APPLICATION OF OPTIMUM WIENER FILTER USING MULTI INPUT IN PREDICTING DISTRIBUTION OF TUFF LITHOLOGY BASED ON 2D SEISMIC DATA ON TOWUTI LAKES

• Main Author: Putri Utami, Adisa
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/40400
• Institution: Institut Teknologi Bandung

The multi-attribute seismic method is a method using a statistical approach, the combination results are more than one seismic attribute to predict subsurface properties. Seismic-based attributes generally have a relatively low frequency than log-base properties, so predicting subsurface properties using seismic attributes produces poor results. Therefore, a step is needed before performing the Multi-Attribute process in predicting this property. One method that can be used is the Optimum Wiener Filter, this filter is commonly used in the deconvolution process to increase the vertical resolution of seismic data. The advantage of Wiener Filter in increasing resolution is that this filter can be used to improve and increase the frequency of data from seismic. Thus, the Wiener Optimum Filter can be used as an additional stage before performing multi-attribute seismic in predicting subsurface properties. In fact, seismic data in this research area can be used for various purposes, in this study two wells and two seismic lines will be used to find lateral distribution of subsurface properties. The subsurface property that you want to propagate is the radioactive element of a lithology that is described by the gamma ray log recording to see the continuity of tuff lithology which has a high gamma ray response, by integrating information from seismic data and well data, as well as additions from the FWO application. In this study synthetic data is used which aims to understand and construct the concept of the Optimum Wiener Filter, and find the most optimal method in the construction of OWF if there are more than one input and output data. The most optimal method will be used in field data applications. The Optimum Wiener Filter application using one Acoustic Impedance attribute as an input value, produces results that are quite good but not optimal. One method used in this study to optimize the results obtained is to use three attribute values as input, namely acoustic impedance, and two complex attributes (quadrature, and reflection strength). The results obtained from this study are that OWF can predict subsurface properties that have a relatively higher frequency by using data with lower frequencies, the use of multi well and multi input in predicting subsurface properties will give more maximal results than just using one well and just one input. The FWO application on multi-attribute also contributes significantly in increasing the correlation and lateral continuity of the prediction of subsurface property distribution.