Support Vector Machines for Classification and Mapping of Reservoir Data

The report deals with the novel application of Support Vector Machines (Support Vectore Classification and Support Vector Regression) for the analysis and modelling of reservoir data. 2 problems are considered: classification and mapping of porosity data. Results are compared with geostatistical models - indicator kriging and ordinary kriging. Variography is widely used to control the performance of the machines. Geostatistical explanations for the SVR hyperparameters are discussed. Obtained results demonstrate flexibility and efficiency of SVM application for the reservoir characterisation.

Related material