To estimate the mechanics parameters of rock mass, the genetic algorithm (GA) is adopted Considering the weakness of GA on convergence performance, a new improved genetic algorithm (IGA) is developed based on the niche algorithm, adaptive probability of crossover and mutation, and elitism strategy. Optimum result of the Shubert function using the proposed algorithm shows that the global convergence performance of genetic algorithms is greatly improved. Based on these improved methods, the process of estimating mechanics parameters is established through the improved genetic algorithms, the finite element method and the theory of displacement back analysis, to estimate the mechanics parameters of rock mass. Moreover, the optimization displacement back analysis program (ODBA) is worked out. Finally, using this program, the mechanics parameters of rock mass in shisanling pumped storage station are estimated, and the results indicate that estimated parameters are compared well with field test mechanics parameters. Consequently, the new IGA should be popularized to estimate the mechanics parameters in geotechnical engineering.
[1]
Giulio Maier,et al.
Optimization Methods for Parametric Identification of Geotechnical Systems
,
1982
.
[2]
R. Clough,et al.
Finite element applications in the characterization of elastic solids
,
1971
.
[3]
Liu Yingxi.
IDENTIFICATION PROCEDURE FOR SHEAR STRENGTH PARAMETERS OF GEOTECHNICAL MATERIALS USING HYBRID GENETIC ALGORITHM
,
2005
.
[4]
Ding De-xin.
THE GENETIC-ALGORITHM-BASED APPROACH TO BACK ANALYSIS OF DISPLACEMENT PARAMETERS OF ELASTO-PLASTIC MODEL FOR ROCK MASS
,
2003
.
[5]
Liu Chuan-xiao.
Research on back-analysis of roadway displacement based on genetic algorithms
,
2004
.
[6]
C. A. Murthy,et al.
Genetic Algorithm with Elitist Model and Its Convergence
,
1996,
Int. J. Pattern Recognit. Artif. Intell..