The increasing electrification of energy systems and the rapid expansion of renewable generation intensify congestion in transmission networks and lead to a growing reliance on redispatch measures. Redispatch ensures system security but entails rising costs and operational complexity, highlighting the need for market-based instruments that better internalize network constraints. One promising approach is the introduction of time- and location-dependent network charges that provide decentralized congestion signals to market participants. This study presents a novel coupling of the agent-based electricity market model PowerACE with an AC-Optimal Power Flow (OPF) network model to investigate whether spatio-temporal network charges can reduce the need for redispatch. The coupling combines the agent-based, zonal market representation of PowerACE with a nodal AC-OPF formulation that captures physical network constraints, losses, and voltage conditions. The models are linked in a sequential, iterative workflow that preserves the strengths of both approaches. Methodologically, an AC-OPF dispatch model is first solved to determine locational marginal prices (LMPs) at each network node for every simulated time step. These LMPs are decomposed to derive a nodal network charge component that reflects congestion -related cost signals. The resulting time- and location-specific network charges are then mapped to generation units and incorporated into variable generation costs within PowerACE. Market agents subsequently use these adjusted variable costs when forming their daily bids in the wholesale market. After market clearing in PowerACE, the resulting dispatch is transferred back to the network model, where the actual physical feasibility of the schedules is evaluated and the necessary redispatch measures are calculated. This procedure enables a direct comparison between scenarios with uniform network charges and scenarios with spatio-temporal network charges derived from LMPs, while holding generation capacities and demand constant. The analysis focuses on how network charges affect bidding behavior, market outcomes, congestion patterns, and redispatch volumes and costs. In particular, the study examines whether agents adapt their bids and dispatch decisions in a way that anticipates network constraints, thereby shifting generation away from congested areas and critical time periods. The results are expected to show that spatio-temporal network charges can partially internalize grid constraints in market outcomes and reduce redispatch needs, although their effectiveness depends on network topology, renewable generation patterns, and the accuracy with which nodal signals can be translated into zonal bidding incentives. By coupling PowerACE with an AC-OPF network model and embedding nodal congestion signals into agent-based bidding decisions, this work contributes to the literature on grid-aware market design and provides a systematic assessment of spatio-temporal network charges as a complement or alternative to redispatch-based congestion management.
Sandmeier et al. (Thu,) studied this question.