Staying Ahead through Built-in Automated Approach for Well Modelling in Gas Fields

Abstract

In oil & gas industry, it is evident that digitalization has become the leading technology to improve interactive data analysis in amultidisciplinary team. Exploitation of data has been a real challenge especially on mature fields with high real-time data frequency extraction on day to day basis. This paper aims to increasethe integration of data and workflow for building well models. An automated workflow through programming has been implemented in offshore Sarawak, Malaysia involving three (3) main processes, which are: procedures to use historical data to select IPR and VLP correlation for well models a workflow to predict well IPR using available surface data a built-in application to auto-match well models to selectthe most representative IPR model based onsufficient production data to be used in forecasting The well performance and technical potential of the field was evaluated in accordance to this workflow, resulting in significant time-saving when analysing the massive pool of data. This enables the proliferation of real-time data usage promptly. Employing this workflow minimizes misinterpretation of frequently ignored erroneous information compared to the conventional method. Greater reliability of well analysis using workflows helps to avoid perpetuating bias tendencies on data used for model calibration or quality check. This will result in better technical confidence needed to make decisions to move forward with development plans for a wide range of production rates and operating conditions. This paper will investigate the impact of using incorrect IPR or VLP versus the results of this workflow based on the same historical dataset. It will also demonstratehow the built-in automated process adds value in improving the understanding of well performance while ensuring that the models are calibrated to the entire historical data as accuratelyas possible.