@inproceedings{41d57437b9ef4034a8836e54845908cb,
title = "Probability evaluation method of gas turbine work-scope based on survival analysis",
abstract = "Maintenance work-scope is the most important guidance when gas turbines visiting workshop. Maintenance cost and after-maintenance performance of gas turbine are affected by the work-scope directly. Thus, the evaluation method is essential in optimization of gas turbine maintenance work-scope. As after-maintenance performance of gas turbine would be influenced by multi-faced factors, it is not easy to predict the precise values of after-maintenance performance. In this paper, a probability evaluation method is proposed for maintenance work-scope assessment. In the proposed evaluation method, before-maintenance performance and repair levels of every module are taken into consideration, and survival analysis is adopted to estimate the reliability that the gas turbine would recover to the target performance state. The probabilistic method would provide a more practical way to evaluate gas turbine maintenance work-scope. Standardization and principal component analysis are adopted in data preprocessing. Finally, maintenance and performance data from a PW4077D gas turbine fleet are used to validate the proposed evaluation method. The experiment results show that the proposed method is much better than traditional method, and the proposed method could provide support for maintenance work-scope optimization.",
keywords = "gas turbine, maintenance, module, survival analysis, work-scope",
author = "Shisheng Zhong and Zhen Li and Lin Lin",
note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 8th IEEE Prognostics and System Health Management Conference, PHM-Harbin 2017 ; Conference date: 09-07-2017 Through 12-07-2017",
year = "2017",
month = oct,
day = "20",
doi = "10.1109/PHM.2017.8079225",
language = "英语",
series = "2017 Prognostics and System Health Management Conference, PHM-Harbin 2017 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
editor = "Bin Zhang and Yu Peng and Haitao Liao and Datong Liu and Shaojun Wang and Qiang Miao",
booktitle = "2017 Prognostics and System Health Management Conference, PHM-Harbin 2017 - Proceedings",
address = "美国",
}