Prediction of production parameter for Mid-term Production Optimization
Abstract
In this Mater thesis, three different methods of prediction were proposed. The aim is to test if with a certain accuracy one can predict the production parameter of a certain well. To do this, several cases with different production strategies were created, out of these two cases were selected due to their promising results. Cases were simulated using Schlumberger Simulation Launcher (ECLIPSE 100) of 2011 and the result from the simulation were analysed to establish which method is preferable for the prediction of a certain well. Different cases were studied and prediction to the future were establish, it has shown that prediction of future production of the wells varied depending on the production strategies and some periods in the production time. Prediction of water cut for reservoir with water injection was established, and it has shown that for wells with water cut between 30-60% the preferred prediction method is extrapolation and although the prediction time is small. For higher water cut i.e. 70% and above, both extrapolation method and prediction by using past production for earlier produced well gives good result. With different if prediction time whereby extrapolation method has longer time. For the better prediction of production parameter (water cut, water production rate and oil production rate), the system need to have the pressure support. Conclusion on the prediction methods are drawn based on the preliminary results.Accurate prediction of the oil rate, water rate and water cut give the information that can be used in production optimization. Being able to predict the future production was one task, but using this information for the betterment of the production was another task, having the idea of the future productions give the idea of how to optimize the system, this was tested where by the objective is to increase the oil production rate while maintaining the water cut of the wells,