Adaptive Trading in Continuous Intraday Electricity Markets for a Storage Unit

Bertrand, Gilles [UCL] Papavasiliou, Anthony [UCL]

The increasing integration of renewable resources in electricity markets has increased the need for producers to correct their trading position close to real time in order to avoid volatile real-time prices. The closest option to delivery time in European markets is to trade in the continuous intraday market. This market is therefore an attractive trading outlet for assets that target at extracting value from their flexibility. Trading in this market is challenging due to the multistage nature of the problem, its high uncertainty and the fact that decisions need to be reached rapidly, in order to lock in profitable trades. We model the trading problem of a storage unit in the Markov Decision Process framework. We present an approach based on policy function approximation for tackling the problem. We provide relevant parameters for defining our policy, and demonstrate the effectiveness of our approach by comparing it to the rolling intrinsic policy on real historical data. Our proposed approach outperforms the rolling intrinsic policy, which is commonly employed in practice for storage units, by increasing profitability by 17.8% on out-of-sample testing for a storage with perfect round-trip efficiency and by 13.6% for a storage unit with a round-trip efficiency of 81%.