论文部分内容阅读
针对混凝土大坝坝体和岩基参数的区间不确定性,构造具有区间分析功能的RNN(粗糙神经网络)模型,并运用该模型反演坝体和岩基区间参数值。应用区间有限元对结构进行正分析,根据区间参数反演的需要选取相应的区间学习样本,利用RNN模型对样本进行模式学习直至网络收敛,最后通过网络回想和反归一法得到坝体和岩基力学参数的区间值。研究结果表明,该方法可用于反演混凝土坝坝体和岩基区间力学参数,反演得到的区间参数值是合理的。此外,基于RNN模型的区间参数反演方法经过一定的拓展和改进,理论上可应用于反演其他类型的区间参数。
According to the interval uncertainty of concrete dam and rock foundation parameters, an RNN (Rough Neural Network) model with interval analysis function is constructed. The model is used to inverse the interval between dam body and rock foundation. The finite element method is used to analyze the structure positively. According to the need of interval parameter inversion, corresponding interval learning samples are selected, and the model is modeled by RNN model until the network converges. Finally, the dam body and rock Interval value of basic mechanics parameters. The results show that this method can be used to inverse the mechanical parameters between the concrete dam body and the rock foundation, and the inversion of the interval parameter values is reasonable. In addition, the method of interval parameter inversion based on RNN model can be used in inversion of other types of interval parameters theoretically through some expansion and improvement.