Finite Element Model Fractional Steps Updating Strategy for Spatial Lattice Structures Based on Generalized Regression Neural Network

Caiwei Liu,Jijun Miao,Changyong Zhao
DOI: https://doi.org/10.1155/2016/5845326
IF: 1.616
2016-01-01
Shock and Vibration
Abstract:In order to get a more accurate finite element model of a spatial lattice structure with bolt-ball joints for health monitoring, a method of modifying the bolt-ball joint stiffness coefficient was proposed. Firstly, the beam element with adjustable stiffness was used in the joint zone in this paper to reveal the semirigid characteristic of the joint. Secondly, the value of stiffness reduction factor ( a r ) was limited in the range of [ 0.2,0.8 ] and the reference value ( a r 0 ) of it was suggested to be 0.5 based on referenced literatures. Finally, the finite element model fractional steps updating strategy based on neural network technique was applied and the limited measuring point information was used to form the network input parameter. A single-layer latticed cylindrical shell model with 157 joints and 414 tubes was used in a shaking TABLE test. Based on the measured modal data, the presented method was verified. The results show that this model updating technique can reflect the true dynamic characters of the shell structure better. Moreover, the neural network can be simplified considerably by using this algorithm. The method can be used for model updating of a latticed shell with bolt-ball joints and has great value in engineering practice.
engineering, mechanical,mechanics,acoustics
What problem does this paper attempt to address?