Research On Matching Method Of 3D Multiwave Seismic Data

The multiwave exploration technology is one of the important development technologies In the field of oil and gas exploration.The matching of multi-wave seismic data is the basis of multi-wave exploration technology.The matching degree of multi-wave seismic data plays an important role in multi-wave joint interpretation and joint inversion.so how to improve the matching accuracy of multi-wave seismic data is one of the core issues of multiwave technology today.At present,the difficulty of multi-wave matching technology is mainly that high-precision matching of PP and PS waves cannot be performed.First of all,the method currently used is basically based on the large horizons to match the data,and the area between hoirzons may not be matched well enough,so that the overall matching of multi-wave data is still lacking.Secondly,high-precision matching cannot be performed in the multi-wave data region that with weak horizon characteristics,and we can only get the horizon matching that is with large errors as the result.In order to solve the above-mentioned problems,this thesis introduces the Dynamic Time Warping(DTW)algorithm in the field of speech recognition to improve the overall matching effect.And the Support Vector Regression(SVR)method is used to solve the problem of poor matching results in weak feature areas.The main work of this thesis is as follows:1.For the problem of poor matching integrity,this thesis combines manual interpretation and automatic matching,and proposes a multi-wave matching algorithm based on horizon control to match the PP and the PS waves directly.Firstly,the horizon position control points are obtained through manual interpretation,and then the control point is brought into the DTW algorithm to constrain the path of the algorithm so that it is within the control range.Not only did that,but also the improved second-order recursive summation function is used instead of the original first-order recursive summation function.Finally,using an adaptive mesh to smoothing the matching result.2.For the problem of inaccurate matching of weak feature regions in multi-wave matching,this thesis proposes a correction method based on support vector machine regression.The results of the horizon matching and the exact matching algorithm above are brought into the vector machine for training,and the regression function is obtained.Finally,the results of other horizon matching are input into the vector machine to obtain the corrected result.This method seeks the connection between data,performs a mapping of inaccurate horizon matching and precise matching results,and it is a new way of correcting multi-wave matching result.3.Designed and implemented a 3D multi-wave seismic data matching software system.Starting from the requirement analysis,combining the implementation of matching and correcting algorithms,the software is divided into functions and modules,and the core algorithm module is explained and analyzed in detail.In this thesis,the multi-wave matching algorithm based on layer control combines the advantages of artificial experience and automatic algorithm,so that the matching result meets the high-precision requirements at a specific horizon and the overall matching effect meets the requirements.In the weak feature region,the multi-wave matching correction method based on SVR can also improve the result of the layer matching.Finally,the algorithm is integrated to provide a complete multi-wave matching solution.