基于随机森林和支持向量机的船舶柴油机故障诊断

摘要

针对船舶低速二冲程柴油机故障的分析问题,提出基于随机森林和支持向量机的船舶柴油机故障诊断方法.对船舶低速二冲程柴油机MAN B&W 6S50MC-C建立故障仿真模型并验证其有效性;在此基础上,通过故障仿真模型生成故障样本.运用基于随机森林的VarSelRF特征选择算法对故障数据进行降维,提出运用支持向量机对降维后的故障数据进行分类的方法.通过仿真试验验证并分析该方法的有效性.%A fault diagnosis method for marine diesel engine based on random forest and support vector machine is proposed.The fault model of MAN B&W 6SS0MC-C marine diesel engine is established and verified with test-bed experiment results.The model is used to generate sample simulations of typical engine faults for processing.The VarSelRF algorithm based on the random forest is used to exclude irrelevant factors and identify effective features to facilitate the diagnosis process,improving the process speed.The processed data is classified by support vector machine classification algorithm.The effectiveness of the proposed method is verified by numerical experiments.