用强化学习优化清洁能源分配，让低收入社区用电更公平。
Optimized Renewable Energy Planning MDP for Socially-Equitable Electricity Coverage in the US
- 构建马尔可夫决策模型，兼顾预算、需求和弱势群体用电需求。
- 使可再生能源占比达32.9%，低收入地区断电人群减少55%。
- 证明公平分配不降低系统性能，适合政策制定者参考。
传统电网阻碍可再生能源接入，加剧用电不平等，低收入社区断电时间更长。本研究提出马尔可夫决策过程（MDP）框架，优化清洁能源分配并明确纳入社会公平考量。模型融合预算约束、用电需求波动及美国八大城市的社经脆弱性指标，评估多种政策方案在清洁电力转型中的公平性表现。数值实验对比了随机分配、贪婪扩张、专家经验规则与蒙特卡洛树搜索基线。结果表明，以公平为导向的优化策略可实现32.9%的可再生能源渗透率，同时使服务不足的低收入人口减少55%。专家策略收益最高，而蒙特卡洛树搜索基线在显著更低预算下仍具竞争力，证明在不牺牲整体性能的前提下，可实现清洁能源资源的公平分配，为气候目标与包容性电力基础设施融合提供可行路径。
原文摘要 · Abstract (English)
Traditional power grid infrastructure presents significant barriers to renewable energy integration and perpetuates energy access inequities, with low-income communities experiencing disproportionately longer power outages. This study develops a Markov Decision Process (MDP) framework to optimize renewable energy allocation while explicitly addressing social equity concerns in electricity distribution. The model incorporates budget constraints, energy demand variability, and social vulnerability indicators across eight major U.S. cities to evaluate policy alternatives for equitable clean energy transitions. Numerical experiments compare the MDP-based approach against baseline policies including random allocation, greedy renewable expansion, and expert heuristics. Results demonstrate that equity-focused optimization can achieve 32.9% renewable energy penetration while reducing underserved low-income populations by 55% compared to conventional approaches. The expert policy achieved the highest reward, while the Monte Carlo Tree Search baseline provided competitive performance with significantly lower budget utilization, demonstrating that fair distribution of clean energy resources is achievable without sacrificing overall system performance and providing ways for integrating social equity considerations with climate goals and inclusive access to clean power infrastructure.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。