基于信息融合技术的图书馆自助借还书机故障诊断研究  

Research on Fault Diagnosis of Self-borrowing Machine Based on Information Fusion Technology

作　　者：费晶[1] FEI Jing(Yulin university,Yulin Shaanxi 719000,China)

机构地区：[1]榆林学院,陕西榆林719000

出　　处：《自动化与仪器仪表》2022年第12期11-15,共5页Automation & Instrumentation

基　　金：榆林市产学研基金项目《陕北地方文献信息资源建设》(CXY-2020-013)。

摘　　要：为了提升传统图书馆自助借还机故障诊断的准确率,提出一种基于经验模态分解(Empirical Mode Decomposition, EMD)与随机森林相结合的故障分类算法,并引用改进后的DS证据理论对诊断结果融合,形成多传感器信息融合诊断方法。实验结果表明:EMD无量纲指标与随机森林结合的故障诊断算法单次测试准确率94%,故障分类测试准确率最高达97.6%;利用改进后的D-S证据理论融合多传感器诊断信息,准确率高达98%,平均差仅为1.01%。可以证明此种融合方法在提高信号准确性与鲁棒性方面具有优越性。In order to improve the accuracy of fault diagnosis of traditional library self-service lending and returning machines, a fault classification algorithm based on EMD(also known as empirical mode decomposition) and random forest is proposed, and the improved DS evidence theory is used to fuse the diagnosis results to form a multi-sensor information fusion diagnosis method. The experimental results show that the single test accuracy rate of the fault diagnosis algorithm based on EMD dimensionless index and random forest is 94%, and the fault classification test accuracy rate is 97.6%;Using the improved D-S evidence theory to fuse the multi-sensor diagnosis information, the accuracy rate is as high as 98%, and the average difference is only 1.01%. It can be proved that this fusion method has advantages in improving signal accuracy and robustness.

关 键 词：故障诊断系统 多传感器信息融合技术 随机森林 EMD无量纲 DS证据理论 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]