A Comprehensive Survey of Prognostics and Health Management Based on Deep Learning for Autonomous Ships

Ellefsen, André Listou; Asoy, Vilmar; Ushakov, Sergey; Zhang, Houxiang

doi:10.1109/tr.2019.2907402

Cited by 71 publications

(29 citation statements)

References 116 publications

(210 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The increased number of successful examples of applications of deep learning in manufacturing, automotive, and aerospace industry has shown that it is a viable tool for condition-based maintenance [5][6][7][8][9]. Current deep learning research for these application areas, however, mostly focuses on changing the model architectures to improve remaining useful life (RUL) or fault prediction accuracy.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning with Dynamically Weighted Loss Function for Sensor-Based Prognostics and Health Management

Rengasamy

Jafari

Rothwell

et al. 2020

Sensors

View full text Add to dashboard Cite

Deep learning has been employed to prognostic and health management of automotive and aerospace with promising results. Literature in this area has revealed that most contributions regarding deep learning is largely focused on the model’s architecture. However, contributions regarding improvement of different aspects in deep learning, such as custom loss function for prognostic and health management are scarce. There is therefore an opportunity to improve upon the effectiveness of deep learning for the system’s prognostics and diagnostics without modifying the models’ architecture. To address this gap, the use of two different dynamically weighted loss functions, a newly proposed weighting mechanism and a focal loss function for prognostics and diagnostics task are investigated. A dynamically weighted loss function is expected to modify the learning process by augmenting the loss function with a weight value corresponding to the learning error of each data instance. The objective is to force deep learning models to focus on those instances where larger learning errors occur in order to improve their performance. The two loss functions used are evaluated using four popular deep learning architectures, namely, deep feedforward neural network, one-dimensional convolutional neural network, bidirectional gated recurrent unit and bidirectional long short-term memory on the commercial modular aero-propulsion system simulation data from NASA and air pressure system failure data for Scania trucks. Experimental results show that dynamically-weighted loss functions helps us achieve significant improvement for remaining useful life prediction and fault detection rate over non-weighted loss function predictions.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Learning with Dynamically Weighted Loss Function for Sensor-Based Prognostics and Health Management

Rengasamy

Jafari

Rothwell

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Hence, in a maintenance perspective, a prognostics and health management (PHM) system, which both include automatic fault detection and associated remaining useful life (RUL) predictions, is crucial in autonomous operations. When the RUL is predicted, the maintenance operation can be scheduled to the next appropriate port of call for the ferry [6]. Nevertheless, the RUL prediction is the available time prior to operational failure after a fault is detected within the engine.…”

Section: Introductionmentioning

confidence: 99%

Automatic Fault Detection for Marine Diesel Engine Degradation in Autonomous Ferry Crossing Operation

Ellefsen

Cheng

Holmeset

et al. 2019

2019 IEEE International Conference on Mechatronics and Automation (ICMA)

Self Cite

View full text Add to dashboard Cite

The maritime industry generally anticipates having semi-autonomous ferries in commercial use on the west coast of Norway by the end of this decade. In order to schedule maintenance operations of critical components in a secure and costeffective manner, a reliable prognostics and health management system is essential during autonomous operations. Any remaining useful life prediction obtained from such system should depend on an automatic fault detection algorithm. In this study, an unsupervised reconstruction-based fault detection algorithm is used to predict faults automatically in a simulated autonomous ferry crossing operation. The benefits of the algorithm are confirmed on data sets of real-operational data from a marine diesel engine collected from a hybrid power lab. During the ferry crossing operation, the engine is subjected to drastic changes in operational loads. This increases the difficulty of the algorithm to detect faults with high accuracy. Thus, to support the algorithm, three different feature selection processes on the input data is compared. The results suggest that the algorithm achieves the highest prediction accuracy when the input data is subjected to feature selection based on sensitivity analysis.

show abstract

“…In the ideal case, the crew members will perform duties other than maintenance, operation, and navigation. Thus, Thus, as the ferries have few or no maintenance personnel on board, a PHM system including automatic fault detection and related residual life prediction (RUL) is crucial in order to schedule maintenance operations to the next appropriate port of call [4]. The PHM system monitors and detects potential faults and predicts future operational trends, aiming to achieve the ideal maintenance policy for the vessel.…”

Section: Introductionmentioning

confidence: 99%

A Step-wise Feature Selection Scheme for a Prognostics and Health Management System in Autonomous Ferry Crossing Operation

Cheng

Ellefsen

et al. 2019

2019 IEEE International Conference on Mechatronics and Automation (ICMA)

Self Cite

View full text Add to dashboard Cite

Developing a reliable algorithm to detect faults automatically within critical components in autonomous ferries is essential for safe and cost-beneficial maritime operations. Autonomous ferries are equipped with hundreds of sensors. Thus, in order to support the algorithm, the input data should be subjected to a feature selection process. This paper introduces a novel step-wise feature selection scheme for prognostics and health management (PHM) system in autonomous ferries. The scheme mainly consists of two steps. The first step is the Pearson correlation analysis to reduce the redundant information among sensors. In order to study the importance of the selected features obtained by correlation analysis and removal of irrelevant features, the second step is sensitivity analysis (SA) based feature selection. The proposed scheme is evaluated on realoperational marine diesel engine data. In the experiments, both fault classification and fault detection demonstrate the feasibility of the proposed approach.

show abstract

A Comprehensive Survey of Prognostics and Health Management Based on Deep Learning for Autonomous Ships

Cited by 71 publications

References 116 publications

Deep Learning with Dynamically Weighted Loss Function for Sensor-Based Prognostics and Health Management

Deep Learning with Dynamically Weighted Loss Function for Sensor-Based Prognostics and Health Management

Automatic Fault Detection for Marine Diesel Engine Degradation in Autonomous Ferry Crossing Operation

A Step-wise Feature Selection Scheme for a Prognostics and Health Management System in Autonomous Ferry Crossing Operation

Contact Info

Product

Resources

About