Article
Engineering, Industrial
Song Fu, Yongjian Zhang, Lin Lin, Minghang Zhao, Shi-sheng Zhong
Summary: The study investigates an effective UDA method called DIDRLSTM, which uses DRLSTM as a feature extractor and integrates domain adaptation and domain confusion modules to improve prognostic performance in RUL prediction.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Engineering, Industrial
Juseong Lee, Mihaela Mitici
Summary: This study proposes a framework that integrates data-driven probabilistic Remaining-Useful-Life (RUL) prognostics with predictive maintenance planning, using aircraft turbofan engines as an example. By employing this framework, the total maintenance cost can be reduced, unscheduled maintenance can be prevented, and the wasted life of engines can be limited.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Engineering, Industrial
Ingeborg de Pater, Arthur Reijns, Mihaela Mitici
Summary: This paper proposes a dynamic, predictive maintenance scheduling framework that takes into account imperfect Remaining Useful Life (RUL) prognostics. Maintenance tasks are scheduled based on periodically updated RUL prognostics and alarms triggered by their evolution over time. A safety factor is used to avoid component failures in the presence of potential errors in the RUL prognostics.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Article
Engineering, Mechanical
Hongqiu Zhu, Ziyi Huang, Biliang Lu, Can Zhou
Summary: This article proposes a method based on dynamic feature construction to predict the remaining useful life of bearings with fatigue failure, using improved convolutional neural networks and long short-term memory networks to achieve prediction.
INTERNATIONAL JOURNAL OF FATIGUE
(2022)
Article
Engineering, Industrial
Yuanfu Li, Yao Chen, Zhenchao Hu, Huisheng Zhang
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
Chemistry, Analytical
Jinsong Yang, Yizhen Peng, Jingsong Xie, Pengxi Wang
Summary: This paper proposes a new method for predicting the remaining useful life (RUL) of rolling bearings based on LSTM and uncertainty quantification. By introducing a fusion metric and an improved dropout method, the estimation accuracy of RUL is enhanced. Verification results demonstrate that the proposed model can accurately predict the point estimation and probability distribution of bearing RUL.
Article
Physics, Multidisciplinary
Cheng Peng, Jiaqi Wu, Qilong Wang, Weihua Gui, Zhaohui Tang
Summary: A dual-channel LSTM neural network model is proposed in this research, which can adaptively select and process time features to improve the accuracy of machinery RUL prediction. Experimental verification shows the effectiveness and stability of this method.
Article
Computer Science, Interdisciplinary Applications
Bernar Tasci, Ammar Omar, Serkan Ayvaz
Summary: This study proposes a machine learning-based predictive maintenance approach to predict the Remaining Useful Life of production lines in manufacturing. By using data from integrated IoT sensors in a real-world factory, the approach aims to predict potential equipment failures on assembly lines in real-time and prevent downtime, resulting in resource savings.
COMPUTERS & INDUSTRIAL ENGINEERING
(2023)
Article
Engineering, Electrical & Electronic
Xinyao Li, Jingjing Li, Lin Zuo, Lei Zhu, Heng Tao Shen
Summary: This article introduces the application of domain adaptation in remaining useful life (RUL) prediction and proposes a novel method that improves model performance by aligning distributions at both the feature level and the semantic level. In addition, the article introduces the use of transformer as the backbone to capture long-term dependencies more efficiently, enhancing the robustness of the model.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Multidisciplinary Sciences
Chenyang Wang, Wanlu Jiang, Yi Yue, Shuqing Zhang
Summary: This paper presents a scheme for predicting the remaining useful life (RUL) of a hydraulic pump (gear pump) using a combination of a deep convolutional autoencoder (DCAE) and a bidirectional long short-term memory (Bi-LSTM) network. Experimental results show that the proposed method has good predictive performance.
Article
Engineering, Electrical & Electronic
Ji-Yan Wu, Min Wu, Zhenghua Chen, Xiao-Li Li, Ruqiang Yan
Summary: The article introduces an algorithm named DELTA, which leverages degradation-aware long short-term memory (LSTM) autoencoder (AE) to enhance the accuracy of RUL prediction. This algorithm dynamically models the degradation factor and explores latent variables to improve RUL prediction accuracy, achieving significant improvements in performance on the FEMTO bearing data set compared to existing algorithms.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Thermodynamics
Zewang Chen, Na Shi, Yufan Ji, Mu Niu, Youren Wang
Summary: This paper proposes a hybrid algorithm combining BLS with RVM for predicting the remaining useful life of lithium-ion batteries. Experimental results show higher prediction accuracy and stronger long-term prediction capabilities for the proposed algorithm.
Article
Mathematics
Xiaojia Wang, Ting Huang, Keyu Zhu, Xibin Zhao
Summary: With the increasing demands on the reliability of industrial equipment due to the transformation of industrial production into intelligent production, accurate prediction of remaining useful life (RUL) plays a pivotal role in intelligent maintenance. To overcome the problems of inadequate feature extraction and poor correlation between prediction results and data, researchers constructed a new fusion model called B-LSTM, which extracts data features based on a broad learning system (BLS) and embeds long short-term memory (LSTM) to process time-series information. Experimental results showed significant improvements compared to several mainstream methods on the C-MAPSS dataset.
Article
Engineering, Mechanical
Eyyup Akcan, Yilmaz Kaya
Summary: This study proposes a novel approach for the remaining useful life (RUL) prediction of bearings. The 1D-TP method is applied to vibration signals and combined with LSTM for accurate RUL assessment. The results demonstrate the successful prediction of bearing life using the 1D-TP + LSTM method.
JOURNAL OF THE BRAZILIAN SOCIETY OF MECHANICAL SCIENCES AND ENGINEERING
(2023)
Article
Engineering, Electrical & Electronic
Yifei Ding, Minping Jia, Yudong Cao
Summary: The article proposes a deep subdomain adaptive regression network (DSARN) for aligning relevant subdomains in source and target domains and demonstrates its effectiveness in the field of prognostics and health management (PHM) of bearings. The DSARN method shows superior performance compared to other state-of-the-art deep learning (DL) and transfer learning (TL) methods.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Automation & Control Systems
Chen Chuang, Lu Ningyun, Jiang Bin, Xing Yin
JOURNAL OF SYSTEMS ENGINEERING AND ELECTRONICS
(2020)
Article
Engineering, Multidisciplinary
Chuang Chen, Cunsong Wang, Ningyun Lu, Bin Jiang, Yin Xing
Summary: This paper presents a novel data-driven predictive maintenance strategy that achieves accurate failure prognosis through degradation feature selection and degradation prognostic modeling modules, outperforming traditional maintenance strategies.
EKSPLOATACJA I NIEZAWODNOSC-MAINTENANCE AND RELIABILITY
(2021)
Article
Computer Science, Interdisciplinary Applications
Chuang Chen, Ningyun Lu, Le Wang, Yin Xing
Summary: An intelligent selection and optimization method of feature variables is proposed to suppress the research octane number (RON) loss in the gasoline refining process. By calculating the importance of main variables and establishing a nonlinear mapping relationship, the optimal values of feature variables are obtained through continuous iterative solution.
COMPUTERS & CHEMICAL ENGINEERING
(2021)
Article
Engineering, Electrical & Electronic
Chuang Chen, Zheng Hong Zhu, Jiantao Shi, Ningyun Lu, Bin Jiang
Summary: This paper presents a dynamic predictive maintenance strategy for modern engineering systems using a deep learning ensemble model. The model, consisting of a deep autoencoder and bidirectional long short-term memory, accurately estimates system health state and remaining useful life. The effectiveness of the proposed strategy is demonstrated by comparing it with recent publications using a dataset from NASA.
IEEE SENSORS JOURNAL
(2021)
Article
Engineering, Electrical & Electronic
Chuang Chen, Jiantao Shi, Mouquan Shen, Lihang Feng, Guanye Tao
Summary: This study combines system failure prediction with maintenance decision-making to develop a predictive maintenance strategy. System failure prediction is achieved by constructing an ensemble model (DAE-LSTMQR-KDE). Replacement cost and ordering cost functions are proposed based on the probability density of system failure time obtained from the ensemble model to support maintenance and inventory decisions. Optimal decisions are determined by minimizing these cost functions. Experimental results show the superiority of the proposed prediction and maintenance method compared to state-of-the-art methods. Different cost structure scenarios are also investigated to demonstrate the flexibility of maintenance decisions based on failure prediction information.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Engineering, Electrical & Electronic
Chuang Chen, Jiantao Shi, Mouquan Shen, Ningyun Lu, Hui Yu, Yukun Chen, Cunsong Wang
Summary: This article proposes a novel deep belief network (DBN) learning method for fault diagnosis of transmitter/receiver (T/R) module. By constructing a sparse DBN based on Gaussian function to learn the relationship between monitoring data and component health conditions, pseudo-labels are generated for unlabeled samples and information entropy is used to reduce pseudo-label noise. The proposed method achieves a mean identification rate of 96.33%, surpassing some DBN-based modeling methods and other intelligent methods.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Engineering, Civil
Lihang Feng, Sui Wang, Jiantao Shi, Pengwen Xiong, Chuang Chen, Di Xiao, Aiguo Song, Peter Xiaoping Liu
Summary: This paper proposes a novel decoupling and calibration method to increase the accuracy of Wheel Force Transducers/Sensors (WFTs). By developing a physical interpretable prime-error framework and utilizing nonlinear error modeling, the conventional decoupling method is improved. Experimental results show that the proposed method achieves good performance in terms of accuracy and computational efficiency.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Chuang Chen, Ningyun Lu, Bin Jiang, Yin Xing, Zheng Hong Zhu
Summary: A novel prediction interval (PI) estimation method is proposed to quantify uncertainties in RUL prediction, combining data clustering, mathematical-statistical analysis, and deep learning techniques in offline and online phases. Experimental results show that the proposed method is a promising tool for providing reliable aeroengine RUL interval estimates.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)