Abstract
Due to the soft stiffness of high-rise buildings in the horizontal direction, strong wind will cause a strenuous structural response. Wind load is one key control load in the design of high-rise buildings. This study analyzes wind-induced fatigue of curtain wall supporting structure of a high-rise building in accordance with dynamic pressure measurement data of wind tunnel, acquiring wind pressure in each part of the structure. The finite element model is established for the curtain wall supporting structure, and the fatigue of corresponding nodes is discussed. Moreover, RBF (radial basis function) neural network regression is introduced to predict the fatigue life of unknown working conditions. Based on the joint distribution model of wind velocity and direction, this study explores the distribution law of fatigue life of supporting structure nodes, proposes a hypothesis of life distribution, and conducts a test. Moreover, working conditions with higher probability life are collected to provide a basis for practical engineering applications. The results show that the average deviation is below 10% by using RBF neural network and the probability life of the sample nodes is between 0 and 1016. Wind velocity is 8~15 m/s and azimuth angles of 50°~100°, 120°~200°, and 260°~300° are found in working conditions with low probability life; about 95% of the fatigue damage takes place in the first 30 conditions, and their fatigue damage values are between 3.5 × 10−3~9.36 × 10−2.
Funder
National Natural Science Foundation of China
Subject
Fluid Flow and Transfer Processes,Computer Science Applications,Process Chemistry and Technology,General Engineering,Instrumentation,General Materials Science
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献