Paper: DETERMINATION OF HISTORICAL FIELD DATA CUT-OFF TIME SUFFICIENT FOR RESERVOIR HISTORY MATCHING

For reservoir history matching, simulated model output is conditioned to observed historical field data by modifying the model parameter so that the simulated data matches the history data. The matched model which is dependent on the historical data utilized for matching is used for reservoir production forecast. The vital question to be answered by reservoir engineers is the waiting period duration in order to have sufficient historical data for matching. In this work an answer is provided for this waiting period. This paper reports cut-off time for sufficient historical data suitable for reservoir history matching. The cut-off time was determined from several history matched scenarios. To calculate the cut-off time, reservoirs with known parameters and twenty four year production data were used as the base case models. Thereafter, simulation models of the actual reservoirs were built. The simulated model was run for 6, 12, 18, 24 and 48 months, respectively, and the simulated production profile matched with their corresponding actual reservoir historical time data to achieve a matched model. The calibrated model was applied to make future reservoir prediction and the simulated model prediction was compared with the base case reservoir production profiles which were known. This approach enabled the determination of historical data cut-off time that is sufficient for good history matching as follows: 1. observed historical data of 18 months are sufficient for a good history match if the simulated model is 75 percent and above close to the actual reservoir description. 2. If the simulated model is between 50