论文部分内容阅读
剩余使用寿命(RUL)预测是预测与健康管理(PHM)中的核心环节。提出一种变工况条件下基于相似性的RUL预测方法。结合相似性预测方法无需进行复杂的退化过程建模而能提供合理预测的优势,引入工况即设备工作时所处的环境或操作载荷等因素的影响来提升设备RUL预测准确性。对参考样本建立多工况的设备退化模型提升模型精度,在服役样本相似性度量预测中进行工况的匹配以实现在变工况下的RUL预测。方法能够更准确地描述实际工程中设备的退化过程和个体差异。依据相同准确度标准完成多组基本相似性方法和本文方法的对比实验结果表明,本文方法能够有效提高RUL预测准确度。
Remaining life expectancy (RUL) prediction is at the core of forecasting and health management (PHM). A similarity-based RUL prediction method under variable working conditions is proposed. Combining with the similarity prediction method, it can provide the advantage of reasonable prediction without complicated degradation process modeling, and the influence of factors such as the working environment or operation load on the working condition is introduced to improve the RUL prediction accuracy. Establishing a multi-condition equipment degradation model for reference samples to improve model accuracy and match working conditions in service sample similarity measurement and prediction to achieve RUL prediction under varying conditions. The method can describe the degradation process and individual differences of equipment more accurately. According to the same accuracy standard, the method of comparing several groups of basic similarities and the comparison of the method in this paper shows that the proposed method can effectively improve the accuracy of RUL prediction.