CAREER: Integrated Engineering and Economic Modeling of Problems in Deregulated Electricity Networks

Georgia Tech Research Corporation, Atlanta GA

Investigators

Abstract

The focus of the proposed research is to develop parsimonious electricity market models and to integrate asset valuation models with probabilistic simulation models of an overall power system. Tremendous market volatility in electricity prices has been witnessed in recently established power markets. Volatile electricity price not only puts pressure on efficient operations of power generation assets, but also increases the importance of having realistic power market models for the purpose of evaluating real assets and performing risk management tasks in quantifying and controlling operational and financial risks. Two competing approaches are now being used for modeling electric power market behaviors for the above mentioned purposes: The "fundamental approach", which emphasizes on the engineering aspects of power markets, relies on system simulation and market operation to determine market prices; and the other so-called "technical approach', which is economically oriented, involves the direct modeling of the stochastic behavior of market prices from historical data and fundamental analysis. While the fundamental approach provides finer details about electricity markets in specific scenarios, it is computationally prohibitive due to the large number of scenarios it entails. The proposed project will combine the strengths of the two approaches by developing specific stochastic models for electricity prices that are calibrated using a probabilistic simulation of an overall electric power system. The methodology and analytical models developed during this project can make a significant impact on the divestiture process and the evaluation of new projects by power system planners and marketers in the deregulated power industry. Costs due to discrepancies in capacity valuation resulting from differing valuation methodologies and assumptions of power price dynamics are estimated to be in the millions. To lower such costs, this approach is expected to yield realistic power asset valuation methodologies that properly reflect the reliability of the power systems.