基于数字化盐田采卤泵的设备预测性维护系统  

Equipment Predictive Maintenance System for Salt Brine Pumps based on Digital Salt Field

作　　者：刘万平 马经黎 谢蓉 LIU Wanping;MA Jingli;XIE Rong(Qinghai Salt Lake Industry Co.,Ltd.,Golmud 816099,China)

机构地区：[1]青海盐湖工业股份有限公司,青海格尔木816099

出　　处：《盐科学与化工》2025年第4期46-49,54,共5页Journal of Salt Science and Chemical Industry

摘　　要：针对盐湖卤水资源需求不断提升与卤水泵分散且故障频发的现状,研究提出并建设了数字化盐田,并在此基础上建立了采卤泵的设备预测性维护系统。首先,介绍盐湖采卤的现状,指出传统巡检与维护的缺点与不足,阐述了基于数字孪生的采卤设备预测性系统基础及原理。然后,从网络建设、数据采集、监测点位设置阐述了系统建设过程。最后,基于数字化盐田提供的网络,通过采集机器设备正常运行时的传感器数据,在设备预测性维护系统上训练了多个数学模型。将训练好的模型接受的实时设备传感器数据作为输入,并预测出设备出现故障的概率。结果表明,该预测性维护系统能够预测卤水泵的设备故障,指导设备的维护工作。In view of the increasing demand for brine water resources in salt lakes and the dispersion and frequent failures of brine pumps,a digital salt field was proposed and constructed,and an equipment predictive maintenance system for salt brine pumps was established on this ba-sis.Firstly,the current situation of brine mining in salt lakes is introduced,the shortcomings and deficiencies of traditional inspection and maintenance are pointed out,and the basis and principle of the predictive system of brine mining equipment based on digital twin are expounded.Then,the process of system construction was expounded from the aspects of network construction,data col-lection and monitoring point setting.Finally,based on the network provided by digital salt field,several mathematical models were trained on the predictive maintenance system by collecting sen-sor data during the normal operation of machinery and equipment.The trained model takes real-time device sensor data as input and predicts the probability of equipment failure.The results showed that the predictive maintenance system can predict the equipment failure of the brine pump and guide the maintenance of the equipment.

关 键 词：数字化盐田 采卤泵 故障预警 预测性维护 

分 类 号：TS341[轻工技术与工程] TH38[机械工程—机械制造及自动化]