Search

Power system decarbonisation is driving the rapid deployment of renewable energy sources (RES) like wind and solar at the transmission and distribution level. Their differences from the synchronous thermal plants they are displacing make secure and efficient grid operation challenging. Frequency stability is of particular concern due to the current lack of provision of frequency ancillary services like inertia or response from RES generators. Furthermore, the weather dependency of RES generation coupled with the proliferation of distributed energy resources (DER) like small-scale solar or electric vehicles permeates future low-carbon systems with uncertainty under which legacy scheduling methods are inadequate. Overly cautious approaches to this uncertainty can lead to inefficient and expensive systems, whilst naive methods jeopardise system security. This thesis significantly advances the frequency-constrained scheduling literature by developing frameworks that explicitly account for multiple new uncertainties. This is in addition to RES forecast uncertainty which is the exclusive focus of most previous works. The frameworks take the form of convex constraints that are useful in many market and scheduling problems. The constraints equip system operators with tools to explicitly guarantee their preferred level of system security whilst unlocking substantial value from emerging and abundant DERs. A major contribution is to address the exclusion of DERs from the provision of ancillary services due to their intrinsic uncertainty from aggregation. This is done by incorporating the uncertainty into the system frequency dynamics, from which deterministic convex constraints are derived. In addition to managing uncertainty to facilitate emerging DERs to provide legacy frequency services, a novel frequency containment service is designed. The framework allows a small amount of load shedding to assist with frequency containment during high RES low inertia periods. The expected cost of this service is probabilistic as it is proportional to the probability of a contingency occurring. The framework optimally balances the potentially higher expected costs of an outage against the operational cost benefits of lower ancillary service requirements day-to-day. The developed frameworks are applied extensively to several case studies. These validate their security and demonstrate their significant economic and emission-saving benefits.