Robust management of Combined Heat and Power systems via linear decision rules

The heat and power outputs of Combined Heat and Power (CHP) units are jointly constrained. Hence, the optimal management of systems including CHP units is a multi-commodity optimization problem. Problems of this type are stochastic, owing to the uncertainty inherent both in the demand for heat and in the electricity prices that owners of CHP units receive for the power they sell in the market. In this work, we model the management problem for a coupled heat-and-power system comprising CHP plants, units solely producing heat as well as heat storages. We propose a robust optimization model including unit commitment, day-ahead power and heat dispatch as well as real-time re-dispatch (recourse) variables. This model yields a solution that is feasible under any realization of the heat demand within a given uncertainty set. Optimal recourse functions for the real-time operation of the units are approximated via linear decision rules to guarantee both tractability and a correct representation of the dynamic aspects of the problem. Numerical results from an illustrative example confirm the value of the proposed approach.