Combining Wavelet Transform and Neural Network to Differentiate the Stratigraphy from Logs of Namorado Oilfield in Campos Basin

On well logging, there is a great interest to improve the vertical resolution of the logs, aiming the identification of different layers or geological formations along the borehole and the construction of a reservoir model. Generally, the identification of hydrocarbon formation lithology from geophysical logs employs several approaches as lithology crossplots (such as ‘‘M–N lithology plot’’ which requires a sonic log, density log, and neutron log) or the combination gamma-ray neutron-density log method. Also, numerous<br>mathematical approaches have been proposed to perform this task computationally, between them, artificial intelligence techniques. In this sense, the purpose of this study was to identify the formation interfaces from geophysical well logs using a combination of wavelet transform and neural network methods. The first technique was applied to smooth the logs, while the second was utilized to fit them to a selected lithological model. The input variables were gamma-ray, resistivity, density, neutron porosity and<br>sonic logs from Namorado Oilfield in Campos Basin. This method is easy to implement in a computer with MATLAB platform and it showed a good performance in the discrimination of main layers.