Supervisors: Athanasios Papakonstantinou, Stefanos Delikaraoglou, Pierre Pinson
in Energy Analytics & Markets group, Centre for Electric Power and Energy, Department of Electrical Engineering
Wind power generators trading their production in the day-ahead (spot) market are exposed to significant uncertainty due to price volatility and partial predictability of their actual output. In order to reduce their risk exposure, they can hedge their positions participating in the futures market which allows them to sell their energy production at fixed prices spanning a pre-specified time period, e.g., one day or one week before their participation in the day-ahead market. However, during periods of high spot market prices this strategy may result in loss of profits. Therefore, the goal of a wind power producer is to find the optimal involvement in each trading floor in order to maximize his expected profit over a given planning horizon, while controlling the risk of profit variability.
In this context, the main objective of this MSc project is for the student to explore the interconnection between the financial (e.g. NASDAQ commodities OMX) and electricity (day-ahead) markets, and to be able to formulate an optimal trading strategy for a wind power producer participating in both those markets.
This problem can be casted as a two-stage stochastic programming model where the first stage represents the participation in the financial market and the second stage represents the day-ahead market offers. This model can be further improved by considering a ‘rolling planning’ type of operation where the position in the futures market will be re-assessed based on updated wind power and price forecasts. The trading strategy will be evaluated using historical data from Nord Pool’s day-ahead market (Elspot) and financial power market contracts from NASDAQ as well as wind power forecasts and measurements optimization models may be implemented in GAMS, Python or Matlab depending upon the background of the successful candidate.
The expected results include relevant literature review, the formulation of the optimization problems and their implementation as well as their evaluation against a benchmark where a wind power producer trades his output only in the day-ahead market.
Power systems operations, energy systems modelling, energy markets, basics of optimization and statistics