Probabilistic geothermal resource assessment using experimental design and response surface methodology: The Leyte geothermal production field

Abstract

Estimating power potential (MWe) with higher accuracy remains a necessity and a challenge confronting the geothermal industry. Reservoir models can predict MWe capacity probabilistically for resource management and decision-making use, but the process remains laborious to implement. This work involves applying the Experimental Design (ED) and Response Surface Methodology (RSM) framework to a calibrated natural-state reservoir model for predicting MWe probabilistically, quantifying uncertainties in the prediction and identifying the key driving factors.This study has chosen six uncertain parameters and implemented the Plackett-Burman design to build twelve versions of the Leyte reservoir model for uncertainty quantification. These uncertain parameters are permeability in the x, y and z-direction, porosity, reinjection enthalpy (RI Enthalpy), and the fraction of reinjection (% RI). Results revealed that all of these parameters except RI Enthalpy are significant predictors of MWe and that the Leyte Geothermal Production Field (LGPF) has an indicative capacity of 670 MWe (P90), 730 MWe (P50) and 790 MWe (P10) if it operates for 50 years.