Abstract
Reservoir simulation models are constructed with a large amount of geophysical, geological, petro physical and dynamic data. Ideally, it should represent the historical dynamic behavior of the reservoir but due to insufficient characterization there is a need to adjust reservoir parameters to calibrate reservoir models. This is achieved by the history matching process in reservoir simulation. Conventionally it is done by iteratively changing the most sensitive global/local parameters with high degree of uncertainty. Most history matching problems are complex, both concerning the number of parameters being adjusted and the level of accuracy required which makes it a time consuming process.
Assisted history matching (AHM) has been a subject of interest in the recent past with many developments in its technology. A loop of changes directs the key reservoir parameters to minimize the mismatch defined inside the assisted history matching application. This aids in decision making by obtaining a number of calibrated reservoir models where each model can be used for performance prediction to obtain representative field development scenario(s) with minimized uncertainties/risks in minimum time and a cost effective manner.
The assisted history matching workflow used for this case study consists of the following components:
Sensitivity analysis
Uncertainty analysis
Tests and improve the reservoir model (Simulation vs Observed matching)
Improve efficiency, speed and quality of the results
Example of history matching of single well field will be compared on performance improvement using manual versus assisted history match. The reservoir model had 90k grid cells cells. Complex near critical nature of fluids were present in the subject field. The well was matched over 6 years of history. Uncertain reservoir parameters (water cut, static pressure) were selected to fine tune the well behavior. The performance difference, between manual history matching and assisted history matching, was measured through total time required for the HM, number of runs and quality of pressure match obtained at the field level.
