Fault Diagnosis of Analog Circuits Using Bayesian Neural Networks with Wavelet Transform as Preprocessor

We have developed an analog circuit fault diagnostic system based on Bayesian neural networks using wavelet transform, normalization and principal component analysis as preprocessors. Our proposed system uses these preprocessing techniques to extract optimal features from the output(s) of an analog circuit. These features are then used to train and test a neural network to identify faulty components using Bayesian learning of network weights. For sample circuits simulated using SPICE, our neural network can correctly classify faulty components with 96% accuracy.     

Diagnostics of Analog Circuits Based on LS-SVM Using Time-Domain Features

Most researchers use wavelet transforms to extract features from a time-domain transient response from analog circuits to train classifiers such as neural networks (NNs) and support vector machines (SVMs) for analog circuit diagnostics. In this paper, we have proposed some new feature selection methods from a time-domain transient response, and compared the diagnostic results based on a least squares SVM (LS-SVM) using different time-domain feature vectors. First, we have improved two traditional feature selection methods: (a) using the mean and standard deviation in wavelet transform features, and (b) using the mean, standard deviation, skewness, kurtosis, and entropy in statistical property features. Then, a conventional time-domain feature vector based on the impulse response properties of a control system has been proposed. The simulation experiments for a leapfrog filter and a nonlinear rectifier show that: (1) the two improved methods have better accuracy than the traditional methods; (2) the proposed conventional time-domain feature vector is effective in the diagnostics of analog circuits—over 99 % for both of the two example circuits; (3) the proposed diagnostic method can diagnose soft faults, hard faults, and multi-faults, regardless of component tolerances and nonlinearity effects.     

基于小波高频系数统计特征的电路故障诊断

对故障电路进行特征提取与分类是模拟电路诊断的两个重要环节。现有方法多对时域响应信号进行小波变换以提取故障特征,并用神经网络或支持向量机方法实现对故障进行分类。为提高模拟电路故障诊断率,提出一种新的特征选取方法:在模拟电路的时域响应中对其进行小波变换,并对变换得到的高频细节系数统计平均值、标准偏差、峭度、熵和偏斜度等统计特征,并建立以支持向量机为分类器的故障诊断系统。以两种常见电路为例,实验结果表明,提出方法对常见电路进行故障诊断,准确率得到提升,精度达到99%以上,优于传统单纯小波系数分析方法,适用于模拟电路的故障诊断。     

基于分数阶相关的模拟集成电路软故障诊断* Deng Yong(邓勇), Shi Yibing(师奕兵)and Zhang Wei(张伟)

针对模拟集成电路软故障诊断的难题,提出了基于分数阶相关的方法。首先,利用分数阶小波包将待测试电路(CUT)的Volterra级数进行分解,计算出分数阶相关函数。然后,用得到的分数阶相关函数构造出待测试电路的故障特征。通过对故障特征的比较,可以将待测试电路的各种软故障状态进行辨识并对故障实现定位。标准电路的仿真实验描述了这一方法并验证了该方法对模拟集成电路软故障诊断的有效性。     

Soft Fault Feature Extraction in Nonlinear Analog Circuit Fault Diagnosis

Aiming at the problem to diagnose soft faults in nonlinear analog circuits, a novel approach to extract fault features is proposed. The approach is based on the Wigner–Ville distribution (WVD) of the subband Volterra model. First, the subband Volterra kernels of the circuit under test are cleared. Then, the subband Volterra kernels are used to obtain the WVD functions. The fault features are extracted from the WVD functions and taken as input data into the hidden Markov model (HMM). Finally, with classification of features using HMMs, the soft fault diagnosis of the nonlinear analog circuit is achieved. The simulations and experiments show that the method proposed in this paper can extract the fault features effectively and improve the fault diagnosis.     

基于小波变换及马氏距离的电路故障定位方法

针对模拟电路中的硬故障,提出了一种基于马氏距离统计学原理的快速定位方法。利用Pspice电路仿真软件,对ITC'97基准电路集中的连续状态可变滤波器电路进行电路建模及故障注入仿真,用Pspice与Matlab软件的数据接口技术,将无故障及各故障状态的仿真数据导入Matlab中,进而在Matlab环境中进行小波变换处理及马氏距离的计算,并以马氏距离作为定位故障元件的依据。经检验,这种故障定位方法对模拟电路硬故障具有快速定位的能力,而且测点少、在线计算量小等优点。     

基于空间特征的非线性电路神经网络诊断

本文提出线性模拟电路的单、双、三故障空间特征,采用分段线性模型（PWL）将非线性电路线性化,通过遗传算法求电路的容羞范围,用神经网络对非线性嘲络进行诊断。本文的方法大火减少了模拟计算量,同时,使神经网络的训练过程加快,训练误差减少,并大大提高了诊断的正确率。     

模拟电路故障诊断的小波包预处理神经网络改进算法

提出了基于小波包预处理的神经网络模拟电路故障诊断方法的两种改进方法:最优小波包变换(OWPT)预处理和不完全小波包变换(IWPT)预处理BP神经网络算法。首先对模拟电路的响应信号用这两种方法进行预处理,然后计算预处理后信号各个频段上的归一化能量,把归一化的能量作为训练样本送给BP网络进行训练,有效减少了BP网络的输入节点和隐层节点的个数,从而减小了神经网络的规模,降低了计算的复杂度,加快了网络的训练和收敛速度。仿真实验表明此方法能够快速有效的对模拟电路的故障进行诊断和定位。     

非线性容差模拟电阻电路故障诊断神经网络方法

将线性电路故障定位 l1 范数最优化算法推广到非线性电路的故障定位 ,由于测后计算是基于神经网络计算机环境 ,所需时间较少 ,能满足现代工业实时性需要。实例和计算机模拟结果表明所提方法是可行的     

Fault dictionary based switched current circuit fault diagnosis using entropy as a preprocessor

A novel method based on a fault dictionary that uses entropy as a preprocessor to diagnose faulty behavior in switched current (SI) circuit is presented in the paper. The proposed method uses a data acquisition board to extract the original signal form the output terminals of the circuit-under-tests. These original data are fed to the preprocessors for feature extraction and finds out the entropies of the signals which are a quantitative measure of the information contained in the signals. The proposed method has the capability to detect and identify faulty transistors in SI circuit by analyzing its output signals with high accuracy. Using entropy of signals to preprocess the circuit response drastically reduces the size of fault dictionary, minimizing fault detect time and simplifying fault dictionary architecture. The result from our examples showed that entropies of the signals fall on different range when the faulty transistors^` Transconductance G_m value varying within their tolerances of 5 or 10%, thus we can identify the faulty transistors correctly when the response do not overlap. The average accuracy of fault recognition achieved is more than 95% although there are some overlapping data when tolerance is considered. The method can classify not only parametric faults but also catastrophic faults. It is applicable to analog circuits as well as SI ones. A low-pass and a band-pass SI filter and a Clock feedthrough cancellation circuit have been used as test beached to verify the effectiveness of the proposed method. A comparison of our work with Yuan et al. (IEEE Trans Instrum Meas 59(3):586–595, 2010), which used entropy and kurtosis as preprocessors, reveals that our method requiring one feature parameter reduces the computation and fault diagnosis time.     

Ensembles of Neural Networks for Fault Diagnosis in Analog Circuits

A new neural network-based analog fault diagnosis strategy is introduced. Ensemble of neural networks is constructed and trained for efficient and accurate fault classification of the circuit under test (CUT). In the testing phase, the outputs of the individual ensemble members are combined to isolate the actual CUT fault. Prominent techniques for producing the ensemble are utilized, analyzed and compared. The created ensemble exhibit high classification accuracy even if the CUT has overlapping fault classes which cannot be isolated using a unitary neural network. Each neural classifier of the ensemble focuses on a particular region in the CUT measurement space. As a result, significantly better generalization performance is achieved by the ensemble as compared to any of its individual neural nets. Moreover, the selection of the proper architecture of the neural classifiers is simplified. Experimental results demonstrate the superior performance of the developed approach.     

Proposal of Fault Diagnosis of Analog Circuits by Combining Operation-Region Model and X–Y Zoning Method: Case Study

We propose a method of diagnosing analog circuits that is achieved by combining an operation-region model and an X–Y zoning method. The X–Y zoning method can be used to detect faults in analog circuits by using the relationship between circuit inputs and outputs. The operation-region model can be used to analyze/model circuit behaviors by utilizing changes in the operation regions of MOS transistors in a circuit. Operation regions are obtained from transistor node voltages at sampling time corresponding to a particular excitation of the input value and the corresponding output value. Since we developed a data processing method to handle data discretely, we could implement a procedure for diagnosis based on the preset test, which is a method of diagnosing digital circuits. We demonstrated the effectiveness of our method by applying it to ITC’97 benchmark circuits with hard and soft faults. We found that the diagnostic resolution is one for every fault.     

Classification of Defective Analog Integrated Circuits Using Artificial Neural Networks

This paper presents a new approach for detecting defects in analog integrated circuits using the feed-forward neural network trained by the resilient error back-propagation method. A feed-forward neural network has been used for detecting catastrophic faults randomly injected in a simple analog CMOS circuit by classification the differences observed in supply current responses of good and faulty circuit. The experimental classification was performed for time and frequency domain, followed by a comparison of results achieved in both domains. It was shown that neural networks might be very efficient and versatile approach for test of analog circuits since an arbitrary fault class or circuit's parameter can be analyzed. Considered defect types and their successful detection by the neural network; and a possible off-chip hardware implementation of the proposed technique are discussed as well. Moreover, optimized hardware architecture of the selected neural network type was designed using VHDL for FPGA realization.     

模拟电路的多频灵敏度故障诊断方法

文章在灵敏度故障诊断方法的基础上提出多频灵敏度参数识别故障诊断方法，并给出选择测试频率的一般原则。该方法能够适用于可及测试节点较少的电路。针对模拟电路中一般只存在部分元件故障的情况，进一步提出只识别部分故障元件参数的多频灵敏度故障诊断方法，使该方法能适用于更大规模的电路。电路仿真结果验证了所提方法的有效性。     

A new swarm-SVM-based fault diagnosis approach for switched current circuit by using kurtosis and entropy as a preprocessor

This paper presents a new fault diagnosis method for switched current (SI) circuits. The kurtoses and entropies of the signals are calculated by extracting the original signals from the output terminals of the circuit. Support vector machine (SVM) is introduced for fault diagnosis using the entropies and kurtoses as inputs. In this technique, a particle swarm optimization is proposed to optimize the SVM to diagnose switched current circuits. The proposed method can identify faulty components in switched current circuit. A low-pass SI filter circuit has been used as test beached to verify the effectiveness of the proposed method. The accuracy of fault recognition achieved is about 97 % although there are some overlapping data when tolerance is considered. A comparison of our work with Long et al. (Analog Integr Circuit Signal Process 66:93–102, 2011), which only used entropy as a preprocessor, reveals that our method performs well in the part of fault diagnostic accuracy.     

基于ELM网络的模拟电路故障诊断

由于模拟电路的非线性、易受外界干扰等因素,模拟电路的故障在设备总故障中占很大的比例。因此,对模拟电路的故障诊断技术进行深入研究具有很重要的意义。文中针对雷达电路的故障进行快速有效的特征提取,采用神经网络中ELM网络建立诊断系统结构,并通过对具体电路的仿真,输出ELM网络的诊断结果。实际应用表明,该系统具有操作简便、诊断精度高的特点,达到了设计要求。     

Fault Diagnosis of Linear Analog Electronic Circuit Based on Natural Response Specification using K-NN Algorithm

This paper reports a novel method for parametric fault diagnosis in linear analog electronic circuits using distance weighted cosine K-Nearest Neighbours (K-NN) algorithm that performs data classification on the basis of cosine similarity between data features or attributes. In this approach the analog electronic Circuit Under Test (CUT) is represented in the form of a transfer function model and natural response specifications of the system such as damping ratio, natural frequency and static gain of the system are extracted as features from this model. For experimentation purpose a second order Sallen-Key band pass filter and a fourth order Chebychev Type 1 low pass filter is considered, the corresponding fault classes are created for each of the circuit. The parameter values of the passive components in the system have been varied to derive the features, and each component whose tolerance varied is labelled with a corresponding fault class. The proposed methodology classifies faulty classes with accuracy greater than 95%.

     

Soft Fault Diagnosis in Analog Circuits Based on Bispectral Models

Aiming at the problem to locate soft faults in analog circuits, a new approach based on bispectral models is proposed. First, the Volterra kernels of the circuit under test (CUT) are calculated. Then, the Volterra kernels are used to construct bispectral models. By comparison with the fault features of the constructed models, soft faults of linear and weak nonlinear components in the analog circuit are identified and the faults are located. Simulations and experiments show the effectiveness of the proposed method in analog circuits.,     

一种基于小波-神经网络的模拟电路故障诊断方法

小波分析具有数据压缩和特征提取的特性,神经网络具有非线性映射和学习推理的优点。结合两者的特点,提出了一种基于小波与神经网络的模拟电路故障诊断方法,该方法用小波变换对电路响应信号进行特征提取,从而简化神经网络的结构,降低计算的复杂度,加快了训练速度。对实例仿真表明,该法能有效地对模拟电路进行故障诊断。     

模拟电路故障诊断中的特征信息提取

在运用BP神经网络进行模拟电路故障诊断过程中,代表故障特征的网络输入至关重要。分析了常见特征信息提取和故障诊断方法,提出一种基于多测试点、多特征信息原始样本集的新方法。运用这种方法构造原始故障特征集,然后作为BP神经网络的输入对网络进行训练,仿真结果表明,通过该方法构造的样本集训练出来的网络对模拟电路故障诊断的正确率优于传统方法,证明了该方法在模拟电路故障诊断中的可行性,为模拟电路的故障诊断提供了一种新方法。