Evaluating Congestion Management Schemes in Liberalized Electricity Markets Applying Agent-based Computational Economics

With the beginning of the liberalization process in the early 1990s, electricity markets headed into extensive changes. Moving away from vertically integrated monopolies, power delivery can nowadays be regarded as consisting of several services, mainly including generation, transmission and distribution. The unbundling of the value creation chain is a common feature of liberalized markets worldwide, although different countries follow different schemes for market organization. While e.g. parts of the United States, Singapore and Chile rely on centralized, pool-based trading concepts, in continental Europe decentralized structures seem to receive higher attention. Nonetheless, crucial to all proposed market designs is the transmission network. Electricity grids exhibit large economies of scale and must be physically interconnected for maximum trading efficiency, making the grid a natural monopoly within a defined region. Although transmission access and tariffs are subject to regulation, there is a growing need for market-based pricing concepts in transmission networks. Ideally, these concepts give not only correct economic incentives, but will also facilitate the physical operation of the network. In this regard, congestion management and pricing methodologies have been subject to intensive research, as these methodologies are crucial for the efficient operation of electricity markets. Looking at the worldwide implementation of congestion management schemes, different approaches can be identified. A pool-based or centralized market design allows for the use of locational marginal pricing, zonal pricing concepts such as market splitting or flow-based market coupling try to increase market liquidity by still relying on centralized elements, whereas explicit auctioning of transmission capacity exhibits the lowest degree of centralization, and thus, provides the greatest freedom to market participants. The different international implementation suggests that there is no definite first-best congestion management system. Every scheme