基于物联网的发电厂设备预测维护分析模型  

A predictive maintenance analysis model for power plant equipment based on the Internet of Things

作　　者：李浩玮 徐鹏 张波涛 韦怡 韩国振 LI Haowei;XU Peng;ZHANG Botao;WEI Yi;HAN Guozhen(Longtan Hydroelectric Development Co.,Ltd.,Longtan Hydroelectric Power Plant,Nanning 530000,China)

机构地区：[1]龙滩水电开发有限公司龙滩水力发电厂,广西南宁530000

出　　处：《电子设计工程》2025年第2期52-56,共5页Electronic Design Engineering

基　　金：2022年龙滩水电开发有限公司龙滩水力发电厂信息化项目《桂冠电力生产运营管控信息(一体化)系统建设项目》(CDT-LTHPC-X-2374)。

摘　　要：准确预测发电厂设备的健康状态对确定设备的可靠性和使用寿命具有重要意义。为了对发电厂设备进行预测维护,文中在长短期记忆模型的基础上,提出了边缘预测维护分析模型。通过考虑设备不同部件的状态来预测设备的剩余使用寿命,从运行设备收集的实时数据来评估设备退化情况。为了标记数据集,使用模糊逻辑来生成维护优先级,这些优先级用于计算设备的实际剩余使用寿命。将提出的模型部署于发电厂设备,并进行性能评估。结果表明,边缘预测维护分析模型在开发和部署方面要容易得多,并且预测维护性能较好。Accurately predicting the health status of power plant equipment is of great significance for determining the reliability and service life of the equipment.In order to perform predictive maintenance on power plant equipment,this paper proposes an edge predictive maintenance analysis model based on short⁃term and short⁃term memory models.Predict the remaining service life of equipment by considering the status of different components,and evaluate equipment degradation based on real⁃time data collected from operating equipment.To label the dataset,fuzzy logic is used to generate maintenance priorities,which are used to calculate the actual remaining service life.Deploy the proposed model to the power plant equipment and conduct performance evaluation.The results indicate that the edge prediction maintenance analysis model is much easier to develop and deploy,and has good predictive maintenance performance.

关 键 词：预测维护 故障检测 长短期记忆模型 边缘预测维护分析模型 

分 类 号：TN98[电子电信—信息与通信工程]