Development of a deep learning-based approach for remaining useful life prediction of turbofan engine
In recent times, there has been a growing interest in predictive maintenance for turbofan engines as it has demonstrated significant potential in improving the maintenance process. The core concept behind predictive maintenance is to predict the remaining useful life (RUL) of a system. With accurate...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/166934 |
| _version_ | 1811681478865584128 |
|---|---|
| author | Ng, Yi Hong |
| author2 | Chen Chun-Hsien |
| author_facet | Chen Chun-Hsien Ng, Yi Hong |
| author_sort | Ng, Yi Hong |
| collection | NTU |
| description | In recent times, there has been a growing interest in predictive maintenance for turbofan engines as it has demonstrated significant potential in improving the maintenance process. The core concept behind predictive maintenance is to predict the remaining useful life (RUL) of a system. With accurate RUL predictions, the maintenance process can be optimised in terms of the maintenance frequency and resources consumed. This leads to considerable cost-savings without compromising the reliability and safety of the system.
Data-driven methods, especially deep learning (DL) models, contribute significantly to the popularity of predictive maintenance as they have achieved unprecedented prediction accuracy. In particular, Graph Neural Network (GNN) has gained rapid interest recently due to its impressive predictive performance in complex applications. Hence, this study aims to develop a novel state-of-the-art DL framework that uses a GNN-based algorithm to predict the RUL of turbofan engines.
The proposed framework includes data processing techniques needed to convert the raw multivariate time series data into compatible inputs for the Long Short-Term Memory (LSTM)-based and GNN-based model. Survival analysis models are also used to extract useful features during data processing. These processed inputs are fed into a novel DL model which is based on LSTM, GNN, attention mechanism, and CatBoost. This model adopts a two-step training strategy where the DL components are trained first, followed by the CatBoost model. In an end-to-end fashion, the DL components are trained using an asymmetric loss function (i.e., LINEX). Subsequently, the CatBoost model replaces the last two fully-connected layers and trains on the features extracted by the DL components.
The repeated experiments have shown that the proposed framework has outperformed many state-of-the-art DL models. The ablation study showed that the key components of the proposed model, including the GNN-based component, are significant contributors to the model performance. Thus, the proposed model serves as a strong benchmark for future state-of-the-art graph-based DL models. |
| first_indexed | 2024-10-01T03:41:35Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/166934 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:41:35Z |
| publishDate | 2023 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1669342023-06-03T16:50:13Z Development of a deep learning-based approach for remaining useful life prediction of turbofan engine Ng, Yi Hong Chen Chun-Hsien School of Mechanical and Aerospace Engineering mchchen@ntu.edu.sg Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Aeronautical engineering::Aircraft motors and engines In recent times, there has been a growing interest in predictive maintenance for turbofan engines as it has demonstrated significant potential in improving the maintenance process. The core concept behind predictive maintenance is to predict the remaining useful life (RUL) of a system. With accurate RUL predictions, the maintenance process can be optimised in terms of the maintenance frequency and resources consumed. This leads to considerable cost-savings without compromising the reliability and safety of the system. Data-driven methods, especially deep learning (DL) models, contribute significantly to the popularity of predictive maintenance as they have achieved unprecedented prediction accuracy. In particular, Graph Neural Network (GNN) has gained rapid interest recently due to its impressive predictive performance in complex applications. Hence, this study aims to develop a novel state-of-the-art DL framework that uses a GNN-based algorithm to predict the RUL of turbofan engines. The proposed framework includes data processing techniques needed to convert the raw multivariate time series data into compatible inputs for the Long Short-Term Memory (LSTM)-based and GNN-based model. Survival analysis models are also used to extract useful features during data processing. These processed inputs are fed into a novel DL model which is based on LSTM, GNN, attention mechanism, and CatBoost. This model adopts a two-step training strategy where the DL components are trained first, followed by the CatBoost model. In an end-to-end fashion, the DL components are trained using an asymmetric loss function (i.e., LINEX). Subsequently, the CatBoost model replaces the last two fully-connected layers and trains on the features extracted by the DL components. The repeated experiments have shown that the proposed framework has outperformed many state-of-the-art DL models. The ablation study showed that the key components of the proposed model, including the GNN-based component, are significant contributors to the model performance. Thus, the proposed model serves as a strong benchmark for future state-of-the-art graph-based DL models. Bachelor of Engineering (Aerospace Engineering) 2023-05-28T12:32:05Z 2023-05-28T12:32:05Z 2023 Final Year Project (FYP) Ng, Y. H. (2023). Development of a deep learning-based approach for remaining useful life prediction of turbofan engine. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/166934 https://hdl.handle.net/10356/166934 en C038 application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Aeronautical engineering::Aircraft motors and engines Ng, Yi Hong Development of a deep learning-based approach for remaining useful life prediction of turbofan engine |
| title | Development of a deep learning-based approach for remaining useful life prediction of turbofan engine |
| title_full | Development of a deep learning-based approach for remaining useful life prediction of turbofan engine |
| title_fullStr | Development of a deep learning-based approach for remaining useful life prediction of turbofan engine |
| title_full_unstemmed | Development of a deep learning-based approach for remaining useful life prediction of turbofan engine |
| title_short | Development of a deep learning-based approach for remaining useful life prediction of turbofan engine |
| title_sort | development of a deep learning based approach for remaining useful life prediction of turbofan engine |
| topic | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Aeronautical engineering::Aircraft motors and engines |
| url | https://hdl.handle.net/10356/166934 |
| work_keys_str_mv | AT ngyihong developmentofadeeplearningbasedapproachforremainingusefullifepredictionofturbofanengine |