-1

Cited by 21 publications

(14 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As explained in Section II, RBM-based and AE-based models can be trained in two steps: layer-wise pretraining, and fine-tuning on the network by stacking previously learned layers. Using this strategy, several researchers built regular DBNs [153]- [159] and AEs [160]- [163] for fault diagnosis; [153]- [155], [160] emphasized network hyperparameter tuning. Note that the layer-wise pretraining step is typically unsupervised, and the pretrained network serves as an initialization to the whole model.…”

Section: ) Structured Datamentioning

confidence: 99%

A Review on Deep Learning Applications in Prognostics and Health Management

et al. 2019

View full text Add to dashboard Cite

Deep learning has attracted intense interest in Prognostics and Health Management (PHM), because of its enormous representing power, automated feature learning capability and best-in-class performance in solving complex problems. This paper surveys recent advancements in PHM methodologies using deep learning with the aim of identifying research gaps and suggesting further improvements. After a brief introduction to several deep learning models, we review and analyze applications of fault detection, diagnosis and prognosis using deep learning. The survey validates the universal applicability of deep learning to various types of input in PHM, including vibration, imagery, time-series and structured data. It also reveals that deep learning provides a one-fits-all framework for the primary PHM subfields: fault detection uses either reconstruction error or stacks a binary classifier on top of the network to detect anomalies; fault diagnosis typically adds a soft-max layer to perform multi-class classification; prognosis adds a continuous regression layer to predict remaining useful life. The general framework suggests the possibility of transfer learning across PHM applications. The survey reveals some common properties and identifies the research gaps in each PHM subfield. It concludes by summarizing some major challenges and potential opportunities in the domain.

show abstract

Section: ) Structured Datamentioning

confidence: 99%

A Review on Deep Learning Applications in Prognostics and Health Management

et al. 2019

View full text Add to dashboard Cite

show abstract

“…HOF is equivalent to being stuck at a max-min deflection position, and float is equivalent to being stuck at 0 rad. Therefore, stuck, HOF, and float for an actuator fault can be regarded as the same action [28], referred to as a stuck condition. Then, the actuator fault model can be designed as:…”

Section: Actuator Fault Modelingmentioning

confidence: 99%

A Novel RFDI-FTC System for Thrust-Vectoring Aircraft Undergoing Control Surface Damage and Actuator Faults During Supermaneuverable Flight

Dong

2019

IEEE Access

Self Cite

View full text Add to dashboard Cite

In this paper, a novel robust fault detection and identification and fault-tolerant control (RFDI-FTC) system is proposed for thrust-vectoring aircraft (TVA) during supermaneuverable flight. To this end, a TVA model that incorporates control surface damage, actuator faults, disturbances, and aerodynamic parameter uncertainty is described, and a novel RFDI-FTC system is designed for the TVA model, which includes 1) an RFDI subsystem with robustness to disturbances and parameter uncertainty and sensitivity to control surface damage and actuator faults by multiple adaptive observers; 2) a command filter FTC subsystem to eliminate the differential expansion in traditional backstepping control and to compensate for control surface damage, actuator faults, disturbances and parameter uncertainty; and 3) a stability analysis for the RFDI-FTC system. The simulation results are given to demonstrate the effectiveness and potential of this system.

show abstract

“…In an actual flight control system, typical actuator faults include the following four types: (1) stuck, (2) a hard‐over failure (HOF), (3) float, and (4) a loss of effectiveness (LOE). The HOF is equivalent to locking at a max‐min deflection position, and the float is equivalent to locking at 0 rad.…”

Section: Problem Formulationmentioning

confidence: 99%

Robust Actuator‐Fault‐Tolerant Control System Based on Sliding‐Mode Observer for Thrust‐Vectoring Aircrafts

Dong

Xiong

2018

Asian Journal of Control

Self Cite

View full text Add to dashboard Cite

In this paper, a robust actuator-fault-tolerant control (FTC) system is proposed for thrust-vectoring aircraft (TVA) control. To this end, a TVA model with actuator fault dynamics, disturbances, and uncertain aerodynamic parameters is described, and a local fault detection and identification (FDI) mechanism is proposed to locate and identify faults, which utilizes an adaptive sliding-mode observer (SMO) to detect actuator faults and two SMOs to identify and estimate their parameters. Finally, a fault-tolerant controller is designed to compensate for these actuator faults, disturbances, and uncertain aerodynamic parameters; the approach combines back-stepping control with fault parameters and a high-order SMO. Furthermore, the stability of the entire control system is validated, and simulation results are given to demonstrate the effectiveness and potential for this robust FTC system.

show abstract

Deep auto-encoder observer multiple-model fast aircraft actuator fault diagnosis algorithm

Cited by 21 publications

References 15 publications

A Review on Deep Learning Applications in Prognostics and Health Management

A Review on Deep Learning Applications in Prognostics and Health Management

A Novel RFDI-FTC System for Thrust-Vectoring Aircraft Undergoing Control Surface Damage and Actuator Faults During Supermaneuverable Flight

Robust Actuator‐Fault‐Tolerant Control System Based on Sliding‐Mode Observer for Thrust‐Vectoring Aircrafts

Contact Info

Product

Resources

About