The Research Of Slope Engineering Parameters Identification Based On Improved Particle Swarm Optimization

One of the difficult problems in analysis and the stress state of the slope deformation is the mechanical parameters and initial estimates of how the appropriate slope stress field. There is no doubt, laboratory test and field test is an effective method to solve this problem, but these two methods have their limitations, such as the slope of the non uniform characteristics, slope mechanical parameters obtained limited field of small sample in laboratory tests or local test exists great randomness based on the experimental results, and a representative is not strong, discrete data, the deviation and the slope mechanical parameters of practical calculation, resulting in the certain degree according to these parameters the results of theoretical analysis and field measurement results have big errors. Inverse analysis method provides an effective way to determine a reasonable slope mechanical parameters. With the development of computer technology, the theory and calculation method of positive analysis of mature gradually, the precision of the survey instruments have gradually increased, according to the field observation data for inversion slope mechanical parameters of the model have good application prospect.In this paper, the main research results are as follows:(1) Analysis of the algebraic and analytic properties of basic particle swarm optimization algorithm.(2) To explore an improved particle swarm optimization algorithm based on autonomous learning, through the autonomy given to particle certain to improve the global search and local search ability breadth depth, analyzed the computational efficiency of the algorithm, and through the actual test functions to verify the algorithm has better searching ability and faster convergence speed than the basic particle swarm algorithm. The algorithm is applied to slope engineering mechanics inverse parameter calculation, the inversion parameters are satisfactory.(3) Based on the improved particle swarm algorithm for active learning, the wheel shaped structure and nonlinear function adjustment parameter weights together, proposed a new improved particle swarm optimization algorithm, and its convergence is discussed. The test functions are optimized analysis function on it, and the improved particle swarm algorithm is applied to construct the parameter inversion problem in slope engineering mechanics, the results show that the method is effective in the slope parameters.