| Popis: |
In recent years, Underground Gas Storage (UGS) has earned itself strategic importance as it guarantees energy sustainability in markets suffering from unpredictable supply. Storage management is a complex activity which faces the challenge of combining the variability of daily commercial client requests along with the capability of the reservoirs to deliver. Gas production and re-injection activity requires standard core competencies of the oil and gas industry, however the process is fast paced as compared to conventional hydrocarbon production activity as time scales from data analysis to decision making is reduced to hours. Stogit Spa, an Italian gas storage company managing 8 depleted gas fields in Italy has implemented an integrated system that dynamically links databases, visualization tools and reservoir/well simulators in order to assist the management process. This system called PERSEO (PErformance Reservoir StoragE Optimization) is aimed at enhancing efficiency by utilising a piece of intelligent fields application. The large amount of data from the 270+ wells of STOGIT equipped with SCADA systems monitoring real time gas rate and well head pressure, provides information for production/injection management. The challenge has been to extract the maximum information from the available database and computerize the process of updating well/reservoir performance automatically. This allows for faster monitoring, analysis and prediction of future well behaviour. A workflow was realised to transform available real time data into calibrated models in the following two steps: ■Design an algorithm capable of filtering stabilised gas rates and wellhead pressures for every injection/production period to extract data representative of the well performance■Automatic updating of well performance (IPR/VLP) model and matching the algorithm filtered data with theoretical and practica1 models (back pressure C,n equation). This paper describes the outline of the implemented PERSEO system, details of the computerized workflow along with sensitivities and the results obtained. |
