| 引用本文: | 李昭昱,艾芊.分时电价下居民用户用电需求响应估计方法[J].电力自动化设备,2023,43(10):121-127 |
| LI Zhaoyu,AI Qian.Demand response estimation method of electricity consumption for residential customer under time of use price[J].Electric Power Automation Equipment,2023,43(10):121-127 |
|
| 摘要: |
| 在组织需求侧资源参与需求响应(DR)的过程中，除了应关注已参加DR的存量资源的管理，还应考虑如何实现对未参加DR的潜在增量用户的潜力分析。针对基于价格的需求响应(PBDR)，关注给定价格下用户响应量估计的关键环节，通过支持向量机回归模型，建立未参加和参加PBDR用户间用电规律的相似性，解决了DR事件起始和持续时间不固定、未参加PBDR用户自身的历史用电信息无法反映其DR特性等难题，实现在假设施加价格信号下，对未参加PBDR用户的用电响应情况及需求弹性的量化估计。基于伦敦地区智能电表实际数据进行实验以验证所提方法的优越性，结果表明所提方法的估计效果优于基于相似日的方法，可为用户筛选提供一定依据。 |
| 关键词: 需求响应 分时电价 居民用户 需求弹性 支持向量机回归 |
| DOI：10.16081/j.epae.202308031 |
| 分类号:TM73 |
| 基金项目:国家重点研发计划项目(2021YFB2401204) |
|
| Demand response estimation method of electricity consumption for residential customer under time of use price |
|
LI Zhaoyu, AI Qian
|
|
Key Laboratory of Control of Power Transmission and Conversion, Ministry of Education, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
|
| Abstract: |
| In the process of organizing demand-side resources participating in demand response(DR),it is essential not only focusing on the management of the existing DR-engaged resources but also considering the potential analysis of potential incremental customers who have not participated in DR. In view of the price-based demand response(PBDR),paying attention to the key levels of the customer response quantity under the given price, the similarity of electricity consumption law between customers not participating and participating PBDR through support vector machine regression model is built. And the challenges associa-ted with the varying initial and duration time of DR events, as well as the historical electricity consumption information of non-participating PBDR customers cannot reflect their DR characteristics are addressed, so that the quantitative estimation of electricity consumption response situation and demand elasticity for non-participating PBDR customers can be achieved under the presumption of applying price signals. The superio-rity of the proposed method is validated through experiments based on the actual smart meter data in London. The results demonstrate that the estimation performance of the proposed method is better than the method based on similar days, thereby providing a solid basis for the selection of customers. |
| Key words: demand response time of use price residential customer demand elasticity support vector regression |
