The use of decentralized Combined Heat and Power (CHP) plants is increasing since the high levels of efficiency they can achieve. Thus, to determine the optimal operation of these systems in dynamic energy-market scenarios, operational constraints and the time-varying price profiles for both electricity and the required resources should be taken into account. In order to maximize the profit during the operation of the CHP plant, this paper proposes an optimization-based controller designed according to the Economic Model Predictive Control (EMPC) approach, which uses a non-constant time step along the prediction horizon to get a shorter step size at the beginning of that horizon while a lower resolution for the far instants. Besides, a softening of related constraints to meet the market requirements related to the sale of electric power to the grid point is proposed. Simulation results show that the computational burden to solve optimization problems in real time is reduced while minimizing operational costs and satisfying the market constraints. The proposed controller is developed based on a real CHP plant installed at the ETA research factory in Darmstadt, Germany.

Peer Reviewed