Research On Iteration Method Using In Potential Field Transformations And Its Filter Characteristic

Data processing and transformation play an important role in gravity-magnetic datainterpretations, appropriate transformation will make the inversion much easier, forexample, the downward continuation can enhance the shallow information, reductionto the pole can eliminate the impact of oblique magnetization, vertical derivative ofthe field can enhance the geological boundary, but some of them are unstable at times,and their transformation factors have a significant amplifying effect, so that the resultsobtained directly from FFT are unstable. As a result, many research work has beendone based on how to increase the stability of the method and how to improve theresult. The iterative method which can solve the problem above to a certain extent isbrought to the forefront recently, but people do not realize its shortcoming. It can gaingood result for some specific problem, but it still has a problem that the result can't beimproved with the increase of the iteration number, and the result is still unstablewhen the iteration is large.This paper carried out study based on the iterative method used in thetransformations of the potential field, the main results are as follow:1.The general iterative formula is deduced based on the study of the iterativemethod, and convergence effect of various factors on the iterative method is analyzed.The results show that the main factor which decides whether the iteration methodswill convergence to the FFT theoretical results is the mapping function from the rawdata to the target data, and the iteration method will converge to the FFT theoreticalcalculation result, if the appropriate mapping function is chosen. The iteration numberdecide not only the compute time but also the compute result.2.Application of iterative method in transformation such as continuation,reduction to the pole, derivation is analysed detailedly, and they are compared withsome methods, about which some new ideas are given.3.Increasing of the iterative number will converge to the FFT theoretical result ifthe appropriate mapping function is chosen, but if the FFT theoretical calculation ofthe transformation is unstable, the iteration will converge to an unstable result. Theconvergences means nothing to the problem, and it will make the error between compute result and real result much larger, so the iteration method can't resolve theinstability of computation fundamentally .