The Promise of the Virtual Plant
By Martin Berutti on 11.09.09
Mimic Simulation

Millions of dollars have been spent in the process industries in the quest for off-line simulation systems that duplicate the environment of the real automation system and process. This quest for the “virtual plant” has not always gone well.

In many cases, the time and money spent developing off-line simulation systems has been wasted due to users and simulation vendors’ focus on model accuracy and perceived fidelity instead of good process dynamics and system integrity. This is an unfortunate reality for many users, but there is a better way. The development of off-line simulation systems that realistically duplicate the automation system and process responses can be accomplished with a modest investment of time and money. And the business results are significant. The off-line “virtual plant” system can be used effectively to:

  • Test automation system application software during normal and upset conditions.
  • Train operators on the effective use of the automation system during normal and upset conditions.
  • Test the automation platform for failure responses during emergency conditions that are too dangerous to test on an operating plant.
  • And finally, test new control strategies, recipes, additions, and other modifications to the automation system.

In many cases the obstacle to implementing an effective simulation environment that provides business and operational benefits can be overcome. It is overcome with a proper approach to the selection of the simulation platform and the manner in which the simulation is developed.

Greg McMillan in the August 2005 issue of Control, clearly articulates this wrong approach, in his article, “What If? Virtual Plant Reality”. Greg states that while “…good physical property data and a high fidelity simulator can yield an accurate process control output for a process input at steady state”, “..empty vessels, zero flows, non-equilibrium conditions, and process non-self-regulation are not properly addressed.” Greg adds, “A new definition of model fidelity is needed that emphasizes the dynamic response of a change in process variables to changes in the manipulated or disturbed variables”. He concludes the article with, “If we define and seek fidelity in what a control system and operator sees the promise of the virtual plant can be fulfilled”.

To summarize, the significant business and operational benefits associated with a “virtual plant” simulation system can be realized if a different approach to simulation system development is taken. This approach changes the focus from simulation fidelity to accurate process dynamics and automation system integrity.

The approach can be summarized as follows.

  • Primary focus on accurate dynamic response of control loops through the use of first order filter plus dead-time lags. This is classic process control modeling but it is imperative that this foundation be implemented and executed correctly.
  • Second, simulation of device characteristics for sensors, pumps, valves that reflect the characteristics of the installed equipment. Valve hysteresis, pump curves, thermal transport lags should display dynamics that match the installed plant equipment.
  • Third, selective application of high-fidelity models to unit operations where they can bring additional operational value. This is usually required in highly-integrated, continuous processes. Selective application allows this to be applied to some unit operations without impacting the cost and complexity of the entire simulation.
  • Finally, the development of the process modeling environment must be done while protecting the integrity of the automation system application software.
Virtual Plant / Control System using Mimic and DeltaV Simulate

If this approach is followed for the implementation of simulation systems, significant operations and business benefits can be achieved. Mimic Simulation Software was developed to allow the user to build a virtual plant and IO system to compliment virtual control systems like Emerson’s DeltaV Simulate System. When used together, as described above, Mimic and DeltaV Simulate deliver the promise of the “virtual plant”.

Other Resources to Review

ARC Position Paper - Using Simulation to Optimize Results of Automation Projects
Control System Integrity and Operator Training
Understanding Simulation Fidelity

I look forward to your comments, questions, or suggestions.
Hope to hear from you soon.

Mart Berutti, 11/09/09