Data Analysis for Oil Production Prediction

Data Analysis for Oil Production Prediction

Christine W. Chan
Copyright: © 2009 |Pages: 8
DOI: 10.4018/978-1-60566-010-3.ch056
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

An economic evaluation of a new oil well is often required, and this evaluation depends heavily on how accurately production of the well can be estimated. Unfortunately, this kind of prediction is extremely difficult because of complex subsurface conditions of reservoirs. The industrial standard approach is to use either curve-fitting methods or complex and timeconsuming reservoir simulations. In this study, we attempted to improve upon the standard techniques by using a variety of neural network and data mining approaches. The approaches differ in terms of prediction model, data division strategy, method, tool used for implementation, and the interpretability of the models. The objective is to make use of the large amount of data readily available from private companies and public sources to enhance understanding of the petroleum production prediction task. Additional objectives include optimizing timing for initiation of advanced recovery processes and identifying candidate wells for production or injection.
Chapter Preview
Top

Main Focus: Variations In Neural Network Modeling Approaches

Prediction Model

The first step of our research was to identify the variables involved in the petroleum production prediction task. The first modeling effort included both production time series and geoscience parameters as input variables for the model. Eight factors that influence production were identified; however, since only data for the three parameters of permeability, porosity, and first shut-in pressure were available, the three parameters were included in the model. The production rates of the three months prior to the target prediction month were also included as input variables. The number of hidden units was determined by trial and error. After training the neural network, a sensitivity test was conducted to measure the impact of each input variable on the output. The results showed that all the geoscience variables had limited (less than 5%) influence on the production prediction.

Therefore, the second modeling effort relied on a model that consists of time series data only. The training and testing error was only slightly different from those of the first model. Hence, we concluded that it is reasonable to omit the geoscience variables from our model. More details on the modeling efforts can be found in (Nguyen et al., 2004).

Complete Chapter List

Search this Book:
Reset