基于听觉模型的印刷机故障诊断技术  

Fault Diagnosis Technology of Printing Press Based on Auditory Model

作　　者：黄勇 邓英媛 Huang Yong;Deng Yingyuan(Sichuan Hina Technology&Engineering Co.,Ltd.,Chengdu 610045,China;Sichuan Xincai PrintCo.,Ltd.,Chengdu 610041,China)

机构地区：[1]四川海纳恒太科技工程有限公司,四川成都610045 [2]四川新财印务有限公司,四川成都610041

出　　处：《科学技术创新》2022年第21期49-52,共4页Scientific and Technological Innovation

摘　　要：印刷机具有结构精密、集成度高等特点,传统依靠人工经验判断的方法难以对其故障做出准确判断,经常会因为零部件故障而导致印刷工作中断。基于听觉模型的印刷机故障诊断可以实现对印刷机运行工况的在线监测,不仅能及时发现异常工况,还能准确诊断故障类型。本文首先概述了滤波器的选型与听觉模型的设计要点,随后设计了基于听觉模型的印刷机墨辊故障诊断实验。结果表明,在声音信号采集与处理、听觉模型参数设置和特征值的提取的基础上,墨辊故障诊断结果的正确率达到了80%,并且故障分类识别精确,较好地满足了故障诊断需要。The printing machine has the characteristics of precise structure and high integration.The traditional method of relying on manual experience judgment is difficult to make accurate judgments on its faults,and the printing work is often interrupted due to the failure of parts and components.The fault diagnosis of the printing press based on the auditory model can realize the online monitoring of the operating conditions of the printing press,which can not only detect abnormal conditions in time,but also accurately diagnose the fault type.This paper firstly summarizes the selection of filters and the design points of the auditory model,and then designs the fault diagnosis experiment of the ink roller of the printing press based on the auditory model.The results show that based on the acquisition and processing of sound signals,the parameter setting of the auditory model and the extraction of eigenvalues,the correct rate of the fault diagnosis results of the ink roller has reached 80%,and the fault classification and identification are accurate,which better meets the needs of fault diagnosis.

关 键 词：听觉模型 印刷机故障 声音信号 支持向量机 

分 类 号：TS803.6[轻工技术与工程] TH165.3[机械工程—机械制造及自动化]