文章摘要
郑伟民,叶承晋,张曼颖,等.基于Softmax概率分类器的数据驱动空间负荷预测[J].电力系统自动化. DOI: 10.7500/AEPS20181220008.
Weimin Zheng,Chengjin Ye,Manying Zhang, et al.Data-driven Spatial Load Forecasting Method Based on Softmax Probabilistic Classifier[J].Automation of Electric Power Systems. DOI: 10.7500/AEPS20181220008.
基于Softmax概率分类器的数据驱动空间负荷预测
Data-driven Spatial Load Forecasting Method Based on Softmax Probabilistic Classifier
DOI:10.7500/AEPS20181220008
关键词: 空间负荷预测  数据挖掘  地块  多元概率分类器  负荷曲线
KeyWords: Spatial load forecasting  data mining  land plot  Softmax  load curve
上网日期:2019-02-01
基金项目:国家自然科学基金资助项目(51807173);中国博士后科学基金资助项目(2018M640558);国家电网公司科技项目 (5211JY170015)
作者单位E-mail
郑伟民 国网浙江省电力有限公司 zheng_weimin@zj.sgcc.com.cn 
叶承晋 浙江大学电气工程学院 yechenjing@zju.edu.cn 
张曼颖 国网浙江省电力有限公司 zhangmanying@zj.sgcc.com.cn 
王蕾 国网浙江省电力有限公司 wang_lei1@zj.sgcc.com.cn 
孙可 国网浙江省电力有限公司 36955370@qq.com 
丁一 浙江大学电气工程学院 yiding@zju.edu.cn 
刘思 国网浙江省电力有限公司 candyls@zju.edu.cn 
摘要:
      本文提出了一种数据驱动空间负荷预测方法。首先,网格化体系下的功能地块被作为空间负荷预测的基本单元,并且通过多维指标体系进行属性描述;基于大量调研数据,通过数据挖掘方法对不同类型地块的空间负荷密度分布规律和负荷曲线典型形态进行提取;建立Softmax多元概率分类模型对未知地块的负荷水平类型进行匹配;自下而上对相邻地块负荷预测结果进行时域叠加,得到更大区域的预测信息,包括其负荷量和预测负荷曲线。算例仿真结果显示,本文提出的空间负荷预测方法在预测精度上有一定提升。
Abstract:
      A data-driven spatial load forecasting (SLF) method based on softmax classifier is proposed in this paper. Land plots are utilized as basic SLF units to describe available multi-attribute data in smart grids. Nonparametric kernel density estimation and adaptive clustering are utilized to aggregate typical values and curves of plot loads. The softmax classifier is introduced to forecast the unknown plot load quantities. Neighbor plot loads are summed up to obtain the estimated loads of larger areas based on clustered daily load curves. In this way, the maximal load, as well as its daily temporal distribution can be obtained at the same time. Case studies demonstrate that the proposed SLF is more applicable than benchmark approaches in accuracy. With the proposed method, net load can be balanced more precisely considering the integration of distributed generations and diversified loads.
查看全文   查看附录   查看/发表评论  下载PDF阅读器