Print this pageAAPG GEO 2010 Middle East
Geoscience Conference & Exhibition
Innovative Geoscience Solutions – Meeting Hydrocarbon Demand in Changing Times
March 7-10, 2010 – Manama, Bahrain
Estimation of Reservoir Properties from Seismic Attributes and Well Log Data Using Artificial 
Neural
Networks
(1) King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia.
Porosity, permeability are key factors to build a 3D geological model for a reservoir. The best method to get these properties would be to measure them on core samples in the laboratory. However, this method is costly and time consuming, and usually only a few out of all wells are cored and even then only a small portion of the well. To fill the gap in the vertical scale, geologists generally use a statistical approach, such as linear or non-linear multiple regressions to correlate reservoir properties with the continuously recorded well log data. Recently, geoscientists have utilized Artificial Intelligence (AI), especially Neural Networks (ANNs), to predict reservoir properties. This talk reports a comparative study of two types of neural networks, a Multiple-Layer Perception MLP, with back propagation neural network, and a General Regression Neural Network GRNN. The viability of these techniques are demonstrated on well log data and seismic attributes from sand stone reservoir in south of Algeria. This study utilizes the basic logs including gamma ray GR, interval transit time DT, shale volume VSH, bulk density RHOB, deep later log LLD and corrected porosity NPHI and five attributes( instantaneous frequency, instantaneous phase, RMS amplitude, half energy and Arc length) to predict porosity, permeability and lithofacies in cored and uncored wells. The agreement between the core data and the predicted values by neural networks demonstrate a successful implementation and validation of the network’s ability to map a complex non-linear relationship between well logs and permeability and porosity. Also the results show that the application of the General Regression Neural
Network GRNN gives a relatively better performance than the Multiple-Layer Perception MLP.