Unit commitment risk evaluation with load uncertainty

In this research I propose both the Single-area Gauss-Markov Load Model and the Multi-area Gauss-Markov Load Model to model the load (ie., demand) in power systems. These load models fully describe the load uncertainty, time-correlation and area-correlation of the load. These models not only can be used in load forecasting but also can be used in operational reliability assessment and production cost simulation. The proposed models are verified by processing load data from a southwestern multi-area power system.;In this dissertation I develop an equivalent assisting unit approach for multi-area power systems using DC power flow and the stochastic load model. This approach provides a quantitative reliability assessment of multi-area power systems. The proposed approach is verified by Monte-Carlo simulation of a four-area power system using DC power flow. The rescheduling policy is extended to multi-area power systems for evaluation of unit commitment risk.;The PJM model for risk evaluation of unit commitment is extended to include a stochastic load model by using the proposed Single-area Gauss-Markov Load Model, which results in a proposed rescheduling policy in this dissertation. The proposed rescheduling policy is verified by simulating operation in the IEEE-RTS.