Article
Automation & Control Systems
Tangfan Xiahou, Zhiguo Zeng, Yu Liu
Summary: This article introduces a MoG-EHMM model that fuses expert knowledge and condition monitoring information for RUL prediction, demonstrating that the performance of RUL prediction can be substantially improved by incorporating expert knowledge with monitoring information.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Computer Science, Information Systems
Sajid Ali, Nabeel Ahmed, Ismail Shah, Showkat Ahmad Lone, Abdelaziz Alsubie
Summary: This study introduces deviation based exponentially weighted moving average control charts for monitoring censored data, considering the impact of shifts, smoothing weights, and censoring rates on different conditional measures. Both simulation and real data sets are used to illustrate the practical applications of the proposed charts.
Article
Chemistry, Analytical
Jorgen F. Pedersen, Rune Schlanbusch, Thomas J. J. Meyer, Leo W. Caspers, Vignesh V. Shanbhag
Summary: This study investigates using acoustic emission (AE) sensors to identify early stages of external leakage initiation in hydraulic cylinders through run to failure studies (RTF). The root mean square (RMS) feature is found to be a potent indicator for understanding leakage initiation.
Article
Engineering, Industrial
Zhenan Pang, Xiaosheng Si, Changhua Hu, Dangbo Du, Hong Pei
Summary: This article proposes a method for estimating the remaining useful life of degrading products by fusing accelerated degradation data and condition monitoring data. The method uses Bayesian inference and Markov Chain Monte Carlo (MCMC) to update posterior distributions of model parameters and approximate the RUL distribution, considering the randomness of model parameters. Validation using a practical case study of accelerometers shows that the method achieves higher RUL estimation accuracy and less uncertainty.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Automation & Control Systems
Shuai Zhao, Shaowei Chen, Fei Yang, Enes Ugur, Bilal Akin, Huai Wang
Summary: By integrating potential failure precursors and optimizing them based on a degradation model, formulating a composite failure precursor can enhance the accurate prediction of remaining useful life in power electronic systems.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Engineering, Electrical & Electronic
Jiri Sova, Petr Kolar, David Burian, Petr Vozabal
Summary: This paper proposes a method for predicting the remaining useful life (RUL) of machine tool spindle bearings using a combined calculation and experimental approach. The calculation model based on the ISO 281 standard uses real loading conditions and operation hours to calculate the spindle lifetime. The RUL is corrected using a bearing condition assessment based on vibration signal and measured data.
Article
Engineering, Industrial
Chao Wang, Tao Zhu, Bing Yang, Minxuan Yin, Shoune Xiao, Guangwu Yang
Summary: A general RUL framework for predicting crack propagation of mechanical system structures based on limited condition monitoring data is proposed. For invisible crack states, a method considering the common characteristics of the object is proposed, using the crack propagation stage as the delay time and combining statistical distribution and hypothesis testing. For visible crack states, methods based on a hypothetical distribution and support vector regression with the Kalman filter, considering the current state and individual degradation characteristics, are proposed. The framework achieves a competitive effect in RUL of internal crack propagation structures of railway wagons, which has important theoretical and practical value for integrity assessment and reliability life prediction.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Engineering, Multidisciplinary
Nabeel Ahmed, Sajid Ali, Ismail Shah
Summary: This article introduces control charts for monitoring censored data using conditional standard deviation (CSD) and compares it with existing censored data charts. The charts are assessed based on average run length and standard deviation of run length, showing that the CSD measure outperforms CEV and CM statistics.
QUALITY AND RELIABILITY ENGINEERING INTERNATIONAL
(2022)
Article
Engineering, Electrical & Electronic
Shuai Zhao, Yingzhou Peng, Fei Yang, Enes Ugur, Bilal Akin, Huai Wang
Summary: Condition monitoring of power devices is crucial for safety and mission-critical systems. Noise and aperiodic degradation measurements can negatively impact health assessment performance, but a proposed method in this article addresses these challenges by incorporating uncertainties and using a stochastic expectation-maximization algorithm for parameter estimation. Numerical analysis and testing on SiC MOSFETs validate the accuracy and robustness of the method.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Chemistry, Multidisciplinary
Jie Li, Yuejin Tan, Bingfeng Ge, Hua Zhao, Xin Lu
Summary: This paper proposes a method for predicting the remaining useful life of a concrete piston in a concrete pump truck using probability statistics and data-driven approaches. It determines the average useful life through probability distribution fitting and establishes regression models based on loading conditions to predict the RUL. Finally, it establishes a replacement warning point based on predictions at different life stages to support inventory management and replacement planning.
APPLIED SCIENCES-BASEL
(2021)
Article
Engineering, Marine
Liang Liu, David Y. Yang, Dan M. Frangopol
Summary: Reliable and continuous operation of ships is crucial, but extending ship service life may come with higher failure risk and additional maintenance costs. Therefore, a cost-benefit evaluation is necessary to determine the optimal duration of service life extension and associated maintenance actions.
Article
Energy & Fuels
Yu Lu, Sida Zhou, Xinan Zhou, Shichun Yang, Mingyan Liu, Xinhua Liu, Heping Ling, Yubo Lian
Summary: In this article, a multi-timescale capacity and lifespan prediction method is proposed. Capacity prediction is done using long short term memory neural network, while remaining useful life prediction is done using Weibull accelerated failure time regression. The proposed methods are validated using two datasets and have shown high accuracy.
JOURNAL OF ENERGY STORAGE
(2023)
Article
Mathematics
Gabriela M. M. Rodrigues, Edwin M. M. Ortega, Gauss M. M. Cordeiro, Roberto Vila
Summary: We propose a new quantile regression model based on a reparameterized exponentiated odd log-logistic Weibull distribution and investigate its structural properties. This model encompasses several known regression models that have wide applications. We employ the maximum likelihood method to estimate the parameters and conduct simulations to examine the finite sample properties of the maximum likelihood estimators. The applicability of the proposed regression model is validated using gastric carcinoma dataset.
Article
Multidisciplinary Sciences
Dongdong Zhao, Feng Liu
Summary: This paper proposes an adversarial-based domain adaptation architecture for estimating the remaining useful life (RU) of bearings. The architecture, called ADACNN, is trained on labeled source data and fine-tuned on similar unlabeled target data. It includes a feature extractor, source domain regressive predictor, and a domain classifier to find domain-invariant features. Experimental results demonstrate the satisfactory generalization performance and practical value of the ADACNN architecture in the industry.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Information Systems
Qibin Zhou, Xin Huang, Bengang Wei, Linlong Ye
Summary: This paper introduces an evaluation method for the impulse life of MOV based on the Weibull distribution and MRR method, and validates the effectiveness of this method by establishing a mathematical model and testing with real MOV samples.