摘要
针对船舶柴油机冷却系统故障诊断中信息和知识具有随机性和不确定性的特点,提出基于贝叶斯网络分类器的船舶柴油机冷却系统故障诊断的NB贝叶斯网络故障诊断模型和TAN故障诊断模型。研究结果表明:这2种故障诊断模型均可通过不断积累完善训练样本,自动修正网络结构参数和概率分布参数,提高诊断效果;采用这2种故障诊断模型,正判率在80.57%以上。
Due to the randomness and uncertainty of fault diagnosis data from cooling system of ship diesel engine,Naive Bayes（NB） and tree augmented naive Bayes（TAN） diagnostic model on cooling system of ship diesel engine were set up based on Bayes network classifier.The results show that the effectiveness of NB model and TAN model can be enhanced by the self-improvement method.When the training samples are accumulated,these models will modify their structure and probability distribution.The correction rates of the proposed models are higher than 80.57%.
出处
《中南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2010年第4期1379-1384,共6页
Journal of Central South University:Science and Technology
基金
湖南省教育厅优秀青年基金资助项目(08B042)
