This website requires JavaScript.

Optimal scheduling of island integrated energy systems considering multi-uncertainties and hydrothermal simultaneous transmission: A deep reinforcement learning approach

Yang LiFanjin BuYuanzheng LiChao Long
Dec 2022
摘要
Multi-uncertainties from power sources and loads have brought significantchallenges to the stable demand supply of various resources at islands. Toaddress these challenges, a comprehensive scheduling framework is proposed byintroducing a model-free deep reinforcement learning (DRL) approach based onmodeling an island integrated energy system (IES). In response to the shortageof freshwater on islands, in addition to the introduction of seawaterdesalination systems, a transmission structure of "hydrothermal simultaneoustransmission" (HST) is proposed. The essence of the IES scheduling problem isthe optimal combination of each unit's output, which is a typical timingcontrol problem and conforms to the Markov decision-making solution frameworkof deep reinforcement learning. Deep reinforcement learning adapts to variouschanges and timely adjusts strategies through the interaction of agents and theenvironment, avoiding complicated modeling and prediction ofmulti-uncertainties. The simulation results show that the proposed schedulingframework properly handles multi-uncertainties from power sources and loads,achieves a stable demand supply for various resources, and has betterperformance than other real-time scheduling methods, especially in terms ofcomputational efficiency. In addition, the HST model constitutes an activeexploration to improve the utilization efficiency of island freshwater.
展开全部
图表提取

暂无人提供速读十问回答

论文十问由沈向洋博士提出,鼓励大家带着这十个问题去阅读论文,用有用的信息构建认知模型。写出自己的十问回答,还有机会在当前页面展示哦。

Q1论文试图解决什么问题?
Q2这是否是一个新的问题?
Q3这篇文章要验证一个什么科学假设?
0
被引用
笔记
问答