摘要
为了降低交通事故发生概率、确保人民生命财产安全,将采集的发动机信号分为8类,基于LMS算法利用线性判别函数构建发动机故障状态识别模型对发动机状态进行快速判断,基于三层BP神经网络构建故障定位分类模型对存在故障的发动机进行故障分类,经过800个实例数据验证,判别准确率高达95.375%,模型诊断效果良好,为发动机故障诊断与定位提供了技术支撑。
In order to reduce the probability of traffic accident,to ensure the people life and property safety,this paper is divided the engine signal acquisition into 8 classes.Based on LMS algorithm we use linear discriminant function to build engine fault state recognition model of engine state for quick judgment,classify fault location based on three layers BP neural network and the model of a faulty engine fault classification.After 800 examples validation,discriminant accuracy as high as 95.375%,the model diagnosis effect is good,for engine fault diagnosis and location it provides technical support.
作者
解淑英
XIE Shuying(Department of Electrical and Mechanical Engineering, Yantai Automobile Engineering Professional College, Yantai 265500, China)
出处
《微型电脑应用》
2021年第12期44-47,共4页
Microcomputer Applications
基金
山东省第三批职业教育技艺技能传承创新平台项目(鲁教师函〔2020〕53号-75)
山东省高等学校应用技术优质协同创新中心项目(鲁教科字〔2020〕6号-28)
山东省高等学校青创人才引育计划(鲁教师函〔2020〕53号-108)
烟台汽车工程职业学院2020年基金项目(2020kj03、2020jg01)。
