On the computation of the probability distribution of the spot market price in a deregulated electricity market

Abstract

In a deregulated electricity market, short- and long term decisions will be based on the spot price of electricity rather than on the load. The hourly spot prices of electricity have so far shown evidence of being highly volatile. In this paper, we propose a stochastic model for the spot market price of electricity in which no historical data for the prices are required. Rather than directly modeling the price, we model the underlying stochastic processes of the marginal unit that determines the price of electricity under market competition. The deregulated market of electricity is still evolving and the set of generating units participating in a particular market can be constantly changing. A model that uses historically observed market prices may thus give inaccurate predictions. Under these circumstances, it is very difficult to collect empirical data to obtain an accurate forecast for the spot market price. Our price model represents the market by a power generation system and captures its fundamental stochastic characteristics. This model highlights the dependence of the price on temperature fluctuations. The model is also used to provide analytical approximations to the bivariate probability distribution of the spot prices at two different hours.