Predictive-Control for challenging industrial processes
Some industrial processes are operated close to constraints and/or have challenging dynamics including strong non-linearities, strong couplings between variables and significant, varying dead times. In case of such challenging plants, the tuning of a PID controller is difficult and its performance limited. In this case, Predictive-Control may provide superior closed loop performance and allow for operation closer to the constraint.
Predictive-Control is a control technique where process model is explicitly used to predict the evolution of controlled variables and this prediction is used to optimize controlled inputs in real-time. The computational cost of the optimizations may be high for more complex plants, however due to recent advances in computer hardware and optimization algorithms, Predictive-Control it is now applicable to a wide range of systems, like the exotic Superfluid Helium Cryogenic Circuit at the Large Hadron Collider (LHC) Arcs (www.predictive-control.com/lhc).
For more information about Predictive-Control and possible applications, please contact me at rafal.noga at predictive-control.org or at the swiss mobile phone 0041 76 268 83 47. I speak English, German, French, Spanish and Polish.
I have graduated in 2007 with a Master's degree in control engineering and since then I have been developing first principles, thermo-hydraulic, distributed parameters, numerical models and optimization algorithms for non-linear estimation and control.