Fault diagnosis in Benes switching networks

A method for the identification of stuck-at bits on a data bus is described. Faulty data words are compared to isolate the stuck-at bits. A similar method can be used to identify an instruction which does not execute correctly in a processor instruction set. Here, faulty programmes, or faulty programme segments, are compared to isolate the faulty instruction. For both bus and instruction fault cases, probabilistic analyses are given with explicit formulas and data for the mean number of faulty samples required for diagnosis, and also the associated standard deviation. Both of these procedures use on-line data and are suitable for the diagnosis of intermittent faults.

This paper proposes a method based on chaotic spread spectrum sequence for synchronous online diagnosis of intermittent faults for complex and cable. networks, such as the aircraft power system. The novelties of the proposed method include: 1) since chaotic spread spectrum signals are sensitive to initial values, the number of chaotic sequences are not limited and the proposed method is not limited by the number of cables under monitoring; 2) with the self-correlation characteristic and cross-correlation characteristic of chaotic sequence, the proposed method can be extended to online diagnosis of multiple cable intermittent faults; and 3) the signal source impedance amplification method is proposed for intermittent arc faults of multiple cables, which improves the reflection coefficient and increases the accuracy of online detection and location of intermittent arc faults. A series of simulations and experiments on multiple cables with different intermittent fault modes are conducted to verify the effectiveness of the proposed approaches.

For avoiding false-alarm and diagnosing intermittent fault, based on the relationship analysis between intermittent fault and false-alarm, the intermittent fault diagnosis model of power system based on HMM (Hidden Markov Model)-SVM (Support Vector Machine) characteristics was constructed in this paper. By means of this model, the intermittent fault diagnosis procedure of power system based on HMM-SVM characteristics was built. Finally, the intermittent fault diagnosis simulation system of power system based on HMM-SVM characteristics was demonstrated and validated by QUEST software. The result showed that the proposed method was available, and can provided guidance for avoiding false-alarm and diagnosing intermittent fault, and support for stable performance in power system.

Controller Area Networks (CANs) play an important role in many safety-critical industrial systems, which places high demands on their reliability performance. However, the intermittent connection (IC) of network cables, a random and transient connectivity problem, is a common but hard troubleshooting fault that can cause network performance degradation, system-level failures, and even safety issues. Therefore, to ensure the reliability of CANs, a fault symptom association model-based IC fault diagnosis method is proposed. Firstly, the symptoms are defined by examining the error records, and the domains of the symptoms are derived to represent the causal relationship between the fault locations and the symptoms. Secondly, the fault probability for each location is calculated by minimizing the difference between the symptom probabilities calculated from the count information and those fitted by the total probability formula. Then, the fault symptom association model is designed to synthesize the causal and the probabilistic diagnostic information. Finally, a model-based maximal contribution diagnosis algorithm is developed to locate the IC faults. Experimental results of three case studies show that the proposed method can accurately and efficiently identify various IC fault location scenarios in networks.

In the field of VLSI testing, fault diagnosis has become necessary. Though there are different types of faults available, the two commonly used faults are bridging and stuck-at faults. The methods that already exist involve the manual modification of the logical circuit like the addition of gates, multiplexers etc. The proposed work is about creating a pattern generation procedure for distinguishing between stuck-at and bridging faults without altering the circuit under test. The faults are considered in pairs and test patterns are generated accordingly to distinguish between the two faults using ATPG tools. The fault inactivation method has been used to generate the test patterns. Diagnosis patterns are generated that will be capable of individually distinguishing each fault pair in the circuit. Experiments have been conducted on the ISCAS’ 85 benchmark circuits and the number of diagnosis patterns obtained for each fault pair were tabulated. Other important parameters like the number of faults detected, fault coverage, CPU run time etc. were also analysed.

With the development of big data, artificial intelligence and other technologies, data-driven aviation equipment fault diagnosis and prediction technology has gradually become a research hotspot in the aviation field. Many typical intelligent algorithm models have been applied to this field. However, limited by the airborne embedded computing environment, there are still some problems in the deployment of intelligent prediction models represented by deep neural networks on aircraft. This paper summarizes and analyzes the research and application of typical deep neural networks such as convolutional neural networks in the field of aircraft fault diagnosis and prediction. Facing the airborne embedded environment, the current difficulties in deploying the deep neural network algorithm model in the airborne environment are analyzed. The development direction of the application of fault prediction and diagnosis algorithms represented by neural networks in the future is discussed.

Diagnosis of incipient parametric faults has been widely addressed and various model-based as well as data-based methods have been reported in the literature. In contrast to incipient faults, it is unknown if, when and for how long intermittent faults will reappear which makes their diagnosis and prognosis difficult. This paper addresses intermittent faults with a magnitude that increases over time. Such a sequence of intermittent faults may reach a failure alarm threshold and may eventually lead to a component or even a system failure. The proposed approach uses a diagnostic Bond Graph (DBG) model for an online detection, isolation and estimation of intermittent faults. Pulses of increasing height and varying width not equally spaced on the time evolution of residuals of Analytical Redundancy Relations (ARRs) obtained from an offline developed DBG indicate a degradation trend of a parameter or a variable. Certain values of this trend are concurrently extrapolated in a repeated failure prognosis resulting in a sequence of Remaining Useful Life (RUL) estimates. The proposed BG-based approach to a failure prognosis in the presence of intermittent faults can be combined with a previously presented BG-approach to incipient fault prognosis and is applicable to mode switching models. For illustration, it is applied in a case study to a small electronic circuit in an offline simulation. The case study also considers intermittent faults of constant magnitude.

The autonomous underwater vehicle (AUV) frequently operates in harsh underwater environments, and timely fault diagnosis of the AUV can prevent mission failure and equipment loss. Data‐driven methods based on a single data source have been widely utilized for fault diagnosis of the AUV because they do not require the construction of complex mechanism models and have high fault diagnosis accuracy. However, these methods face challenges in accomplishing complex fault diagnosis tasks because the single data source provides very restricted fault features. To address this issue, an attention mechanism‐based multisensor data fusion neural network (MDFNN) for AUV fault diagnosis is proposed in this work. First, a feature extraction layer based on the two‐dimensional (2D) convolutional method with a 1D kernel is introduced to extract features from each sensor data separately, significantly optimizing the model architecture. Second, an efficient channel attention mechanism‐based feature fusion layer is proposed to reassign weights to the features of each sensor data, enabling the model to focus more on crucial features. Finally, the fused features are input to the fully connected layers and softmax layer to realize the fault diagnosis of multisensor data. In the end, the diagnostic performance of the proposed MDFNN is evaluated utilizing real AUV experimental data. The experiment shows that the proposed MDFNN has a very fast convergence speed and 98.37% fault diagnosis accuracy, demonstrating its excellent fault diagnosis performance. The proposed MDFNN provides a generalized and simply structured fault diagnosis framework for the AUV with multiple types of sensor data, providing significant engineering value.

Deep model integrated with data correlation analysis for multiple intermittent faults diagnosis

Fault diagnosis is useful in helping technicians detect, isolate, and identify faults, and troubleshoot. Bayesian network (BN) is a probabilistic graphical model that effectively deals with various uncertainty problems. This model is increasingly utilized in fault diagnosis. This paper presents bibliographical review on use of BNs in fault diagnosis in the last decades with focus on engineering systems. This work also presents general procedure of fault diagnosis modeling with BNs; processes include BN structure modeling, BN parameter modeling, BN inference, fault identification, validation, and verification. The paper provides series of classification schemes for BNs for fault diagnosis, BNs combined with other techniques, and domain of fault diagnosis with BN. This study finally explores current gaps and challenges and several directions for future research.

Fault diagnosis, including fault detection and isolation, is a critical task for computer networks. Among the various techniques used for online system-level diagnosis, we are interested in the approach based on temporal information processing. The delays of the computer networks are inevitable, and the fault localization process has to take into account bounded delays or the temporal constraints. Temporal information is fundamental in model-based diagnosis. There can be cycles or loops in a computer network, but the fault reasoning methods for such cases are seldom considered in the literature. This paper provides an analytic model based on the cyclic temporal constraint network (CTCN), which aims at the fault diagnosis of cyclic computer networks using temporal information. The goal of the proposed framework is twofold: given the network structures and the predetermined candidate fault causes, the CTCN model corresponding the computer network under test is formulated; based on the CTCN model, given the alarms sequences with timestamps, the fault diagnosis process is executed to determine the most likely fault cause(s) with its/their time interval(s) of occurrence(s). The reasoning method is dependent on time point and time distance information, with which the fault motivators (i.e., actors) and fault responders (i.e., victims) can be identified. The calculation process consists of three steps: 1) establishment of the objective function; 2) determination of the fault propagation paths; and 3) determination of the expected states with a given fault hypothesis. Finally, the proposed method is demonstrated via an application study, and the effectiveness of our proposed method is verified.

To solve the problem of fault prediction and diagnosis of household air conditioning, an improved Bayesian network (BN) fault diagnosis model is proposed. Firstly, the orthogonal defect classification (ODC) is used to make statistics and analysis of air conditioning fault data, and the structure of BN fault diagnosis model is built based on the analysis results. Then, genetic algorithm (GA) is used to optimize the conditional probability of network nodes and determine the network parameters. Finally, the cooling and heating failure data of household air conditioning were taken as an example to diagnose. Compared with the traditional BN model, the accuracy of fault diagnosis increases from 81.13% to 92.83%, which verifies the effectiveness of the model.

Diagnosis of intermittent scan chain failures still remains a hard problem. In this article, we demonstrate that the use of artificial neural networks (ANNs) can lead to significantly higher accuracy. The key of this method is a multistage process incorporating ANNs with gradually refined focuses. During this process, the final fault suspect is elected through multiple rounds of ANN inference, instead of just one round. At each stage, identification of a proper Affine Group, used as the “candidate set of scan cells for the next round of ANN inference,” will influence the final diagnostic accuracy. Thus, we propose a validation-based learning procedure for Affine Group derivation to further boost the final diagnostic accuracy. The experimental results on benchmark circuits have shown that this method is, on the average, 17.46% more accurate than a state-of-the-art commercial tool for intermittent stuck-at-0 faults.

The Internet and associated network technologies are an increasingly integral part of modem day working practices. With this increase in use comes an increase in dependence. For some time, commentators have noted that given the level of reliance on data networks, there is a paucity of monitoring tools and techniques to support them. As this area is addressed, more data regarding network performance becomes available. However, a need to automatically analyse and interpret this performance data now becomes imperative. The paper takes one-way latency as an example performance metric. The term 'data exception' is employed to describe delay data that is unusual or unexpected. Data exceptions can be used by network operators to assess the effect of network modifications, failures and usage, and can also help in the diagnosis of intermittent network faults. Automating the detection of data exceptions is a non-trivial process that is not well suited to a rule-based solution. The paper shows that data exceptions can be identified by the use of a two-stage approach. The Kolmogorov-Smirnov test can initially be applied to detect general changes in the delay distribution, a neural network can then be used to categorise the change. The approach is evaluated using a network simulation.

A novel fault diagnosis method for turbo-generator set based on fractal exponent theory and wavelet network is presented. When faults occur, they usually produce nonstationary vibration signals. The wavelet transform is used to localizes the characteristics of vibration signal in the time frequency domains and in a view of the inter relationship of wavelet transform between fractal theory, the whole and local fractal exponents obtained from wavelet transform coefficients as features are presented for extracting fault signals, which are inputted into radial basis function for fault pattern recognition. The improved Levenberg-Marquardt (LM) optimization technique is used to complete the network structure parameters. By means of choosing enough samples to train the fault diagnosis network and the information representing the faults is input into the trained wavelet network, and according to the output result the type of fault can be determined. The practical diagnosis for stator temperature fluctuation and rotor vibration demonstrates that the wavelet fractal network can provide an effective way to diagnosis faults for turbo-generator set in power system.