S3GRAF

Introduction

S3connect couples surface network models in GAP to one or more reservoir simulation models. The link uses full two-way coupling between surface and subsurface, allowing the creation of production forecasts that include the full complexity of the surface pipeline dynamics and detailed reservoir modelling, including heterogeneity.

S3connect provides a very simple interface for linking the GAP model to reservoir models, including the connection of corresponding wells in the two systems. For fully compositional systems, the mapping between the components can also be easily specified. Setup of a coupled system is therefore very straightforward, allowing engineers to concentrate solely on engineering problems.

Benefits

  • Ease of use for both production and reservoir engineers
  • Couples to multiple simulators
  • Cost effectiveness: couples to standard Eclipse and does not require the purchase of additional simulators
  • Efficient use of simulator licences
  • Robust
  • Straightforward IT setup, with minimal IT overhead

Applications

  • Production forecasting and optimisation
  • More accurate reservoir simulation with realistic surface constraints
  • Gas lift optimisation studies
  • Backout analysis
  • Waterflood management
  • Asset management and planning

Key Features

  • Supports Eclipse, E300, FrontSim and VIP
  • Connects multiple reservoir models
  • Remote runs under UNIX/LINUX
  • LSF job scheduling
  • Automatic IPR curve generation
  • Supports all GAP well types (including gas lifted)
  • Supports fitted and table-lookup GAP IPR types

Technical Details

Full bi-directional coupling between surface and sub-surface is achieved by transferring the reservoir conditions to GAP in the form of well inflow performance data, which is used in the solution of the surface network. The surface network solution is transferred to the reservoir models in the form of well controls for the next timestep. Thus, the run proceeds by taking data from the reservoir simulation(s) at each schedule step, passing this data to GAP and solving the surface network (with or without optimisation). The network solve results in constraints on rates, BHPs or THPs which are then passed back to the simulator: a restart of the simulator including the new constraints is initiated, running up to the next schedule step (GAP network solve). This process is repeated for each step in the prediction schedule.

The key to coupling a reservoir simulation model to a surface network is the generation of inflow performance data for each well that is coupled. S3connect has a number of options for controlling the generation of IPR data, and can provide the IPR data to GAP in several formats (three-phase tables, single-phase tables or parametric forms). It is important to note that S3connect always couples at the sandface (IPR data); wellhead performance curves are not generated.

A number of options are provided for controlling the parameters passed between GAP and the reservoir simulator, and it is possible to modify the parameters that are passed from the simulator to GAP to account for known errors in the reservoir model that may have a negative impact on the accuracy of the surface network calculation (such as a less than perfect history match giving rise to errors in water cut for some wells in the simulation). This ‘well calibration’ functionality is also useful for performing simple ‘what if’ scenarios without the need to make changes to the simulation model -  for example, the effect of watering out of wells over different timescales can be investigated.

 

S3GRAF - text to go here

John Logie Baird Programme WinnerQueens Award Winner