基于KNN算法的大型压缩机组设备在用油油质性能评估方法

Oil Quality Performance Evaluation Based on KNN Algorithm

针对大型压缩机组设备检修与用油安全问题,考虑压缩机组设备运行特点及用油情况,从基本理化指标和污染性能指标两个维度,选择16个具体指标构建了基于K邻近(K-Nearest Neighbor,KNN)算法的数据处理与评价模型。利用该模型对设备在用油品进行油质性能分析评估可得到5种不同等级的润滑油劣化结果,以投运中的3台压缩机组为例对所构建的模型进行验证,结果与润滑油实际检测评价结果吻合。表明所提出的方法适用于压缩机组在用油评价,可供判断油品的品质,为科学换油提供数据支持,辅助决策机组检修。

Aiming at the problem of equipment maintenance and oil safety of large compressor units,considering the operation characteristics and oil use of compressor units,16 specific indexes were selected from the two dimensions of basic physical and chemical indexes and pollution performance indexes,and a data processing and evaluation model based onK-Nearest Neighbor(KNN)algorithm was constructed.This model was applied to the analysis and evaluation of the oil quality performance of the in-service oil of the equipment,the deterioration results of five different grades of lubricating oil was obtained.Three compressor units in operation were taken as examples to verify the proposed KNN model,the results were consistent with the actual test and evaluation results of lubricating oil.The results show that the proposed method is suitable for evaluating the in-use oil of compressor unit,judging the oil quality,and provides data support for reasonable oil replacement,and support the unit maintenance decision.