A MATHEMATICAL MODEL FOR WIRELINE LOGS FILTERING USING DIFFERENTIAL FUZZY PARAMETERS

Abstract

Signal processing is a set of techniques used to extract data from any signal. A signal may be mathematically characterized as a function of several variables (parameters), such as time, distance, resistivity, or radiance. A signal is usually acquired using one or more analog or digital devices, such as a temperature sensor, a digital camera, or a resistivity probe. After acquisition, the processing depends on the nature of the signal and its information. In addition to the information being studied, the recorded signal may contain noise, which hinders the extraction of information, leading to ambiguous or erroneous results. The objective of this study is to present a non-linear technique for noise attenuation in well logs using fuzzy sets and fuzzy logic. The proposed filter evaluates the continuity of the log measurements through the use of two differential parameters. Sudden leaps in the measurements may indicate the presence of noise; therefore, for each point in the log, the filter evaluates the degree of discontinuity and provides a correction value to be applied. Differential fuzzy filtering is applied to data from synthetic and real well logs to conduct a performance evaluation using the MSE (mean square error).