In deregulated electricity markets, Generation Companies (GENCOs) are exposed to substantial financial risk due to volatile and uncertain electricity prices. Traditional generation asset valuation approaches, which rely primarily on expected profit, fail to adequately capture downside risk under market uncertainty. This study proposes an integrated risk-aware framework for generation asset valuation by embedding Value-at-Risk (VaR) into a Price-Based Unit Commitment (PBUC) model. VaR is employed to quantify potential profit losses at different confidence levels, enabling GENCOs to explicitly assess downside exposure associated with electricity price fluctuations. Spot price uncertainty is modeled using the Delta-Normal approach based on historical PJM market data. The resulting nonlinear mixed-integer optimization problem is solved using an Improved Immune Algorithm (IIA) enhanced with the Taguchi Method to improve convergence stability and solution diversity. Case studies on the IEEE 15-unit system demonstrate that the proposed IIA consistently outperforms conventional evolutionary algorithms in terms of profitability, robustness, and convergence reliability. The VaR analysis further reveals pronounced left-tail risk in profit distributions, particularly during peak-load periods, highlighting the importance of risk-adjusted commitment strategies. The proposed framework provides a practical decision-support tool for GENCOs to balance profitability and downside risk in competitive electricity markets.
Chen et al. (2026) studied this question.