Optimal Control of Power Generation Systems using Realistic Object-Oriented Modelica Models

Equation-based, object-oriented dynamic modelling of power generation systems using Modelica is now well-established for simulation purposes. To which extent it can be used for optimal control purposes is still an open research question. In this paper, a realistic Modelica model of a combined cycle plant is developed using Modelica, and then used to formulate and solve the nonlinear control problem of optimal start-up with thermal stress constraints using three different methodologies, all based on direct collocation, implemented in two different tools. The model used in this paper is much more accurate than the one used in previous similar studies. The positive result obtained allow to gain confidence on potential of direct use of Modelica models for optimal power plant control studies.