Inclusion of frequency stability constraints in unit commitment using separable programming

Abstract

In this paper we address the problem of frequency stability in the unit commitment (UC) optimisation process. We include a set of appropriately defined frequency stability constraints in the UC problem formulation for operational planning scenarios in advance of real-time operation. Consequently, we cover the system against the loss of the largest infeed under the N−1 security criterion.The main contribution of our work consists of using the method of separable programming to incorporate a linearised Frequency Nadir constraint into the UC problem. In our work, the UC problem is formulated as a mixed-integer linear program (MILP). This renders a fast convergence in the solution for a system such as the the three area IEEE RTS-96 system. Meanwhile, we have included the possibility of synthetic inertia provision from the wind farms, which helps to increase the available inertia in the system before a generation outage. The simulations are run using an extended version of MATPOWER tailored for solving UC problems (MOST) which is run in MATLAB.