The paper proposes a two-stage adaptive robust optimization model for the security-constrained unit commitment (SCUC) problem, addressing the challenges posed by increasing supply and demand uncertainties due to the integration of variable generation resources and price-responsive demand. The model is designed to be more practical compared to conventional stochastic programming approaches, as it only requires a deterministic uncertainty set rather than a probability distribution. The proposed model ensures robustness against all possible realizations of the modeled uncertainty and incorporates critical constraints such as network, ramp rate, and transmission security constraints. A practical solution methodology based on a combination of Benders decomposition and outer approximation techniques is developed. Extensive numerical experiments on a large-scale power system operated by the ISO New England demonstrate the economic and operational advantages of the proposed model over traditional reserve adjustment approaches. The results show that the adaptive robust model achieves lower average dispatch and total costs, significantly reduces cost volatility, and is more robust to different probability distributions of load. The advantages of the adaptive robust model are further highlighted in numerical tests with higher levels of net injection uncertainty.The paper proposes a two-stage adaptive robust optimization model for the security-constrained unit commitment (SCUC) problem, addressing the challenges posed by increasing supply and demand uncertainties due to the integration of variable generation resources and price-responsive demand. The model is designed to be more practical compared to conventional stochastic programming approaches, as it only requires a deterministic uncertainty set rather than a probability distribution. The proposed model ensures robustness against all possible realizations of the modeled uncertainty and incorporates critical constraints such as network, ramp rate, and transmission security constraints. A practical solution methodology based on a combination of Benders decomposition and outer approximation techniques is developed. Extensive numerical experiments on a large-scale power system operated by the ISO New England demonstrate the economic and operational advantages of the proposed model over traditional reserve adjustment approaches. The results show that the adaptive robust model achieves lower average dispatch and total costs, significantly reduces cost volatility, and is more robust to different probability distributions of load. The advantages of the adaptive robust model are further highlighted in numerical tests with higher levels of net injection uncertainty.