微型电脑应用2025，Vol.41Issue(3)：144-148,5.
基于BP神经网络的电力系统多渠道异构数据融合方法
Multi-channel Heterogeneous Data Fusion Method of Power System Based on BP Neural Network
林敏辉 1王超 2万祥虎 2陶鹏 2王卫华2
作者信息
- 1. 国网福建省电力有限公司物资分公司,福建,福州 350000
- 2. 安徽继远软件有限公司,安徽,合肥 230093
- 折叠
摘要
Abstract
Aimed at the problem of poorer data fusion effects caused by missing and inconsistent data across tables in power sys-tem heterogeneous data fusion,a multi-channel heterogeneous data fusion method based on BP neural network is proposed.Ac-cording to the extracted characteristics of multi-channel heterogeneous data,its visual adjustability characteristics are used to complete the adaptive mining of multi-channel heterogeneous data.The dimension and quantity of the mined data are reduced through noise elimination,error inspection,format review and other operations.After normalizing and unifying the dimensions of the data collected from each channel,the processing results are input as BP neural network model to obtain the characteristic function that meets the expected value and expected value error of multi-channel heterogeneous data fusion,and multi-channel heterogeneous data fusion is realized.The experiment shows that using this method to extract multi-channel heterogeneous data features is more stable than the initial heterogeneous data features with a higher anti-interference ability.The amplitude varia-tion range is 0.09 dB～0.10 dB.The decision coefficient R2 is 0.9293,which indicates that the actual prediction results are very close to the expected results.The maximum number of reads and writes is 6 × 106 times,which can achieve complex multi-channel heterogeneous data fusion and improve the accuracy of multi-channel heterogeneous data analysis.关键词
电力系统/BP神经网络/异构数据/多渠道/数据融合Key words
power system/BP neural network/heterogeneous data/multi-channel/data fusion分类
机械制造引用本文复制引用
林敏辉,王超,万祥虎,陶鹏,王卫华..基于BP神经网络的电力系统多渠道异构数据融合方法[J].微型电脑应用,2025,41(3):144-148,5.
