基于深度置信网络的车载电源故障诊断方法
|
摘要
针对车载电源故障机理复杂且知识经验不足,传统浅层神经网络诊断效果难能满意的问题,研究了基于深度置信网络的车载电源故障诊断方法.该方法借助于30 kW车载电源仿真系统采集的几种常见故障数据,通过对深度置信网络进行预训练与反向微调,构建了车载电源相应故障的深度诊断神经网络,从而实现了车载电源几类常见故障的有效智能诊断.该方法的优势在于能够将车载电源的故障特征提取与故障诊断有机融合,摆脱了传统浅层故障诊断方法对大量信号处理技术与诊断经验的依赖,仿真试验也进一步昭示出文中方法在车载电源故障诊断中的有效性和适宜性.
Aimed at the problem that the failure mechanism vehicle-carried power supply is complex, lack of its knowledge and experience, and unsatisfactory diagnosis effect of traditional shallow neural network, a diagnosis method of fault in vehicle-carried power supply is studied in this article based on deep belief networks. In this method, the data of several common faults gathered with the help of simulation system of 30 kW vehicle-carried power supply and the deep belief network are used to carry out pre-training and reverse fine-tuning of the deep belief network and construct a neural network of deep diagnosis of corresponding faults of vehicle-carried power supply, thus realizing effective intelligent diagnosis of several common faults of vehicle-carried power supply. The advantage of this method lies in its capability of merging organically the fault feature extraction with fault diagnosis and getting rid of the dependence of traditional shallow fault diagnosis method on massive signal processing techniques and diagnosis experiences. Simulation test further shows also the validity and suitability of this method for fault diagnosis of vehicle-carried power supply.
Aimed at the problem that the failure mechanism vehicle-carried power supply is complex, lack of its knowledge and experience, and unsatisfactory diagnosis effect of traditional shallow neural network, a diagnosis method of fault in vehicle-carried power supply is studied in this article based on deep belief networks. In this method, the data of several common faults gathered with the help of simulation system of 30 kW vehicle-carried power supply and the deep belief network are used to carry out pre-training and reverse fine-tuning of the deep belief network and construct a neural network of deep diagnosis of corresponding faults of vehicle-carried power supply, thus realizing effective intelligent diagnosis of several common faults of vehicle-carried power supply. The advantage of this method lies in its capability of merging organically the fault feature extraction with fault diagnosis and getting rid of the dependence of traditional shallow fault diagnosis method on massive signal processing techniques and diagnosis experiences. Simulation test further shows also the validity and suitability of this method for fault diagnosis of vehicle-carried power supply.
引文
[1] 周东华,胡艳艳.动态系统的故障诊断技术 [J].自动化学报,2009,35(6):748-758.
[2] GAO Z,CECATI C,DING S X.A survey of fault diagnosis and fault-tolerant techniques-part Ⅱ:fault diagnosis with knowledge-based and hybrid/active approaches [J].IEEE Transactions on Industrial Electronics,2015,62(6):3768-3774.
[3] WANG H,CHEN J,DONG G.Feature extraction of rolling bearing's early weak fault based on EEMD and tunable Q-factor wavelet transform [J].Mechanical Systems & Signal Processing,2014,48(1/2):103-119.
[4] VAN M,KANG H J.Bearing defect classification based on individual wavelet local fisher discriminant analysis with particle swarm optimization [J].IEEE Transactions on Industrial Informatics,2016,12(1):124-135.
[5] HINTON G E,SALAKHUTDINOV R R.Reducing the dimensionality of data with neural networks [J].Science,2006,313(5786):504-507.
[6] 孙志军,薛磊,许阳明,等.深度学习研究综述 [J].计算机应用研究,2012,29(8):2806-2810.
[7] 林鹏.基于深度图像学习的人体部位识别 [D].上海:上海交通大学,2012.
[8] YU D,DENG L.Deep learning and its applications to signal and information processing [J].IEEE Signal Processing Magazine,2011,28(1):145-154.
[9] LEE H,GROSSE R,RANGANATH R,et al.Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations [C]//ICML 2009.Montreal,Quebec,Canada:DBLP,2009:609-616.
[10] 赵光权,葛强强,刘小勇,等.基于DBN的故障特征提取及诊断方法研究 [J].仪器仪表学报,2016,37(9):1946-1953.
[11] TAMILSELVAN P,WANG P.Failure diagnosis using deep belief learning based health state classification [J].Reliability Engineering & System Safety,2013,115(7):124-135.
[12] TRAN V T,ALTHOBIANI F,BALL A.An approach to fault diagnosis of reciprocating compressor valves using Teager-Kaiser energy operator and deep belief networks [J].Expert Systems with Applications,2014,41(9):4113-4122.
[13] 雷亚国,贾峰,周昕,等.基于深度学习理论的机械装备大数据健康监测方法 [J].机械工程学报,2015,51(21):49-56.
[14] 贾丽会,张修如.BP算法分析与改进 [J].计算机技术与发展,2006,16(10):101-103.
[15] LECUN Y,BENGIO Y,HINTON G E.Deep learning [J].Nature,2015,521(7553):436-444.
[16] 段艳杰,吕宜生,张杰,等.深度学习在控制领域的研究现状与展望 [J].自动化学报,2016,42(5):643-654.
[2] GAO Z,CECATI C,DING S X.A survey of fault diagnosis and fault-tolerant techniques-part Ⅱ:fault diagnosis with knowledge-based and hybrid/active approaches [J].IEEE Transactions on Industrial Electronics,2015,62(6):3768-3774.
[3] WANG H,CHEN J,DONG G.Feature extraction of rolling bearing's early weak fault based on EEMD and tunable Q-factor wavelet transform [J].Mechanical Systems & Signal Processing,2014,48(1/2):103-119.
[4] VAN M,KANG H J.Bearing defect classification based on individual wavelet local fisher discriminant analysis with particle swarm optimization [J].IEEE Transactions on Industrial Informatics,2016,12(1):124-135.
[5] HINTON G E,SALAKHUTDINOV R R.Reducing the dimensionality of data with neural networks [J].Science,2006,313(5786):504-507.
[6] 孙志军,薛磊,许阳明,等.深度学习研究综述 [J].计算机应用研究,2012,29(8):2806-2810.
[7] 林鹏.基于深度图像学习的人体部位识别 [D].上海:上海交通大学,2012.
[8] YU D,DENG L.Deep learning and its applications to signal and information processing [J].IEEE Signal Processing Magazine,2011,28(1):145-154.
[9] LEE H,GROSSE R,RANGANATH R,et al.Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations [C]//ICML 2009.Montreal,Quebec,Canada:DBLP,2009:609-616.
[10] 赵光权,葛强强,刘小勇,等.基于DBN的故障特征提取及诊断方法研究 [J].仪器仪表学报,2016,37(9):1946-1953.
[11] TAMILSELVAN P,WANG P.Failure diagnosis using deep belief learning based health state classification [J].Reliability Engineering & System Safety,2013,115(7):124-135.
[12] TRAN V T,ALTHOBIANI F,BALL A.An approach to fault diagnosis of reciprocating compressor valves using Teager-Kaiser energy operator and deep belief networks [J].Expert Systems with Applications,2014,41(9):4113-4122.
[13] 雷亚国,贾峰,周昕,等.基于深度学习理论的机械装备大数据健康监测方法 [J].机械工程学报,2015,51(21):49-56.
[14] 贾丽会,张修如.BP算法分析与改进 [J].计算机技术与发展,2006,16(10):101-103.
[15] LECUN Y,BENGIO Y,HINTON G E.Deep learning [J].Nature,2015,521(7553):436-444.
[16] 段艳杰,吕宜生,张杰,等.深度学习在控制领域的研究现状与展望 [J].自动化学报,2016,42(5):643-654.