A fast and flexible stochastic dynamic programming model of the electricity market: VTT-EMM - structure and use

The Nordic power system and electricity market consists of Finland, Sweden, Norway and Denmark, which have a common market place. Most of the electricity produced and consumed goes through the Nord Pool Spot exchange. Our main objective was to develop a fast, agile, robust and transparent, mediumto long-term mathematical model of the stochastic development of the spot price of electricity on the Nord Pool market. In this report we present the Electricity Market Model of VTT Technical Research Centre of Finland, VTT-EMM. The model is a mathematical representation, based on stochastic dynamic programming, of the dynamics of the demand, production and market price of electricity in the power system of the Nordic countries. The power system is versatile and dominated by hydropower, and modelling the market is a demanding task. There are structural simplifications (one hydro reservoir, time aggregation) as well as data input generalisations (index series for power and CHP heat demand, modelled hydro inflow, and capacity aggregation) in the VTT-EMM model, which affect the results. Simplification and aggregation cut the computing time and data acquisition effort required and improve the flexibility of the model. Can we obtain reasonable market price and production estimates using a greatly simplified representation of the hydro system and the other power plants? Test results are compared with both spot prices and forward prices on the market. The results are considered to be good enough empirically, especially on an annual level. One weakness is that short-term spot price fluctuations are not captured very well. Further model development could, thus, include methods to integrate the short-term intermittency of wind power into the model and modelling of intra-market price areas. One strength can be found in the assessment of a more long-term future using multiple scenarios, even up to several hundred.