The Rayleigh Surface Wave Dispersion Curve Is Inverted Improve Intelligent Optimization Algorithm Research

The Rayleigh surface wave exploration method is widely used in various shallow underground space fields such as urban geological survey.One of its cores is surface wave dispersion inversion.Dispersion curve inversion methods are divided into two categories:linear and nonlinear inversion methods.The nonlinear inversion method has been paid more and more attention because it does not rely too much on the initial model and has a strong global optimization ability.Particle swarm optimization is widely used in the inversion of surface wave dispersion curves,but conventional particle swarm optimization is prone to premature maturity and falls into local extreme values.In this paper,three optimization strategies of nonlinear adaptive inertia weight,introduction of compression factor and boundary condition constraints are used to improve it,and a hybrid particle swarm optimization algorithm(HPSO)is proposed to invert the dispersion curve.In addition,the barnacle mating optimization algorithm was introduced into the surface wave dispersion curve inversion for the first time,and the adaptive genetic barnacle mating optimization algorithm(AGBMO)was proposed by referring to the optimization strategy of the genetic algorithm.Two improved intelligent optimization algorithms are used to invert the fundamentalorder dispersion curve of the three-layer classical geological model,and compared with the inversion results of the particle swarm optimization algorithm with boundary constraints,it is found that the inversion of the two improved intelligent optimization algorithms proposed in this paper is The accuracy is higher,reflecting the superiority of the improved method.In addition,the anti-noise ability test and the joint inversion ability of multi-modal dispersion curves of the two methods were tested and analyzed.The adaptive genetic barnacle mating optimization algorithm(AGBMO)has good anti-noise ability and stability,which lays the foundation for the subsequent research on the hybrid intelligent optimization algorithm of the two.Finally,by using two methods to invert the measured dispersion curve,the inversion results are consistent with the prior results,which shows the practicability of the two improved methods proposed in the paper.