Sonic Tomography Method Research Based On Simulated Annealing Algorithm

Tomographic imaging can reflect the internal structure of the research object intuitively in the form of image intuitive,even for some research object which cannot be sliced,so it has a wide range of applications in the medical CT,geological exploration and drilling.Sonic tomography can use sound which has a good penetration by applying to the external of the subjects,can operate easy,has more practical research value.But the existing methods are limited by the characteristics problem which is not only one solving,always easily plunged into local solution or have a low computational efficiency.This paper aims to improve the operation efficiency of tomographic imaging method through improving the simulated annealing which is a nonlinear inversion method,and can also be very good to avoid local solutions.The main contents of the paper are as follows:Firstly,the research purpose and meaning of seismic tomography are introduced,the domestic and foreign history and present situation of seismic tomography is introduced,also introduce the method of seismic tomography by model parametrization,forward,inversion and evaluation of solution quality.Emphasis on introducing the principle and methods of inversion.Secondly,using simulated annealing method which is a nonlinear inversion method to calculate the model,and study the convergence of the objective function values,analysis the shortcomings when it is calculating, improve the way to select speed parameters and to wave the model by aiming the disadvantages of simulated annealing algorithm,put forward a improved simulated annealing algorithm.Finally,applie the improved simulated annealing algorithm in simple and complex models, compare the result images with the general simulated annealing algorithm when running the same time and the model,and then expand the scope of the velocity parameter selection, through comparative prove that the improved simulated annealing approach does not only keep the the advantage of previous algorithm which can avoid the local solution,but also can improve the operation efficiency of sonic tomography method, and even apply to the situation that model has a lot of speed parameter or the modelparameters selection is large range, so the quality of the sound wave tomography is improved.