The growing share of renewable generation increasingly pushes controllable generators out of the market. At the same time, the need for controllable capacity during periods of low renewable output remains. Capacity mechanisms have therefore become a common policy instrument to secure resource adequacy. However, procuring enough capacity in total does not ensure that it is located where the transmission network can deliver it during scarcity. This is especially relevant in zonal electricity markets, where internal network constraints are ignored during market clearing and corrected afterwards through redispatch. A locational signal in the capacity mechanism could improve the deliverability of procured capacity but may also increase capacity market expenditure. This thesis investigates how locational signals shape the trade-off between capacity procurement costs and congestion outcomes in a zonal electricity market.
The thesis develops a sequential two-stage model consisting of a market-equilibrium stage followed by a cost-based redispatch stage. The first stage combines a zonal energy market with a capacity mechanism, while the second stage restores network feasibility through cost-based redispatch. Because several market-equivalent investment patterns can lead to different congestion outcomes, the analysis does not rely on a single solver-selected equilibrium. Instead, PTDF-based exploration weights are used to identify spatially different siting outcomes with the same first-stage market result. The model is first verified on a small proof-of-concept system and then applied to a 34-node representation of the Dutch transmission network using eight representative days. Two case studies are considered: one in which most controllable capacity must still be built, and one in which the existing generation fleet already covers most of the system-wide requirement.
The results show that the locational signal changes where capacity is built more clearly than how much capacity is procured. In both case studies, zonal procurement redirects investment towards locally deficient areas and improves the ability of the system to restore network feasibility during scarcity. The effect on redispatch volume and cost is less consistent, while grid-related load shedding responds more clearly. The impact on capacity market expenditure depends strongly on the inherited generation fleet. When most capacity is newly built, the increase in expenditure remains limited. When only a small residual deficit remains, the locational design becomes considerably more expensive because procurement must take place in the deficient zone and accepted legacy capacity also receives the zonal clearing price. Cross-zonal participation can reduce these costs, but at the same time weakens the locational signal and increases reliance on the transmission network.
The findings show that aggregate resource adequacy is not sufficient when internal network constraints limit the delivery of procured capacity. Locational capacity procurement can improve deliverability, but it does not remove congestion and should not be treated as a substitute for grid expansion, bidding-zone reform, or other congestion-management measures. The design problem is therefore not only how much capacity should be procured, but how much of it must be local to remain useful during scarcity.
... Read more