摘要
针对岸桥起升减速箱故障频发的现状,提出一种基于 k -均值聚类的载荷分类准则和健康状态识别方法。建立减速箱振动模型,通过振动信号来反映减速箱的运行状态。根据聚类中心的变化得到减速箱健康状态的发展趋势,并对实际故障数据进行分析。应用 k -均值聚类对振动信号载荷状态进行分类,实现工况的可视化并对岸桥的载重量进行合理的安排。结果表明:该方法不仅能够将振动信号的载荷状态进行合理的分类,而且能够实现对减速箱各个运行状态的诊断与识别。
A method based on k -means clustering is proposed to improve the healthy condition recognition for quayside container crane reducers. The model of the reducer vibration is established to monitor the running conditions through observing the vibration signals. The k -means clustering is used to classify the vibration signals implying the load conditions. The clustering results are visualized to indicate the working conditions. The changes of the cluster center are monitored to supervise the development trend of the health conditions. The design finds out the actual fault if it happens, and adjusts the load distribution of the crane.
作者
侯美慧
胡雄
王冰
HOU Meihui;HU Xiong;WANG Bing(Logistics Engineering College, Shanghai Maritime University, Shanghai 201306, China)
出处
《中国航海》
CSCD
北大核心
2019年第3期105-109,共5页
Navigation of China
基金
国家高技术研究发展计划(863计划)(2013AA041106)
国家自然科学基金(31300783)
上海海事大学创新基金(2016ycx063)
关键词
岸桥减速箱
振动模型
K-均值聚类
状态识别
quayside container crane reducer
vibration model
k-means clustering
condition recognition
