TY - THES

T1 - Development of a Comprehensive Real-Time Production Performance Monitoring Workflow

AU - Stoyanoff, Viktoria Milene

N1 - embargoed until null

PY - 2006

Y1 - 2006

N2 - In the field of hydrocarbon production, information from individual wells is the main performance indicator. Although many wells get instrumented with sensors and gauges, and data becomes available in real-time by using supervisory control and data acquisition systems, online rate measurements are still an exception. Since, back allocation procedures take days to months and therefore do not provide the necessary rate information in time, a workflow was developed to make advantage of real-time measured data in the field and provide the key performance indicators for evaluating production performance of wells in time. The required workflow includes the following steps: data quality control and integrity test, well test validation, continuous well rate estimation using real-time measurements, downtime detection, production loss calculations, calculating rate compliance, evaluating well productivity, and (semi) automated update of well models for a large number of wells. Further, the workflow should be applicable for artificial lifted wells and flexible to cope with missing parameters offering alternatives for the necessary calculations. Automation of the workflow itself or certain steps within the workflow and the required user interaction shall be evaluated. Conventional well models (e.g. PIPESIM) as well as data-driven methods like Neural Networks enable integrity of the system. Schlumbergers software packages DECIDE! and PIPESIM are used in this work.

AB - In the field of hydrocarbon production, information from individual wells is the main performance indicator. Although many wells get instrumented with sensors and gauges, and data becomes available in real-time by using supervisory control and data acquisition systems, online rate measurements are still an exception. Since, back allocation procedures take days to months and therefore do not provide the necessary rate information in time, a workflow was developed to make advantage of real-time measured data in the field and provide the key performance indicators for evaluating production performance of wells in time. The required workflow includes the following steps: data quality control and integrity test, well test validation, continuous well rate estimation using real-time measurements, downtime detection, production loss calculations, calculating rate compliance, evaluating well productivity, and (semi) automated update of well models for a large number of wells. Further, the workflow should be applicable for artificial lifted wells and flexible to cope with missing parameters offering alternatives for the necessary calculations. Automation of the workflow itself or certain steps within the workflow and the required user interaction shall be evaluated. Conventional well models (e.g. PIPESIM) as well as data-driven methods like Neural Networks enable integrity of the system. Schlumbergers software packages DECIDE! and PIPESIM are used in this work.

KW - DECIDE PIPESIM Backallocation forward rate estimation Network

KW - neural workflow

KW - production SCADA

KW - Echtzeit PIPESIM DECIDE Datenqualitätskontrolle Netzwerk

KW - neural Rückzuteilung Backallocation

M3 - Diploma Thesis

ER -