arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.09785eess.SYcs.AIcs.LGcs.SYmath.OC
收藏

多能源系统的模型预测控制和强化学习案例研究

Model-predictive control and reinforcement learning in multi-energy system case studies

ABB(ABB集团);Vrije Universiteit Brussel (VUB)(布鲁塞尔自由大学(VUB));KU Leuven(鲁汶大学);EnergyVille(能源城研究所)

Glenn Ceusters, Román Cantú Rodríguez, Alberte Bouso García, Rüdiger Franke, Geert Deconinck, Lieve Helsen, Ann Nowé, Maarten Messagie, Luis Ramirez Camargo

摘要:

模型预测控制(MPC)提供了一种最优控制技术,以建立和确保多能源系统的总运行成本保持在最小值,同时满足所有系统约束。然而,这种方法假定了一个适当的模型的基础系统动态,这是容易建模错误,不一定是自适应的。这就产生了相关的初始和持续的项目特定工程费用。在本文中,我们提出了一种基于策略和非策略的多目标强化学习(RL)方法,该方法不假设先验模型,将其与线性MPC(LMPC -反映当前实践,尽管非线性MPC表现更好)进行基准测试-两者都来自一般的最优控制问题,突出了它们的差异和相似之处。在一个简单的多能源系统(MES)配置案例研究中,我们表明,一个双延迟深度确定性政策梯度(TD 3)RL代理提供了匹配和超越完美的前瞻LMPC基准(101.5%)的潜力。而现实的LMPC,即不完美的预测,仅达到98%。而在更复杂的MES系统配置中,RL代理的性能通常较低(94.6%),但仍优于现实的LMPC(88.9%)。在这两个案例研究中,RL代理在使用季度与环境交互的2年训练期后表现优于现实LMPC。我们的结论是,强化学习是一种可行的多能源系统的最优控制技术,给予足够的约束处理和预训练,以避免不安全的相互作用和长时间的训练,在未来的基本工作中提出。

英文摘要:

Model-predictive-control (MPC) offers an optimal control technique to establish and ensure that the total operation cost of multi-energy systems remains at a minimum while fulfilling all system constraints. However, this method presumes an adequate model of the underlying system dynamics, which is prone to modelling errors and is not necessarily adaptive. This has an associated initial and ongoing project-specific engineering cost. In this paper, we present an on- and off-policy multi-objective reinforcement learning (RL) approach, that does not assume a model a priori, benchmarking this against a linear MPC (LMPC - to reflect current practice, though non-linear MPC performs better) - both derived from the general optimal control problem, highlighting their differences and similarities. In a simple multi-energy system (MES) configuration case study, we show that a twin delayed deep deterministic policy gradient (TD3) RL agent offers potential to match and outperform the perfect foresight LMPC benchmark (101.5%). This while the realistic LMPC, i.e. imperfect predictions, only achieves 98%. While in a more complex MES system configuration, the RL agent's performance is generally lower (94.6%), yet still better than the realistic LMPC (88.9%). In both case studies, the RL agents outperformed the realistic LMPC after a training period of 2 years using quarterly interactions with the environment. We conclude that reinforcement learning is a viable optimal control technique for multi-energy systems given adequate constraint handling and pre-training, to avoid unsafe interactions and long training periods, as is proposed in fundamental future work.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑