基于变权累加生成GM(1,1)模型预测应力集中系数的方法研究

On the Prediction Methods of Stress Concentration Coefficient Based on Variable Weighting Accumulated Generating GM(1,1) Model

应力集中是工程实际中影响构件破坏的主要原因之一,由于种种限制,不能直接测得应力集中处的最大应变值或应力值。本研究将灰色系统理论中的GM(1,1)模型应用于应力集中问题的数据预测处理,探讨粒子算法的变权累加生成GM(1,1)模型求出应力集中处的最大应变值,并根据实际预测结果,以带小孔的拉板为例,对采样点的数据建立模型,实例表明预测结果是可信的,为工程测试的数据处理提供了一种比较实用的新方法。

Stress concentration is one of the main reasons that destroy components in engineering, and we can not measure the maximum strain values or stress values directly in it because of some limits. By applying the GM （ 1,1 ） from the grey model in data processing predicted from stress concentration, this paper is to explore the variable weighting accumulated generating GM （ 1,1 ） and obtain the maximum stress values in stress concentration. Taking the steropeic tie - plate as an example and constructing the models with sample data, this paper will prove the predicted results believable and offer a new practical way for data processing in engineering testing.