Continuous reservoir model calibration with time-dependent reservoir properties diagnosed from long-term down-hole transient pressure data

Abstract

Reservoir properties may change during the long-term production, such as time-dependent skin factor and permeability. The reservoir modeling with constant reservoir properties cannot match the field performance, which will bring huge uncertainties for future forecasting. Therefore, for the reservoir with time-dependent reservoir properties, reservoir model needs to be calibrated continuously. However, diagnosis of changes in reservoir properties is difficult. This paper proposes a method of real-time reservoir simulation model update using time-dependent reservoir properties diagnosed from transient pressure with wavelet transform (WT). A key diagnostic function can identify the time and degree of changes in reservoir properties, and then the sliding window technique is applied to divide pressure history into different time windows, and in each time window reservoir properties change so little that they can be treated to be constant. When the changes in reservoir properties are large, time window will slide to the subsequent one and near-wellbore model (NWM) will be selected from the full-field reservoir model and then updated with new reservoir properties. An effective workflow of calibrating reservoir model by updating NWM with time-dependent reservoir properties has been proposed and demonstrated with a case study. This method can improve reliability of long-term reservoir modeling and reduce uncertainties of future forecasting, especially for the reservoir with time-dependent reservoir properties.