The invention discloses an escalator unsupervised diagnosis method and system, and the method comprises the steps: installing sensors at a plurality of measurement points of an escalator, and obtaining the vibration data of the escalator; setting a plurality of key time periods for collecting vibration data, extracting data collected at a certain time period of a certain measuring point, and obtaining single-measuring-point single-time-period data; performing multi-dimensional feature extraction on the single-measurement-point single-time-period data to obtain single-dimensional feature data of multiple different dimensions of the single-measurement-point single-time-period data; the method comprises the following steps: extracting single-dimensional feature data of a single measuring point and a single time period of multiple days in a normal operation date of the escalator, constructing a health data set of the dimension, extracting single-dimensional feature data of the same dimension of the same measuring point and the same time period of multiple days in a to-be-diagnosed date, and constructing a corresponding to-be-diagnosed data set; and performing anomaly diagnosis on the to-be-diagnosed data set and the health data set, and outputting different levels of diagnosis results. The acquired data feature system is comprehensive, the diagnosis accuracy is high, and the abnormal phenomenon can be accurately captured.
本发明公开了一种自动扶梯无监督诊断方法及系统,在自动扶梯的多个测点安装传感器,获得自动扶梯的振动数据;设定采集振动数据的多个关键的时间段,提取某一测点的某一时间段采集的数据,获得单测点单时段数据;对单测点单时段数据进行多维度特征提取,获得单测点单时段数据的多个不同维度的单维特征数据;提取自动扶梯正常运行日期内多日的单测点单时段的单维特征数据,构建该维度的健康数据集,提取待诊断日期内多日的同一测点同一时间段的同一维度的单维特征数据,构建与之对应的待诊断数据集;对待诊断数据集和健康数据集进行异常诊断,输出不同层次的诊断结果。本发明采集的数据特征体系全面,诊断准确率高,能够精准捕捉到异常现象。
Unsupervised diagnosis method and system for escalator
一种自动扶梯无监督诊断方法及系统
2024-08-06
Patent
Electronic Resource
Chinese
Escalator control method and system and escalator system
European Patent Office | 2021
|Bearing diagnosis device, bearing diagnosis method and escalator
European Patent Office | 2020
|