Best Practices in Calibration of Integrated Production Modeling for Tengiz Field

摘要

Petroleum Experts Integrated Production Modeling (IPM) toolkit is widely used in Tengizchevroil (TCO) for short and long term production forecasting in the Tengiz field. The short and long term models each have their own strategic focus. The short term IPM model is used for day-to-day management of field activities to ensure business plan production volumes are delivered. The long term IPM model is used to manage base business and optimize scenarios for various future development projects. The short term IPM model is validated and history matched continuously and forms the basis of the long term model to meet business objectives. Because of the complexity of the Tengiz gathering system, close collaboration among multiple teams, systematic updates and continuous improvement of the model is required. It is also important to have an accurate temperature prediction across the entire field network to conduct flow assurance studies, optimize plant inlet temperatures and maximize plant throughput. As a result of the implemented calibration efforts, the model predicts field production rates within 1% of the actual field performance. The model provides more granularity and higher confidence in ensuring that the plant can produce at its maximum capacity when wells are shut-in for surveillance activities, pipeline repairs and meter station (MS) shutdowns for gathering line repairs. In a long term perspective, it's crucial to have an accurate model to plan MS shutdowns for upgrades and the successful startup of major capital projects. This paper describes model improvement initiatives, the best practices and lessons learned during the calibration process of the complex IPM model. A structured workflow for the calibration of the IPM model is also provided. The calibration steps are described in detail with relevant examples throughout the paper. This workflow can be used by IPM practitioners in other fields to construct and maintain integrated production system models.