Stochastic risk-averse energy and reserve scheduling and pricing schemes with non-convexities and revenue caps

As weather-dependent renewables become the main source of generation in power systems, their intermittent supply nature requires system operators to account for uncertainty in market price formation and scheduling. In fact, renewable generation forecast errors cause high system-balancing costs. Conventional generators, on the other hand, have non-convex cost structures which incentivise misalignment with marginal pricing-based market outcomes. Out-of-market mechanisms, which blur price signals, are used to reimburse generators for their opportunity costs. Moreover, a surge in fuel prices may cause windfall profits for low-cost technologies and, to address this, the European Union recently considered introducing tailor-designed revenue caps. Leveraging on a risk-averse two-stage stochastic programming framework, this work enhances state-of-the-art models by providing a robust decision-making tool for system operators. Based on a primal-dual pricing, this tool endogenously balances energy and reserve scheduling related system costs with a stand-alone generator cost recovery under various schemes. The resulting market-clearing procedure is formulated as a Mixed Integer Bi-Linear Programming problem. A hybridisation of McCormick envelopes and binary expansion is proposed to enhance the computational capability to solve the problem. Three case studies illustrate the effectiveness of the proposed procedure.