层级式机械装备健康指数模型及管理系统构建

Establishment of Hierarchical Health Degree Model and Management System for Mechanical Equipment

大型机械高端装备智能化是国民经济快速发展和国家安全稳定的重要保障。为解决零件、部件和装备健康指数量化问题,结合零件特征参数和集群装备大数据,使用长短时记忆网络(Long Short-Term Memory,LSTM)和互相关系数均值方法构建了零件健康指数量化模型,采用模糊C均值聚类和欧几里得范数(L2范数)融合优化了健康指数。以零部件自身权重为依据,提出了健康指数传递模型,解决了部件和装备健康指数量化问题。针对数据库和健康管理系统实现问题,以矿井提升机为例,构建了基于工业互联网的层级式闭环系统,能够实现数据采集、边缘协同、状态监测、故障诊断、量化评估和维修决策等方面的一体化集成。通过在洛阳中信重工示范应用,为大型机械高端装备健康、稳定、智能运行奠定了基础。

The large-scale mechanical high-end equipment is an important guarantee for the rapid development of the social economy and national security.To address the quantification issue of the health degree for parts,components,and equipment,a quantitative model for the health degree was proposed by combining parts feature parameters and cluster equipment big data,using Long Short-Term Memory network(LSTM)and cross-correlation coefficient mean method.The health degree was optimized by fusing fuzzy C-means clustering and Euclidean norm(L2 norm)method.Based on the weight of each part,the health degree transfer model is proposed to realize the quantification of health degrees for components and equipment.In response to the implementation issues in databases and health management systems,taking the example of mine hoists,a hierarchical closed-loop system based on the industrial internet is established,which can achieve the integration of data acquisition,edge collaboration,status monitoring,fault diagnosis,quantitative evaluation,and maintenance decision.Through the demonstration application in Luoyang CITIC Heavy Industries,a foundation has been laid for the healthy,stable,and intelligent operation of large-scale,high-end mechanical equipment.