Identification of structure and parameters of rheological constitutive model for rocks using differential evolution algorithm

Abstract

To determine structure and parameters of a rheological constitutive model for rocks, a new method based on differential evolution (DE) algorithm combined with FLAC^3D (a numerical code for geotechnical engineering) was proposed for identification of the global optimum coupled of model structure and its parameters. At first, stochastic coupled mode was initialized, the difference in displacement between the numerical value and in-situ measurements was regarded as fitness value to evaluate quality of the coupled mode. Then the coupled-mode was updated continually using DE rule until the optimal parameters were found. Thus, coupled-mode was identified adaptively during back analysis process. The results of applications to Jinping tunnels in China show that the method is feasible and efficient for identifying the coupled-mode of constitutive structure and its parameters. The method overcomes the limitation of the traditional method and improves significantly precision and speed of displacement back analysis process.

Keywords

rheological constitutive model / rocks / differential evolution algorithm / identification / FLAC^3D

References

[1] ChengZ.-j., KangW.-f., HuangJ.-fan.. On the locked in stress, creep and dilatation of rocks and the constitutive equations [J]. Chin J Rock Mech Eng, 1991, 10(2): 299-312

[2] ZhongL.-xun.. Enlightenments from the accident of violent landslide in Italy [J]. Chin J Geol Hazard Control, 1994, 5(4): 77-84

[3] FengX.-T., YangC.-xiang.. Genetic evolution of non-linear material constitutive models [J]. Comput Methods Appl Mech Eng, 2001, 190(45): 5957-5973

[4] GaoW., ZhengY.-ren.. Identification of the geo-material constitutive model based on genetic algorithm [J]. Chin J Rock Mech Eng, 2002, 21(1): 8-11

[5] FurukawaT., SugataT., YoshimuraS., HoffmanM.. An automated system for simulation and parameter identification of inelastic constitutive models [J]. Comput Methods Appl Mech Eng, 2002, 191(21/22): 2235-2260

[6] FengX.-t., ChenB.-r., YangC.-x., ZhouH., DingX.-li.. Identification of visco-elastic models for rocks using genetic programming coupled with the modified particle swarm optimization algorithm [J]. Int J Rock Mech and Min Sci, 2006, 43(5): 789-801

[7] ZhuZ.-d., XuW.-ya.. Visco-elastic constitutive model recognition of rock mass and its engineering application [J]. Chin J Rock Mech Eng, 2002, 21(11): 1605-1609

[8] StornR., PriceK.. Differential evolution, a simple and efficient heuristic strategy for global optimization over continuous Spaces [J]. Journal of Global Optimization, 1997, 11(4): 341-359

[9] BergeyP. K., RagsdaleC.. Modified differential evolution: a greedy random strategy for genetic recombination [J]. Omega, 2005, 33(3): 255-265

[10] KaeloP., AlimM.. A numerical study of some modified differential evolution algorithms [J]. European Journal of Operational Research, 2006, 169(3): 1176-1184

[11] SuG.-s., FengX.-ting.. Parameter identification of constitutive model for hard rock under high situ stress condition using particle swarm optimization algorithm [J]. Chin J Rock Mech Eng, 2005, 24(17): 3029-3034

[12] Itasca Consulting Group Inc. FLAC3D User’s Manual (Version 3.0) [R]. Itasca Consulting Group Inc, 2005.