Data-driven prognostic framework for remaining useful life prediction

Journal article

Asmaa Motrani, Rachid Noureddine
International Journal of Industrial and Systems Engineering, vol. 43, Inderscience Publishers (IEL), 2023, pp. 210--221

DOI: 10.1504/IJISE.2023.128666

Read

Cite

APA Click to copy
Motrani, A., & Noureddine, R. (2023). Data-driven prognostic framework for remaining useful life prediction. International Journal of Industrial and Systems Engineering, 43, 210–221. https://doi.org/10.1504/IJISE.2023.128666

Chicago/Turabian Click to copy
Motrani, Asmaa, and Rachid Noureddine. “Data-Driven Prognostic Framework for Remaining Useful Life Prediction.” International Journal of Industrial and Systems Engineering 43 (2023): 210–221.

MLA Click to copy
Motrani, Asmaa, and Rachid Noureddine. “Data-Driven Prognostic Framework for Remaining Useful Life Prediction.” International Journal of Industrial and Systems Engineering, vol. 43, Inderscience Publishers (IEL), 2023, pp. 210–21, doi:10.1504/IJISE.2023.128666.

BibTeX Click to copy

@article{motrani2023a,
  title = {Data-driven prognostic framework for remaining useful life prediction},
  year = {2023},
  journal = {International Journal of Industrial and Systems Engineering},
  pages = {210--221},
  publisher = {Inderscience Publishers (IEL)},
  volume = {43},
  doi = {10.1504/IJISE.2023.128666},
  author = {Motrani, Asmaa and Noureddine, Rachid}
}

Abstract

Industrial prognostic, based on data resulting from a monitoring up stream, is considered as a crucial stage in making complex industrial systems more reliable. The purpose of the industrial prognostic is to predict the future state of the monitored system, and to give, more specifically, an estimation of its remaining useful lifetime (RUL). Among the used approaches, data-driven prognostic is the most promising when dealing with multitude heterogeneous data. The aim of this work is to present a data-driven prognostic framework implementation, where the RUL is determined through the association of statistical and artificial intelligence methods. This framework is based on the relevance vector machine (RVM) technique to build the predictive degradation model in the offline part, and on the similarity-based interpolation (SBI) technique for the prediction of the remaining useful life in the online part. The different steps of the proposed framework are described and implemented through a case study.

Keywords

Prognostic and health management, PHM, data-driven prognostic, sparse Bayesian learning, SBL, relevance vector machine, RVM, sparse Bayesian interpolation, SBI, remaining useful life, RUL

Data-driven prognostic framework for remaining useful life prediction

Journal article

Cite

Abstract

Keywords

Follow this website