摘要
船舶的正常运行有赖于智能缺陷故障的诊断和识别能力。此次研究提出基于关联规则算法的船舶故障智能诊断方法,利用频繁项集更新策略改进经典Apriori算法,通过时间序列模式和BP神经网络完成故障数据挖掘和分类以实现故障智能诊断。算法性能测试发现改进后的Apriori算法的运行效率提高48.26%,船舶油机排气温度过高和气门漏气两者的关系最为密切,其支持度和置信度分别为53.3333%和94.1176%。新的船舶故障诊断方法对故障分类的准确性提高约20%,节省故障分类时间约4分钟。
The normal operation of ship depends on the ability of intelligent fault diagnosis and recognition.In this study,an intelligent fault diagnosis method based on association rules algorithm is proposed,which improves the classical Apriori al-gorithm by using frequent itemset updating strategy,and fault data mining and classification through time series pattern and BP neural network is completed to realize intelligent fault diagnosis.The algorithm performance test shows that the operation effi-ciency of the improved Apriori algorithm is improved by 48.26%,and the relationship between the high exhaust temperature of the marine engine and the air leakage of the valve is the closest,with the support and confidence of 53.3333%and 94.1176%respectively.The new method can improve the accuracy of fault classification by about 20%and save about 4 minutes.
作者
刘宪珍
LIU Xianzhen(Maritime college,Qingdao Harbour Vocational and Technical College,Qingdao Shandong 266404,China)
出处
《自动化与仪器仪表》
2021年第6期100-103,共4页
Automation & Instrumentation
基金
船舶动力工程技术交通运输行业重点实验室开放基金(No.KLMPET2018-07)。