Estimation of Static Formation Temperatures in Geothermal and Oil Well by Inversion of Logged Temperatures
- Key words
- Formation Temperature; Numerical Simulation; Inverse Problem; Levenberg-Marquardt; Cognitive Models; Artificial Intelligence
- Conference
- Geothermal Resources Council Transactions
- Year
- 2008
- Session
- Well logging; Reservoir modeling; Computer modeling
- Language
- English
Abstract
Three inverse methods are used to obtain static formation temperatures in geothermal and oil wells. The methods include: (a) A least squares optimization algorithm or Levenberg- Marquardt method (MLM); (b) A Proportional Integral Control method (PI), and (c) An Artificial Intelligence method (IA). Results obtained using data from an oil well show that it is feasible to predict SFT reasonably by these methods however the IA method predicts lower temperatures. SFT´s obtained by the MLM and PI methods also compare well with simpler methods like the Horner and spherical-radial flow method.
Copyright 2008, Geothermal Resources Council. Readers who download papers from this site should honour the copyright of the original authors, and may not copy or distribute the work further without the permission of the original publisher.