Assessing Uncertainty in Future Performance Predictions of Lumped Parameter Models Using the Randomized Maximum Likelihood Method

Abstract

In this work, we present a methodology within the context of stochastic simulation for assessing uncertainty in future pressure changes simulated by using history-matched lumped models for low-temperature geothermal systems. Specifically, we consider the randomized maximum likelihood method (RML) for the assessment of uncertainty. We show that this methodology allows us to incorporate into the performance predictions any uncertainties in both the model and the measured data. In this way, we are able to characterize or appraise the uncertainty in the predicted future pressure changes. Once the uncertainty in predicted performance is characterized or assessed, it is possible to make reservoir management decisions that account for an incomplete knowledge of the actual geothermal system. One synthetic example application is presented to show the use of the methodology proposed in this work.