A PETROLEUM CHEMISTRY PERSPECTIVE ON INTEGRATING IN-SITU SENSOR DATA WITH RESERVOIR MODELS FOR ENHANCED OIL RECOVERY USING KALMAN FILTER

One of the significant benefits of in-situ sensor data is its integration with reservoir simulation models, which helps to optimize Enhanced Oil Recovery (EOR) processes. Located in the reservoir or wellbore, these sensors provide realtime data on pressure, temperature, and fluid composition that can help to improve dynamic updates of the models. This paper presents an approach to assimilate in-situ data into reservoir simulations, combining the Kalman Filter, a powerful technique for data assimilation. It fuses the actual sensor data with the state predictions iteratively, thus reducing uncertainty and growing accuracy. In other words, the task is to predict reservoir states with a model obtained and refine these predictions by using sensor measurements. Rapid development of EOR deployment enables continuous integration that can adjust steaming rates, chemical utilization, or stimulation accordingly. The methodology is validated by simulation results, showing an encouraging performance for various EOR processes in a more innovative and practical way.