During the past decades the challenges involved in the petroleum exploration and production have dramatically increased. Extreme operating conditions such as highly deviated wells, deep water drilling or narrow pressure windows force the necessity for automatic and smart systems in order to successfully and safely accomplish all tasks. Solutions for such issues are provided by automatic monitoring and well control event detection systems which can significantly simplify the drilling process and unfold a new level of response time. The major challenge in hydraulic modeling is mainly due to the fact, that representing real conditions in a mathematical model is difficult to achieve. Careful parameter selection and required mathematical approximations is critical for the success of the model.

This paper presents a new approach, supporting the driller in monitoring the hydraulic system and in the recognition of well control events at an early stage such, that proper counteractions can be initiated before any damage occurs. In this respect, a hydraulic real-time monitor based on sensor data has been developed, using artificial neural networks to compute, recognize and predict abnormal events in the wellbore. Preceding the numerical simulation, a data reliability and quality routine is initiated which assures the best use of the provided source data.

The results of these advanced modeling techniques are presented to the driller, providing a detailed real time wellbore hydraulic overview including information on pump requirements, fracture and pore pressure limitations and hole cleaning requirements. Additionally the stand pipe pressure readings are overlaid on the calculated safe operating window, thus giving clear and accurate information of the hydraulic status in real time. The sufficient level of accuracy combined with a clear interface enables the driller to initiate immediate counteractions. In addition, the abnormal event recognition option contributes significantly to the drilling performance optimization.

This content is only available via PDF.
You can access this article if you purchase or spend a download.