摘要
为实现在不解体状况下诊断发动机汽缸故障,通过分析发动机缸盖振动信号的频谱,提出一种基于频带能量划分的特征提取方法。对发动机振动信号的频谱进行频带划分,将各频段信号的能量累加组成的向量作为故障分类的特征向量,应用BP神经网络算法对发动机活塞不同程度的故障进行识别。试验结果表明:基于频带能量划分和BP神经网络识别发动机活塞故障的方法是可行的。
To diagnose engine cylinder fault without disassembly,it firstly proposes a feature extraction method based on frequency-band energy division by analyzing the frequency spectra of engine cylinder head vibration signal in this paper.Then,it divides the frequency spectra of the engine vibration signal into frequency bands,and takes the vector accumulated by energy of each frequency-band as the eigenvector. Finally,it identifies the fault of engine piston in varying degrees with BP neural network algorithm. The experimental result shows that the fault of the engine piston in varying degrees can be identified based on frequency-band energy division and BP neural network.
引文
[1]贾继德,吴礼林.活塞敲缸故障的诊断研究[J].农业机械学报,2000,31(2):79-81.
[2]胡志勇,牛家骅.基于时域能量划分和PSO-SVM的发动机故障诊断[J].汽车工程,2016,38(1):86-90.
[3]王珍,马孝江.局域波时频法在柴油机缸套活塞磨损诊断中的应用研究[J].内燃机学报,2002,20(2):157-160.
[4]杜灿谊,喻菲菲.发动机活塞敲击故障仿真与诊断分析[J].车用发动机,2013(2):79-84.
[5]肖云魁,李会梁.基于双谱的柴油发动机活塞销故障诊断[J].内燃机学报,2008,26(4):369-373.
[6]CHEN J,RANDALL R B,PEETERS B.Advanced diagnostic system for piston slap faults in IC engines,based on the nonstationary characteristics of the vibration signals[J].Mechanical Systems and Signal Processing,2016,75(1):434-454.
[7]LUDWIG O,NUNES U.Novel maximum-margin training algorithms for supervised neural networks[J].IEEE Trans.Neural Netw.,2010,21(6):972-983.
[8]TAN P N,STEINBACH M.Introduction to Data Mining[J].Pearson Education,Inc,Publishing as Addison Wesley,2013,30(4):101-111.
[9]常春.基于振动分析的柴油机汽缸压力不解体检测研究[D].天津:军事交通学院,2016.
[10]HAYKIN S.Neural network:a comprehensive foundation[M].Bei Jing:Machinery Industry Press,2004:71-80.
[11]WANG J P,LI X M,Yong Hang.Motor failure diagnosis based on ant colnoy algorithm and BP Neural Network[J].Noise&Vibration Worldwide,2011,40(11):51-56.
[12]毛亚龙,钱晓东.基于BP神经网络的发动机故障诊断研究[J].工业控制计算机,2015,28(6):80-81.
[13]张玲玲,肖静.基于MATLAB的机械故障诊断技术案例教程[M].北京:高等教育出版社,2016:215-216.
[14]司景萍,牛家骅.基于小波包神经网络的发动机振动信号分析[J].振动测试与诊断,2013,33(5):763-769.
[15]徐礼超,马从兵.基于BP神经网络的汽车发动机故障诊断方法研究[J].拖拉机与农用运输车,2007,34(6):101-104.
[16]高隽.人工神经网络原理及仿真实例[M].北京:机械工业出版社,2003:17-18.