电力通信链路模拟数据智能融合方法  

Intelligent data fusion method for power communication link simulation

作　　者：李勇[1] 韩俊飞[1] 李秀芬[1] 王鹏[1] 王蓓[1] LI Yong;HAN Junfei;LI Xiufen;WANG Peng;WANG Bei(Institute of Information and Communication Technology,Inner Mongolia Electric Power Research Institute,Hohhot 010020,Inner Mongolia,China)

机构地区：[1]内蒙古电力科学研究院信息通信技术研究所,内蒙古呼和浩特010020

出　　处：《沈阳工业大学学报》2023年第6期710-715,共6页Journal of Shenyang University of Technology

基　　金：国家自然科学基金项目(61033013)。

摘　　要：为了提高电力通信系统数据融合时的节点存活率,提高网络连通性,提出了基于异构数据源的电力通信链路模拟数据智能融合方法。利用最小二乘残差估计法辨别噪声数据,识别异构数据中的不良数据。通过k-means聚类获取目标平均值,实现数据离散化并消除噪声。使用频繁项集的关联规则确定置信度阈值并挖掘不低于该阈值的数据。采用深度受限玻尔兹曼机算法将不同类型模拟数据映射到同一矢量空间内,实现智能融合。仿真实验结果表明,该数据融合方法的平均系统能量消耗为66.35 J,网络连通度范围为0.85~1,达到了提高节点存活率以及提升网络连通性的目的。In order to improve the survivability and network connectivity of data fusion in power communication system,an intelligent data fusion method based on heterogeneous data sources was proposed.The least squares residual estimation method was used to identify noisy data and bad data in heterogeneous data sources.Through k-means clustering,the target average was obtained,the data were discretized and the noise was eliminated.The association rules of frequent itemsets were used to determine the confidence threshold and mine the data no less than the threshold.Different types of simulated data were mapped into the same vector space by using depth-restricted Boltzmann machine algorithm to realize intelligent fusion.Simulation results show that the average system energy consumption is 66.35 J,and the network connectivity range is within 0.85 to 1,confirming the improved node survival rate and network connectivity.

关 键 词：异构数据源 电力通信 链路模拟 智能融合 波尔曼兹机 数据融合 节点存活率 

分 类 号：TP318.2[自动化与计算机技术—计算机软件与理论]