Quantifying Spinning Reserve in Systems With Significant Wind Power Penetration

The traditional unit commitment and economic dispatch approaches with deterministic spinning reserve requirements are inadequate given the intermittency and unpredictability of wind power generation. Alternative power system scheduling methods capable of aggregating the uncertainty of wind power, while maintaining reliable and economic performance, need to be investigated. In this paper, a probabilistic model of security-constrained unit commitment is proposed to minimize the cost of energy, spinning reserve and possible loss of load. A new formulation of expected energy not served considering the probability distribution of forecast errors of wind and load, as well as outage replacement rates of various generators is presented. The proposed method is solved by mixed integer linear programming. Numerical simulations on the IEEE Reliability Test System show the effectiveness of the method. The relationships of uncertainties and required spinning reserves are verified.