Nonlinear Global Optimization Methods In The Residual Statics Problem

As well known, static correction in rugged areas is an important problem in seiemic exploration, but it is also hard and is not resolved well at present, and related research and referring material are little. In this paper, this problem is further diccussed, and some new progresses are made.Following the field static correction,the residual static measurements are able to be determined through the traditional residual static estimation.However,in case the measurements are larger than half of the period or the S/N ratio is too low,a periodical wiggling of the events will occur with the static application in the traditional way.A new method to solve this problem must be found.In fact the static measurement is a nonlinear question that could be determined by hybrid optimization.Simulated annealing(SA),genetic algorithm(GA) and Tabu search(TS) ,which is developed in 1980s and 1990s,come from simulation of properties on nature.Three kinds of optimization methods are extensively pay attention to solve nonlinear optimal questions and be characterized by global convergence.This paper analyses and studies a great deal of modeling test on residual static correction using SA,GA and TS on the basis of fully absorbing and digesting research achievements of domestic and overseas scholars on nonlinear optimization technique.In this paper,some achievements have been made as following:(1) Tabu search is used in residual staic correction at first time.The method to solve the question of residual static correction by using Tabu search has been studied deeply.(2) A kind of hybrid parallel search strategy (namely hybrid parallel genetic algorithm)has been developed forward in the light of idea of parallel calculation and diversifacation of biological evolution.This algorithm further slow down the pace of premature convergence of GA.(3) The algorithm which combined GA and SA and the algorithm which combined TS and GA has been studied deeply.The advantages and disadvantages of them has been pointed.(4) A algorithm which combined the advantages of GA,SA,TS and Stack power maximization has been developed.Moreover,we have applied it to models and processing practice.We secceeded at applying this algorithm on the processing of the field data from YanQi area and Santanghu area of Xinjiang autonomous ,Bayinhaote...