EVOLUTIONARY IDENTIFICATION OF RHEOLOGICAL BEHAVIOR OF CLAYSTONE
In this paper,we develop a hybrid evolutionary identification scheme to model the creep behavior of clayey stones.The genetic programming evolves the structures of rheological models automatically using function tree representation.An interpreter is designed to construct the mathematical expression for each function tree (model structure) and introduce necessary model parameters.A genetic algorithm is employed to evolve the best values of the internal model parameters.Then,all obtained models are compared with experiment results and selected for further evolution.The evolution process is repeated by using genetic operators and the principle of survival of the fittest to perform the process of so-called symbol regression until find the global satisfied model.Applications to the claystones relating to the Three Gorges Project permanent shiplock are used to demonstrate that the present method can efficiently evolve mathematical models for predicting the rheological behavior of rock-like materials.
rheological behaivor claystone modeling genetic algorithms genetic programming
Y.F.Zhu C.X.Yang B.R.Chen X.T.Feng
School of Resources & Civil Engineering,Northeastern University,Shengyang,Liaoning 110004,China Institute of Rock and Soil Mechanics,The Chinese Academy of Sciences,Wuhan,Hubei 430071,China
国际会议
第二届国际非均质材料力学会议(The Second International Conference on Heterogeneous Material Mechanics)
安徽黄山
英文
1397
2008-06-03（万方平台首次上网日期，不代表论文的发表时间）