基于空调自适应修正模型的户用微电网能量优化

发布时间：2018-05-14 17:00

本文选题：户用微电网 + 空调负荷　；参考：《电力系统自动化》2017年15期

【摘要】：空调负荷作为居民用户中的主要负荷,成为户用微电网能量管理的重要对象。以空调负荷为研究重点,结合户用微电网中其他可控负荷及储能设备,提出一种计及空调模型自适应修正的户用微电网能量管理策略。针对空调所属建筑物热力学模型中空调能效比系数、建筑物参数不易获取的问题,提出根据空调在当前场景下运行的历史数据,用遗传算法拟合得到适应当前场景的更为准确的热力学模型参数。针对室内人员活动、其他冷热型负荷工作状态改变等环境动态变化因素导致室内温度偏离优化温度的问题,用Q学习算法对空调设置温度进行在线自适应修正。最后,通过算例和实验证明了所提策略可以适应不同场景和环境的动态变化,提高了能量管理的适用性和准确性。
[Abstract]:Air conditioning load, as the main load of residential users, has become an important object of energy management in household microgrid. With the emphasis on air conditioning load and other controllable load and energy storage equipment in household microgrid, an energy management strategy for household microgrid considering adaptive modification of air conditioning model is proposed. In order to solve the problem that the coefficient of energy efficiency ratio and building parameters are difficult to obtain in the thermodynamic model of the building, the paper puts forward the historical data of air conditioning running in the current scene. A more accurate thermodynamic model parameter suitable for the current scene was obtained by genetic algorithm fitting. In order to solve the problem that indoor temperature deviates from the optimal temperature caused by dynamic environmental factors such as indoor personnel activities and other changes in the working state of cooling and heating load, the Q learning algorithm is used to modify the air conditioning setting temperature on line. Finally, examples and experiments show that the proposed strategy can adapt to the dynamic changes of different scenarios and environments, and improve the applicability and accuracy of energy management.
【作者单位】：东南大学电气工程学院;南京师范大学电气与自动化工程学院;国网江苏省电力公司电力科学研究院;
【基金】：江苏省科技计划项目(BE2015012-1) 国家电网公司科技项目(SGTYHT/14-JS-188) 江苏省普通高校研究生科研创新计划项目(SJLX15_0050)~~
【分类号】：TM727

【参考文献】