A practical reliability design method considering the compound weight and load-sharing

Li, Y.; Coolen, F.P.A.; Zhu, C.

Abstract

Reliability design is an important work in the early design stage of offshore wind turbines. Due to the incomplete considerations and poor feasibility of the drawbacks for existing methods, a set of the practical reliability design method is proposed in this paper. The time characteristics and many influential factors of units are considered in the design process. The influential factors of the system’s units are scored by several experts with extensive engineering experience. Based on this, the reliability allocation and the maintainability prediction of the repairable system are performed using different methods. To realistically evaluate the reliability level of each unit obtained by three different methods, a fuzzy reliability evaluation method is developed to rank the reliability level of each unit with considerations of the mean time between failure (MTBF), mean time to repair (MTTR), failure frequency and availability using the compound weight and fuzzy membership function. Following this, redundant design is used to eliminate weaknesses to keep the system reliability at a high level. Using the reliability data obtained above, a time-dependent reliability model of the system considering load sharing is built to explore the influences of reliability allocation on the system reliability in the 20-year service life. The effectiveness and feasibility of the proposed approaches are demonstrated with a 5MW offshore wind turbine.

Citation

Li, Y., Coolen, F., & Zhu, C. (2020). A practical reliability design method considering the compound weight and load-sharing. International Journal of Approximate Reasoning: Uncertainty in Intelligent Systems, 127, 17-32. https://doi.org/10.1016/j.ijar.2020.09.001

• Journal Article Type: Article
• Acceptance Date: Sep 23, 2020
• Online Publication Date: Oct 1, 2020
• Publication Date: 2020-12
• Deposit Date: Sep 29, 2020
• Publicly Available Date: Oct 1, 2021
• Journal: International Journal of Approximate Reasoning
• Print ISSN: 0888-613X
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 127
• Pages: 17-32
• DOI: https://doi.org/10.1016/j.ijar.2020.09.001
• Public URL: https://durham-repository.worktribe.com/output/1261526