摘要
针对舰船发电用柴油机增压器的润滑失效问题,提出一种改进多阈值方法来实现对多工况异常情况的实时预测。该方法包含两部分,首先参考滑动平均滤波法对比分析各机组的运行数据,实现对异常数据的提取;其次采用一种基于外部特性法的箱型图法对分析后提取的异常数据进行多阈值表达式的设定。为了验证方法的有效性,基于某发电用柴油机在发电机工作过程中的实际故障数据进行验证,利用C++和Qt建立了可视化的增压器润滑失效故障在线监测与预测系统。验证表明,依据改进多阈值方法建立的系统可以实现对增压器在发电机全负载工作过程中异常情况的判断和对润滑失效的预测。
In this paper,based on the lubrication failure of marine diesel generator supercharger,an improved multithreshold method is proposed for solving the real-time prediction of abnormal situations.This method contains two points,firstly,the operating data of each unit were compared and analyzed by referring to the moving average filtering method,so as to achieve the extraction of abnormal data;secondly,a box graph method based on external characteristic method is used to set multi-threshold expressions for the extracted abnormal data after analysis.In order to verify the validity of the method,based on the actual failure data in the working process,the visually online monitoring and prediction system of the supercharger’s lubrication failure is established by using C++and Qt.The verification shows that the system established by the multi-threshold method can realize the real-time judgment on abnormal circumstances and the prediction of the supercharger’s lubrication failure in the whole load working process of the generator.
作者
杨天诣
程垠钟
杜剑维
刘世生
YANG Tian-yi;CHENG Yin-zhong;DU Jian-wei;LIU Shi-sheng(China Ship Research and Development Academy,Beijing 100101,China)
出处
《舰船科学技术》
北大核心
2022年第2期145-150,共6页
Ship Science and Technology
关键词
发电用柴油机
增压器润滑失效
多阈值
故障预测
diesel generator
lubrication failure of supercharger
multi-threshold
fault prediction
