A unified fatigue reliability-based design optimization framework for aircraft turbine disk. (November 2021)

• Record Type: Journal Article
• Authors: Song, Lu-Kai; Bai, Guang-Chen; Li, Xue-Qin; Wen, Jie
• Is Part Of: International journal of fatigue. Volume 152(2021)
• Publication Date: 2021-11
• Subjects: Reliability-based design optimization -- Low cycle fatigue -- Surrogate model -- Artificial neural network; Materials -- Fatigue -- Periodicals; Materials -- Fatigue; Periodicals; 620.1122
• DOI: 10.1016/j.ijfatigue.2021.106422
• Languages: English
• ISSNs: 0142-1123

Abstract

Highlights: Propose a hierarchical fuzzy-neuro (HFN) surrogate method to improve simulation efficiency. Develop a multi-level collaboration (MLC) model for fatigue reliability-based optimization. A unified fatigue reliability-based design optimization framework is built with HFN and MLC. Complete the low cycle fatigue reliability optimal design of aircraft turbine disk. The proposed HFN and MLC are verified to hold high precision and efficiency. Abstract: To improve the modeling efficiency and optimization accuracy for fatigue reliability-based design optimization (FRBDO) of aircraft turbine disk, a unified framework integrating by hierarchical fuzzy-neuro (HFN) surrogate method and multi-level collaboration (MLC) optimization model are presented. The HFN method is developed with absorbing fuzzy-neuro surrogate into hierarchical modeling strategy. The MLC model is proposed by considering influencing factors and constraint conditions in multiple layers and multiple cycles. The presented framework was applied to the FRBDO of a high-pressure turbine disk. The optimization results show that the presented framework holds high computational efficiency and accuracy in FRBDO of turbine disk.