New Approach For Fault Diagnosis Research Articles

A new approach for fault diagnosis in TCSC compensated transmission lines using FWHT and machine learning techniques

In this research a new approach is introduced for detecting and classifying faults in a Thyristor-Controlled Series Capacitor (TCSC) compensated power system. Our proposed scheme utilizes both the Fast Walsh Hadamard Transform (FWHT) and machine learning algorithms. The FWHT is employed to extract fault features from current data obtained from the TCSC compensated transmission line, while machine learning algorithms such as K-Nearest Neighbor (K-NN) and Support Vector Machine (SVM) are used to classify the extracted features for the purpose of fault detection and identification. To evaluate the performance of our proposed scheme, simulation studies were conducted on a test system under various fault conditions. The simulation results demonstrate that our approach is highly effective in accurately and quickly detecting and classifying faults, even when noise and TCSC compensation are present. This scheme has the potential to enhance the reliability and efficiency of power transmission systems.

Sound-aided fault feature extraction method for rolling bearings based on stochastic resonance and time-domain index fusion

The vibration signals of rolling bearings contain a large amount of running information, and have been widely applied in fault diagnosis and status monitoring. However, the fault features decay significantly in the transmission paths, which reduces the detection efficiency. This paper fuses the time-domain vibration and sound signals, and proposes a fault feature extraction method based on the fusion signal. A new indicator is set to show the fusion performance, and a sliding Hanning window is put on the time axis for the optimal fusion signal. The ant colony algorithm is used for the optimization of parameters in stochastic resonance system, and the final output signal is obtained at the highest signal-to-noise ratio. Experimental results show that the fault feature extraction performance gets greatly increased through the proposed method, and the feature enhancement is closely related with the parameters of the Hanning window. The phase difference between the vibration and sound signals is eliminated in the time-domain signal fusion, making the method capable of processing vibration and sound signals collected with time difference. This study proposes a new approach for fault diagnosis, and helps in the status monitoring of bearing systems.

A new approach for fault diagnosis with full-scope simulator based on state information imaging in nuclear power plant

In this paper, a new approach aimed at the Fault Diagnosis with Full-scope Simulator based on the State Information Imaging (FDFSSII) in NPP is proposed. The FDFSSII approach first constructs a series of gray-image which presents the operating transient (included normal and fault condition) according to the real time monitoring data. Furthermore, the Machine Learning (ML) technology is employed to achieve image feature extraction and classification by analyzing and learning from massive amounts of historical and synthetic gray-image data – the image feature is extracted by the Kernel Principal Component Analysis (KPCA) and classified by the designed classifiers in different learning methods. Finally, diagnosis effect is evaluated by the F1 score. The simulation result shows that the FDFSSII approach has achieved good effect for the fault diagnosis in NPP. Meanwhile, it simplifies the process of nuclear reactor with the large monitoring data and provides useful support information to the operators.

Synchronous Diagnosis of Discrete-Event Systems

In general, systems are formed by the composition of several modules, local components, or subsystems, and may exhibit a large number of states. The growth of the composed model with the number of system components leads to high-computational costs for diagnosis techniques based on the composed plant model. We propose in this article a new approach for fault diagnosis that avoids the direct use of the composed system model for the diagnoser implementation, reducing the computational cost for diagnosis. The diagnosis strategy is based on the observation of the fault-free behavior of the system components. In this regard, we introduce the definition of synchronous diagnosability of the language of a discrete-event system (DES) with respect to the languages of its components and provide a method to verify this property. An algorithm that efficiently computes the fault-free behavior model of the system components is also proposed. We extend this approach to a decentralized architecture and introduce the definition of synchronous codiagnosability. Moreover, a comparison between the classical definition of diagnosability, synchronous diagnosability, synchronous codiagnosability, and modular diagnosability of DESs is established. The verification of synchronous diagnosability for a didactic automated system is presented to show the results of the article. Note to Practitioners —In the traditional fault diagnosis approaches, in order to design a diagnoser for a discrete-event system, it is necessary to obtain its model, including both fault-free and postfault behaviors. Since, in general, complex systems are formed by several modules, components, or subsystems, then the composed system model may have a large number of states, which leads to diagnosers, obtained by using the traditional techniques, with a huge number of states, increasing the computational cost for their implementation. In order to circumvent this problem, we propose in this article the implementation of a diagnoser that is computed from the fault-free behavior models of the system components, avoiding the direct use of the composed system model for diagnosis. We also present a decentralized architecture in order to enable its use for large systems where the sensor information is distributed. The verification of the diagnosability of the system considering the fault diagnosis scheme proposed in this article is applied to a didactic automated plant.

Fault Diagnosis in the Field of Additive Manufacturing (3D Printing) Using Bayesian Networks

In this work, a new approach for fault diagnosis in the field of additive manufacturing (3d printing) using artificial intelligence will be given. This approach is based on the marriage of the Bayesian Networks theory and data acquisition techniques. Bayesian Networks are well known for their ability to infer probabilities and to give decisional support under uncertainty. In order to do so, these probability engines must be constructed and maintained by a big amount of data and information using learning algorithms. This work provides a methodology that uses sensors based data acquisition and processing to construct such networks. Some of these sensors are already available in most of the 3d printers available in the market, while other sensors were additionally embedded in a studied 3d printer in order to enrich the number of observational variables to gain a high level of fault diagnosis accuracy and support.

A fault diagnosis and quality prediction method of ball valves based on state transition probability matrix in Markov chain

Product quality is very important because many faults have a real impact on users directly. It is significative that Markov chain theory is applied in product quality prediction. A new method of state transition probability matrix in Markov chain is used to provide fault diagnosis and quality prediction in this paper. The aim of this study is to deliver a quality prediction method for mechanical and electrical product by using the data collection and analysis techniques. Firstly, a new approach of fault diagnosis was presented by using the theory of state transition probability matrix. Then, Markov chain theory is used to build the mathematical patterns in the quality prediction with the state transition probability matrix. Finally, we illustrate the veracity rationality and science of the approved method using a numerical example of ball valves.

A Real-Time Method for the Diagnosis of Multiple Switch Faults in NPC Inverters Based on Output Currents Analysis

This paper presents a new approach for fault diagnosis in three-level neutral point clamped inverters. The proposed method is based on the average values of the positive and negative parts of normalized output currents. This method is capable of detecting and locating multiple open-circuit faults in the controlled power switches of converters in half of a fundamental period of those currents. The implementation of this diagnostic approach only requires two output currents of the inverter. Therefore, no additional sensors are needed other than the ones already used by the control system of a drive based on this type of converter. Moreover, through the normalization of currents, the diagnosis is independent of the load level of the converter. The performance and effectiveness of the proposed diagnostic technique are validated by experimental results obtained under steady-state and transient conditions.

Fault feature extraction of rolling element bearings using sparse representation

Influenced by factors such as speed fluctuation, rolling element sliding and periodical variation of load distribution and impact force on the measuring direction of sensor, the impulse response signals caused by defective rolling bearing are non-stationary, and the amplitudes of the impulse may even drop to zero when the fault is out of load zone. The non-stationary characteristic and impulse missing phenomenon reduce the effectiveness of the commonly used demodulation method on rolling element bearing fault diagnosis. Based on sparse representation theories, a new approach for fault diagnosis of rolling element bearing is proposed. The over-complete dictionary is constructed by the unit impulse response function of damped second-order system, whose natural frequencies and relative damping ratios are directly identified from the fault signal by correlation filtering method. It leads to a high similarity between atoms and defect induced impulse, and also a sharply reduction of the redundancy of the dictionary. To improve the matching accuracy and calculation speed of sparse coefficient solving, the fault signal is divided into segments and the matching pursuit algorithm is carried out by segments. After splicing together all the reconstructed signals, the fault feature is extracted successfully. The simulation and experimental results show that the proposed method is effective for the fault diagnosis of rolling element bearing in large rolling element sliding and low signal to noise ratio circumstances.

Generic scalable fault diagnosis system for multimachine power grids

A new approach for fault diagnosis in power grids is presented in this paper. The approach is capable of detecting, classifying and locating, with high speed and accuracy, any type of faults anywhere in the grid using bus voltage or line current measurements only. Fault detection and classification is accomplished by applying logic on measured data while fault location is determined by fault location identification functions generated off-line by using a least-square support vector machine. The high speed and accuracy aspects of the proposed approach are verified through case studies in three test power systems.

Fault diagnosis of induction motors utilizing local binary pattern-based texture analysis

Abstract Fault diagnosis of induction motors in the practical industrial fields is always a challenging task due to the difficulty that lies in exact identification of fault signatures at various motor operating conditions in the presence of background noise produced by other mechanical subsystems. Several signal processing approaches have been adopted so far to mitigate the effect of this background noise in the acquired sensor signal so that fault-related features can be extracted effectively. Addressing this issue, this paper proposes a new approach for fault diagnosis of induction motors utilizing two-dimensional texture analysis based on local binary patterns (LBPs). Firstly, time domain vibration signals acquired from the operating motor are converted into two-dimensional gray-scale images. Then, discriminating texture features are extracted from these images employing LBP operator. These local feature descriptors are later utilized by multi-class support vector machine to identify faults of induction motors. The efficient texture analysis capability as well as the gray-scale invariance property of the LBP operators enables the proposed system to achieve impressive diagnostic performance even in the presence of high background noise. Comparative analysis reveals that LBP8,1 is the most suitable texture analysis operator for the proposed system due to its perfect classification performance along with the lowest degree of computational complexity.

Application of Algorithm of Hidden Markov Model and High-Order Spectrum in Fault Diagnosis of Power Electronic Circuit

This paper presents a method of applying the discrete hidden markov model and high-order spectrum to fault diagnosis of power electronic circuit.With bispectrum analysis, an ARMA model parametric bispectrum estimation is presented for Fault feature extraction firstly,and then fault modes were trained and recognized by Discrete Hidden Markov Model. Finally, electric locomotive main converter of SS8 type is used as an example to illustrate the new approach of fault diagnosis. The experimental results show that the proposed method has a high correct rate.The correct rate of the proposed method is 100%in the case of no noise or 5% noise whicn is 16.11% and 23.79% higher respectively than that of DHMM and BP neural network methods. So the method has practical value.

A SVDD approach of fuzzy classification for analog circuit fault diagnosis with FWT as preprocessor

In this paper, a new approach of fault diagnosis in analog circuits, which employs the Fractional Wavelet Transform (FWT) to extract fault features and adopts a fuzzy multi-classifier based on the Support Vector Data Description (SVDD) to diagnose circuit faults, is proposed. Firstly, a discrete FWT algorithm by the fractional kernel matrix is performed to preprocess fault samples. To obtain the optimal fractional order, two methods trained with the genetic algorithm are introduced. One approach is performed by the best diagnostic result, and the other is based on the maximum among-cluster center distance by the Kernel Fuzzy C-Means (KFCM) algorithm. In this paper, a threshold value is used to decrease the fuzzy region which in the overlap between hyperspheres of SVDD. Then, a SVDD fuzzy multi-classifier is applied to diagnose faults in analog circuit, and fuzzy faults are diagnosed in fuzzy sets by the relative distance. The simulation results show that the FWT succeeds in extracting local fault features and the classifier effectively detects faults.

Simplified Interval Observer Scheme: A New Approach for Fault Diagnosis in Instruments

There are different schemes based on observers to detect and isolate faults in dynamic processes. In the case of fault diagnosis in instruments (FDI) there are different diagnosis schemes based on the number of observers: the Simplified Observer Scheme (SOS) only requires one observer, uses all the inputs and only one output, detecting faults in one detector; the Dedicated Observer Scheme (DOS), which again uses all the inputs and just one output, but this time there is a bank of observers capable of locating multiple faults in sensors, and the Generalized Observer Scheme (GOS) which involves a reduced bank of observers, where each observer uses all the inputs and m-1 outputs, and allows the localization of unique faults. This work proposes a new scheme named Simplified Interval Observer SIOS-FDI, which does not requires the measurement of any input and just with just one output allows the detection of unique faults in sensors and because it does not require any input, it simplifies in an important way the diagnosis of faults in processes in which it is difficult to measure all the inputs, as in the case of biologic reactors.

FAULT DIAGNOSIS AND LOCATION ON MULTI-LOAD DISTRIBUTION NETWORK BY A CURRENT CLARKE–CONCORDIA PATTERN APPROACH

A new approach for fault diagnosis and location on electrical power multi-load distribution network is presented. Normally, the bus voltage in the distribution networks is almost constant even in presence of a line fault. Therefore, it is possible to identify and locate different types of faults only through the analysis of current lines. Under this consideration it is possible to convert this problem into a current pattern recognition analysis. Consequently, using a Clarke-Concordia transformation it is possible to obtain a different pattern for each fault and location. To identify those patterns an eigenvalue approach is used. A polynomial function is also used to establish the nonlinear relationship existing between fault location and line current patterns. Results are presented which show the effectiveness of the proposed algorithm for a correct fault identification and location on an electrical power multi-load distribution network.

Fault Diagnosis in Mixed-Signal Low Testability System

This paper describes a new approach for fault diagnosis of analog multi-phenomenon systems with low testability. The developed algorithms include identification of ambiguity groups, fault diagnosis methodology and solving low testability equations. Our aim is to identify a minimum number of faulty parameters that satisfy the test equations called a minimum form solution. An algorithm to find a minimum form solution is presented, which is based on the solution invariant matrix and an identification of singular cofactors of this matrix. System simulation using a developed Cpp and Matlab programs was performed to test different faulty circuits. Test examples are discussed and simulation results are presented.

Signal Processing Based Quantitative Schemes for Fault Detection/Diagnosis in Dynamic Systems

Two signal processing based techniques are presented in this paper, one for fault detection, the other for fault diagnosis. In the first scheme, a recursive least squares based adaptive lattice filter is used to monitor the performance of the system and to detect the occurrence of faults. The fault detection is carried out by performing statistical analysts on adaptive filter residuals. The advantage of this scheme is that neither a priori knowledge of the model, nor the order of the system is required for fault detection. The second part of the paper introduces a new approach for fault diagnosis which combines the linear predictive coding (both forward and backward) algorithms with the root loci representation of the variation of the physical system parameters. However, the detailed model of the fault-free system is required for this part. The distinctive feature of this approach is that the fault can be diagnosed as variations of the physical system parameters, therefore, the post-fault model of the system can be obtained easily. Such information may be of importance for fault compensation, such as during the controller reconfiguration process.