| 引用本文: | 傅质馨,李潇逸,朱俊澎,袁越.基于马尔科夫决策过程的家庭能量管理智能优化策略[J].电力自动化设备,2020,40(7): |
| FU Zhixin,LI Xiaoyi,ZHU Junpeng,YUAN Yue.Intelligent optimization strategy of home energy management based on Markov decision process[J].Electric Power Automation Equipment,2020,40(7): |
|
| 摘要: |
| 在迅速发展的通信技术和泛在电力物联网建设的背景下，结合多种信息交互方式和人工智能技术可为提高家庭能量管理的智能化程度提供新的思路。提出一种结合实时信息交互的家庭能量管理智能优化策略。首先，给出了以用户用能费用为基础的马尔科夫决策过程模型，采用动态规划方法求解模型，重点在家庭用电设备调度过程中考虑实时电价信息和用户的随机行为等不确定因素的影响；在此基础上，结合事件触发机制有效提高家庭能量管理系统的运行效率，进而给出从家庭能量管理控制中心到用电设备的智能优化调度方法；最后，通过仿真算例证实了所提方法的有效性，结果表明其能在减少用户用电费用的同时给出满足用户用电需求的优化用电策略。 |
| 关键词: 家庭能量管理系统 马尔科夫决策过程 随机动态规划 实时电价 泛在电力物联网 智能优化 |
| DOI：10.16081/j.epae.202006029 |
| 分类号:TM732 |
| 基金项目:国家重点研发计划项目(2016YFB0900100) |
|
| Intelligent optimization strategy of home energy management based on Markov decision process |
|
FU Zhixin1,2, LI Xiaoyi1, ZHU Junpeng1,2, YUAN Yue1,2
|
|
1.College of Energy and Electrical Engineering, Hohai University, Nanjing 211100, China;2.Research Center for Renewable Energy Generation Engineering of Ministry of Education, Hohai University, Nanjing 211100, China
|
| Abstract: |
| Under the background of the rapid development of communication technology and the construction of Ubiquitous Power Internet of Things, the combination of multiple information interaction modes and artificial intelligence technologies can provide a new idea for improving the intelligence degree of home energy management. An intelligent optimization strategy of home energy management considering real-time information interaction is proposed. Firstly, the Markov decision process model based on user’s energy consumption cost is given and the dynamic programming method is used to solve the model. The influence of uncertain factors, such as real-time electricity price information, user’s random behavior, and so on, is consi-dered in the scheduling process of household electrical equipment. On this basis, combining with the event trigger mechanism, the operation efficiency of home energy management system is effectively improved, and then the intelligent optimization scheduling method from the home energy management control center to the electrical equipment is proposed. Finally, the effectiveness of the proposed method is proved by a simulation example, and the results show that it can provide the optimal power consumption strategy to meet the user’s demand while reducing the user’s power consumption cost. |
| Key words: home energy management system Markov decision process stochastic dynamic planning real-time electricity price Ubiquitous Power Internet of Things intelligent optimization |
