基于SSA-Hurst-ARIMA组合模型的船舶柴油发电机组故障特征短期预测
Short-term Prediction of Fault Characteristics of Marine Diesel Generator Set Based on SSA-Hurst-ARIMA Combinatorial Model
梁清政 1王浩 1程垠钟 1杨天诣 1姚钦博1
作者信息
摘要
为提高船舶柴油发电机组故障特征短期预测精度,建立基于奇异谱分析(Singular Spectrum Analysis,SSA)、Hurst指数、自回归移动平均(Auto-Regressive Integrated Moving Average,ARIMA)的组合预测模型.以某试验中船舶柴油发电机组运行数据为基础,选取增压器滑油压强数据,对比分析单一ARIMA模型、SSA主成分-ARIMA组合模型和SSA-Hurst-ARIMA组合模型的预测效果.结果表明,SSA-Hurst-ARIMA组合模型的预测效果优于单一ARIMA模型和SSA主成分-ARIMA组合模型,更适合应用于船舶柴油发电机组故障特征的短期预测.
Abstract
In order to improve the short-term prediction accuracy of fault characteristics of marine diesel generator sets,a combined prediction model based on Singular Spectrum Analysis(SSA),Hurst index and Auto-Regressive Integrated Moving Average(ARIMA)is established.Based on the operating data of a marine diesel generator set in an experiment,the data of supercharger lubricating oil pressure are selected to compare and analyze the forecasting effects of a single ARIMA model,SSA principal component-ARIMA combination model and SSA-Hurst-ARIMA combination model.The results show that the prediction effect of SSA-Hurst-ARIMA combined model is better than that of the single ARIMA model and SSA principal component-ARIMA combined model,which is more suitable for the short-term prediction of fault characteristics of marine diesel generator sets.
关键词
船舶柴油发电机组/故障特征/短期预测/奇异谱分析(SSA)/Hurst指数/自回归移动平均(ARIMA)模型Key words
marine diesel generator set/fault characteristics/short-term forecast/Singular Spectrum Analysis(SSA)/Hurst index/Auto-Regressive Integrated Moving Average(ARIMA)model引用本文复制引用
出版年
2024
NETL
NSTL
万方数据