WAVELET DATA PROCESSING IN THE PROBLEMS OF ALLOCATION IN RECOVERY WELL LOGGING

摘要

Methods of Geophysical Research (GRM) are widely used in order to construct valid models of oil and gas fields. The essence of these methods is the process of logging sounding, that means the following: in the oil well falls instrument for measuring various characteristic values, then these findings are processed. One of the commonly used methods are the method of high-frequency induction logging isoparametric sounding (HILIS) and the method of lateral logging sounding (LLS). General interpretation of the scheme used in these systems, as follows: in the first stage well logs analyzed, the data is processed and allocated beds - areas with homogeneous characteristics of the signal. Then, in the resulting layers are determined value of the electric resistance. These models are then visualized in the GUI. To ensure proper construction of models and acceptable computing speed is a very important task of the initial treatment, as in the results of measurements is often present apparatus and geological noise, complicating analysis. This noise is a high frequency signal fluctuations, which often tend to smooth out the filtration transformation curve. In the same class of problems includes the correct alignment of the geological formations. Correctness is defined geo electric theory and empirical experiences geophysics specialists who often produce alignment layers boundaries manually, based on the image of the curves in the diagram. The paper was presented to data processing algorithm based on the wavelet transform signal. Main research questions are define digital signal of well and processing methods for considered signals. This work is aimed at optimizing the transfer of digital signals in an oil well logging. Prior to that, the process was carried out by other methods, but the method presented me wavelet applied to this problem for the first time.The basic idea was to minimize the sent data - signals for subsequent processing. These results indicate that the proposed algorithm ensures real efficiency in processing signals for the primary task allocation layers [1].