Application of Artificial Neural Networks to Downhole Fluid Analysis

Reservoir characterization and asset management require<br>comprehensive information about formation fluids. Obtaining<br>this information at all stages of the exploration and<br>development cycle is essential for field planning and<br>operation. Traditionally, fluid information has been obtained<br>by capturing samples and then measuring the<br>pressure/volume/temperature (PVT) properties in a laboratory.<br>More recently, downhole fluid analysis (DFA) during<br>formation testing has provided real-time fluid information.<br>However, the extreme conditions of the downhole<br>environment limit the DFA tools to measuring just a small<br>subset of the fluid properties provided by a laboratory.<br>Nevertheless, these tools are valuable in predicting other PVT<br>properties from the measured data. These predictions can be<br>used in real time to optimize the sampling program, help<br>evaluate completion decisions, and understand flow assurance<br>issues.