基于高斯过程的提升机轴承性能评测方法研究
|
摘要
随着科学技术的进步和工业需求的发展，提升机一方面不断向大型、复杂、高速、高效的方向发展，另一方面却又面临更加苛刻的工作和运行环境。提升机轴承作为矿井提升设备的关键装备，是整个提升系统的重要组成部分和核心关键部件，同时也是最易出现故障的薄弱部件，它的运行状态好坏直接影响到整个系统的运行。一旦提升机轴承部件发生故障，就可能破坏整台设备甚至影响整个生产过程，造成巨大经济损失，还可能导致灾难性的人员伤亡和形成严重的社会影响。因此，如何有效评测提升机轴承的运行性能，研究制定设备维修计划，借助智能信息处理的手段实行设备智能维护，确保设备安全可靠运行，避免或减少各种恶性事故突发，对煤矿企业安全生产、减灾救灾和煤炭工业的健康持续发展来说是当前迫切需要解决的问题。
本论文的主要研究工作包括：
（1）针对设备性能退化数据中的高维特性会导致后续评测算法效率低下的问题，提出了一个基于谱回归（SR）的轴承性能特征提取方法。该方法首先通过标记和非标记样本构造仿射近邻图，以揭示给定数据的内在结构信息，随后根据所学得的响应值，再用普通的回归方法学习嵌入函数，最后采用最小二乘法获得最佳映射方向，避免了计算特征空间稠密矩阵的问题。通过多组公测轴承实验数据和提升机轴承实验数据证实了SR方法比PCA、FA以及其他常用的流形学习技术更有效。
（2）针对有监督的机器学习方法在训练评估模型时需要大量正确标记的训练样本以及训练样本类不均衡的问题，提出了一个基于半监督高斯过程的分类算法。该算法首先根据类不均衡数据的特性进行数据预处理，然后利用少量的标记数据进行高斯过程分类训练，选取预测概率置信度最高的未标记数据，向该未标记数据注入合理的类标记信息，并且自动地将新标记过的数据样本加入到原有的训练集中，用扩充后的训练集再次进行高斯过程分类，最后采用自训练迭代执行，直至构造出性能最优的半监督高斯过程分类器。通过仿真数据和公测轴承数据进行了验证，并成功用于提升机轴承性能退化评估实验中。
（3）针对设备性能退化数据中自然存在的时序性难以用于标准的高斯过程回归的问题，提出了一个基于隐马尔可夫高斯过程的预测算法。该算法很好地结合了HMM的时序预测特性和高斯过程回归的核方法预测优势，建立了一个HMMGP模型，将每一个设备性能退化状态对应成一个高斯过程，利用EM算法学习模型参数，并使用共轭梯度法求得似然函数极大时的参数，为了避免求逆矩阵运算，采用Cholesky分解法降低了算法复杂度。论文用多组轴承全寿命实验数据进行了验证，并成功用于提升机轴承性能退化预测实验中，与其他预测方法进行了对比，表明HMMGP预测模型具有更高的预测精度。
本文研究的内容涉及到了提升机轴承性能特征提取、轴承性能退化评估和性能退化预测三方面的问题，主要采用了高斯过程机器学习方法及相关改进算法。在研究了相关领域前期工作的基础上对已有的算法进行了分析和改进，使用了轴承故障诊断领域中常用的公测数据集进行了实验，并与相关算法进行了性能对比，验证了本文提出的算法的有效性，最后在提升机轴承性能退化评测中对文中所提出的算法进行了应用。应用结果表明，本文的研究成果在提升机轴承性能评测中取得了不错效果，对提高煤矿生产及设备可靠性具有一定的指导意义。
The key equipment of mine hoisting is the hoist bearing, it is also the importantpart of the lifting system and the core components; meanwhile, it is the weak part ofthe system and easy to cause fault; thus its operating status directly affects theoperation of the whole system. Therefore, how to effectively assess the operationalstatus of hoist bearing, how to study and formulate equipment maintenance program,how to use the means of intelligent information processing to implement intelligentmaintenance of the equipments, which is an urgent problem to be solved to make surethe operation of the equipments is safe and reliable, so that a variety of unexpectedfatal accidents can be avoided or reduce. In the field of coal production, it is also animportant topic to make the process of production safely and the coal industryhealthily.
The main research work is elaborated as following:
(1) In order to improve the efficiency of assessment and prediction algorithmwhen dealing with the data degradation in the high-dimensional feature space, amethod, which is based on the spectral regression, is proposed to capture the featuresof the hoist bearing. Through multiple sets of experimental data of bearing, it isconfirmed that the method of spectral regression is more effective than PCA, FA orthe other common manifold learning techniques.
(2) As the methods of supervised learning normally need a larger number ofcorrectly per-labeled training examples and the distribution of training examples isimbalance, a method of Semi-supervised Gaussian process is proposed in the field ofassessing the degradation of the hoist bearing. The effectiveness of such method isproven through several experiments in practice.
(3) To improve the efficient of Gaussian process when dealing with equipmentperformance degradation data, a method of Hidden Markov Gaussian process isproposed in the field of assessing the degradation of the hoist bearing. The method iswell combined the kernel-based Gaussian process regression with the characteristicsof HMM in predicting the performance degradation data and can be used to build up aHMMGP model. The comparing tests are carried out by using the some sets ofexperimental data of full life of the bearings. HMMGP prediction model has higherprediction accuracy.
The contents in this thesis include capturing the features of the hoist bearing, assessing and predicting the degradation of the bearing by using the Gaussianprocesses learning and the related improved methods. The proposed algorithm is usedin a real case to assess the degradation of the hoist bearing, the result of theapplication shows that the effectiveness of the new algorithm when it is used to assessthe degradation of the hoist bearing, and it has great influence on improving thereliability of mining equipment and process of the mining production.
本论文的主要研究工作包括：
（1）针对设备性能退化数据中的高维特性会导致后续评测算法效率低下的问题，提出了一个基于谱回归（SR）的轴承性能特征提取方法。该方法首先通过标记和非标记样本构造仿射近邻图，以揭示给定数据的内在结构信息，随后根据所学得的响应值，再用普通的回归方法学习嵌入函数，最后采用最小二乘法获得最佳映射方向，避免了计算特征空间稠密矩阵的问题。通过多组公测轴承实验数据和提升机轴承实验数据证实了SR方法比PCA、FA以及其他常用的流形学习技术更有效。
（2）针对有监督的机器学习方法在训练评估模型时需要大量正确标记的训练样本以及训练样本类不均衡的问题，提出了一个基于半监督高斯过程的分类算法。该算法首先根据类不均衡数据的特性进行数据预处理，然后利用少量的标记数据进行高斯过程分类训练，选取预测概率置信度最高的未标记数据，向该未标记数据注入合理的类标记信息，并且自动地将新标记过的数据样本加入到原有的训练集中，用扩充后的训练集再次进行高斯过程分类，最后采用自训练迭代执行，直至构造出性能最优的半监督高斯过程分类器。通过仿真数据和公测轴承数据进行了验证，并成功用于提升机轴承性能退化评估实验中。
（3）针对设备性能退化数据中自然存在的时序性难以用于标准的高斯过程回归的问题，提出了一个基于隐马尔可夫高斯过程的预测算法。该算法很好地结合了HMM的时序预测特性和高斯过程回归的核方法预测优势，建立了一个HMMGP模型，将每一个设备性能退化状态对应成一个高斯过程，利用EM算法学习模型参数，并使用共轭梯度法求得似然函数极大时的参数，为了避免求逆矩阵运算，采用Cholesky分解法降低了算法复杂度。论文用多组轴承全寿命实验数据进行了验证，并成功用于提升机轴承性能退化预测实验中，与其他预测方法进行了对比，表明HMMGP预测模型具有更高的预测精度。
本文研究的内容涉及到了提升机轴承性能特征提取、轴承性能退化评估和性能退化预测三方面的问题，主要采用了高斯过程机器学习方法及相关改进算法。在研究了相关领域前期工作的基础上对已有的算法进行了分析和改进，使用了轴承故障诊断领域中常用的公测数据集进行了实验，并与相关算法进行了性能对比，验证了本文提出的算法的有效性，最后在提升机轴承性能退化评测中对文中所提出的算法进行了应用。应用结果表明，本文的研究成果在提升机轴承性能评测中取得了不错效果，对提高煤矿生产及设备可靠性具有一定的指导意义。
The key equipment of mine hoisting is the hoist bearing, it is also the importantpart of the lifting system and the core components; meanwhile, it is the weak part ofthe system and easy to cause fault; thus its operating status directly affects theoperation of the whole system. Therefore, how to effectively assess the operationalstatus of hoist bearing, how to study and formulate equipment maintenance program,how to use the means of intelligent information processing to implement intelligentmaintenance of the equipments, which is an urgent problem to be solved to make surethe operation of the equipments is safe and reliable, so that a variety of unexpectedfatal accidents can be avoided or reduce. In the field of coal production, it is also animportant topic to make the process of production safely and the coal industryhealthily.
The main research work is elaborated as following:
(1) In order to improve the efficiency of assessment and prediction algorithmwhen dealing with the data degradation in the high-dimensional feature space, amethod, which is based on the spectral regression, is proposed to capture the featuresof the hoist bearing. Through multiple sets of experimental data of bearing, it isconfirmed that the method of spectral regression is more effective than PCA, FA orthe other common manifold learning techniques.
(2) As the methods of supervised learning normally need a larger number ofcorrectly per-labeled training examples and the distribution of training examples isimbalance, a method of Semi-supervised Gaussian process is proposed in the field ofassessing the degradation of the hoist bearing. The effectiveness of such method isproven through several experiments in practice.
(3) To improve the efficient of Gaussian process when dealing with equipmentperformance degradation data, a method of Hidden Markov Gaussian process isproposed in the field of assessing the degradation of the hoist bearing. The method iswell combined the kernel-based Gaussian process regression with the characteristicsof HMM in predicting the performance degradation data and can be used to build up aHMMGP model. The comparing tests are carried out by using the some sets ofexperimental data of full life of the bearings. HMMGP prediction model has higherprediction accuracy.
The contents in this thesis include capturing the features of the hoist bearing, assessing and predicting the degradation of the bearing by using the Gaussianprocesses learning and the related improved methods. The proposed algorithm is usedin a real case to assess the degradation of the hoist bearing, the result of theapplication shows that the effectiveness of the new algorithm when it is used to assessthe degradation of the hoist bearing, and it has great influence on improving thereliability of mining equipment and process of the mining production.
引文
[1]陈维健.矿山大型固定装备测试技术[M].徐州:中国矿业大学出版社,2007.
[2]《矿井提升机故障处理和技术改造》编委会.矿井提升机故障处理和技术改造(第二版)[M].北京:机械工业出版社,2013.
[3]陈进.机械故障诊断理论及应用[M].北京:高等教育出版社,2010.
[4]雷金波.基于逻辑回归和支持向量机的设备状态退化评估与趋势预测研究[D].上海:上海交通大学,2008.
[5]王增才,王树梁,任锴胜等.基于EEMD的提升机天轮轴承故障诊断方法[J].煤炭学报,2012,37(4):689-694.
[6]王树梁.基于HHT的提升机天轮轴承故障诊断方法研究[D].济南:山东大学,2011.
[7] J. Lee. Intelligent Maintenance Systems (IMS) Technologies. www.imscenter.net,2010.
[8]钟鑫.基于逻辑回归和高斯混合模型的设备故障诊断技术研究与应用[D].北京:北京化工大学,2010.
[9]郭磊.基于核模式分析方法的旋转机械性能退化评估技术研究[D].上海:上海交通大学,2009.
[10]潘玉娜,陈进,李兴林.基于模糊c-均值的设备性能退化评估方法[J].上海交通大学学报,2009,43(11):1794-1797.
[11] Debin Zhao, Jihong Yan. Performance prediction methodology based on patternrecognition[J]. Signal Processing,2011,91:2194-2203.
[12] Linxia Liao, Jay Lee. Design of a reconfigurable prognostics platform for machine tools[J].Expert Systems with Applications,2010,37:240-252.
[13] Ilias Bilionis, Nicholas Zabaras. Multi-output local Gaussian process regression: Applicationsto uncertainty quantification[J]. Journal of Computational Physics,2012,231(17):5718-5746.
[14] Anjishnu Banerjee, David B. Dunson, and Surya T. Tokdar.Efficient Gaussian processregression for large datasets[J]. Biometrika,2013,100(1):75-89.
[15] Dongbing Gu, Huosheng Hu. Spatial Gaussian process regression with mobile sensornetworks[J]. IEEE Transactions on Neural Networks and Learning Systems,2012,23(8):1279-1290.
[16] Michael Kemmler, Erik Rodner, Esther-Sabrina Wacker. One-class classification withGaussian processes[J]. Pattern Recognition,2013,46:3507-3518.
[17] Deisenroth M.P., Turner R.D., Huber M.F., et al. Robust filtering and smoothing withGaussian processes[J]. Automatic Control, IEEE Transactions on,2012,57(7):1865-1871.
[18] Julka N, Thirunavukkarasu A, Lendermann P, et al. Making use of prognostics healthmanagement information for aerospace spare components logistics network optimisation[J].Computers in Industry,2011,62(6):613-622.
[19] S Ferreiro, A Arnaiz, B Sierra, I Irigoien.A Bayesian network model integrated in aprognostics and health management system for aircraft line maintenance[J]. Journal ofAerospace Engineering,2011,225(8):886-901.
[20] Musallam M, Johnson C M. Real-time compact thermal models for health management ofpower electronics[J]. Power Electronics, IEEE Transactions on,2010,25(6):1416-1425.
[21] Janoyan K D, Whelan M J. Intelligent transportation infrastructure technologies for conditionassessment and structural health monitoring of highway bridges[M]//Sustainable andResilient Critical Infrastructure Systems. Springer Berlin Heidelberg,2010:159-184.
[22] Teixeira E L S, Tjahjono B, Alfaro S C A. A novel framework to link prognostics and healthmanagement and product–service systems using online simulation[J]. Computers in Industry,2012,63(7):669-679.
[23] Qiu H, Lee J, Lin J, et al. Wavelet filter-based weak signature detection method and itsapplication on rolling element bearing prognostics[J]. Journal of Sound and Vibration,2006,289(4):1066-1090.
[24] Huang R, Xi L, Li X, et al. Residual life predictions for ball bearings based on self-organizingmap and back propagation neural network methods[J]. Mechanical Systems and SignalProcessing,2007,21(1):193-207.
[25] Lee J, Ni J, Djurdjanovic D, et al. Intelligent prognostics tools and e-maintenance[J].Computers in industry,2006,57(6):476-489.
[26] Marichal G N, Artes M, Garcia-Prada J C. An intelligent system for faulty-bearing detectionbased on vibration spectra[J]. Journal of Vibration and Control,2011,17(6):931-942.
[27] Ao Y H. A review on development and trend of Intelligent maintenance system[J]. AdvancedMaterials Research,2011,314:2365-2369.
[28] Zhou X, Xi L, Lee J. Reliability-centered predictive maintenance scheduling for acontinuously monitored system subject to degradation[J]. Reliability Engineering&SystemSafety,2007,92(4):530-534.
[29] Jardine A K S, Lin D, Banjevic D. A review on machinery diagnostics and prognosticsimplementing condition-based maintenance[J]. Mechanical systems and signal processing,2006,20(7):1483-1510.
[30] Heng A, Zhang S, Tan A C C, et al. Rotating machinery prognostics: State of the art,challenges and opportunities[J]. Mechanical Systems and Signal Processing,2009,23(3):724-739.
[31] Isermann R. Fault-diagnosis applications[M]. Springer,2011.
[32] Vachtsevanos G, Wang P. Fault prognosis using dynamic wavelet neuralnetworks[C]//AUTOTESTCON Proceedings,2001. IEEE Systems Readiness TechnologyConference. IEEE,2001:857-870.
[33] Nefti S, Oussalah M. A neural network approach for railway safety prediction[C]//Systems,Man and Cybernetics,2004IEEE International Conference on. IEEE,2004,4:3915-3920.
[34] Lou X, Loparo K A. Bearing fault diagnosis based on wavelet transform and fuzzyinference[J]. Mechanical systems and signal processing,2004,18(5):1077-1095.
[35] Hajnayeb A, Ghasemloonia A, Khadem S E, et al. Application and comparison of anANN-based feature selection method and the genetic algorithm in gearbox fault diagnosis[J].Expert Systems with Applications,2011,38(8):10205-10209.
[36]朱永生,王成栋,张优云.基于证据加权调整方法的神经网络及在故障诊断中的应用[J],机械工程学报,2002,38(6):66-71.
[37]梅杰.基于人工神经网络的旋转机械故障诊断专家系统[D].武汉:武汉理工大学,2011.
[38] Yu J. A hybrid feature selection scheme and self-organizing map model for machine healthassessment[J]. Applied Soft Computing,2011,11(5):4041-4054.
[39]李明华,屈彦明,周孟戈等.基于多Agent及Petri网的变压器故障诊断系统[J].西安交通大学学报,2006,40(2):223-227.
[40] Nicholas NC. A fuzzy Petri net based mode identification algorithm for fault diagnosis ofcomplex systems[J]. System Diagnosis and Prognosis: Security and Condition MonitoringIssues Iii,2003,5107:44-53.
[41] Jiroveanu G, Boel RK. A distributed approach for fault detection and diagnosis based onTime Petri Nets[J]. Mathematics and Computers in Simulation,2006,70(5-6):287-313.
[42] Yang SK. An experiment of state estimation for predictive maintenance using Kalman filteron a DC motor[J]. Reliability Engineering&System Safety,2002,75(1):103-111.
[43] Pandya Y, Parey A. Failure path based modified gear mesh stiffness for spur gear pair withtooth root crack[J]. Engineering Failure Analysis,2012,3:128-149.
[44] Sondkar P, Kahraman A. A dynamic model of a double-helical planetary gear set[J].Mechanism and Machine Theory,2013,70:157-174.
[45] Gasperin M, Juricic D, Boskoski P, et al. Model-based prognostics of gear health usingstochastic dynamical models[J]. Mechanical Systems and Signal Processing,2011,25(2):537-548.
[46] Tran V T, Thom Pham H, Yang B S, et al. Machine performance degradation assessment andremaining useful life prediction using proportional hazard model and support vectormachine[J]. Mechanical Systems and Signal Processing,2012,32:320-330.
[47] Bacha K, Souahlia S, Gossa M. Power transformer fault diagnosis based on dissolved gasanalysis by support vector machine[J]. Electric Power Systems Research,2012,83(1):73-79.
[48] Widodo A, Yang B S. Support vector machine in machine condition monitoring and faultdiagnosis[J]. Mechanical Systems and Signal Processing,2007,21(6):2560-2574.
[49]王红军,徐小力,张建民,设备状态趋势的SVM预示技术研究[J],机械科学与技术,2006,25(4):379-381.
[50] Huang G, Chen H, Zhou Z, et al. Two-class support vector data description[J]. PatternRecognition,2011,44(2):320-329.
[51]潘玉娜,陈进.小波包-支持向量数据描述在轴承性能退化评估中的应用研究[J].振动与冲击,2009,28(4):164-167.
[52] Shen Z, He Z, Chen X, et al. A monotonic degradation assessment index of rolling bearingsusing fuzzy support vector data description and running time[J]. Sensors,2012,12(8):10109-10135.
[53] Zhu X, Zhang Y, Zhu Y. Bearing performance degradation assessment based on the roughsupport vector data description[J]. Mechanical Systems and Signal Processing,2013,34(1-2):203-217.
[54]刘美芳,余建波,尹纪庭.基于贝叶斯推论和自组织映射的轴承性能退化评估方法[J].计算机集成制造系统,2012,18(010):2237-2244.
[55] Jin S, Liu Y, Lin Z. A Bayesian network approach for fixture fault diagnosis in launch of theassembly process[J]. International Journal of Production Research,2012,50(23):6655-6666.
[56] Sankararaman S, Mahadevan S. Bayesian methodology for diagnosis uncertaintyquantification and health monitoring[J]. Structural Control and Health Monitoring,2013,20(1):88-106.
[57] Barber D. Bayesian reasoning and machine learning[M]. Cambridge University Press,2012.
[58] Tipping M E. Sparse Bayesian learning and the relevance vector machine[J]. The Journal ofMachine Learning Research,2001,1:211-244.
[59] Caesarendra W, Widodo A, Yang B S. Application of relevance vector machine and logisticregression for machine degradation assessment[J]. Mechanical Systems and SignalProcessing,2010,24(4):1161-1171.
[60] Widodo A, Kim E Y, Son J D, et al. Fault diagnosis of low speed bearing based on relevancevector machine and support vector machine[J]. Expert Systems with Applications,2009,36(3):7252-7261.
[61] Carl Edward Rasmussen and Christopher K.I.Williams. Gaussian processes for machinelearning[M]. The MIT Press,2006.
[62] Murphy K P. Machine learning: a probabilistic perspective[M]. The MIT Press,2012.
[63] Kapoor A, Grauman K, Urtasun R, et al. Gaussian processes for object categorization[J].International Journal of Computer Vision,2010,88(2):169-188.
[64] Lin Y Y, Liu T L, Fuh C S. Multiple kernel learning for dimensionality reduction[J]. PatternAnalysis and Machine Intelligence, IEEE Transactions on,2011,33(6):1147-1160.
[65] Savitsky T, Vannucci M, Sha N. Variable selection for nonparametric Gaussian process priors:Models and computational strategies[J]. Statistical Science,2011,26(1):130-149.
[66] Li Q, Salman R, Test E, et al. Parallel multitask cross validation for support vector machineusing GPU[J]. Journal of Parallel and Distributed Computing,2013,73(3):293-302.
[67] Franc V, Zien A, Scholkopf B. Support vector machines as probabilisticmodels[C]//Proceedings of the28th International Conference on Machine Learning (ICML).2011:665-672.
[68] Diosan L, Rogozan A, Pecuchet J P. Improving classification performance of support vectormachine by genetically optimising kernel shape and hyper-parameters[J]. AppliedIntelligence,2012,36(2):280-294.
[69] Gibson N P, Aigrain S, Roberts S, et al. A Gaussian process framework for modellinginstrumental systematics: application to transmission spectroscopy[J]. Monthly Notices of theRoyal Astronomical Society,2012,419(3):2683-2694.
[70] Paul R, Triebel R, Rus D, et al. Semantic categorization of outdoor scenes with uncertaintyestimates using multi-class Gaussian process classification[C]//Intelligent Robots andSystems (IROS),2012IEEE/RSJ International Conference on. IEEE,2012:2404-2410.
[71] Khan E, Mohamed S, Murphy K. Fast Bayesian inference for non-conjugate Gaussianprocess regression[C]//Advances in Neural Information Processing Systems25.2012:3149-3157.
[72] Xu C, Tao D, Li Y, et al. Large-margin multi-view Gaussian process for imageclassification[C]//Proceedings of the Fifth International Conference on Internet MultimediaComputing and Service. ACM,2013:7-12.
[73] Lutz A. Robust classification and semi-supervised object localization with Gaussianprocesses[M]//Pattern Recognition. Springer Berlin Heidelberg,2011:456-461.
[74] Kapoor A, Grauman K, Urtasun R, et al. Active learning with Gaussian processes for objectcategorization[C]//Computer Vision,2007. ICCV2007. IEEE11th International Conferenceon. IEEE,2007:1-8.
[75] Rodner E, Wacker E S, Kemmler M, et al. One-class classification for anomaly detection inwire ropes with Gaussian processes in a few lines of code[J]. Training,2011,1:1-5.
[76] Polajnar T, Rogers S, Girolami M. Classification of protein interaction sentences via Gaussianprocesses[M]//Pattern Recognition in Bioinformatics. Springer Berlin Heidelberg,2009:282-292.
[77] Zhong M, Lotte F, Girolami M, et al. Classifying EEG for brain computer interfaces usingGaussian processes[J]. Pattern Recognition Letters,2008,29(3):354-359.
[78]王华忠.高斯过程及其在软测量建模中的应用[J].化工学报,2007,58(11):2840-2845.
[79]王磊,邹北骥,彭小宁等.基于高斯过程的表情动作单元跟踪技术[J].电子学报,2007,35(11):2087-2091.
[80]熊志化.高斯过程模型及其在工业过程软测量中的应用研究[D].上海:上海交通大学,2006.
[81]陈世敏.山区道路降雨量推估模式建立之研究—以阿里山公路为例[D].台湾:台湾国立科技大学,2005.
[82]陈凤.基于HRRP和JEM信号的雷达目标识别技术研究[D].西安:西安电子科技大学,2009.
[83]姚伏天,基于高斯过程的高光谱图像分类研究[D]，杭州:浙江大学,2010.
[84]孙欣尧,王雪,王晟.无线传感网络协同概率多模识别方法[J].通信学报,2011,32(6):141-147.
[85] Urtasun R, Darrell T. Discriminative Gaussian process latent variable model forclassification[C]//Proceedings of the24th international conference on Machine learning.ACM,2007:927-934.
[86] Jakab H, Csato L. Improving Gaussian process value function approximation in policygradient algorithms[M]//Artificial Neural Networks and Machine Learning–ICANN2011.Springer Berlin Heidelberg,2011:221-228.
[87] Mukai T, Kuriyama S. Geostatistical motion interpolation[C]//ACM Transactions on Graphics(TOG). ACM,2005,24(3):1062-1070.
[88] Zhu X, Ghahramani Z, Lafferty J. Semi-supervised learning using Gaussian fields andharmonic functions[C]//ICML.2003,3:912-919.
[89] Csato L, Opper M. Sparse representation for Gaussian process models[J]. Advances in neuralinformation processing systems,2001:444-450.
[90] Wiesel A. Unified framework to regularized covariance estimation in scaled Gaussianmodels[J]. Signal Processing, IEEE Transactions on,2012,60(1):29-38.
[91] Honkela A, Gao P, Ropponen J, et al. tigre: Transcription factor inference through Gaussianprocess reconstruction of expression for bioconductor[J]. Bioinformatics,2011,27(7):1026-1027.
[92] Gramacy R B, Taddy M. Categorical inputs, sensitivity analysis, optimization and importancetempering with tgp version2, an R package for treed Gaussian process models[J]. Journal ofStatistical Software,2012,33:1-48.
[93] Behm J E, Edmonds D A, Harmon J P, et al. Multilevel statistical models and the analysis ofexperimental data[J]. Ecology,2013,94(7):1479-1486.
[94]刘开云,刘保国,徐冲.基于遗传-组合核函数高斯过程回归算法的边坡非线性变形时序分析智能模型[J].岩石力学与工程学报,2009,28(10):2128-2134.
[95]刘开云,方昱,刘保国.基于进化高斯过程回归算法的隧道工程弹塑性模型参数反演[J].2011,33(6):883-889.
[96]冯爱明,方利民,林敏.近红外光谱分析中的高斯过程回归方法[J],光谱学与光谱分析,2011,31(6):1514-1517.
[97]单志辉.基于高斯过程回归的翼型快速设计研究[D].南京:南京航空航天大学,2011.
[98]徐冲.分岔隧道设计施工优化与稳定性评价[D].北京:北京交通大学,2011.
[99]杨晓晶.基于EMD和高斯过程回归组合模型的短期电力负荷预测方法研究[D].昆明:昆明理工大学,2010.
[100]夏益民.基于传感器信息融合的移动机器人定位与地图创建研究[D].广州:广东工业大学,2011.
[101]张健中.一类连续化工生产过程的模型辨识及非线性预测控制研究[D].哈尔滨:哈尔滨工业大学,2010.
[102] Rasmussen C E, Nickisch H. Gaussian processes for machine learning (GPML) toolbox[J].The Journal of Machine Learning Research,2010,9999:3011-3015.
[103] Yu J. Bearing performance degradation assessment using locality preserving projections andGaussian mixture models[J]. Mechanical Systems and Signal Processing,2011,25(7):2573-2588.
[104] Yang Z, Cai L, Gao L, et al. Adaptive redundant lifting wavelet transform based on fittingfor fault feature extraction of roller bearings[J]. Sensors,2012,12(4):4381-4398.
[105] van Wyk B J, van Wyk M A, Qi G. Difference histograms: A new tool for time seriesanalysis applied to bearing fault diagnosis[J]. Pattern Recognition Letters,2009,30(6):595-599.
[106] Siegel D, Al-Atat H, Shauche V, et al. Novel method for rolling element bearing healthassessment—A tachometer-less synchronously averaged envelope feature extractiontechnique[J]. Mechanical Systems and Signal Processing,2012,29:362-376.
[107] Pan Y, Chen J, Li X. Bearing performance degradation assessment based on lifting waveletpacket decomposition and fuzzy c-means[J]. Mechanical Systems and Signal Processing,2010,24(2):559-566.
[108] Yang L, Lv J, Xiang Y. Underdetermined blind source separation by parallel factor analysisin time-frequency domain[J]. Cognitive computation,2013:1-8.
[109] Liao L, Lee J. A novel method for machine performance degradation assessment based onfixed cycle features test[J]. Journal of Sound and Vibration,2009,326(3):894-908.
[110] Widodo A, Yang B S. Machine health prognostics using survival probability and supportvector machine[J]. Expert Systems with Applications,2011,38(7):8430-8437.
[111] C me E, Oukhellou L, Den ux T, et al. Fault diagnosis of a railway device usingsemi-supervised independent factor analysis with mixing constraints[J]. Pattern Analysisand Applications,2012,15(3):313-326.
[112] Yu J B. Bearing performance degradation assessment using locality preserving projections[J].Expert Systems with Applications,2011,38(6):7440-7450.
[113] Cai D. Spectral regression: a regression framework for efficient regularized subspacelearning[M]. ProQuest,2009.
[114] Cai D, He X, Han J. Speed up kernel discriminant analysis[J]. The VLDB Journal,2011,20(1):21-33.
[115] Wang C, Chen J, Sun Y. Sensor network localization using kernel spectral regression[J].Wireless Communications and Mobile Computing,2010,10(8):1045-1054.
[116] Lin G F, Zhu H, Fan Y D, et al. Human action recognition based on random spectralregression[M]//Artificial Intelligence and Computational Intelligence. Springer BerlinHeidelberg,2011:451-461.
[117] Zhang B, Gao Y. Spectral regression dimension reduction for multiple features facial imageretrieval[J]. International Journal of Biometrics,2012,4(1):77-101.
[118] Garg G, Singh V, Gupta J R P, et al. Wrapper based wavelet feature optimization for EEGsignals[J]. Biomedical Engineering Letters,2012,2(1):24-37.
[119] Yaqub M F, Gondal I, Kamruzzaman J. Inchoate fault detection framework: Adaptiveselection of wavelet nodes and cumulant orders[J]. Instrumentation and Measurement,IEEE Transactions on,2012,61(3):685-695.
[120] Rafiee J, Tse P W, Harifi A, et al. A novel technique for selecting mother wavelet functionusing an intelligent fault diagnosis system[J]. Expert Systems with Applications,2009,36(3):4862-4875.
[121] Yu J. Health condition monitoring of machines based on hidden markov model andcontribution analysis[J]. Instrumentation and Measurement, IEEE Transactions on,2012,61(8):2200-2211.
[122] Loparo, K. Bearings vibration data set, case western reserve university. Available online:http://www.eecs.case.edu/laboratory/bearing/welcome_overview.htm,2012,3.
[123] Catanzaro B, Sundaram N, Keutzer K. Fast support vector machine training andclassification on graphics processors[C]//Proceedings of the25th international conferenceon Machine learning. ACM,2008:104-111.
[124] Tohmé M, Lengellé R. Maximum margin one class support vector machines for multiclassproblems[J]. Pattern Recognition Letters,2011,32(13):1652-1658.
[125] Miller D J, Raghuram J, Kesidis G, et al. Improved generative semisupervised learningbased on finely grained component-conditional class labeling[J]. Neural Computation,2012,24(7):1926-1966.
[126] Zhang T, Oles F. The value of unlabeled data for classification problems[C]//Proceedings ofthe Seventeenth International Conference on Machine Learning,(Langley, P., ed.).2000:1191-1198.
[127] Joachims T. Learning to classify text using support vector machines: methods, theory andalgorithms[M]. Kluwer Academic Publishers,2002.
[128] Zhu X, Goldberg A B. Introduction to semi-supervised learning[J]. Synthesis lectures onartificial intelligence and machine learning,2009,3(1):1-130.
[129] Zhu X, Lafferty J, Ghahramani Z. Combining active learning and semi-supervised learningusing Gaussian fields and harmonic functions[C]//ICML2003workshop on the continuumfrom labeled to unlabeled data in machine learning and data mining.2003:58-65.
[130] Cai D, Wang X, He X. Probabilistic dyadic data analysis with local and globalconsistency[C]//Proceedings of the26th Annual International Conference on MachineLearning. ACM,2009:105-112.
[131] Weston J, Ratle F, Mobahi H, et al. Deep learning via semi-supervisedembedding[M]//Neural Networks: Tricks of the Trade. Springer Berlin Heidelberg,2012.
[132] Luo Y, Tao D, Geng B, et al. Manifold regularized multi-task learning for semi-supervisedmulti-label image classification[J]. Image Processing, IEEE Transactions on,2013,22(2):523-536.
[133]李宏伟,刘扬,卢汉清,等.结合半监督核的高斯过程分类[J].自动化学报,2009,35(7):888-895.
[134] Zhao Z K, Qian J S, Cheng J, et al. Manifold regularization based semi-supervisedregression on multivariate time series[J]. Journal of China University of Mining&Technology,2011,40(3):492-498.
[135] Feng H. Gene classification using parameter-free semi-supervised manifold learning[J].IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB),2012,9(3):818-827.
[136] Kapoor A, Ahn H, Qi Y, et al. Hyperparameter and kernel learning for graph basedsemi-supervised classification[C]//Advances in Neural Information Processing Systems.2005:627-634.
[137] Sindhwani V, Chu W, Keerthi S S. Semi-supervised Gaussian process classifiers[C]//IJCAI.2007:1059-1064.
[138] Brinker K, Fürnkranz J, Hüllermeier E. A unified model for multilabel classification andranking[C]//Proceedings of the2006conference on ECAI2006:17th European Conferenceon Artificial Intelligence. IOS Press,2006:489-493.
[139] Patel I, Rao Y S. Speech recognition using hidden Markov model with MFCC-subbandtechnique[C]//Recent Trends in Information, Telecommunication and Computing (ITC),2010International Conference on. IEEE,2010:168-172.
[140] Azeem S A, Ahmed H. Effective technique for the recognition of offline Arabic handwrittenwords using hidden Markov models[J]. International Journal on Document Analysis andRecognition (IJDAR),2013:1-14.
[141] Wan J, Ruan Q, An G, et al. Gesture recognition based on hidden Markov model from sparserepresentative observations[C]//Signal Processing (ICSP),2012IEEE11th InternationalConference on. IEEE,2012,2:1180-1183.
[142] Lin J C, Wu C H, Wei W L. Error weighted semi-coupled hidden Markov model foraudio-visual emotion recognition[J]. Multimedia, IEEE Transactions on,2012,14(1):142-156.
[143] Boutros T, Liang M. Detection and diagnosis of bearing and cutting tool faults using hiddenMarkov models[J]. Mechanical systems and signal processing,2011,25(6):2102-2124.
[144] Soualhi A, Clerc G, Razik H, et al. Fault detection and diagnosis of induction motors basedon hidden Markov model[C]//Electrical Machines (ICEM),2012International Conferenceon. IEEE,2012:1693-1699.
[145] Zhang R G, Tan Y H. A novel fault diagnosis for rolling element bearings based oncontinuous wavelet transform and hidden Markov model[J]. Applied Mechanics andMaterials,2012,109:313-317.
[146] Liu T, Chen J, Dong G. Zero crossing and coupled hidden Markov model for a rollingbearing performance degradation assessment[J]. Journal of Vibration and Control,2013,7:107-119.
[147] Tobon-Mejia D A, Medjaher K, Zerhouni N, et al. A data-driven failure prognostics methodbased on mixture of Gaussians hidden Markov models[J]. Reliability, IEEE Transactions on,2012,61(2):491-503.
[148] Liu Z, Li Q, Liu X, et al. A hybrid LSSVR/HMM-based prognostic approach[J]. Sensors,2013,13(5):5542-5560.
[149]肖文斌.基于耦合隐马尔可夫模型的滚动轴承故障诊断与性能退化评估研究[D].上海:上海交通大学,2011.
[150]李航.统计学习方法[M].北京:清华大学出版社,2012.
[151]朱义.基于CHMM的设备性能退化评估方法研究[D].上海:上海交通大学,2009.
[152] Liu T, Chen J, Zhou X N, et al. Bearing performance degradation sssessment using lineardiscriminant analysis and coupled HMM[C]//Journal of Physics: Conference Series. IOPPublishing,2012,364(1):012028.
[153] Wang M, Wang J. CHMM for tool condition monitoring and remaining useful lifeprediction[J]. The International Journal of Advanced Manufacturing Technology,2012,59(5-8):463-471.
[154] Kocijan J, Tanko V. Prognosis of gear health using Gaussian processmodel[C]//EUROCON-International Conference on Computer as a Tool (EUROCON).IEEE,2011:1-4.
[155] Boskoski P, Gasperin M, Petelin D. Bearing fault prognostics based on signal complexityand Gaussian process models[C]//Prognostics and Health Management (PHM),2012IEEEConference on. IEEE,2012:1-8.
[156] Hong S, Zhou Z. Remaining useful life prognosis of bearing based on Gauss processregression[C]//Biomedical Engineering and Informatics (BMEI),20125th InternationalConference on. IEEE,2012:1575-1579.
[157] Hong S, Zhou Z. Application of Gaussian process regression for bearing degradationassessment[C]. Information Science and Service Science and Data Mining (ISSDM),2012:644-648.
[158]胡明.基于高斯过程的非线性预测控制[D].广州:华南理工大学,2012.
[159] Kim M, De la Torre F. Multiple instance learning via Gaussian processes[J]. Data Miningand Knowledge Discovery,2013:1-29.
[160] Roberts S, Osborne M, Ebden M, et al. Gaussian processes for time-series modelling[J].Philosophical Transactions of the Royal Society A: Mathematical, Physical and EngineeringSciences,2013,371(1984).
[161] Del Moral P, Doucet A, Jasra A. An adaptive sequential Monte Carlo method forapproximate Bayesian computation[J]. Statistics and Computing,2012,22(5):1009-1020.
[162] Anitescu M, Chen J, Wang L. A matrix-free approach for solving the parametric Gaussianprocess maximum likelihood problem[J]. SIAM Journal on Scientific Computing,2012,34(1): A240-A262.
[163] Alvarez M A, Lawrence N D. Computationally efficient convolved multiple output Gaussianprocesses[J]. Journal of Machine Learning Research,2011,12(5):1459-1500.
[164] Lázaro-Gredilla M, Qui onero-Candela J, Rasmussen C E, et al. Sparse spectrum Gaussianprocess regression[J]. The Journal of Machine Learning Research,2010,99:1865-1881.
[165] J. Lee, H. Qiu, G. Yu, et al. Rexnord technical services,"Bearing data set", IMS, Universityof Cincinnati.,2012.
[166] Tang B, Dong S, Song T. Method for eliminating mode mixing of empirical modedecomposition based on the revised blind source separation[J]. Signal Processing,2012,92(1):248-258.
[167] Dong S, Tang B, Zhang Y. A repeated single-channel mechanical signal blind separationmethod based on morphological filtering and singular value decomposition[J].Measurement,2012,45(8):2052-2063.
[168] Yan J, Guo C, Wang X. A dynamic multi-scale Markov model based methodology forremaining life prediction[J]. Mechanical Systems and Signal Processing,2011,25(4):1364-1376.
[2]《矿井提升机故障处理和技术改造》编委会.矿井提升机故障处理和技术改造(第二版)[M].北京:机械工业出版社,2013.
[3]陈进.机械故障诊断理论及应用[M].北京:高等教育出版社,2010.
[4]雷金波.基于逻辑回归和支持向量机的设备状态退化评估与趋势预测研究[D].上海:上海交通大学,2008.
[5]王增才,王树梁,任锴胜等.基于EEMD的提升机天轮轴承故障诊断方法[J].煤炭学报,2012,37(4):689-694.
[6]王树梁.基于HHT的提升机天轮轴承故障诊断方法研究[D].济南:山东大学,2011.
[7] J. Lee. Intelligent Maintenance Systems (IMS) Technologies. www.imscenter.net,2010.
[8]钟鑫.基于逻辑回归和高斯混合模型的设备故障诊断技术研究与应用[D].北京:北京化工大学,2010.
[9]郭磊.基于核模式分析方法的旋转机械性能退化评估技术研究[D].上海:上海交通大学,2009.
[10]潘玉娜,陈进,李兴林.基于模糊c-均值的设备性能退化评估方法[J].上海交通大学学报,2009,43(11):1794-1797.
[11] Debin Zhao, Jihong Yan. Performance prediction methodology based on patternrecognition[J]. Signal Processing,2011,91:2194-2203.
[12] Linxia Liao, Jay Lee. Design of a reconfigurable prognostics platform for machine tools[J].Expert Systems with Applications,2010,37:240-252.
[13] Ilias Bilionis, Nicholas Zabaras. Multi-output local Gaussian process regression: Applicationsto uncertainty quantification[J]. Journal of Computational Physics,2012,231(17):5718-5746.
[14] Anjishnu Banerjee, David B. Dunson, and Surya T. Tokdar.Efficient Gaussian processregression for large datasets[J]. Biometrika,2013,100(1):75-89.
[15] Dongbing Gu, Huosheng Hu. Spatial Gaussian process regression with mobile sensornetworks[J]. IEEE Transactions on Neural Networks and Learning Systems,2012,23(8):1279-1290.
[16] Michael Kemmler, Erik Rodner, Esther-Sabrina Wacker. One-class classification withGaussian processes[J]. Pattern Recognition,2013,46:3507-3518.
[17] Deisenroth M.P., Turner R.D., Huber M.F., et al. Robust filtering and smoothing withGaussian processes[J]. Automatic Control, IEEE Transactions on,2012,57(7):1865-1871.
[18] Julka N, Thirunavukkarasu A, Lendermann P, et al. Making use of prognostics healthmanagement information for aerospace spare components logistics network optimisation[J].Computers in Industry,2011,62(6):613-622.
[19] S Ferreiro, A Arnaiz, B Sierra, I Irigoien.A Bayesian network model integrated in aprognostics and health management system for aircraft line maintenance[J]. Journal ofAerospace Engineering,2011,225(8):886-901.
[20] Musallam M, Johnson C M. Real-time compact thermal models for health management ofpower electronics[J]. Power Electronics, IEEE Transactions on,2010,25(6):1416-1425.
[21] Janoyan K D, Whelan M J. Intelligent transportation infrastructure technologies for conditionassessment and structural health monitoring of highway bridges[M]//Sustainable andResilient Critical Infrastructure Systems. Springer Berlin Heidelberg,2010:159-184.
[22] Teixeira E L S, Tjahjono B, Alfaro S C A. A novel framework to link prognostics and healthmanagement and product–service systems using online simulation[J]. Computers in Industry,2012,63(7):669-679.
[23] Qiu H, Lee J, Lin J, et al. Wavelet filter-based weak signature detection method and itsapplication on rolling element bearing prognostics[J]. Journal of Sound and Vibration,2006,289(4):1066-1090.
[24] Huang R, Xi L, Li X, et al. Residual life predictions for ball bearings based on self-organizingmap and back propagation neural network methods[J]. Mechanical Systems and SignalProcessing,2007,21(1):193-207.
[25] Lee J, Ni J, Djurdjanovic D, et al. Intelligent prognostics tools and e-maintenance[J].Computers in industry,2006,57(6):476-489.
[26] Marichal G N, Artes M, Garcia-Prada J C. An intelligent system for faulty-bearing detectionbased on vibration spectra[J]. Journal of Vibration and Control,2011,17(6):931-942.
[27] Ao Y H. A review on development and trend of Intelligent maintenance system[J]. AdvancedMaterials Research,2011,314:2365-2369.
[28] Zhou X, Xi L, Lee J. Reliability-centered predictive maintenance scheduling for acontinuously monitored system subject to degradation[J]. Reliability Engineering&SystemSafety,2007,92(4):530-534.
[29] Jardine A K S, Lin D, Banjevic D. A review on machinery diagnostics and prognosticsimplementing condition-based maintenance[J]. Mechanical systems and signal processing,2006,20(7):1483-1510.
[30] Heng A, Zhang S, Tan A C C, et al. Rotating machinery prognostics: State of the art,challenges and opportunities[J]. Mechanical Systems and Signal Processing,2009,23(3):724-739.
[31] Isermann R. Fault-diagnosis applications[M]. Springer,2011.
[32] Vachtsevanos G, Wang P. Fault prognosis using dynamic wavelet neuralnetworks[C]//AUTOTESTCON Proceedings,2001. IEEE Systems Readiness TechnologyConference. IEEE,2001:857-870.
[33] Nefti S, Oussalah M. A neural network approach for railway safety prediction[C]//Systems,Man and Cybernetics,2004IEEE International Conference on. IEEE,2004,4:3915-3920.
[34] Lou X, Loparo K A. Bearing fault diagnosis based on wavelet transform and fuzzyinference[J]. Mechanical systems and signal processing,2004,18(5):1077-1095.
[35] Hajnayeb A, Ghasemloonia A, Khadem S E, et al. Application and comparison of anANN-based feature selection method and the genetic algorithm in gearbox fault diagnosis[J].Expert Systems with Applications,2011,38(8):10205-10209.
[36]朱永生,王成栋,张优云.基于证据加权调整方法的神经网络及在故障诊断中的应用[J],机械工程学报,2002,38(6):66-71.
[37]梅杰.基于人工神经网络的旋转机械故障诊断专家系统[D].武汉:武汉理工大学,2011.
[38] Yu J. A hybrid feature selection scheme and self-organizing map model for machine healthassessment[J]. Applied Soft Computing,2011,11(5):4041-4054.
[39]李明华,屈彦明,周孟戈等.基于多Agent及Petri网的变压器故障诊断系统[J].西安交通大学学报,2006,40(2):223-227.
[40] Nicholas NC. A fuzzy Petri net based mode identification algorithm for fault diagnosis ofcomplex systems[J]. System Diagnosis and Prognosis: Security and Condition MonitoringIssues Iii,2003,5107:44-53.
[41] Jiroveanu G, Boel RK. A distributed approach for fault detection and diagnosis based onTime Petri Nets[J]. Mathematics and Computers in Simulation,2006,70(5-6):287-313.
[42] Yang SK. An experiment of state estimation for predictive maintenance using Kalman filteron a DC motor[J]. Reliability Engineering&System Safety,2002,75(1):103-111.
[43] Pandya Y, Parey A. Failure path based modified gear mesh stiffness for spur gear pair withtooth root crack[J]. Engineering Failure Analysis,2012,3:128-149.
[44] Sondkar P, Kahraman A. A dynamic model of a double-helical planetary gear set[J].Mechanism and Machine Theory,2013,70:157-174.
[45] Gasperin M, Juricic D, Boskoski P, et al. Model-based prognostics of gear health usingstochastic dynamical models[J]. Mechanical Systems and Signal Processing,2011,25(2):537-548.
[46] Tran V T, Thom Pham H, Yang B S, et al. Machine performance degradation assessment andremaining useful life prediction using proportional hazard model and support vectormachine[J]. Mechanical Systems and Signal Processing,2012,32:320-330.
[47] Bacha K, Souahlia S, Gossa M. Power transformer fault diagnosis based on dissolved gasanalysis by support vector machine[J]. Electric Power Systems Research,2012,83(1):73-79.
[48] Widodo A, Yang B S. Support vector machine in machine condition monitoring and faultdiagnosis[J]. Mechanical Systems and Signal Processing,2007,21(6):2560-2574.
[49]王红军,徐小力,张建民,设备状态趋势的SVM预示技术研究[J],机械科学与技术,2006,25(4):379-381.
[50] Huang G, Chen H, Zhou Z, et al. Two-class support vector data description[J]. PatternRecognition,2011,44(2):320-329.
[51]潘玉娜,陈进.小波包-支持向量数据描述在轴承性能退化评估中的应用研究[J].振动与冲击,2009,28(4):164-167.
[52] Shen Z, He Z, Chen X, et al. A monotonic degradation assessment index of rolling bearingsusing fuzzy support vector data description and running time[J]. Sensors,2012,12(8):10109-10135.
[53] Zhu X, Zhang Y, Zhu Y. Bearing performance degradation assessment based on the roughsupport vector data description[J]. Mechanical Systems and Signal Processing,2013,34(1-2):203-217.
[54]刘美芳,余建波,尹纪庭.基于贝叶斯推论和自组织映射的轴承性能退化评估方法[J].计算机集成制造系统,2012,18(010):2237-2244.
[55] Jin S, Liu Y, Lin Z. A Bayesian network approach for fixture fault diagnosis in launch of theassembly process[J]. International Journal of Production Research,2012,50(23):6655-6666.
[56] Sankararaman S, Mahadevan S. Bayesian methodology for diagnosis uncertaintyquantification and health monitoring[J]. Structural Control and Health Monitoring,2013,20(1):88-106.
[57] Barber D. Bayesian reasoning and machine learning[M]. Cambridge University Press,2012.
[58] Tipping M E. Sparse Bayesian learning and the relevance vector machine[J]. The Journal ofMachine Learning Research,2001,1:211-244.
[59] Caesarendra W, Widodo A, Yang B S. Application of relevance vector machine and logisticregression for machine degradation assessment[J]. Mechanical Systems and SignalProcessing,2010,24(4):1161-1171.
[60] Widodo A, Kim E Y, Son J D, et al. Fault diagnosis of low speed bearing based on relevancevector machine and support vector machine[J]. Expert Systems with Applications,2009,36(3):7252-7261.
[61] Carl Edward Rasmussen and Christopher K.I.Williams. Gaussian processes for machinelearning[M]. The MIT Press,2006.
[62] Murphy K P. Machine learning: a probabilistic perspective[M]. The MIT Press,2012.
[63] Kapoor A, Grauman K, Urtasun R, et al. Gaussian processes for object categorization[J].International Journal of Computer Vision,2010,88(2):169-188.
[64] Lin Y Y, Liu T L, Fuh C S. Multiple kernel learning for dimensionality reduction[J]. PatternAnalysis and Machine Intelligence, IEEE Transactions on,2011,33(6):1147-1160.
[65] Savitsky T, Vannucci M, Sha N. Variable selection for nonparametric Gaussian process priors:Models and computational strategies[J]. Statistical Science,2011,26(1):130-149.
[66] Li Q, Salman R, Test E, et al. Parallel multitask cross validation for support vector machineusing GPU[J]. Journal of Parallel and Distributed Computing,2013,73(3):293-302.
[67] Franc V, Zien A, Scholkopf B. Support vector machines as probabilisticmodels[C]//Proceedings of the28th International Conference on Machine Learning (ICML).2011:665-672.
[68] Diosan L, Rogozan A, Pecuchet J P. Improving classification performance of support vectormachine by genetically optimising kernel shape and hyper-parameters[J]. AppliedIntelligence,2012,36(2):280-294.
[69] Gibson N P, Aigrain S, Roberts S, et al. A Gaussian process framework for modellinginstrumental systematics: application to transmission spectroscopy[J]. Monthly Notices of theRoyal Astronomical Society,2012,419(3):2683-2694.
[70] Paul R, Triebel R, Rus D, et al. Semantic categorization of outdoor scenes with uncertaintyestimates using multi-class Gaussian process classification[C]//Intelligent Robots andSystems (IROS),2012IEEE/RSJ International Conference on. IEEE,2012:2404-2410.
[71] Khan E, Mohamed S, Murphy K. Fast Bayesian inference for non-conjugate Gaussianprocess regression[C]//Advances in Neural Information Processing Systems25.2012:3149-3157.
[72] Xu C, Tao D, Li Y, et al. Large-margin multi-view Gaussian process for imageclassification[C]//Proceedings of the Fifth International Conference on Internet MultimediaComputing and Service. ACM,2013:7-12.
[73] Lutz A. Robust classification and semi-supervised object localization with Gaussianprocesses[M]//Pattern Recognition. Springer Berlin Heidelberg,2011:456-461.
[74] Kapoor A, Grauman K, Urtasun R, et al. Active learning with Gaussian processes for objectcategorization[C]//Computer Vision,2007. ICCV2007. IEEE11th International Conferenceon. IEEE,2007:1-8.
[75] Rodner E, Wacker E S, Kemmler M, et al. One-class classification for anomaly detection inwire ropes with Gaussian processes in a few lines of code[J]. Training,2011,1:1-5.
[76] Polajnar T, Rogers S, Girolami M. Classification of protein interaction sentences via Gaussianprocesses[M]//Pattern Recognition in Bioinformatics. Springer Berlin Heidelberg,2009:282-292.
[77] Zhong M, Lotte F, Girolami M, et al. Classifying EEG for brain computer interfaces usingGaussian processes[J]. Pattern Recognition Letters,2008,29(3):354-359.
[78]王华忠.高斯过程及其在软测量建模中的应用[J].化工学报,2007,58(11):2840-2845.
[79]王磊,邹北骥,彭小宁等.基于高斯过程的表情动作单元跟踪技术[J].电子学报,2007,35(11):2087-2091.
[80]熊志化.高斯过程模型及其在工业过程软测量中的应用研究[D].上海:上海交通大学,2006.
[81]陈世敏.山区道路降雨量推估模式建立之研究—以阿里山公路为例[D].台湾:台湾国立科技大学,2005.
[82]陈凤.基于HRRP和JEM信号的雷达目标识别技术研究[D].西安:西安电子科技大学,2009.
[83]姚伏天,基于高斯过程的高光谱图像分类研究[D]，杭州:浙江大学,2010.
[84]孙欣尧,王雪,王晟.无线传感网络协同概率多模识别方法[J].通信学报,2011,32(6):141-147.
[85] Urtasun R, Darrell T. Discriminative Gaussian process latent variable model forclassification[C]//Proceedings of the24th international conference on Machine learning.ACM,2007:927-934.
[86] Jakab H, Csato L. Improving Gaussian process value function approximation in policygradient algorithms[M]//Artificial Neural Networks and Machine Learning–ICANN2011.Springer Berlin Heidelberg,2011:221-228.
[87] Mukai T, Kuriyama S. Geostatistical motion interpolation[C]//ACM Transactions on Graphics(TOG). ACM,2005,24(3):1062-1070.
[88] Zhu X, Ghahramani Z, Lafferty J. Semi-supervised learning using Gaussian fields andharmonic functions[C]//ICML.2003,3:912-919.
[89] Csato L, Opper M. Sparse representation for Gaussian process models[J]. Advances in neuralinformation processing systems,2001:444-450.
[90] Wiesel A. Unified framework to regularized covariance estimation in scaled Gaussianmodels[J]. Signal Processing, IEEE Transactions on,2012,60(1):29-38.
[91] Honkela A, Gao P, Ropponen J, et al. tigre: Transcription factor inference through Gaussianprocess reconstruction of expression for bioconductor[J]. Bioinformatics,2011,27(7):1026-1027.
[92] Gramacy R B, Taddy M. Categorical inputs, sensitivity analysis, optimization and importancetempering with tgp version2, an R package for treed Gaussian process models[J]. Journal ofStatistical Software,2012,33:1-48.
[93] Behm J E, Edmonds D A, Harmon J P, et al. Multilevel statistical models and the analysis ofexperimental data[J]. Ecology,2013,94(7):1479-1486.
[94]刘开云,刘保国,徐冲.基于遗传-组合核函数高斯过程回归算法的边坡非线性变形时序分析智能模型[J].岩石力学与工程学报,2009,28(10):2128-2134.
[95]刘开云,方昱,刘保国.基于进化高斯过程回归算法的隧道工程弹塑性模型参数反演[J].2011,33(6):883-889.
[96]冯爱明,方利民,林敏.近红外光谱分析中的高斯过程回归方法[J],光谱学与光谱分析,2011,31(6):1514-1517.
[97]单志辉.基于高斯过程回归的翼型快速设计研究[D].南京:南京航空航天大学,2011.
[98]徐冲.分岔隧道设计施工优化与稳定性评价[D].北京:北京交通大学,2011.
[99]杨晓晶.基于EMD和高斯过程回归组合模型的短期电力负荷预测方法研究[D].昆明:昆明理工大学,2010.
[100]夏益民.基于传感器信息融合的移动机器人定位与地图创建研究[D].广州:广东工业大学,2011.
[101]张健中.一类连续化工生产过程的模型辨识及非线性预测控制研究[D].哈尔滨:哈尔滨工业大学,2010.
[102] Rasmussen C E, Nickisch H. Gaussian processes for machine learning (GPML) toolbox[J].The Journal of Machine Learning Research,2010,9999:3011-3015.
[103] Yu J. Bearing performance degradation assessment using locality preserving projections andGaussian mixture models[J]. Mechanical Systems and Signal Processing,2011,25(7):2573-2588.
[104] Yang Z, Cai L, Gao L, et al. Adaptive redundant lifting wavelet transform based on fittingfor fault feature extraction of roller bearings[J]. Sensors,2012,12(4):4381-4398.
[105] van Wyk B J, van Wyk M A, Qi G. Difference histograms: A new tool for time seriesanalysis applied to bearing fault diagnosis[J]. Pattern Recognition Letters,2009,30(6):595-599.
[106] Siegel D, Al-Atat H, Shauche V, et al. Novel method for rolling element bearing healthassessment—A tachometer-less synchronously averaged envelope feature extractiontechnique[J]. Mechanical Systems and Signal Processing,2012,29:362-376.
[107] Pan Y, Chen J, Li X. Bearing performance degradation assessment based on lifting waveletpacket decomposition and fuzzy c-means[J]. Mechanical Systems and Signal Processing,2010,24(2):559-566.
[108] Yang L, Lv J, Xiang Y. Underdetermined blind source separation by parallel factor analysisin time-frequency domain[J]. Cognitive computation,2013:1-8.
[109] Liao L, Lee J. A novel method for machine performance degradation assessment based onfixed cycle features test[J]. Journal of Sound and Vibration,2009,326(3):894-908.
[110] Widodo A, Yang B S. Machine health prognostics using survival probability and supportvector machine[J]. Expert Systems with Applications,2011,38(7):8430-8437.
[111] C me E, Oukhellou L, Den ux T, et al. Fault diagnosis of a railway device usingsemi-supervised independent factor analysis with mixing constraints[J]. Pattern Analysisand Applications,2012,15(3):313-326.
[112] Yu J B. Bearing performance degradation assessment using locality preserving projections[J].Expert Systems with Applications,2011,38(6):7440-7450.
[113] Cai D. Spectral regression: a regression framework for efficient regularized subspacelearning[M]. ProQuest,2009.
[114] Cai D, He X, Han J. Speed up kernel discriminant analysis[J]. The VLDB Journal,2011,20(1):21-33.
[115] Wang C, Chen J, Sun Y. Sensor network localization using kernel spectral regression[J].Wireless Communications and Mobile Computing,2010,10(8):1045-1054.
[116] Lin G F, Zhu H, Fan Y D, et al. Human action recognition based on random spectralregression[M]//Artificial Intelligence and Computational Intelligence. Springer BerlinHeidelberg,2011:451-461.
[117] Zhang B, Gao Y. Spectral regression dimension reduction for multiple features facial imageretrieval[J]. International Journal of Biometrics,2012,4(1):77-101.
[118] Garg G, Singh V, Gupta J R P, et al. Wrapper based wavelet feature optimization for EEGsignals[J]. Biomedical Engineering Letters,2012,2(1):24-37.
[119] Yaqub M F, Gondal I, Kamruzzaman J. Inchoate fault detection framework: Adaptiveselection of wavelet nodes and cumulant orders[J]. Instrumentation and Measurement,IEEE Transactions on,2012,61(3):685-695.
[120] Rafiee J, Tse P W, Harifi A, et al. A novel technique for selecting mother wavelet functionusing an intelligent fault diagnosis system[J]. Expert Systems with Applications,2009,36(3):4862-4875.
[121] Yu J. Health condition monitoring of machines based on hidden markov model andcontribution analysis[J]. Instrumentation and Measurement, IEEE Transactions on,2012,61(8):2200-2211.
[122] Loparo, K. Bearings vibration data set, case western reserve university. Available online:http://www.eecs.case.edu/laboratory/bearing/welcome_overview.htm,2012,3.
[123] Catanzaro B, Sundaram N, Keutzer K. Fast support vector machine training andclassification on graphics processors[C]//Proceedings of the25th international conferenceon Machine learning. ACM,2008:104-111.
[124] Tohmé M, Lengellé R. Maximum margin one class support vector machines for multiclassproblems[J]. Pattern Recognition Letters,2011,32(13):1652-1658.
[125] Miller D J, Raghuram J, Kesidis G, et al. Improved generative semisupervised learningbased on finely grained component-conditional class labeling[J]. Neural Computation,2012,24(7):1926-1966.
[126] Zhang T, Oles F. The value of unlabeled data for classification problems[C]//Proceedings ofthe Seventeenth International Conference on Machine Learning,(Langley, P., ed.).2000:1191-1198.
[127] Joachims T. Learning to classify text using support vector machines: methods, theory andalgorithms[M]. Kluwer Academic Publishers,2002.
[128] Zhu X, Goldberg A B. Introduction to semi-supervised learning[J]. Synthesis lectures onartificial intelligence and machine learning,2009,3(1):1-130.
[129] Zhu X, Lafferty J, Ghahramani Z. Combining active learning and semi-supervised learningusing Gaussian fields and harmonic functions[C]//ICML2003workshop on the continuumfrom labeled to unlabeled data in machine learning and data mining.2003:58-65.
[130] Cai D, Wang X, He X. Probabilistic dyadic data analysis with local and globalconsistency[C]//Proceedings of the26th Annual International Conference on MachineLearning. ACM,2009:105-112.
[131] Weston J, Ratle F, Mobahi H, et al. Deep learning via semi-supervisedembedding[M]//Neural Networks: Tricks of the Trade. Springer Berlin Heidelberg,2012.
[132] Luo Y, Tao D, Geng B, et al. Manifold regularized multi-task learning for semi-supervisedmulti-label image classification[J]. Image Processing, IEEE Transactions on,2013,22(2):523-536.
[133]李宏伟,刘扬,卢汉清,等.结合半监督核的高斯过程分类[J].自动化学报,2009,35(7):888-895.
[134] Zhao Z K, Qian J S, Cheng J, et al. Manifold regularization based semi-supervisedregression on multivariate time series[J]. Journal of China University of Mining&Technology,2011,40(3):492-498.
[135] Feng H. Gene classification using parameter-free semi-supervised manifold learning[J].IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB),2012,9(3):818-827.
[136] Kapoor A, Ahn H, Qi Y, et al. Hyperparameter and kernel learning for graph basedsemi-supervised classification[C]//Advances in Neural Information Processing Systems.2005:627-634.
[137] Sindhwani V, Chu W, Keerthi S S. Semi-supervised Gaussian process classifiers[C]//IJCAI.2007:1059-1064.
[138] Brinker K, Fürnkranz J, Hüllermeier E. A unified model for multilabel classification andranking[C]//Proceedings of the2006conference on ECAI2006:17th European Conferenceon Artificial Intelligence. IOS Press,2006:489-493.
[139] Patel I, Rao Y S. Speech recognition using hidden Markov model with MFCC-subbandtechnique[C]//Recent Trends in Information, Telecommunication and Computing (ITC),2010International Conference on. IEEE,2010:168-172.
[140] Azeem S A, Ahmed H. Effective technique for the recognition of offline Arabic handwrittenwords using hidden Markov models[J]. International Journal on Document Analysis andRecognition (IJDAR),2013:1-14.
[141] Wan J, Ruan Q, An G, et al. Gesture recognition based on hidden Markov model from sparserepresentative observations[C]//Signal Processing (ICSP),2012IEEE11th InternationalConference on. IEEE,2012,2:1180-1183.
[142] Lin J C, Wu C H, Wei W L. Error weighted semi-coupled hidden Markov model foraudio-visual emotion recognition[J]. Multimedia, IEEE Transactions on,2012,14(1):142-156.
[143] Boutros T, Liang M. Detection and diagnosis of bearing and cutting tool faults using hiddenMarkov models[J]. Mechanical systems and signal processing,2011,25(6):2102-2124.
[144] Soualhi A, Clerc G, Razik H, et al. Fault detection and diagnosis of induction motors basedon hidden Markov model[C]//Electrical Machines (ICEM),2012International Conferenceon. IEEE,2012:1693-1699.
[145] Zhang R G, Tan Y H. A novel fault diagnosis for rolling element bearings based oncontinuous wavelet transform and hidden Markov model[J]. Applied Mechanics andMaterials,2012,109:313-317.
[146] Liu T, Chen J, Dong G. Zero crossing and coupled hidden Markov model for a rollingbearing performance degradation assessment[J]. Journal of Vibration and Control,2013,7:107-119.
[147] Tobon-Mejia D A, Medjaher K, Zerhouni N, et al. A data-driven failure prognostics methodbased on mixture of Gaussians hidden Markov models[J]. Reliability, IEEE Transactions on,2012,61(2):491-503.
[148] Liu Z, Li Q, Liu X, et al. A hybrid LSSVR/HMM-based prognostic approach[J]. Sensors,2013,13(5):5542-5560.
[149]肖文斌.基于耦合隐马尔可夫模型的滚动轴承故障诊断与性能退化评估研究[D].上海:上海交通大学,2011.
[150]李航.统计学习方法[M].北京:清华大学出版社,2012.
[151]朱义.基于CHMM的设备性能退化评估方法研究[D].上海:上海交通大学,2009.
[152] Liu T, Chen J, Zhou X N, et al. Bearing performance degradation sssessment using lineardiscriminant analysis and coupled HMM[C]//Journal of Physics: Conference Series. IOPPublishing,2012,364(1):012028.
[153] Wang M, Wang J. CHMM for tool condition monitoring and remaining useful lifeprediction[J]. The International Journal of Advanced Manufacturing Technology,2012,59(5-8):463-471.
[154] Kocijan J, Tanko V. Prognosis of gear health using Gaussian processmodel[C]//EUROCON-International Conference on Computer as a Tool (EUROCON).IEEE,2011:1-4.
[155] Boskoski P, Gasperin M, Petelin D. Bearing fault prognostics based on signal complexityand Gaussian process models[C]//Prognostics and Health Management (PHM),2012IEEEConference on. IEEE,2012:1-8.
[156] Hong S, Zhou Z. Remaining useful life prognosis of bearing based on Gauss processregression[C]//Biomedical Engineering and Informatics (BMEI),20125th InternationalConference on. IEEE,2012:1575-1579.
[157] Hong S, Zhou Z. Application of Gaussian process regression for bearing degradationassessment[C]. Information Science and Service Science and Data Mining (ISSDM),2012:644-648.
[158]胡明.基于高斯过程的非线性预测控制[D].广州:华南理工大学,2012.
[159] Kim M, De la Torre F. Multiple instance learning via Gaussian processes[J]. Data Miningand Knowledge Discovery,2013:1-29.
[160] Roberts S, Osborne M, Ebden M, et al. Gaussian processes for time-series modelling[J].Philosophical Transactions of the Royal Society A: Mathematical, Physical and EngineeringSciences,2013,371(1984).
[161] Del Moral P, Doucet A, Jasra A. An adaptive sequential Monte Carlo method forapproximate Bayesian computation[J]. Statistics and Computing,2012,22(5):1009-1020.
[162] Anitescu M, Chen J, Wang L. A matrix-free approach for solving the parametric Gaussianprocess maximum likelihood problem[J]. SIAM Journal on Scientific Computing,2012,34(1): A240-A262.
[163] Alvarez M A, Lawrence N D. Computationally efficient convolved multiple output Gaussianprocesses[J]. Journal of Machine Learning Research,2011,12(5):1459-1500.
[164] Lázaro-Gredilla M, Qui onero-Candela J, Rasmussen C E, et al. Sparse spectrum Gaussianprocess regression[J]. The Journal of Machine Learning Research,2010,99:1865-1881.
[165] J. Lee, H. Qiu, G. Yu, et al. Rexnord technical services,"Bearing data set", IMS, Universityof Cincinnati.
[166] Tang B, Dong S, Song T. Method for eliminating mode mixing of empirical modedecomposition based on the revised blind source separation[J]. Signal Processing,2012,92(1):248-258.
[167] Dong S, Tang B, Zhang Y. A repeated single-channel mechanical signal blind separationmethod based on morphological filtering and singular value decomposition[J].Measurement,2012,45(8):2052-2063.
[168] Yan J, Guo C, Wang X. A dynamic multi-scale Markov model based methodology forremaining life prediction[J]. Mechanical Systems and Signal Processing,2011,25(4):1364-1376.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |