用组合优化+神经网络，高效筛选美国核电站候选地。

As clean energy demand grows to meet sustainability and net-zero goals, nuclear energy emerges as a reliable option. However, high capital costs remain a challenge for nuclear power plants (NPP), where repurposing coal power plant sites (CPP) with existing infrastructure is one way to reduce these costs. Additionally, Brownfield sites-previously developed or underutilized lands often impacted by industrial activity-present another compelling alternative. This study introduces a novel multi-objective optimization methodology, leveraging combinatorial search to evaluate over 30,000 potential NPP sites in the United States. Our approach addresses gaps in the current practice of assigning pre-determined weights to each site attribute that could lead to bias in the ranking. Each site is assigned a performance-based score, derived from a detailed combinatorial analysis of its site attributes. The methodology generates a comprehensive database comprising site locations (inputs), attributes (outputs), site score (outputs), and the contribution of each attribute to the site score. We then use this database to train a neural network model, enabling rapid predictions of nuclear siting suitability across any location in the United States. Our findings highlight that CPP sites are highly competitive for nuclear development, but some Brownfield sites are able to compete with them. Notably, four CPP sites in Ohio, North Carolina, and New Hampshire, and two Brownfield sites in Florida and California rank among the most promising locations. These results underscore the potential of integrating machine learning and optimization techniques to transform nuclear siting, paving the way for a cost-effective and sustainable energy future.