Long-term uncertainty evaluation of pool electricity markets

Pool electricity markets are cleared under the strong assumption of having a perfectly known future; in real life, this is anything but true. The inability to predict the random parameters of the supply and the demand function introduces risk into the market clearing process. Therefore, the main interest is to minimize such risk by means of a trade-off of the mean and the variance of the social cost function. This paper considers random variations on the levels and on the slopes of the quadratic supply and demand functions. Correlation is considered between the corresponding coefficients of the supply and demand curves. By means of the mean-variance Markowitz theory, the risk introduced by these random variations is analyzed. A comprehensive analysis on the effects that the mean-variance Markowitz theory has on the nodal spot prices and on the point-elasticities of the supply and demand curves is made. The non-linear optimization model presented in this paper is validated through a three-, a six-, and a 21-node system.