Short-term planning and bidding of hydropower production in the electricity market

The deregulation of the Nordic electricity market has caused new challenges for all the parties in the electricity production market. The continuous and rapid variation of the electricity price has caused difficulties since production planning is based on the price and optimal production planning is important to succeed in this market. Over 50 percent of the electricity in the Nordic Region is produced by hydropower which is the most important supply factor affecting the electricity price. In addition, hydropower is a very flexible method of production although dependent on the current hydrological situation. Production planning for hydropower can be divided into time periods of different length. The short-term planning covers the time period from one day to the following month and it is also affected by the mid-term planning parameters. It also involves the real time operation of the plants. The most important question for the short-term hydropower planning is how to divide the energy produced in the best possible way over the required time horizon. The aim of this thesis is to develop a model for the short-term hydropower production planning through which it is possible to optimize energy production, based on the price forecast and the predicted water level, within the next week. The river system covered in this work is owned by several parties and hydropower plants are operated by a third party. In addition to the normal restrictions the river system has additional restrictions listed in an agreement between the parties. The optimization model developed in this work is a linearized description of a nonlinear river system. The model finds an optimal way to discharge water between days and to allocate energy between hours within a planning horizon of one week, based on the price forecasts. In addition the thesis compares different heuristic bidding by which the energy produced is offered to the Nord Pool Spot models in order to overcome uncertainties related to the price forecast. The restrictions of the river system also limit the options when making bids. The results of the optimization program are compared with the results of the previous program and it was concluded that the program developed adds profitability and productivity. The new program allocates the discharges within a week considerably more aggressive and in more detail so that the results are economically optimal. The comparison of the bidding models is restricted mainly to hourly bids by which it is possible to get more profitable results compared to those with the previous model. /Kir1