计算机应用与软件2026，Vol.43Issue(5)：219-226,8.DOI:10.3969/j.issn.1000-386x.2026.05.029
面向热电联产经济调度优化的自适应噪声注入PPO强化学习算法
ADAPTIVE NOISE INJECTION PPO REINFORCEMENT LEARNING ALGORITHM FOR ECONOMIC SCHEDULING OPTIMIZATION OF COGENERATION
摘要
Abstract
In order to deal with the current urgent energy crisis,improve the utilization rate of energy,this paper proposes an adaptive noise injection-based proximal policy optimization algorithm for economic optimization scheduling in cogeneration systems.By introducing adaptive noise in the clipping range,this method allowed the policy to dynamically adjust the corresponding clipping range during the policy training,enhancing the exploration capability and improving the economic efficiency of the model.The experiments show that compared with the traditional proximal policy optimization algorithm,the improved algorithm in this paper has lower energy consumption and lower operating cost in the overall energy scheduling task.关键词
热电联产系统/优化调度/强化学习/近端策略优化Key words
Cogeneration system/Optimal scheduling/Reinforcement learning/Proximal policy optimization分类
信息技术与安全科学引用本文复制引用
朱平,黄云峰..面向热电联产经济调度优化的自适应噪声注入PPO强化学习算法[J].计算机应用与软件,2026,43(5):219-226,8.基金项目
上海市2021年度"科技创新行动计划"科技支撑碳达峰碳中和专项(21DZ1207302) （21DZ1207302）
国家自然科学基金青年科学基金项目(51607111). （51607111）
