Operation and Control of Prosumers to Provide Flexibility and Capacity Firming

Abstract

The global energy transition, driven by decarbonisation targets and the growth of distributed energy resources (DERs), is turning passive consumers into prosumers with rooftop PV–battery systems. These prosumers enhance flexibility but complicate the task of maintaining reliability and affordability. This thesis develops a bilevel framework that links system-wide decisions (upper level) with prosumer responses (lower level) to coordinate prosumer flexibility under market conditions, focusing on tariff design, system operation and capacity firming. Using a simplified 2025 Australian National Electricity Market model, it shows how prosumer virtual power plants can reshape load profiles, improve dispatchability and provide cost-effective decentralised firm capacity.

First, the thesis proposes a bilevel tariff model that co-designs dynamic import and export tariffs under net billing, where exports are credited below the retail import rate. Solved with a decomposition algorithm, the model shows that optimised asymmetric tariffs smooth net load, reduce coincident PV export peaks and better align prosumer actions with system needs.

Second, a bilevel production cost model (PCM) assesses the operational impacts of these tariffs. Unlike conventional PCMs that assume centralised control or symmetric tariffs, it represents autonomous prosumers driven by incentives for self-consumption and quantifies how policy-driven responses affect flexibility, highlighting reductions under strict net billing as battery capacities scale.

Finally, the thesis extends the bilevel PCM to evaluate the capacity firming potential of prosumer batteries. By introducing firming constraints at the upper level, the model allows operators to tap battery capacity after self-consumption, reducing reliance on gas generation and utility-scale storage. Overall, the framework links decentralised prosumer decision-making with system planning and offers tools for policymakers and planners in high-DER grids.