Reservoir engineering datasets and simple modelling in early steamfield development

Abstract

In order to expedite development of the Indonesian geothermal potential, the government plans to drill deep exploration wells before tendering out new geothermal working areas. Success of such activities is subject to the proper location of the exploration wells and thorough testing of the wells during drilling, recovery and testing. Reservoir engineering datasets comprise a large fraction of these data and their analysis is the subject of this report. Volumetric models being developed at the surface exploration stage can give an indication of the power potential at hand. Here a few examples are given, from three different countries. The Corbetti resource in Ethiopia, infers a reserve area of 50-100 km2 , 1-2 km of thickness and 260-360°C temperature. A Monte Carlo based volumetric model for these parameters yields a probable generating capacity at or above 970 MWe. Similarly, the already drilled Rantau Dedap resource in Indonesia is characterized by 3-17 km2 of a 1.3 km thick reservoir at 230-290°C. This yields some 50 MWe of proven generating potential using the volumetric model approach and a single-flash heat to power conversion. A bottoming binary system may provide another 25 MWe.

Example can also be taken from Iceland. Analysis of downhole pressure and temperature data in a 2500 m deep well SV-26 in Svartsengi, Iceland, results in a static water table at 425 m, a boiling point with depth zone to 700 m, and near isothermal 240°C single-phase water reservoir to total depth. Two major feedzones are identified at 1225 and 2200 m TVD (total vertical depth) and the well pressure pivot point is located at 1350 m TVD and at 77 bar-g. Multistage injection and production test analysis result in a cold well injectivity index of 11 kg/s/bar and a hot well productivity index of 28 kg/s/bar. Both indicate a very good reservoir. Detailed hydrological models of injection and production transient pressure data show well skin in the negative territory and reservoir permeabilities in the 10-1000 mD range. Higher model values arise from models calibrated against well data collected before running the 7” slotted liner. This may suggest that the lower cold water injectivity index relates to liner holes viscosity effects. A numerical flowing wellbore model was able to capture these properties; the bottomhole and wellhead pressure with flow and the dynamic temperature and pressure profiles with depth. The model predicts a shallow to deep feedzone ratio at 2:1. The modelled wellhead output curve infers a much higher maximum flowrate than the tested maximum of 47 kg/s, thereby taking well capacity close to 10 MWe.