基于优化机器学习算法的模拟电路故障诊断研究
|
摘要
模拟电路故障诊断理论和方法的研究是目前研究的热门课题。现代电子技术正在日新月异的高速发展，电路规模和结构日趋模块化和功能化，给模拟电路故障诊断提出了新的更具挑战性的要求。从本质上来看，模拟电路的故障诊断问题，属于一种模式识别问题。当前，模拟电路故障诊断研究中的两大问题是如何提取信号特征和如何建立诊断机的问题。小波理论的出现和发展，机器学习算法的日益成熟，使得利用小波对故障信号进行预处理，再利用机器学习算法来进行故障诊断成为有效和值得研究的解决方法，为模拟电路的故障诊断提供了新的有效途径。神经网络和支持向量机均为机器学习领域非常流行的方法。神经网络的优化原理是基于经验风险最小化，缺点是易陷入局部最优，但在处理大样本情况时性能很好；而支持向量机有严格的理论和数学基础，基于结构风险最小化原则,泛化能力优于前者，且算法具有全局最优性,但目前主要是一种针对小样本统计的理论算法。因此本文在对遗传免疫算法、蚁群算法、小波分析、神经网络和支持向量机进行研究的基础上，深入探讨了模拟电路故障诊断神经网络方法和向量机方法存在的问题，提出一些新的融合算法用于对故障诊断分类器进行优化。而利用小波分析作为神经网络的预处理手段，进行信号的前置处理也是目前研究的热点，可以更好的提取故障信号的特征向量。适用合适的特征向量输入分类器进行故障诊断，可以提高诊断速度并得到更高的诊断正确率。基于此，本文也提出一种新的故障特征小波提取方法，并在实例中将得到的故障特征值输入神经网络完成故障诊断。本文最后还介绍了基于模拟电路故障诊断神经网络理论而研发的自动测试与诊断系统装备的相关技术方案，完成了模拟电路故障诊断神经网络方法从理论到实践的完整过程。本文的主要内容和创新点主要体现在以下几个方面：
1．对模拟电路故障诊断的神经网络方法进行了研究。在分析和阐述了神经网络各种学习规则及用算法原理的基础上，着重介绍了BP网络和RBF网络，以及使用神经网络进行模拟电路故障诊断的方法，并详细分析和比较了两者各自的优缺点和适用的场合。
2.对小波神经网络的模拟电路故障诊断方法进行了研究，并提出新的故障特征提取方法。小波的突出优点，是有良好的时频局部性，因此，可将小波作为故障诊断信号的预处理器，先利用小波变换来对电路测试节点的电压信号进行消噪和分解。本文针对分解后的小波系数，提出一种提取故障特征的方法，即小波系数各分量均方根法。具体操作方法是对消噪和分解后的信号进行小波变换，分别计算其小波系数各分量的均方根，然后再进行主元分析与归一化处理，将得到的数据作为故障特征，对神经网络进行训练，再用训练后的神经网络来进行故障分类，完成故障诊断过程。本文通过诊断实例对此方法进行了详细阐述，并通过诊断结果验证了方法的有效性和实用性。
3.提出一种新的模拟电路故障诊断的优化神经网络方法。本文针对RBF网络核函数参数难以选择的缺点，提出一种新的优化算法——免疫蚁群算法，并将其用于RBF网络参数的寻优过程。接下来将通过优化得到的RBF网络用于模拟电路故障诊断，通过实例给出详细的训练算法。该方法在对免疫算法、蚁群算法进行深入研究的基础上，提出使用注入免疫因子的蚁群来实现对RBF网络核函数参数的寻优，新算法引入免疫算法的“抗体浓度”概念，使算法既具有全局搜索的能力，又提高了收敛性能，在充分搜索寻优空间的同时，提高算法的运行时间。本文对免疫蚁群算法和传统算法进行了比较，并通过诊断实例将免疫蚁群算法用于优化RBF神经网络。实例证明,同遗传神经网络进行比较，这种新的免疫蚁群RBF网络有较少的计算工作量，可以在更快的收敛速度下得到更高的诊断正确率。
4.研究了支持向量机（Support Vector Machine,SVM）用做分类器的一般方法和过程，并提出了一种基于改进蚁群算法对向量机核函数进行优化的算法。支持向量机有扎实的理论基础，和传统的神经网络相比，算法的效率和精度都比较高。但SVM目前在处理海量数据和多分类问题时还存在一些困难，尤其是目前尚无成熟完善的理论来指导选择对分类器性能起关键作用的参数。基于此，本章提出了利用改进蚁群算法进行SVM核函数参数的寻优，利用蚁群算法的良好优化性能，减少SVM的训练量，再用SVM形成故障分类器，达到对故障进行快速诊断的目的。本章最后给出了将优化后的SVM应用于模拟电路故障诊断的仿真实例，并与使用传统双线性搜索法得到参数的SVM故障诊断进行了比较，证明了此方法的有效性、合理性和性能优势。
5.描述基于DSP控制的自动测试与诊断系统(Automatic Test and DiagnosisSystem, ATS)的技术方案。在模拟电路故障诊断理论与方法的指导下，本文介绍了系统的测试原理，硬件结构和软件实现，给出了基于DSP控制的主板模块、激励源模块、数字测试模块、PCB模块的基本原理以及硬软件实现方案。
Analog circuits fault diagnosis theory and method is currently a hot researchtopic. The rapid development of modern electronic technology and modular design ofcircuit size and structure proposed a more challenging requirement for analog circuitfault diagnosis. Basically, the problem of analog circuits fault diagnosis is a patternrecognition problem. The two major problems in this study are how to extract thesignal characteristics and how to create a diagnostic machine.
The appearance and development of wavelet theory and machine learningalgorithms made the study move into a new level. Currently, it’s becoming a verypopular method to use wavelet analysis to pre-process the test signals and usemachine learning algorithm to do diagnosis. This solution provides a new andeffective way to solve the problem of analog circuit fault diagnosis.
Neural network (NN) and support vector machine (SVM) both are effective andpopular methods in machine learning algorithm. The optimization principal of neuralnetwork is based on Empirical Risk Minimization (ERM). NN has a good performancein dealing with large sample case, but its disadvantage is easy to fall into localoptimum. The optimization principal of SVM is based on Structural RiskMinimization（SRM）,which made SVM has a strict mathematical theory basis. Also,SVM has global optimality and better generalization than NN. But at present, SVM ismainly a theoretically algorithm focus on small sample situation.
Therefore, based on research about genetic algorithm, immune algorithm, antcolony system, wavelet analysis, NN and SVM, difficulties of NN-based analog faultdiagnosis approach and SVM-based analog fault diagnosis approach are discussed inthis dissertation. Next, some new optimization algorithms of NN and new waveletmethod of analog fault diagnosis are presented. At the last, a technical solutions ofrelevant automatic test and diagnosis system (ATDS) based on DSP，NN and expertsystem is introduced.The main contents and achievements of the paper are as follows:
1. Neural-network-based fault diagnosis of analog circuits is proposed. Based onresearch and analysis of NN learning principle and application, back propagation (BP)neural network and radial basis function (RBF) neural network are introduced indetail. Described the application of the two networks in fault diagnosis, and compared the advantages and disadvantages of the two. Also, the applicable condition isdiscussed in detail in this paper.
2. A new fault features extraction methods is proposed based on research aboutwavelet theory. The outstanding advantage of wavelet is good time-frequencylocalization. Therefore, we can use wavelet as a preprocessor to do de-noising anddecomposition of test node voltages of analog circuits. Then use the components ofwavelet coefficients to extract fault features. In this paper, a new method based on theroot mean square(RMS) value of the components of wavelet coefficients is proposed.After principal component analysis (PCA) and normalization, the RMS value is usedas the inputs of neural network to identify fault classes. The detailed steps are listedthrough diagnosis examples, and correctness of the methods is also verified.
3. A new optimized neural network based fault diagnosis method of analogcircuits is proposed. Aimed to resolved the problem of hard to choose parameters forRBF neural network(RBFNN), a new method based on immune algorithm and antcolony system(ACS-IP) is presented. This method based on the research of immunealgorithm, ant colony system and genetic algorithm, which bring the concept ofimmune factor into ant colony. ACS-IP introduced ‘Antibody concentration’ toimprove global optimization and convergence of RBFNN. This method cansignificantly shorten the running time, at the same time, it can enlarge optimizationsearching space. The specific steps and application results of the new method andtraditional method are discussed in a comparing way in detail in this paper. Andsuperiority of ACS-IP is proved by the comparison. It has less computational work,faster convergence speed and higher diagnostic accuracy.
4. A new optimization algorithm is proposed to select Support Vector Machine(SVM) parameters for analog circuit fault diagnosis. The theory and algorithm ofSVM, especially the application in fault diagnosis, is researched in this paper. Withsolid theoretical foundation, SVM has better performance than NN. But SVM hasdefects on dealing with mass data and multi-classification problems. Particularly,there is no developed method to choose parameter of SVM. An improved ant colonysystem is proposed to solve this problem. To get better performance with betterparameters, SVM can be designed as classifiers for fault diagnosis with less trainingand shorter running time. The effectiveness and superiority the proposed method isfurther verified by examples.
5. Described the technical solution of Automatic Test and Diagnosis System (ATS). With the theoretical guide of neural network method to do analog circuit faultdiagnosis, the experimental device of DSP-based ATDS is built. This paper introducedthe basic design principles, hardware structure and software realization. The designrules and solution of communication between DSP and PC, DSP-based mainboard,excitement module, analog&digital testing module and printed circuit board (PCB)testing module are all presented in this paper.
1．对模拟电路故障诊断的神经网络方法进行了研究。在分析和阐述了神经网络各种学习规则及用算法原理的基础上，着重介绍了BP网络和RBF网络，以及使用神经网络进行模拟电路故障诊断的方法，并详细分析和比较了两者各自的优缺点和适用的场合。
2.对小波神经网络的模拟电路故障诊断方法进行了研究，并提出新的故障特征提取方法。小波的突出优点，是有良好的时频局部性，因此，可将小波作为故障诊断信号的预处理器，先利用小波变换来对电路测试节点的电压信号进行消噪和分解。本文针对分解后的小波系数，提出一种提取故障特征的方法，即小波系数各分量均方根法。具体操作方法是对消噪和分解后的信号进行小波变换，分别计算其小波系数各分量的均方根，然后再进行主元分析与归一化处理，将得到的数据作为故障特征，对神经网络进行训练，再用训练后的神经网络来进行故障分类，完成故障诊断过程。本文通过诊断实例对此方法进行了详细阐述，并通过诊断结果验证了方法的有效性和实用性。
3.提出一种新的模拟电路故障诊断的优化神经网络方法。本文针对RBF网络核函数参数难以选择的缺点，提出一种新的优化算法——免疫蚁群算法，并将其用于RBF网络参数的寻优过程。接下来将通过优化得到的RBF网络用于模拟电路故障诊断，通过实例给出详细的训练算法。该方法在对免疫算法、蚁群算法进行深入研究的基础上，提出使用注入免疫因子的蚁群来实现对RBF网络核函数参数的寻优，新算法引入免疫算法的“抗体浓度”概念，使算法既具有全局搜索的能力，又提高了收敛性能，在充分搜索寻优空间的同时，提高算法的运行时间。本文对免疫蚁群算法和传统算法进行了比较，并通过诊断实例将免疫蚁群算法用于优化RBF神经网络。实例证明,同遗传神经网络进行比较，这种新的免疫蚁群RBF网络有较少的计算工作量，可以在更快的收敛速度下得到更高的诊断正确率。
4.研究了支持向量机（Support Vector Machine,SVM）用做分类器的一般方法和过程，并提出了一种基于改进蚁群算法对向量机核函数进行优化的算法。支持向量机有扎实的理论基础，和传统的神经网络相比，算法的效率和精度都比较高。但SVM目前在处理海量数据和多分类问题时还存在一些困难，尤其是目前尚无成熟完善的理论来指导选择对分类器性能起关键作用的参数。基于此，本章提出了利用改进蚁群算法进行SVM核函数参数的寻优，利用蚁群算法的良好优化性能，减少SVM的训练量，再用SVM形成故障分类器，达到对故障进行快速诊断的目的。本章最后给出了将优化后的SVM应用于模拟电路故障诊断的仿真实例，并与使用传统双线性搜索法得到参数的SVM故障诊断进行了比较，证明了此方法的有效性、合理性和性能优势。
5.描述基于DSP控制的自动测试与诊断系统(Automatic Test and DiagnosisSystem, ATS)的技术方案。在模拟电路故障诊断理论与方法的指导下，本文介绍了系统的测试原理，硬件结构和软件实现，给出了基于DSP控制的主板模块、激励源模块、数字测试模块、PCB模块的基本原理以及硬软件实现方案。
Analog circuits fault diagnosis theory and method is currently a hot researchtopic. The rapid development of modern electronic technology and modular design ofcircuit size and structure proposed a more challenging requirement for analog circuitfault diagnosis. Basically, the problem of analog circuits fault diagnosis is a patternrecognition problem. The two major problems in this study are how to extract thesignal characteristics and how to create a diagnostic machine.
The appearance and development of wavelet theory and machine learningalgorithms made the study move into a new level. Currently, it’s becoming a verypopular method to use wavelet analysis to pre-process the test signals and usemachine learning algorithm to do diagnosis. This solution provides a new andeffective way to solve the problem of analog circuit fault diagnosis.
Neural network (NN) and support vector machine (SVM) both are effective andpopular methods in machine learning algorithm. The optimization principal of neuralnetwork is based on Empirical Risk Minimization (ERM). NN has a good performancein dealing with large sample case, but its disadvantage is easy to fall into localoptimum. The optimization principal of SVM is based on Structural RiskMinimization（SRM）,which made SVM has a strict mathematical theory basis. Also,SVM has global optimality and better generalization than NN. But at present, SVM ismainly a theoretically algorithm focus on small sample situation.
Therefore, based on research about genetic algorithm, immune algorithm, antcolony system, wavelet analysis, NN and SVM, difficulties of NN-based analog faultdiagnosis approach and SVM-based analog fault diagnosis approach are discussed inthis dissertation. Next, some new optimization algorithms of NN and new waveletmethod of analog fault diagnosis are presented. At the last, a technical solutions ofrelevant automatic test and diagnosis system (ATDS) based on DSP，NN and expertsystem is introduced.The main contents and achievements of the paper are as follows:
1. Neural-network-based fault diagnosis of analog circuits is proposed. Based onresearch and analysis of NN learning principle and application, back propagation (BP)neural network and radial basis function (RBF) neural network are introduced indetail. Described the application of the two networks in fault diagnosis, and compared the advantages and disadvantages of the two. Also, the applicable condition isdiscussed in detail in this paper.
2. A new fault features extraction methods is proposed based on research aboutwavelet theory. The outstanding advantage of wavelet is good time-frequencylocalization. Therefore, we can use wavelet as a preprocessor to do de-noising anddecomposition of test node voltages of analog circuits. Then use the components ofwavelet coefficients to extract fault features. In this paper, a new method based on theroot mean square(RMS) value of the components of wavelet coefficients is proposed.After principal component analysis (PCA) and normalization, the RMS value is usedas the inputs of neural network to identify fault classes. The detailed steps are listedthrough diagnosis examples, and correctness of the methods is also verified.
3. A new optimized neural network based fault diagnosis method of analogcircuits is proposed. Aimed to resolved the problem of hard to choose parameters forRBF neural network(RBFNN), a new method based on immune algorithm and antcolony system(ACS-IP) is presented. This method based on the research of immunealgorithm, ant colony system and genetic algorithm, which bring the concept ofimmune factor into ant colony. ACS-IP introduced ‘Antibody concentration’ toimprove global optimization and convergence of RBFNN. This method cansignificantly shorten the running time, at the same time, it can enlarge optimizationsearching space. The specific steps and application results of the new method andtraditional method are discussed in a comparing way in detail in this paper. Andsuperiority of ACS-IP is proved by the comparison. It has less computational work,faster convergence speed and higher diagnostic accuracy.
4. A new optimization algorithm is proposed to select Support Vector Machine(SVM) parameters for analog circuit fault diagnosis. The theory and algorithm ofSVM, especially the application in fault diagnosis, is researched in this paper. Withsolid theoretical foundation, SVM has better performance than NN. But SVM hasdefects on dealing with mass data and multi-classification problems. Particularly,there is no developed method to choose parameter of SVM. An improved ant colonysystem is proposed to solve this problem. To get better performance with betterparameters, SVM can be designed as classifiers for fault diagnosis with less trainingand shorter running time. The effectiveness and superiority the proposed method isfurther verified by examples.
5. Described the technical solution of Automatic Test and Diagnosis System (ATS). With the theoretical guide of neural network method to do analog circuit faultdiagnosis, the experimental device of DSP-based ATDS is built. This paper introducedthe basic design principles, hardware structure and software realization. The designrules and solution of communication between DSP and PC, DSP-based mainboard,excitement module, analog&digital testing module and printed circuit board (PCB)testing module are all presented in this paper.
引文
[1]杨士元.模拟系统的故障诊断与可靠性设计.北京：清华大学出版社，1993
[2]王淑娟，陈博，赵国良.基于小波包变换预处理的模拟电路故障诊断方法.电工技术学报，2003，18(4)：118-122
[3]彭敏放.容差模拟电路故障诊断屏蔽理论与信息融合方法研究:[湖南大学博士论文].湖南：湖南大学，2006
[4]谭阳红.基于小波和神经网络的大规模模拟电路故障诊断研究：[湖南大学博士论文].湖南：湖南大学，2005
[5] Farzan Aminian, Mehran Aminian, H.W.Collins. Analog fault diagnosis ofactual circuits using neural networks. IEEE Transactions on Instrumentationand Measurement,2002,51(3):544-550
[6] Robert Spina, Shambhu Upadhyaya. Linear circuit fault diagnosis usingneuromorphic analyzers. IEEE Transactions on Circuits and Systems-II:Analog and Digital Signal Processing,1997,44(3):188-196
[7]彭敏放，何怡刚，王耀南.模拟电路的融合智能故障诊断.中国电机工程学报，2006，26(3):19-24
[8] X.Rong Li, Yunmin Zhu, Jiu Wang. Optimal linear estimation fusions—part I:Unified fusion rules. IEEE Transactions on Information Theory,2003,49(9):2192-2208
[9] Z.F. Huang, C.S. Lin, R.W. Liu. Topological Condition on Multiple FaultTestability of Analog Circuits. In Prceedings of IEEE Int. Symp. CAS.1982,1152-1155.
[10]尚勇，闫春江，严璋.基于信息融合的大型油浸电力变压器故障诊断.中国电机工程学报，2002，22(7):114-118
[11] Ren C. Luo, Chih-Chen Yih, Kuo Lan Su. Multisensor fusion and integration:approaches, applications, and future research directions. IEEE Sensors Journal,2002,2(9):107-119
[12]王淑娟,陈博,赵国良.基于小波包变换预处理的模拟电路故障诊断方法.电工技术学报,2003,18(4):118-122
[13]孙义闯，何怡刚．模拟电路可测拓扑条件和可测性分析与设计．湖南大学学报(自然科学版)，2002，29(1)：85-92
[14]谭阳红,何怡刚,陈洪云等.大规模电路故障诊断神经网络方法.电路与系统学报,2001,6(4):25-28
[15]彭羽中,樊锐,刘强.基于人工神经网络的电路故障诊断专家系统.系统工程与电子技术,2002,24(10):116-119
[16]李杏梅,张友纯.大规模电路故障诊断BP神经网络新方法.自动化与仪器仪表,2002,104(6):56-62
[17] Wenji Zhu, Yigang He. A neural-network-based fault diagnosis approach foranalog circuits by using wavelet transformation and fractal dimension as apreprocessor. International journal of electrical and computer engineering,2010,5(6):350-358
[18]李春明,王勇.基于小波神经网络的模拟电路故障诊断.微计算机信息,2007,23(1):204-205
[19] Salama A E，Starzyk J A．A Unified decomposition approach for
[20] fault location in large analog networks．IEEE trans．CAS，1984，31(7)：609-621
[21] R.S. Berkowitz. Condition for Network-element-value Solvablity. IRE Trans.on Circuit Theory,1962,15(9):25-29
[22] N.Navid, A.Willson. A theory and an Algorithm for Analog Circuit FaultDiagnosis. In: IEEE Trans. on CAS,1979,26(7):440-457.
[23] T.Ozawa, M.Yamada. Conditions for Determining Parameter Values in aLinear Active Network from Node Voltage Measurements. In: Proceeding ofIEEE Int. Symp. CAS,1982,274-277.
[24] Z.F.Huang, C.S.Lin and R.W.Liu. Node-fault Diagnosis and a Design ofTestability. IEEE Trans. On CAS,1983,30(5):257-265.
[25] Z.F.Huang, C.S.Lin, R.W.Liu. Topological Condition on Multiple FaultTestability of Analog Circuits. In: Proceedings of IEEE Int. Symp. CAS.1982:1152-1155.
[26] He Yigang, Tan Yanghong, Sun Yichuang. Wavelet neural network approachfor fault diagnosis of analogue circuits. IEE proceedings-circuits devices andsystems,2004,151(4):379-384
[27] J． W． Bandler． Recent Advances in Fault Location of AnalogNetworks．In：Proc．IEEE Int．Symp．CAS．，Canada：1984，660-663
[28] Bandler J W, Salama A E．Fault diagnosis of analog circuits．Proccedings ofthe IEEE，1985，73(8)：1279-1325．
[29]刘丹.模拟电路故障诊断中故障字典应用的研究：[华中科技大学硕士论文].武汉：华中科技大学，2006
[30]王宏.模拟电路故障诊断字典法研究：[西安电子科技大学硕士论文].西安：西安电子科技大学，2006
[31]王玲.大规模模拟电路故障诊断理论与方法研究：[湖南大学硕士论文].湖南：湖南大学，2006
[32]李传亮.基于SVDD和参数辨识的模拟电路故障诊断方法研究：[航空航天大学硕士论文].江苏：南京航空航天大学，2011
[33]袁海英.基于时频分析和神经网络的模拟电路故障诊断及可测性研究[电子科技大学博士论文].四川：电子科技大学，2006
[34]宋小安.模拟电路故障诊断的专家系统法与BP神经网络法研究：[河海大学硕士论文].江苏：河海大学，2005
[35] Maidon Y, Jervis B.W. Using Artificial Neural Networks or LagrangeInterpolation to Charaetcrize the Faults in an Analog Circuit: An ExperimentalStudy. IEEE Transactions on Instrumentation and Measurement,1999,48(5):932-938.
[36] Moharnmadi K, Monfared A.R.M, Nejad A.M. Fault diagnosis of analogcircuits with tolerances by using RBF and BP neural networks. StudentConference on Research and Development,2002,3:317-321.
[37]尉乃红,杨士元,童诗白.基于BP网络的线性电路故障诊断.计算机学报,1997,20(4):360-366
[38] Fanni A, Guiua A. A neural network diagnosis approach for analog circuits.Applied Intelligence, Kluwer Academic Publishers,1999,2:169-186.
[39] Catelani M, Fort A, et a1. Application of radial basis function network to thepreventive maintenance of electronic analog circuits．In: Proceedings of the16th1EEE Instrumentation and Measurement Technology Conference,1999,1:510-513.
[40] E1-Gamal MA, Abu M F. A combined clustering and neural network approachfor analog multiple hard fault classification.1999Kluwer AcademicPublishers．Journal of Electronic Testing: Theory and Applications，1999,14(3):207-217.
[41] Catelani M, Fort A, Nosi G. Application of radial basis function network to thepreventive maintenance of electronic analog circuits. In: Proc of the16th IEEEInstrumentation and Measurement Technology Conference.1999:510-513
[42]王勇.基于小波神经网络的模拟电路故障诊断的研究：[内蒙古工业大学硕士论文].内蒙古：内蒙古工业大学，2006
[43]王鹏宇,黄智刚.模糊理论与神经网络结合对模拟电路进行分层故障诊断.电子测量技术,2002,1:7-9
[44]陈晓娟.模拟电路神经网络故障诊断方法的研究：[吉林大学博士论文].吉林：吉林大学，2006
[45] L. Lu, J. Wei. Research of circuit fault diagnosis based on fuzzy neuralnetwork. Computer Measurement&Control,2002,10(12):789-795.
[46] Liang G C，He Y G．A fault identification approach for analog circuits usingfuzzy neural network mixed with genetic algorithms． In: IEEEProc．Intelligent Systems and Signal Processing，Changsha China，2003，2：1267-1272．
[47]陈治明，罗飞，曹建忠.基于小波分析的多RBF神经网络轧制力设定模型华南理工大学学报(自然科学版)，2010，38(2):142-148
[48]刘泓，莫玉龙．小波神经网络逼近非线性函数算法．应用科学学报，1998，16(2)：163-169
[49] Kalayci T， Ozdamar O， Erdol N． The use of wavelet transform aspreprocessor for the neural network detection of EEG spikes．In:Proc IEEESoutheast Conference，1994:1-3
[50] Q.Zhang, A.Benveniste. Wavelet networks. IEEE Trans.on Neural Network,1992,3:889–898.
[51]黎建平．基于小波分析与数据挖掘的旋转机械故障诊断研究：[江西理工大学硕士论文]．江西：江西理工大学，2010
[52]虞国全．基于支持向量机的智能故障诊断技术研究：[南昌大学硕士论文]．南昌：南昌大学，2007
[53] S Sathiya Keerthi, Efficient Tuning of SVM Hyperparameters UsingRadius/Margin Bound and Iterative Algorithms. IEEE Transactions on NeuralNetworks,2002,13(5):1225-1229
[54] Weston J, Watkins C. Multi-class Support Vector Machines. Technical ReportCSD-TR-98-04, Royal Holloway, University of London, Department ofComputer Science,1998
[55] Bottou L, Cortes C. Comparison of Classifier Methods: A Case Study inHandwriting Digit Recognition. In:International Conference on PatternRecognition, IEEE Computer Society Press,1994:77-87
[56] Slamani M, Kaminska B. Analog circuit fault diagnosis based on sensitivitycomputation and functional testing. IEEE Trans. on Design＆Test ofComputers,1992,9(1):30-39.
[57] Centus, Fanni A., Marchesi M. Wavelet analysis for diagnostic problems.8thMediterranean Electro technical Conference,1996,3:1571-1574.
[58] Robert Spain，Shambhu Upadhyaya．Linear Circuit Fault Diagnosis UsingNeuromorphic Analyzers， IEEE Transaction on Circuits and SystemsII，Analog and Digital Signal Processing，1997，44(3)：188-196．
[59] He Y, Sun Y. Neural network-based L1-norm optimization approach for faultdiagnosm of nonlinear circuits with tolerance. IEE Proceedings on Circuits,Devices and Systems,2001,148(4):223-228.
[60] Bernieri A, D Apuzzo M, Sansone L, et al. A neural network approach foridentification and fault diagnosis on dynamic systems. IEEE transactions oninstrumentation and measurement,1994,43(6):867-873
[61] Catelani M, Fort A, et a1. Application of radial basis function network to thepreventive maintenance of electronic analog circuits．Proceedings of the16th1EEE Instrumentation and Measurement Technology Conference,1999,1:510-513.
[62]孙义闯．模拟电路故障诊断理论与方法综述．大连海运学院学报，1989，15(4)：68-75
[63]李巍华，张盛刚.基于改进证据理论及多神经网络融合的故障分类.机械工程学报，2010，46(9):93-99
[64] J. Meador, A. Wu, C. T. Tseng. Fast diagnosis of integrated circuits faultsusing feed forward neural networks. in Proc. IEEE INNS Int. Joint Conf.Neural Networks, Seattle, WA, vol.1, July1991, pp:269–272.
[65] B. Kagle, J. Murphy. Neural network diagnosis of multiple-fault conditions inelectronic circuit boards. in Proc. First Workshop Neural NetworksAcademic/Industrial/NASA/Defense, Auburn Univ., June1990.
[66]庞伟区.数模混合信号芯片的测试与可测性设计研究：[湖南大学硕士论文].湖南：湖南大学,2004
[67]何怡刚．非线性容差模拟电阻电路故障诊断神经网络方法．固体电子学研究发展，2002，22(3)：297-299
[68]虞和济，陈长征，张省.基于神经网络的智能诊断.北京：冶金工业出版社，2002
[69] Y. Tan, Y. He, C. Cui. A novel method for analog fault diagnosis based onneural networks and genetic algorithms. IEEE Trans. Instrum. Meas.,2008,57(11):2631-2639.
[70] Vapnik V N．Statistical Learning Theory．许建华，张学工译．统计学习理论．北京：电子工业出版社．2004．
[71] Cortes C，Vapnik V. Support-Vector Networks．Machine Learning，1995，20(3)：273-297
[72] Cristianini N，Shawe-Taylor J．An Introduction to Support Vector Machinesand Other Kernel-based Learning Mechods．李国正，王猛，曾华军译．支持向量机导论．北京：电子工业出版社，2000．
[73] Vapnik V, Golowich S， Smola A． Support Vector Method for FunctionApproximation，Regression Estimation,and Signal Processing．Advances inNeural Information Processing Systems，1997，(9)：281-287．
[74] Scholkopf B， Smola A J． New Support Vector Algorithms． neuralComputation，2000，12(5)：1207-1245．
[75] Suykens J., Vandewalle J． Least Squares Support Vector MachineClassifiers．Neural Processing Letters，1999，9(3)：293-300．
[76] Cherkassky V, Ma Y. Practical selection of SVM parameters and noiseestimation for SVM regression．Neural networks，2004, l7(1)：113—126．
[77] Joachime T. Estimating the Generalization Performance of a SVM Efficiently.Informatik LS VⅢ, University Dortmund,2001.
[78] Lin H T, Lin C J, A study on sigmoid Kernels for SVM and the Training ofNon-PSD Kernels by SMO-step Methods. Technical Report,2003
[79] Keerthi S S, Lin C J. Tasymptotic Behaviors of Support Vector Machines withGaussian Kernel. Neural Computation,2003,15:1667-1689
[80] Bottou L, Cortes C, Dcnker et al. Comparison of Classifier Methods: A CaseStudy in Handwriting Digit Recognition. In: International Conference onPattern Recognition, IEEE Computer Society Press,1994:77~87
[81] Bredeusteiner E J， Bennett K P. Multicategory Classification by SupportVector Machines． Computational Optimization and Applications，1999，12(1)：53-79．
[82] Ceperic V, Baric A．Modeling of analog circuits by using support vectorregression machines． Proceedings of the2004llth IEE InternationalConference011Electronics，Circuits and Systems，2004：391-394．
[83] Ding M， Vemur R I． An active learning scheme using support vectormachines for analog circuit feasibility classification．18th InternationalConference on VLSI Design，2005：528-534．
[84] Vapnik V N. An overview of statistical learning theory. IEEE Trans on NeuralNetworks,1999,10(5):988-999
[85] Wang A, Qiu Z, Liu B．Fault Diagnosis of Circuits with Tolerance Based onSupport Vector Machines． Proceedings of Intemafional Conference onCommunications，Circuits and Systems,2006：2235-2238．
[86]吴洪兴，彭宇，彭喜元．基于支持向量机多分类方法的模拟电路故障诊断研究．电子测量与仪器学报，2007，2l(4)：27-31
[87] Krebel U, Pairwise.Classification and support vector machines. Cambridge,USA: The MIT Press,1999:255-268
[88]王承．基于神经网络的模拟电路故障诊断方法研究：[电子科技大学博士学位论文]．成都：电子科技大学，2005．
[89]田雨波.混合神经网络技术.北京:科学出版社,2009
[90]蒋宗礼.人工神经网络导论.北京:高等教育出版社,2003
[91] Rutkowski J. A two stage neural network DC fault dictionary． IEEEtrans．CAS，l995：299-302
[92] Moody J, Darken C. Fast learning in networks of locally-tuned processingunits. Neural Computation,1989,(1):281-294
[93] Mulgrew Bernard. Applying radial basis functions. IEEE Signal Processing Magazine,1996,13(2):50-65.
[94] Sing J.K, Basu D K, Nasipuri M. Self-Adaptive RBF Neural Network-BasedSegmentation of Medical Images of the Brain. Proceedings of ICISIP,2005,18(7):447-452
[95] Chiu S L．Fuzzy model identification based on cluster estimation．Journal ofIntelligent and Fuzzy System，1994,2（3）
[96] Benoudjit N, Verleysen M. On the kernel widths in radial basis functionnetworks. Neural Processing Letters,2003,18(2):139–154
[97] Barry Gomm J, Ding Li Yu.Selecting radial basis function networks centerswith recursive orthogonal least squares training. IEEE Transactions on NeuralNetworks,2000,11(2):306-314.
[98] Guo J J, Luh P B. Selecting Input Factors for Clusters of Gaussian Radial BasisFunction Networks to Improve Market Clearing Price Prediction. IEEETransactions on Power Systems,2003,18(2):665-672.
[99] Lee S，Y Szu H．Fractional Fourier Transform，wavelet transforms，andadaptive neural networks．Optical Engineering，l994，33(7)：2326-2330
[100]崔锦泰.小波分析导论．西安：西安交通大学出版社,1995
[101]胡昌华,张军波,夏华等.基于MATLAB的系统分析与设计—小波分析.第1版.西安：西安电子科技大学出版社,1999
[102] Daubechies I． The Wavelet Transform， Time-Frequency Localization andSignal Analysis．IEEE Trans．On Information Theory，l990，36(5)：961-1005
[103] Mallat S． A theory of multiresolution signal decomposition： the waveletrepresentation．IEEE Trans Pattern Anal, Machine Intel,1989,11：674-693
[104] Mallat S． Multifrequency channel decomposition of images and waveletmodels．IEEE Trans, On ASSP，1989，37(12)：2091-2110
[105]李建平．小波分析方法的应用．重庆：重庆大学出版社，1999
[106]何炜.基于Daubechies小波的水下图像去噪方法研究:[哈尔滨工程大学硕士论文]．黑龙江：哈尔滨工程大学，2006
[107]焦李成,杨淑媛.自适应多尺度网络理论与应用.北京:科学出版社,2008
[108] Y.He, Y.Tan, Y.Sun.Wavelet neural network approach for fault diagnosis ofanalog circuits. IEE Proc.-Circuits Devices and Systems,2004,Vol.151(4):379-384
[109] M.Aminian, F.Aminian. Neural-network based analog-circuit fault diagnosisusing wavelet transform as preprocessor. IEEE Trans.Circuits Syst-II,2000,47(2):151-156
[110]杨福生.小波变换的工程分析与应用.北京：科学出版社，2006
[111]王小平，曹立明.遗传算法理论、应用与软件实现．西安：西安交通大学出版社，2002
[112]周明，孙树栋.遗传算法原理及其应用，国防工业出版社，l999
[113] Von John H. Holland. Adaptation in Natural and Artificial Systems: AnIntroductory Analysis with Applications to Biology, Control and ArtificialIntelligence (Complex Adaptive Systems),1975
[114]李敏强，寇纪淞，李丹等．遗传算法的基本理论与应用.北京：科学出版社，2002
[115]周明，孙树栋.遗传算法原理及其应用，国防工业出版社，l999
[116]莫宏伟.人工免疫系统原理与应用.哈尔滨：哈尔滨工业大学出版社，2003
[117] Wang Lei，Pan Jin，Jiao Licheng．The immune algorithm．ACTA ElectronicSinica，2000，28(7)：74-78．
[118] Dorigo M, Vittorio Maniezzo, Alberto Colorni. The Ant System: optimizationby a colony of cooperating agents. IEEE Transactions on systems, Man andCybernetics-Part B,1996,26(1)：1-13.
[119] Dorigo M, Gambardella L. M. Ant Colorues for the Traveling SalesmanProblem. BioSystems,1997,43(2):73-81.
[120] Dorigo M, Gianni Di Caro,Thomas Stutzle. Ant algorithms. Future GenerationCompuer System,2000,16:5-7.
[121] Gutjahr J. W. A Graph Based Ant System and it s Convergence. FutureGeneration Computer Systems,2000,16(9):873-888.
[122] Wu Q H, Zhang J H, Xu X H. An ant colony algorithm with mutation features.Journal of Computer Research and Development,1999,36(10):1240-1245.
[123]王颖,谢剑英.一种自适应蚁群算法及其仿真研究.系统仿真学报,2002,14(1):31-33.
[124]张素兵,吕国英,刘泽民,等.基于蚂蚁算法的QoS路由调度方法.电路与系统学报,2000,3(1):1-5.
[125] Chang C S, Tian L, Wen F.A new approach to fault section estimation in powersystem using ant system. Electric Power System Researc,1999,49:63-70.
[126] Von John H. Holland. Adaptation in Natural and Artificial Systems: AnIntroductory Analysis with Applications to Biology, Control and ArtificialIntelligence (Complex Adaptive Systems),1975
[127] Maniezzo V. Genetic evolution of the topology and weight distribution ofneural networks. IEEE Trans on Neural Networks,1994,5(1):39-53.
[128] Maniezzo V, Caebonaro A. An ants Heuristic for the Frequency AssignmentProblem. Future Generatoin Computer Systems,2000，16(8):927-935.
[129] Yin Guiliang，Xiao Liping,Wu Changqi．Squirrel-cage motors fault diagnosisusing immulogy principles．Proceedings of the CSEE，2003，23(3)：132-236
[130]宫新保，周希朗.基于免疫聚类和遗传算法的RBF网络设计方法.应用科学报,2004,22(1)
[131] Steinwart I．On the influence of the kernel on the generalization ability ofsupport vector machines. http://www.minet.uni-jenade/Math-Net/reports/sources/200I/01-01report.ps.
[132]李行善，左毅，孙杰．自动测试系统集成技术．北京：电子工业出版社，2004
[133]谭阳红，蒋文科，何怡刚.基于OrCAD10.5的电子电路分析与设计.北京：国防工业出版社，2005
[134]时万春．现代集成电路测试技术．第一版．北京：化学工业出版社，2006
[2]王淑娟，陈博，赵国良.基于小波包变换预处理的模拟电路故障诊断方法.电工技术学报，2003，18(4)：118-122
[3]彭敏放.容差模拟电路故障诊断屏蔽理论与信息融合方法研究:[湖南大学博士论文].湖南：湖南大学，2006
[4]谭阳红.基于小波和神经网络的大规模模拟电路故障诊断研究：[湖南大学博士论文].湖南：湖南大学，2005
[5] Farzan Aminian, Mehran Aminian, H.W.Collins. Analog fault diagnosis ofactual circuits using neural networks. IEEE Transactions on Instrumentationand Measurement,2002,51(3):544-550
[6] Robert Spina, Shambhu Upadhyaya. Linear circuit fault diagnosis usingneuromorphic analyzers. IEEE Transactions on Circuits and Systems-II:Analog and Digital Signal Processing,1997,44(3):188-196
[7]彭敏放，何怡刚，王耀南.模拟电路的融合智能故障诊断.中国电机工程学报，2006，26(3):19-24
[8] X.Rong Li, Yunmin Zhu, Jiu Wang. Optimal linear estimation fusions—part I:Unified fusion rules. IEEE Transactions on Information Theory,2003,49(9):2192-2208
[9] Z.F. Huang, C.S. Lin, R.W. Liu. Topological Condition on Multiple FaultTestability of Analog Circuits. In Prceedings of IEEE Int. Symp. CAS.1982,1152-1155.
[10]尚勇，闫春江，严璋.基于信息融合的大型油浸电力变压器故障诊断.中国电机工程学报，2002，22(7):114-118
[11] Ren C. Luo, Chih-Chen Yih, Kuo Lan Su. Multisensor fusion and integration:approaches, applications, and future research directions. IEEE Sensors Journal,2002,2(9):107-119
[12]王淑娟,陈博,赵国良.基于小波包变换预处理的模拟电路故障诊断方法.电工技术学报,2003,18(4):118-122
[13]孙义闯，何怡刚．模拟电路可测拓扑条件和可测性分析与设计．湖南大学学报(自然科学版)，2002，29(1)：85-92
[14]谭阳红,何怡刚,陈洪云等.大规模电路故障诊断神经网络方法.电路与系统学报,2001,6(4):25-28
[15]彭羽中,樊锐,刘强.基于人工神经网络的电路故障诊断专家系统.系统工程与电子技术,2002,24(10):116-119
[16]李杏梅,张友纯.大规模电路故障诊断BP神经网络新方法.自动化与仪器仪表,2002,104(6):56-62
[17] Wenji Zhu, Yigang He. A neural-network-based fault diagnosis approach foranalog circuits by using wavelet transformation and fractal dimension as apreprocessor. International journal of electrical and computer engineering,2010,5(6):350-358
[18]李春明,王勇.基于小波神经网络的模拟电路故障诊断.微计算机信息,2007,23(1):204-205
[19] Salama A E，Starzyk J A．A Unified decomposition approach for
[20] fault location in large analog networks．IEEE trans．CAS，1984，31(7)：609-621
[21] R.S. Berkowitz. Condition for Network-element-value Solvablity. IRE Trans.on Circuit Theory,1962,15(9):25-29
[22] N.Navid, A.Willson. A theory and an Algorithm for Analog Circuit FaultDiagnosis. In: IEEE Trans. on CAS,1979,26(7):440-457.
[23] T.Ozawa, M.Yamada. Conditions for Determining Parameter Values in aLinear Active Network from Node Voltage Measurements. In: Proceeding ofIEEE Int. Symp. CAS,1982,274-277.
[24] Z.F.Huang, C.S.Lin and R.W.Liu. Node-fault Diagnosis and a Design ofTestability. IEEE Trans. On CAS,1983,30(5):257-265.
[25] Z.F.Huang, C.S.Lin, R.W.Liu. Topological Condition on Multiple FaultTestability of Analog Circuits. In: Proceedings of IEEE Int. Symp. CAS.1982:1152-1155.
[26] He Yigang, Tan Yanghong, Sun Yichuang. Wavelet neural network approachfor fault diagnosis of analogue circuits. IEE proceedings-circuits devices andsystems,2004,151(4):379-384
[27] J． W． Bandler． Recent Advances in Fault Location of AnalogNetworks．In：Proc．IEEE Int．Symp．CAS．，Canada：1984，660-663
[28] Bandler J W, Salama A E．Fault diagnosis of analog circuits．Proccedings ofthe IEEE，1985，73(8)：1279-1325．
[29]刘丹.模拟电路故障诊断中故障字典应用的研究：[华中科技大学硕士论文].武汉：华中科技大学，2006
[30]王宏.模拟电路故障诊断字典法研究：[西安电子科技大学硕士论文].西安：西安电子科技大学，2006
[31]王玲.大规模模拟电路故障诊断理论与方法研究：[湖南大学硕士论文].湖南：湖南大学，2006
[32]李传亮.基于SVDD和参数辨识的模拟电路故障诊断方法研究：[航空航天大学硕士论文].江苏：南京航空航天大学，2011
[33]袁海英.基于时频分析和神经网络的模拟电路故障诊断及可测性研究[电子科技大学博士论文].四川：电子科技大学，2006
[34]宋小安.模拟电路故障诊断的专家系统法与BP神经网络法研究：[河海大学硕士论文].江苏：河海大学，2005
[35] Maidon Y, Jervis B.W. Using Artificial Neural Networks or LagrangeInterpolation to Charaetcrize the Faults in an Analog Circuit: An ExperimentalStudy. IEEE Transactions on Instrumentation and Measurement,1999,48(5):932-938.
[36] Moharnmadi K, Monfared A.R.M, Nejad A.M. Fault diagnosis of analogcircuits with tolerances by using RBF and BP neural networks. StudentConference on Research and Development,2002,3:317-321.
[37]尉乃红,杨士元,童诗白.基于BP网络的线性电路故障诊断.计算机学报,1997,20(4):360-366
[38] Fanni A, Guiua A. A neural network diagnosis approach for analog circuits.Applied Intelligence, Kluwer Academic Publishers,1999,2:169-186.
[39] Catelani M, Fort A, et a1. Application of radial basis function network to thepreventive maintenance of electronic analog circuits．In: Proceedings of the16th1EEE Instrumentation and Measurement Technology Conference,1999,1:510-513.
[40] E1-Gamal MA, Abu M F. A combined clustering and neural network approachfor analog multiple hard fault classification.1999Kluwer AcademicPublishers．Journal of Electronic Testing: Theory and Applications，1999,14(3):207-217.
[41] Catelani M, Fort A, Nosi G. Application of radial basis function network to thepreventive maintenance of electronic analog circuits. In: Proc of the16th IEEEInstrumentation and Measurement Technology Conference.1999:510-513
[42]王勇.基于小波神经网络的模拟电路故障诊断的研究：[内蒙古工业大学硕士论文].内蒙古：内蒙古工业大学，2006
[43]王鹏宇,黄智刚.模糊理论与神经网络结合对模拟电路进行分层故障诊断.电子测量技术,2002,1:7-9
[44]陈晓娟.模拟电路神经网络故障诊断方法的研究：[吉林大学博士论文].吉林：吉林大学，2006
[45] L. Lu, J. Wei. Research of circuit fault diagnosis based on fuzzy neuralnetwork. Computer Measurement&Control,2002,10(12):789-795.
[46] Liang G C，He Y G．A fault identification approach for analog circuits usingfuzzy neural network mixed with genetic algorithms． In: IEEEProc．Intelligent Systems and Signal Processing，Changsha China，2003，2：1267-1272．
[47]陈治明，罗飞，曹建忠.基于小波分析的多RBF神经网络轧制力设定模型华南理工大学学报(自然科学版)，2010，38(2):142-148
[48]刘泓，莫玉龙．小波神经网络逼近非线性函数算法．应用科学学报，1998，16(2)：163-169
[49] Kalayci T， Ozdamar O， Erdol N． The use of wavelet transform aspreprocessor for the neural network detection of EEG spikes．In:Proc IEEESoutheast Conference，1994:1-3
[50] Q.Zhang, A.Benveniste. Wavelet networks. IEEE Trans.on Neural Network,1992,3:889–898.
[51]黎建平．基于小波分析与数据挖掘的旋转机械故障诊断研究：[江西理工大学硕士论文]．江西：江西理工大学，2010
[52]虞国全．基于支持向量机的智能故障诊断技术研究：[南昌大学硕士论文]．南昌：南昌大学，2007
[53] S Sathiya Keerthi, Efficient Tuning of SVM Hyperparameters UsingRadius/Margin Bound and Iterative Algorithms. IEEE Transactions on NeuralNetworks,2002,13(5):1225-1229
[54] Weston J, Watkins C. Multi-class Support Vector Machines. Technical ReportCSD-TR-98-04, Royal Holloway, University of London, Department ofComputer Science,1998
[55] Bottou L, Cortes C. Comparison of Classifier Methods: A Case Study inHandwriting Digit Recognition. In:International Conference on PatternRecognition, IEEE Computer Society Press,1994:77-87
[56] Slamani M, Kaminska B. Analog circuit fault diagnosis based on sensitivitycomputation and functional testing. IEEE Trans. on Design＆Test ofComputers,1992,9(1):30-39.
[57] Centus, Fanni A., Marchesi M. Wavelet analysis for diagnostic problems.8thMediterranean Electro technical Conference,1996,3:1571-1574.
[58] Robert Spain，Shambhu Upadhyaya．Linear Circuit Fault Diagnosis UsingNeuromorphic Analyzers， IEEE Transaction on Circuits and SystemsII，Analog and Digital Signal Processing，1997，44(3)：188-196．
[59] He Y, Sun Y. Neural network-based L1-norm optimization approach for faultdiagnosm of nonlinear circuits with tolerance. IEE Proceedings on Circuits,Devices and Systems,2001,148(4):223-228.
[60] Bernieri A, D Apuzzo M, Sansone L, et al. A neural network approach foridentification and fault diagnosis on dynamic systems. IEEE transactions oninstrumentation and measurement,1994,43(6):867-873
[61] Catelani M, Fort A, et a1. Application of radial basis function network to thepreventive maintenance of electronic analog circuits．Proceedings of the16th1EEE Instrumentation and Measurement Technology Conference,1999,1:510-513.
[62]孙义闯．模拟电路故障诊断理论与方法综述．大连海运学院学报，1989，15(4)：68-75
[63]李巍华，张盛刚.基于改进证据理论及多神经网络融合的故障分类.机械工程学报，2010，46(9):93-99
[64] J. Meador, A. Wu, C. T. Tseng. Fast diagnosis of integrated circuits faultsusing feed forward neural networks. in Proc. IEEE INNS Int. Joint Conf.Neural Networks, Seattle, WA, vol.1, July1991, pp:269–272.
[65] B. Kagle, J. Murphy. Neural network diagnosis of multiple-fault conditions inelectronic circuit boards. in Proc. First Workshop Neural NetworksAcademic/Industrial/NASA/Defense, Auburn Univ., June1990.
[66]庞伟区.数模混合信号芯片的测试与可测性设计研究：[湖南大学硕士论文].湖南：湖南大学,2004
[67]何怡刚．非线性容差模拟电阻电路故障诊断神经网络方法．固体电子学研究发展，2002，22(3)：297-299
[68]虞和济，陈长征，张省.基于神经网络的智能诊断.北京：冶金工业出版社，2002
[69] Y. Tan, Y. He, C. Cui. A novel method for analog fault diagnosis based onneural networks and genetic algorithms. IEEE Trans. Instrum. Meas.,2008,57(11):2631-2639.
[70] Vapnik V N．Statistical Learning Theory．许建华，张学工译．统计学习理论．北京：电子工业出版社．2004．
[71] Cortes C，Vapnik V. Support-Vector Networks．Machine Learning，1995，20(3)：273-297
[72] Cristianini N，Shawe-Taylor J．An Introduction to Support Vector Machinesand Other Kernel-based Learning Mechods．李国正，王猛，曾华军译．支持向量机导论．北京：电子工业出版社，2000．
[73] Vapnik V, Golowich S， Smola A． Support Vector Method for FunctionApproximation，Regression Estimation,and Signal Processing．Advances inNeural Information Processing Systems，1997，(9)：281-287．
[74] Scholkopf B， Smola A J． New Support Vector Algorithms． neuralComputation，2000，12(5)：1207-1245．
[75] Suykens J., Vandewalle J． Least Squares Support Vector MachineClassifiers．Neural Processing Letters，1999，9(3)：293-300．
[76] Cherkassky V, Ma Y. Practical selection of SVM parameters and noiseestimation for SVM regression．Neural networks，2004, l7(1)：113—126．
[77] Joachime T. Estimating the Generalization Performance of a SVM Efficiently.Informatik LS VⅢ, University Dortmund,2001.
[78] Lin H T, Lin C J, A study on sigmoid Kernels for SVM and the Training ofNon-PSD Kernels by SMO-step Methods. Technical Report,2003
[79] Keerthi S S, Lin C J. Tasymptotic Behaviors of Support Vector Machines withGaussian Kernel. Neural Computation,2003,15:1667-1689
[80] Bottou L, Cortes C, Dcnker et al. Comparison of Classifier Methods: A CaseStudy in Handwriting Digit Recognition. In: International Conference onPattern Recognition, IEEE Computer Society Press,1994:77~87
[81] Bredeusteiner E J， Bennett K P. Multicategory Classification by SupportVector Machines． Computational Optimization and Applications，1999，12(1)：53-79．
[82] Ceperic V, Baric A．Modeling of analog circuits by using support vectorregression machines． Proceedings of the2004llth IEE InternationalConference011Electronics，Circuits and Systems，2004：391-394．
[83] Ding M， Vemur R I． An active learning scheme using support vectormachines for analog circuit feasibility classification．18th InternationalConference on VLSI Design，2005：528-534．
[84] Vapnik V N. An overview of statistical learning theory. IEEE Trans on NeuralNetworks,1999,10(5):988-999
[85] Wang A, Qiu Z, Liu B．Fault Diagnosis of Circuits with Tolerance Based onSupport Vector Machines． Proceedings of Intemafional Conference onCommunications，Circuits and Systems,2006：2235-2238．
[86]吴洪兴，彭宇，彭喜元．基于支持向量机多分类方法的模拟电路故障诊断研究．电子测量与仪器学报，2007，2l(4)：27-31
[87] Krebel U, Pairwise.Classification and support vector machines. Cambridge,USA: The MIT Press,1999:255-268
[88]王承．基于神经网络的模拟电路故障诊断方法研究：[电子科技大学博士学位论文]．成都：电子科技大学，2005．
[89]田雨波.混合神经网络技术.北京:科学出版社,2009
[90]蒋宗礼.人工神经网络导论.北京:高等教育出版社,2003
[91] Rutkowski J. A two stage neural network DC fault dictionary． IEEEtrans．CAS，l995：299-302
[92] Moody J, Darken C. Fast learning in networks of locally-tuned processingunits. Neural Computation,1989,(1):281-294
[93] Mulgrew Bernard. Applying radial basis functions. IEEE Signal Processing Magazine,1996,13(2):50-65.
[94] Sing J.K, Basu D K, Nasipuri M. Self-Adaptive RBF Neural Network-BasedSegmentation of Medical Images of the Brain. Proceedings of ICISIP,2005,18(7):447-452
[95] Chiu S L．Fuzzy model identification based on cluster estimation．Journal ofIntelligent and Fuzzy System，1994,2（3）
[96] Benoudjit N, Verleysen M. On the kernel widths in radial basis functionnetworks. Neural Processing Letters,2003,18(2):139–154
[97] Barry Gomm J, Ding Li Yu.Selecting radial basis function networks centerswith recursive orthogonal least squares training. IEEE Transactions on NeuralNetworks,2000,11(2):306-314.
[98] Guo J J, Luh P B. Selecting Input Factors for Clusters of Gaussian Radial BasisFunction Networks to Improve Market Clearing Price Prediction. IEEETransactions on Power Systems,2003,18(2):665-672.
[99] Lee S，Y Szu H．Fractional Fourier Transform，wavelet transforms，andadaptive neural networks．Optical Engineering，l994，33(7)：2326-2330
[100]崔锦泰.小波分析导论．西安：西安交通大学出版社,1995
[101]胡昌华,张军波,夏华等.基于MATLAB的系统分析与设计—小波分析.第1版.西安：西安电子科技大学出版社,1999
[102] Daubechies I． The Wavelet Transform， Time-Frequency Localization andSignal Analysis．IEEE Trans．On Information Theory，l990，36(5)：961-1005
[103] Mallat S． A theory of multiresolution signal decomposition： the waveletrepresentation．IEEE Trans Pattern Anal, Machine Intel,1989,11：674-693
[104] Mallat S． Multifrequency channel decomposition of images and waveletmodels．IEEE Trans, On ASSP，1989，37(12)：2091-2110
[105]李建平．小波分析方法的应用．重庆：重庆大学出版社，1999
[106]何炜.基于Daubechies小波的水下图像去噪方法研究:[哈尔滨工程大学硕士论文]．黑龙江：哈尔滨工程大学，2006
[107]焦李成,杨淑媛.自适应多尺度网络理论与应用.北京:科学出版社,2008
[108] Y.He, Y.Tan, Y.Sun.Wavelet neural network approach for fault diagnosis ofanalog circuits. IEE Proc.-Circuits Devices and Systems,2004,Vol.151(4):379-384
[109] M.Aminian, F.Aminian. Neural-network based analog-circuit fault diagnosisusing wavelet transform as preprocessor. IEEE Trans.Circuits Syst-II,2000,47(2):151-156
[110]杨福生.小波变换的工程分析与应用.北京：科学出版社，2006
[111]王小平，曹立明.遗传算法理论、应用与软件实现．西安：西安交通大学出版社，2002
[112]周明，孙树栋.遗传算法原理及其应用，国防工业出版社，l999
[113] Von John H. Holland. Adaptation in Natural and Artificial Systems: AnIntroductory Analysis with Applications to Biology, Control and ArtificialIntelligence (Complex Adaptive Systems),1975
[114]李敏强，寇纪淞，李丹等．遗传算法的基本理论与应用.北京：科学出版社，2002
[115]周明，孙树栋.遗传算法原理及其应用，国防工业出版社，l999
[116]莫宏伟.人工免疫系统原理与应用.哈尔滨：哈尔滨工业大学出版社，2003
[117] Wang Lei，Pan Jin，Jiao Licheng．The immune algorithm．ACTA ElectronicSinica，2000，28(7)：74-78．
[118] Dorigo M, Vittorio Maniezzo, Alberto Colorni. The Ant System: optimizationby a colony of cooperating agents. IEEE Transactions on systems, Man andCybernetics-Part B,1996,26(1)：1-13.
[119] Dorigo M, Gambardella L. M. Ant Colorues for the Traveling SalesmanProblem. BioSystems,1997,43(2):73-81.
[120] Dorigo M, Gianni Di Caro,Thomas Stutzle. Ant algorithms. Future GenerationCompuer System,2000,16:5-7.
[121] Gutjahr J. W. A Graph Based Ant System and it s Convergence. FutureGeneration Computer Systems,2000,16(9):873-888.
[122] Wu Q H, Zhang J H, Xu X H. An ant colony algorithm with mutation features.Journal of Computer Research and Development,1999,36(10):1240-1245.
[123]王颖,谢剑英.一种自适应蚁群算法及其仿真研究.系统仿真学报,2002,14(1):31-33.
[124]张素兵,吕国英,刘泽民,等.基于蚂蚁算法的QoS路由调度方法.电路与系统学报,2000,3(1):1-5.
[125] Chang C S, Tian L, Wen F.A new approach to fault section estimation in powersystem using ant system. Electric Power System Researc,1999,49:63-70.
[126] Von John H. Holland. Adaptation in Natural and Artificial Systems: AnIntroductory Analysis with Applications to Biology, Control and ArtificialIntelligence (Complex Adaptive Systems),1975
[127] Maniezzo V. Genetic evolution of the topology and weight distribution ofneural networks. IEEE Trans on Neural Networks,1994,5(1):39-53.
[128] Maniezzo V, Caebonaro A. An ants Heuristic for the Frequency AssignmentProblem. Future Generatoin Computer Systems,2000，16(8):927-935.
[129] Yin Guiliang，Xiao Liping,Wu Changqi．Squirrel-cage motors fault diagnosisusing immulogy principles．Proceedings of the CSEE，2003，23(3)：132-236
[130]宫新保，周希朗.基于免疫聚类和遗传算法的RBF网络设计方法.应用科学报,2004,22(1)
[131] Steinwart I．On the influence of the kernel on the generalization ability ofsupport vector machines. http://www.minet.uni-jenade/Math-Net/reports/sources/200I/01-01report.ps.
[132]李行善，左毅，孙杰．自动测试系统集成技术．北京：电子工业出版社，2004
[133]谭阳红，蒋文科，何怡刚.基于OrCAD10.5的电子电路分析与设计.北京：国防工业出版社，2005
[134]时万春．现代集成电路测试技术．第一版．北京：化学工业出版社，2006