首页> 外国专利> STATE IDENTIFICATION METHOD BY CHARACTERISTIC ANALYSIS OF HISTOGRAM IN TIME REGION AND FREQUENCY REGION

STATE IDENTIFICATION METHOD BY CHARACTERISTIC ANALYSIS OF HISTOGRAM IN TIME REGION AND FREQUENCY REGION

机译:时域和频域直方图特征分析的状态识别方法

摘要

To provide an effective and common state identification method for monitoring and diagnosing various facilities, in consideration of reflecting the characteristic of the state of the facility reflected by the histogram of the time domain and frequency region of a measurement signal in order to effectively perform state identification and pattern recognition, for example, facility state monitoring and abnormality diagnosis.SOLUTION: A state identification method comprises a step in which a signal of an object is collected at a predetermined measurement frequency and a measurement time in a method for state diagnosis and pattern recognition; a step of obtaining waveform data of a signal; and a step of obtaining a histogram of the time domain or frequency region of the waveform data; and a step of integrating a segment value of the histogram by a multivariate analysis method and a technique of artificial intelligence and performing a state identification and a pattern recognition of the object. By using at least one of histograms in the time domain and histogram in the frequency domain, the histogram segment values are combined by a multivariate analysis method and a technique of artificial intelligence, and the state identification and pattern recognition of the object are carried out.SELECTED DRAWING: Figure 5
机译:考虑到反映测量信号的时域和频域的直方图所反映的设施状态的特征,以便提供一种有效且通用的状态识别方法来监视和诊断各种设施,以便有效地进行状态识别解决方案:状态识别方法包括以下步骤:在用于状态诊断和模式识别的方法中,以预定的测量频率和测量时间收集对象的信号。 ;获得信号的波形数据的步骤;获得所述波形数据的时域或频域的直方图的步骤;通过多元分析方法和人工智能技术对直方图的分割值进行积分,并对物体进行状态识别和图案识别的步骤。通过使用时域中的直方图和频域中的直方图中的至少一个,通过多元分析方法和人工智能技术来组合直方图段值,并且执行对象的状态识别和模式识别。选定的图纸:图5

著录项

  • 公开/公告号JP2018205292A

    专利类型

  • 公开/公告日2018-12-27

    原文格式PDF

  • 申请/专利权人 SONG LIUYANG;JINYAMA HO;

    申请/专利号JP20170207617

  • 发明设计人 SONG LIUYANG;JINYAMA HO;

    申请日2017-10-10

  • 分类号G01M99/00;G05B23/02;G01H17/00;

  • 国家 JP

  • 入库时间 2022-08-21 12:21:22

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号