Neural Network Model for Permeability Prediction from Reservoir Well Logs
نویسندگان
چکیده
The estimation of the formation permeability is considered a vital process in assessing reservoir deliverability. prediction such rock property with use minimum number inputs mandatory. In general, porosity and are independent petrophysical properties. Despite these observations, theoretical relationships have been proposed, as that by Kozeny–Carmen theory. This theory, however, treats highly complex porous medium very simple manner. Hence, this study proposes comprehensive ANN model based on back propagation learning algorithm using FORTRAN language to predict from available well logs. proposed uses weight visualization curve technique optimize hidden neurons layers. Approximately 500 core data points were collected generate model. These data, including gamma ray, sonic travel time, bulk density, numerous wells drilled Western Desert Gulf areas Egypt. results show order accurately, set must be divided into 60% for training, 20% testing, validation 25 neurons. yielded correlation coefficient (R2) 98% training 96.5% an average absolute percent relative error (AAPRE) 2.4%. To validate model, two published correlations (i.e., dual water Timur’s models) calculating used achieve target. addition, had lowest mean square (MSE) 0.035 AAPRE 0.024, while highest MSE 0.84 APPRE 0.645 compared data. indicate robust has strong capability predicting wireline log
منابع مشابه
A Co-Evolutionary Fuzzy System for Reservoir Well Logs Interpretation
Well log data are routinely used for stratigraphic interpretation of the earth’s subsurface. This paper investigates using a co-evolutionary fuzzy system to generate a well log interpreter that can automatically process well log data and interpret reservoir permeability. The methodology consists of 3 steps: 1) transform well log data into fuzzy symbols which maintain the character of the origin...
متن کاملA Fuzzy Symbolic Representation for Intelligent Reservoir Well Logs Interpretation
Well log data are routinely used for stratigraphic interpretation of the earth’s subsurface. This paper presents an automatic blocking scheme that transforms numerical well log data into a fuzzy symbolic representation. This representation maintains the character of the original log curves, which is essential for the stratigraphic interpretation, while making the interpretation task easier. Add...
متن کاملAn Integrated Application of Neural Network and Markov Chain Techniques to Prediction of Lithofacies from Well Logs
The Permian Council Grove Group in the Panoma Field of southwest Kansas has yielded 80 x 10 meter of gas from approximately 2600 wells from a 60-meter interval at depths of 800-1,000 meters since its discovery in the 1960’s. Initial gas saturation, production rates and cumulative production in the Panoma Field are controlled by the distribution of porosity and permeability in the field, which a...
متن کاملNanofluid Thermal Conductivity Prediction Model Based on Artificial Neural Network
Heat transfer fluids have inherently low thermal conductivity that greatly limits the heat exchange efficiency. While the effectiveness of extending surfaces and redesigning heat exchange equipments to increase the heat transfer rate has reached a limit, many research activities have been carried out attempting to improve the thermal transport properties of the fluids by adding more thermally c...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Processes
سال: 2022
ISSN: ['2227-9717']
DOI: https://doi.org/10.3390/pr10122587