An integrated petrophysical study using well logging data for evaluating a gas field in the Gulf of Thailand | |
| Author | Thoedpong Witthayapradit |
| Call Number | AIT Thesis no.GE-08-17 |
| Subject(s) | Gas fields--Thailand, Gulf of Neural networks (Computer science) Gas wells--Thailand, Gulf of |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Geosystem Exploration and Petroleum Geoengineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. GE-08-17 |
| Abstract | Permeability, porosity and water saturation are the most three important petrophysical properties used for hydrocarbon reservoir evaluation. The permeability and porosity can be directly measured from the core samples in laboratory. However, only some depth intervals of a well are cored or coring could not be done of at all. Therefore, one has always to rely on well logging data for petrophysical analysis. With such measurement as density, neutron and acoustic logging, porosity can be routinely evaluated. Estimation of permeability is more difficult. The Artificial Neural Network (ANN) can be an additional help to estimate permeability and porosity from well logging data. This study has the main objective as performing a petrophysical analysis to evaluate formation of three open boreholes, i.e. Well-1, Well-2 and Well-3, located in the North Malay basin. First, a well logging interpretation was using Interactive Petrophysics (IP) software as donated by Schlumberger to AIT to determine the clay volume, porosity and water saturation of reservoir zones using Archie’s equation which a factor and cementation exponent were derived from the interpretation. The permeability was estimated from the permeability-porosity relationship of core analysis. There were nine reservoirs zones with total thickness 1688 ft in Well-1, 21 reservoir zones with total thickness 661.5 ft in Well-2 and 17 reservoir zones with total thickness 443.5 ft in Well-3. In addition, the study of flow unit characterization was done from available core analysis data. Three distinct flow units were found having FZI equal to 1.5, 4.5 and 13.8. In the final part, the back propagation ANN was studied and applied for porosity and permeability prediction using training and testing data set from Well-1 and Well-2. The training and testing data were selected from all well logging and core analysis data, from reservoir zone interpreted by well logging and from flow unit characterization. The reduction of ANN input data was studied to enhance prediction result. From the study, the reduction of GR log input improved the ANN generalization and gave a better permeability and porosity prediction. The optimal ANN for porosity prediction composed of 15 hidden neurons (MSE=15.66) and the optimal ANN for permeability prediction composes of 9 hidden neurons (MSE=11,736). |
| Year | 2009 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis ;no. GE-08-17 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Civil and Infrastucture Engineering (DCIE) |
| Academic Program/FoS | Geotechnical Engineering (GE) |
| Chairperson(s) | Pham Huy Giao; |
| Examination Committee(s) | Noppadol Phien-wej;Le Hai An; |
| Scholarship Donor(s) | RTG Fellowship; |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 2009 |