演习
疲劳极限
极限(数学)
组分(热力学)
结构工程
极限状态设计
工程类
材料科学
汽车工程
机械工程
数学
物理
数学分析
热力学
作者
Felix Weber,Vitali Züch,Volker P. Schulz,Dennis Bosse,Georg Jacobs,Christoph Broeckmann
标识
DOI:10.1007/s10010-025-00779-1
摘要
Abstract With modern wind turbines facing continuous upscaling in terms of size and power, challenges in manufacturing, logistics, assembly, and total cost arise. To account for these challenges, an increase in the power density, e.g., of the drive train, is required. While modifications of the component geometry allow for potential optimization, significant potential is currently locked up by safety factors accounting for uncertainty in local material properties. To unlock this potential, the uncertainty of local material properties, especially the local fatigue strength, needs to be reduced. This study presents an experimental approach to analyze and quantify the local fatigue strength of a cast component by testing specimens manufactured from hollow drill samples taken from the cast component. Therefore, the change in temperature due to internal dissipation effects is observed during fatigue testing of the specimen in a load increase test. Analysis of the stress dependent change in temperature results in virtual test points that can be evaluated to derive local fatigue strengths. The temperature change within the specimen is determined by low-cost Negative Temperature Coefficient (NTC) thermistors placed on the surface of the specimen. The potential of the method is demonstrated for cast iron grade EN-GJS-500-14. The testing of different casts of EN-GJS-500-14 demonstrates the robustness of uncertainty estimation of the proposed method. Especially for heavy-section castings, this procedure allows for the determination of a component specific local material fatigue strength without an effect on the overall component integrity. Implementing this testing procedure as a fingerprint of a cast component could reduce the uncertainty in the local material properties and consequently reduce the applied safety factors.
科研通智能强力驱动
Strongly Powered by AbleSci AI