相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Engineering, Industrial
Yuanfu Li et al.
Summary: This paper proposes a novel approach that combines knowledge and deep learning models for the remaining useful life (RUL) prediction. By representing the sensor relationships as flow charts and transforming them into embedding vectors for clustering, the proposed approach guides the arrangement of sensor data and the construction of hybrid deep learning models. The robustness and reliability of the approach are demonstrated on the NASA open dataset C-MAPSS, showing improved prediction accuracy compared to existing methods. The interpretable deep learning model constructed using knowledge highlights the feasibility and reliability of fusing knowledge and deep learning models.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Computer Science, Hardware & Architecture
Ikram Remadna et al.
Summary: This article proposes a new hybrid deep architecture for predicting machine failure time, which has better data analysis and dimensionality reduction capability, improving feature space distribution and interpretability. The performance of the model has been tested and it outperforms existing models.
IEEE TRANSACTIONS ON RELIABILITY
(2023)
Article
Engineering, Industrial
Yupeng Wei et al.
Summary: Prediction of state-of-health and remaining useful life is crucial for lithium-ion batteries. Existing methods fail to reveal the correlation among features and lack the ability to use the most relevant part of time-series data. To address these issues, a two-stage optimization model is introduced to construct an undirected graph with optimal graph entropy. Graph convolutional networks with attention mechanisms are developed based on the graph to accurately predict the state-of-health and remaining useful life of batteries. The proposed method outperforms existing data-driven methods and achieves a minimum root-mean-squared-error of 0.0104 and 5.80 for state-of-health and remaining useful life prediction, respectively.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Engineering, Industrial
Sayaka Kamei et al.
Summary: The current prognostics approaches for a network of assets are centralized and reliant on the availability of assets' sensors, failures, and anomaly data. One solution is decentralized Federated Learning (FL). The current paper aims to compare the performance of a centralized model with two decentralized FL algorithms to predict the remaining useful life (RUL) of an asset. The results show that FedProx performed better than FedAvg generally, and Transformer architecture performed better overall than LSTM across all datasets in the centralized and decentralized scenarios.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Automation & Control Systems
Yu Huang et al.
Summary: This paper proposes a novel sequence-to-sequence predictive model based on a variational autoencoder for predicting future performance progression and remaining useful life of systems. The model, trained with generative adversarial networks, is capable of handling uncertainties and providing probabilistic predictions. Validation using real-world health monitoring data shows significant performance improvement in long-term degradation progress and RUL prediction tasks.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Engineering, Industrial
Danyang Xu et al.
Summary: This paper proposes a novel data-driven method called Dual-Stream Self-Attention Neural Network (DS-SANN) for RUL estimation. By employing the multi-head self-attention mechanism and dual-stream structure network, DS-SANN can better capture the correlations and internal differences of monitoring data, leading to improved estimation performance for RUL.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Article
Engineering, Industrial
Jiusi Zhang et al.
Summary: This paper introduces a novel bidirectional GRU model with temporal self-attention mechanism for predicting remaining useful life (RUL). Experimental results demonstrate its superiority over existing machine learning and deep learning methods.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Article
Engineering, Industrial
Tongyang Pan et al.
Summary: This paper proposes a multi-head attention network coupled with adaptive meta-transfer learning for accurate prediction of the remaining useful life (RUL) of cryogenic bearings in rocket engines under the steady stage. The proposed method is compared with typical benchmark algorithms using real monitoring data, and the results indicate better performance in multiple evaluation indexes.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Article
Engineering, Industrial
Linchuan Fan et al.
Summary: This paper introduces a new method for equipment health status monitoring that utilizes deep learning and attention mechanisms to accurately select and utilize useful signals, thereby improving prediction performance. The author conducted a series of experiments to demonstrate the effectiveness and advanced performance of the method and proposed an interpretability analysis method.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Article
Automation & Control Systems
Zhenghua Chen et al.
Summary: This article proposes an attention-based deep learning framework for the prediction of machine's remaining useful life (RUL). By integrating handcrafted features with automatically learned features and developing a feature fusion framework, the performance of RUL prediction can be improved.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2021)
Article
Engineering, Industrial
Junqiang Liu et al.
Summary: The study introduces an improved multi-stage Long Short Term Memory network with clustering (ILSTMC) for predicting the Remaining Useful Life (RUL) of an aero-engine. Experimental results demonstrate that this method effectively reduces prediction errors in comparison to traditional models.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Engineering, Industrial
Yudong Cao et al.
Summary: The study introduces a new deep learning framework TCN-RSA for predicting the remaining useful life of mechanical systems, and demonstrates its superiority in experiments.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Review
Engineering, Mechanical
Sung Wook Kim et al.
Summary: Despite the rapid development of AI, limitations such as lack of robustness and interpretability have hindered its widespread adoption. To overcome these limitations, new branches of deep learning such as physics-informed neural networks have emerged, providing new opportunities for advancement in the field.
JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY
(2021)
Article
Engineering, Mechanical
Jun Xia et al.
Summary: This paper proposes a new method based on LSTM and MLSA mechanism to improve the accuracy and computational efficiency of remaining useful life (RUL) estimation in mechanical systems. By designing multilayer MLSA mechanism and LSTM to extract and process degradation data features, the method is validated to have high computational efficiency, accuracy, and robustness in RUL estimation for aeroengines compared to other methods.
ENGINEERING FAILURE ANALYSIS
(2021)
Article
Automation & Control Systems
Yi Qin et al.
Summary: A new neural network, GDAU, was proposed to predict the RUL of rolling bearings, showing higher accuracy and convergence speed compared to conventional methods.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Computer Science, Artificial Intelligence
Gavneet Singh Chadha et al.
Summary: This paper presents a novel approach for multivariate time series data analysis using deep convolutional neural networks and a generalized dilation technique to extract arbitrary patterns in the input feature space. Experimental results show superior prediction capability of the proposed method for remaining useful lifetime (RUL) estimation problems, setting a new state of the art compared to previous results in literature.
Article
Computer Science, Artificial Intelligence
Mohamed Ragab et al.
Summary: This study discusses the accurate estimation of remaining useful life of industrial equipment and proposes a new attention-based sequence to sequence model to address the limitations of existing deep learning methods. By optimizing two loss functions and using attention mechanism to handle long sequences, experimental results show superior performance of the proposed method on real datasets.
Article
Computer Science, Information Systems
Yan Song et al.
Summary: Collecting massive industrial data from IIoT assets improves data-driven methods for prognostics and health management systems. However, current approaches for bearing RUL prediction do not effectively weigh the contributions of data from different sensors and time steps, reducing efficiency in the big data era. A proposed deep learning-based method with an attention mechanism shows high accuracy and efficiency in practical applications.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Computer Science, Artificial Intelligence
Wei Ming Tan et al.
Summary: This paper proposes a Multi-variable Time Series (MTS) focused approach to prognostics, utilizing a lightweight Convolutional Neural Network (CNN) to enhance accuracy and training speed for the estimation of Remaining Useful Life (RUL). The deployment of the network on a lightweight hardware platform demonstrates compactness and efficiency in resource-restricted environments.
Article
Automation & Control Systems
Jun Wu et al.
Article
Computer Science, Information Systems
Hao Zhang et al.
Article
Engineering, Industrial
Chen Jinglong et al.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2019)
Article
Computer Science, Artificial Intelligence
Shun-Yao Shih et al.
Article
Computer Science, Artificial Intelligence
Kaixiang Peng et al.
Article
Automation & Control Systems
Yaping Deng et al.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2019)
Article
Computer Science, Information Systems
Jialin Li et al.
Article
Computer Science, Artificial Intelligence
Yuting Wu et al.
Article
Engineering, Industrial
Xiang Li et al.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2018)
Article
Computer Science, Information Systems
Ran Zhang et al.
Article
Engineering, Mechanical
Runqing Huang et al.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2007)
