Toward Design of Risk-Based Real-Time Dispatch at Value

The increasing presence of uncertainties in today's power systems has brought new risks in its daily operation. In addition to the uncertainty of demand and renewable resources, the slow generator's deviation from the system operator's command in real-time dispatch has recently been identified as a new type of uncertainty which may lead to risks of increased stress on AGC as well as market inefficiency. Instead of the centralized framework previously proposed, in this paper we develop a risk-based market structure to manage the risk in a distributed way. The generators with deviations are charged with the corresponding redispatch cost caused by their deviations based on market-based prices. As a result, the slow generators internalize the potential risks of deviations in their simple bid functions submitted in real-time energy market. We show through simulations that the proposed risk-based market structure 1) rewards the right stakeholders/technologies for what they do by allocating the cost of risk to the causation; 2) reduces overall generation cost as well as risk in online power balancing.