A multi-stage stochastic optimization approach to optimal bidding on energy markets

One of the most challenging tasks for an energy producer is represented by the optimal bidding on energy markets. Each eligible plant has to submit bids for the spot market one day before the delivery time and bids for the ancillary services provision. Allocating the optimal amount of energy, jointly minimizing the risk and maximizing profits is not a trivial task, since one has to face several sources of stochasticity, such as the high volatility of energy prices and the uncertainty of the production, due to the deregulation and to the growing importance of renewable sources. In this paper the optimal bidding problem is formulated as a multi-stage optimization problem to be solved in a receding horizon fashion, where at each time step a risk measure is minimized in order to obtain optimal quantities to bid on the day ahead market, while reserving the remaining production to the ancillary market. Simulation results show the optimal bid profile for a trading day, based on stochastic models identified from historical data series from the Italian energy market.