引用本文:李彬,明雨,郝一浩,陈宋宋,王隗东.基于融合FCN-TCN-LSTM的工业大用户可调节潜力分析模型[J].电力自动化设备,2023,43(7):
LI Bin,MING Yu,HAO Yihao,CHEN Songsong,WANG Weidong.Adjustable potential analysis model for large industrial users based on FCN-TCN-LSTM fusion[J].Electric Power Automation Equipment,2023,43(7):
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 4296次   下载 861  
基于融合FCN-TCN-LSTM的工业大用户可调节潜力分析模型
李彬1, 明雨1, 郝一浩1, 陈宋宋2,3, 王隗东2,3
1.华北电力大学 电气与电子工程学院,北京 102206;2.需求侧多能互补优化与供需互动技术北京市重点实验室,北京 100192;3.中国电力科学研究院有限公司,北京 100192
摘要:
综合考虑调控成本和价格激励的影响,开展工业大用户双向可调节潜力的分时段分析是提升新型电力系统负荷管理能力的关键。建立一种基于融合全卷积网络、时域卷积网络、长短期记忆网络的模型,以分析工业大用户可调节潜力。建立全卷积网络数据集重构模型,并基于典型负荷特性指标实现对具有高可调节潜力负荷数据的工业大用户的遴选;以高可调节潜力数据集为基础,建立改进时域卷积网络模型,对工业大用户进行调控成本影响下的可调节潜力分析测算。基于实际数据对所提模型进行验证,算例结果表明,所提模型可分析出工业大用户典型设备的可调节潜力,且模型的稳定性与精确度较高。
关键词:  需求响应  可调节潜力  工业设备调控  全卷积网络  时间卷积网络  长短期记忆网络
DOI:10.16081/j.epae.202209028
分类号:TM73
基金项目:国家电网有限公司总部科技项目(分布式“源荷储”资源聚合调控通信技术研究及应用)(5700-202258216A-1-1-ZN)
Adjustable potential analysis model for large industrial users based on FCN-TCN-LSTM fusion
LI Bin1, MING Yu1, HAO Yihao1, CHEN Songsong2,3, WANG Weidong2,3
1.School of Electrical and Electronic Engineering, North China Electric Power University, Beijing 102206, China;2.Beijing Key Laboratory of Demand Side Multi-energy Carriers Optimization and Interaction Technique, Beijing 100192, China;3.China Electric Power Research Institute, Beijing 100192, China
Abstract:
It is critical to promote the load management ability of new style power system to carry out time-phased analysis of bi-directional adjustable potential for large industrial users comprehensively considering the influence of regulation cost and price incentive. A model based on the fusion of fully convolutional network(FCN),temporal convolutional network(TCN) and long short-term memory network(LSTM) is established to analyze the adjustable potential of large industrial users. A dataset reconstruction model of fully convolutional network is established, and the selection of large industrial users with high adjustable potential load data is realized based on typical load characteristic indicators. Based on high adjustable potential data set, an improved temporal convolutional network is established, the adjustable potential of large industrial users is analyzed and measured under the influence of regulation cost. The proposed model is verified based on real data, and case results show that the proposed model can analyze the adjustable potential of typical equipments of large industrial users, and the model has high stability and accuracy.
Key words:  demand response  adjustable potential  industrial equipment regulation  FCN  TCN  LSTM

用微信扫一扫

用微信扫一扫