Font Size: a A A

The Application Of An Improved Particle Swarm Optimization In Inversion Of Mechanical Parameter Of Slope Engineering

Posted on:2008-04-20Degree:MasterType:Thesis
Country:ChinaCandidate:W WangFull Text:PDF
GTID:2132360215983938Subject:Geotechnical engineering
Abstract/Summary:PDF Full Text Request
The calculation of the mechanical model parameters correctly is one of the difficult problem in analyzing the deformation and stress condition of slope. Optimal estimation of rock mechanic model parameters is obtained by comparing the difference of the data from field observation and theoretical model. The inverse problem of parameter identification is translated into optimization problem by defining objective function. As the development of computer technology, the analysis theory and calculation method, as well as the precision of observation instrument, is more and more mature, the inversion of parameter of rock mechanic model according to the field observation data has nice application foreground. The achievements are as follows:(1) For applying the Particle Swarm Optimization to settling the inverse problem of mechanical parameters in slope engineering, the restraint process of invisible walls to PSO is put forward according to the fundamentals of inversion of parameter. The feasibility of theintelligent inversion method is proved by analyzing the displacement of a two dimension slope.(2) For improve the calculation efficiency of optimizing inversion and reduce the calculation work, the SPSO is further improved. The differential particles and optimum particles are adopted to additional time flying strategy and keeping the optimum position strategy individually, and the direct search method is cited to fabricated a new Fast Particle Swarm Optimization with direct search method (FPSO-DS). Base on the emulation test using Schaffer function, and the comparison of the optimizing results of the SPSO and the Particle Swarm Optimization with linearly Decreasing weight, the validity and stability of improved method is proved. The optimizing efficiency of improved method is estimated quantificationally by the average truncated generations and the distribution entropy of truncated generations.(3) The FPSO-DS method of inversion of parameter is carried out by VC program and validated by checking case, it can be found that the FPSO-DS has a nice result to enhance the optimum solution rapidity of convergence and calculation precision by comparing with the SPSO method.(4) The FPSO-DS method is applied to the inverse analysis of Three Gorge high rock slope, considering the relief zone of slope excavation, the elastic-plastic constitutive model of FLAC is used to the inversion of mechanic parameters of rock slope. The two dimension emulation analysis of the Three Gorge high rock slope deformation is done with the parameters which gained from inversion, and the rationality of inversion parameters is testified through comparing to the monitoring results.
Keywords/Search Tags:slpoe engineering, inversion of parameter, Particle Swarm Optimization, additional time flying strategy, Direct search mothed
PDF Full Text Request
Related items