| 引用本文: | 吴迪,王韵楚,郁春雷,刘晟源,林振智,杨莉.基于高斯过程回归的工业用户需求响应潜力评估方法[J].电力自动化设备,2022,42(7): |
| WU Di,WANG Yunchu,YU Chunlei,LIU Shengyuan,LIN Zhenzhi,YANG Li.Demand response potential evaluation method of industrial users based on Gaussian process regression[J].Electric Power Automation Equipment,2022,42(7): |
|
| 摘要: |
| 精确的用户需求响应潜力评估有助于电力公司或者负荷聚合商合理调用需求侧资源，提高需求响应的实施成效，以及降低电力系统的负荷峰谷差。在此背景下，针对用电容量大且负荷规律性强的工业用户，提出了一种基于高斯过程回归的需求响应潜力评估方法。首先，构建了基于时序分解算法的工业用户负荷分解模型，并提出了负荷趋势性和周期性分量的可中断负荷特征提取方法以及工业用户的需求响应意愿特征提取方法；然后，基于提取的特征，构建了基于高斯过程回归的工业用户需求响应潜力评估模型；最后，以浙江省工业用户的实际需求响应数据为例，对所提需求响应潜力评估方法的有效性进行验证。仿真结果表明所提方法可以较为准确地评估工业用户的需求响应潜力，为电力公司或者负荷聚合商制定需求响应方案提供参考。 |
| 关键词: 需求响应 潜力评估 时序分解算法 可中断负荷 负荷特征 高斯过程回归 工业用户 |
| DOI：10.16081/j.epae.202206001 |
| 分类号:TM73 |
| 基金项目:国家自然科学基金联合基金重点支持项目(U2166206) |
|
| Demand response potential evaluation method of industrial users based on Gaussian process regression |
|
WU Di1, WANG Yunchu1, YU Chunlei2, LIU Shengyuan1, LIN Zhenzhi1, YANG Li1
|
|
1.College of Electrical Engineering, Zhejiang University, Hangzhou 310027, China;2.Zhejiang Huayun Information Technology Co.,Ltd.,Hangzhou 310012, China
|
| Abstract: |
| Accurate demand response potential evaluation of users is helpful for power companies or load aggregators to reasonably allocate demand-side resources, improve the implementation effect of demand res-ponse, and reduce the load peak-valley difference of power system. In this context, a demand response potential evaluation method based on Gaussian process regression is proposed for the industrial users with large electricity consumption and strong load regularity. Firstly, a load decomposition model of industrial users based on STL(Seasonal and Trend decomposition using Loess) algorithm is constructed, and the feature extraction method of interruptible load for load trend and periodic components is proposed as well as the feature extraction method of industrial users’ demand response willingness. Then, based on the extracted features, the demand response potential evaluation model of industrial users based on Gaussian process regression is constructed. Finally, taking the actual demand response data of industrial users in Zhejiang Pro-vince as an example, the validity of the proposed demand response potential evaluation method is verified. Simulative results show that the proposed method can accurately evaluate the demand response potential of industrial users and provide a reference for power companies or load aggregators to formulate demand res-ponse plans. |
| Key words: demand response potential evaluation STL algorithm interruptible load load features Gaussian process regression industrial users |
