Design and Implementation of a Digital Twin for Live Petroleum Production Optimization
Fig . 10 : Field data process flow diagram and its corresponding simulation schematic .
Simulation parameter sensitivity was assessed using box plots similar to the examples shown in Figure 6 . It was observed that the spread of change ( 0-30 psi ) in separator pressure was higher than that of bottom-hole pressure ( 0-3 psi ) during set-point changes . This exploratory data analysis was helpful in the design of the simulation setup that led to some crucial decisions . For example , reduction of bottom hole pressure is considered to be one of the primary objectives 10 of artificial lift . In the nodal analysis performed through simulation software it is common to find literature with separator or the wellhead specified to be the end node 11 .
Such systems assume the end node pressure to be constant during a set-point change . In a previous version of this work , several challenges were observed in mimicking the out pressures of physical system due to the assumption that the simulated wellhead / separator pressures were to be held constant during a gas injection change . Based on the data from the box plots , it was demonstrated to be an incorrect assumption . T
o mimic the physical system that accommodates separator pressures to change with changes in gas injection rates , the gas sales and compressor nodes downstream of the separator were set to be the end nodes . This resulted in a better correlation between simulation output and physical sensor output during gas injection changes .
11
Camargo , Edgar & Aguilar , Jose & Rios , Addison & Rivas , Francklin & Aguilar-Martin , Joseph . ( 2008 ). Nodal analysis-based design for improving gas lift wells production .
IIC Journal of Innovation - 49 -