基于混沌循环测量矩阵的新型能源互联网的同域采样方法  

Unified domain sampling method for new energy internet based on chaotic circulant measurement matrix

作　　者：杨杉 谭博 郭静波[1] Yang Shan;Tan Bo;Guo Jingbo(Department of Electrical Engineering,Tsinghua University,Beijing 100084,China)

机构地区：[1]清华大学电机工程与应用电子技术系,北京100084

出　　处：《可再生能源》2023年第3期370-376,共7页Renewable Energy Resources

基　　金：国家电网有限公司总部管理科技项目资助(面向5G的信息能源融合技术与装备研制5206002000DB)。

摘　　要：随着新型能源互联网监测单元数量和类型的不断增多,使得采样率和采样数据量不断增加。为减少采样数据量,减轻传统数据采样负担,文章基于压缩感知理论,提出一种同域采样方法(Unified Domain Sampling,UDS),其本质在于利用压缩感知理论,实现不同类型数据的同域空间投影,通过基于同域空间的投影压缩采样来降低采样数据量。在基于压缩感知的同域空间投影中,利用混沌的伪随机性和循环矩阵的易构性,建立了混沌循环测量矩阵,并基于猫(Cat)映射实现了具体的混沌循环测量矩阵,以及对应的同域采样方法。实验结果表明,文章同域采样方法具有优秀的数据压缩和恢复能力,并且所采用的混沌循环测量矩阵优于传统的高斯循环测量矩阵。As the number and type of monitoring units increase in new energy internet,the sampling rate and amount of sampling data also tends to increase.Therefore,we propose unified domain sampling(UDS),a data sampling method based on compressive sensing principle,in order to reduce the amount of sampling data and relieve the burden of traditional data sampling.The UDS tries to take advantage of compressive sensing principle by projecting the different types of data into the unified space for compression sampling,which requires fewer samples being taken.This method constructs a chaotic circulant measurement matrix by incorporating pseudorandom chaotic sequence with manageable circulant matrix based on unified domain spatial projection of compressed sensing.Construction method of chaotic circulant measurement matrix based on Cat chaotic mapping and the unified domain sampling method related to this measurement matrix are studied.Experiments results show that the proposed UDS method has excellent performance in data compression and recovery,the adopted chaotic circulant measurement matrix is also advantageous to conventional gaussian circulant measurement matrix.

关 键 词：新型能源互联网 数据采样 循环矩阵 混沌原理 CAT映射 

分 类 号：TK51[动力工程及工程热物理—热能工程]