基于维纳过程的船舶柴油机增压器寿命预测

Residual life prediction for marine diesel turbocharger based on Wiener process

[目的]针对船舶柴油机增压器难以收集到全生命周期性能退化数据的问题,提出一种基于维纳过程的寿命预测模型。[方法]首先,采用K-Means模型对增压器实际运行工况进行聚类,提取出典型工况数据;然后,使用贝叶斯突变点检测模型识别增压器的缺陷点;最后,建立基于维纳过程的退化模型,并以某型船用柴油机增压器为应用对象,预测增压器的剩余使用寿命。[结果]结果显示,基于维纳过程的寿命预测方法能够在不需要同类设备历史退化数据的情况下对增压器的剩余寿命进行预测。[结论]所提方法对缺少故障样本的船舶柴油机增压器寿命预测具有一定的参考价值。

[Objectives]In view of the difficulty of collecting the full life cycle performance degradation data of a marine diesel turbocharger,a life prediction model based on the Weiner process is proposed to predict the residual life of the turbocharger.[Methods]First,the K-Means model is used to cluster the actual operating conditions of the turbocharger and extract the typical operating condition data,then the Bayesian change-point detection model is used to identify the defect points of the turbocharger,and finally,the residual life of a certain type of turbocharger is predicted by the Weiner process degradation model.[Results]The results prove that the proposed method can predict the residual lifetime of the turbocharger without the historical degradation data of similar equipment.[Conclusions]The method proposed herein is valuable for marine diesel turbochargers without fault samples.