广东电力2016，Vol.29Issue(12)：10-14,5.DOI:10.3969/j.issn.1007-290X.2016.12.003
基于信息融合的电力大数据可视化预处理方法
Visualization Pretreatment Method for Electric Power Big Data Based on Information Fusion
摘要
Abstract
On the basis of studying fundamental features of electric power big data,this paper analyzes efficiency and performance in visualization pretreatment for electric power big data.By introducing information fusion method and in allusion to features of large quantities and various types of electric power big data,it uses big data example classification and fusion model based on combination of Map-Reduce and neural network to solve the problem of heterogeneous data fusion in coexistence of structuring,semi-structuring and non-structuring of big data.By using clustering and non-linear mapping capability of neural network and forward neural network model based on improved Hermite orthogonal polynomial,and combining Map-Reduce model,it is able to realize parallel processing for large-scale dataset and blend heterogeneous data of the same target into a unified description view so as to finish classification and fusion for multi-source heterogeneous data.Finally,example simulation is finished on Hadoop platform.关键词
电力大数据/数据可视化/信息融合/神经网络/模糊数学Key words
electric power big data/data visualization/information fusion/neural network/fuzzy mathematics分类
信息技术与安全科学引用本文复制引用
李刚,刘燕,宋雨,尹军..基于信息融合的电力大数据可视化预处理方法[J].广东电力,2016,29(12):10-14,5.基金项目
国家自然科学基金资助项目(51407076) （51407076）
河北省自然科学基金资助项目(F2014502050) （F2014502050）
河北省高等学校科学研究项目(Z2013007) （Z2013007）
中央党校基本科研业务费专项资金(2015ZD28) （2015ZD28）
