Valuation and investment of generation assets

Abstract

The re-regulation of electric power industry around the world has raised many new challenges for all stakeholders. This research is to valuate generation assets within reregulated electricity markets, both in short-term and long-term. The focus is to valuate operation flexibility under market uncertainties from the viewpoint of a Generation Company (GENCO). This research proposes to model the movements of electricity markets with Hidden Markov Model (HMM) driven by underlying market forces. An electricity market is modeled as a dynamic system evolving over time according to Markov processes. At any time interval, the electricity market can be in one state and transit to another state in the next time interval. The true market states are hidden from a market participant behind the incomplete observation. The observations, such as market-clearing price and quantity, are modeled to follow multiple probabilistic distributions. This research proposes to further decompose the market forces into physical and economic drivers if a specific electricity market employs Location Marginal Price (LMP) mechanism. The physical drivers include transmission network topology and generation technology. The economic drivers include fuel prices, demand uncertainties, and profit maximization of market participants with incomplete information. The decomposition captures the strengths of engineering-based production cost approach and mark-to-market stochastic approach. This research valuates generation assets with real option analysis. The value of generation assets is maximized based on the Hidden Markov Model (HMM) and newest observation of electricity markets. Such an optimization problem is formulated as Partially Oberserable Markov Decision Problem (POMDP). The solution of a POMDP provides a GENCO both the optimal operating policy and values of generation assets. The value of perfect and imperfect information is also identified. Investment in generation assets is also analyzed with real option. This research incorporates fuzzy sets and numbers to capture the fuzziness and possibilities of long-term electricity markets movements. Fuzzy sets and numbers provide the modeler flexibilities to