Detection Of Different Faults Research Articles

Fault Diagnosis in Induction Motors through Infrared Thermal Images Using Convolutional Neural Network Feature Extraction

The development of diagnostic systems for rotating machines such as induction motors (IMs) is a task of utmost importance for the industrial sector. Reliable diagnostic systems allow for the accurate detection of different faults. Different methods based on the acquisition of thermal images (TIs) have emerged as diagnosis systems for the detection of IM faults to prevent the further generation of faults. However, these methods are based on artisanal feature selection, so obtaining high accuracy rates is usually challenging. For this reason, in this work, a new system for fault detection in IMs based on convolutional neural networks (CNNs) and thermal images (TIs) is presented. The system is based on the training of a CNN using TIs to select and extract the most salient features of each fault present in the IM. Subsequently, a classifier based on a decision tree (DT) algorithm is trained using the features learned by the CNN to infer the motor conditions. The results of this methodology show an improvement in the accuracy, precision, recall, and F1-score metrics for 11 different conditions.

An Intelligent Fault Detection and Classification Scheme for Distribution Lines Using Machine Learning

The current paper focuses on the development and deployment of Machine Learning (ML) based algorithms for the classification and detection of different faults in the electrical distribution system. The methodology adapted using ML has higher computational accuracy than traditional computational algorithms. The parameters involved in developing ML for fault detection/classification are fundamental frequency, fault voltage, and current components at fault situations. During faults, the current and voltage waveforms consist of high-frequency transient signals. The Wavelet Decomposition (WD) technique is used to break down transient signals to obtain the required information. To investigate the performance of the ML-based algorithms, an IEEE 33 bus system is utilized, and a fault is generated in Matlab/Simulink environment. The methodologies used for fault detection and classification are K Nearest Neighbor (KNN), Decision Tree (DT), and Support Vector Machine (SVM). The performance of the designed algorithm is assessed by employing a confusion matrix, and the results demonstrated extraordinarily high accuracy.

Diagnosis of induction motor faults using the motor current normalized residual harmonic analysis method

Induction motors are susceptible to many types of faults that can become catastrophic and cause production stoppages. To predict emerging faults and avoid an unexpected failure, an induction motor can be diagnosed using the proposed motor current normalized residual harmonic analysis method. This method is simple and efficient. It is based on a more sensitive fault signature and a single frequency of the harmonic for the different faults. It is based also on the analysis of residual harmonics, which can be measured between the normalized linear fast Fourier transform spectrum of the healthy motor current and that of the faulty motor current. The proposed method can mitigate the task of the fault detection system by monitoring the occurrence of a residual harmonic fixed at a precise frequency. This study focuses on the rotor fault of broken bars and inter-turn short circuits in stator windings. In-depth experimental results demonstrate the rapid detection of different faults at the characteristic frequency of 50 Hz. It is not necessary to conduct a wide spectral sweep to search each time for different faults that have variable characteristic frequencies depending on the type of fault. This method can be useful for diagnosing other types of faults such as eccentricities, bearings and magnetic circuits.

Can this fault be detected: A study on fault detection via automated test generation

Automated test generation can reduce the manual effort in improving software quality. A test generation method employs code coverage, such as the widely-used branch coverage, to guide the inference of tests. These tests can be used to detect hidden faults. An automatic tool takes a specific type of code coverage as a configurable parameter. Given an automated tool of test generation, a fault may be detected by one type of code coverage, but omitted by another. In frequently released software projects, the time budget of testing is limited. Configuring code coverage for a testing tool can effectively improve the quality of projects. In this paper, we conduct a study on whether a fault can be detected by specific code coverage in automated test generation. We build predictive models with 60 metrics of faulty source code to identify detectable faults under eight types of code coverage. In the experiment, an off-the-shelf tool, EvoSuite is used to generate test data. Experimental results based on four research questions show that different types of code coverage result in the detection of different faults; a code coverage can be used as a supplement to increase the number of detected faults if another coverage is applied first; for each coverage, the number of detected faults increases with its cutoff time in test generation. Our result shows that the choice of code coverage can be learned via multi-objective optimization from sampled faults and directly applied to new faults. This study can be viewed as a preliminary result to support the configuration of code coverage in the application of automated test generation.

Robust fault detection for wind turbines using reference model-based approach

This paper presents a reference model-based approach for detection of different faults in a wind turbine. Stochastic uncertainty has been considered in the model of wind turbine. The fault detection scheme is so designed that the generated residual is robust against the uncertainty. For residual evaluation purpose, generalized likelihood ratio (GLR) test has been performed. Threshold is computed using the table of chi-square distribution with one degree of freedom. Occurrence of a fault is concluded whenever evaluated residual crosses the threshold. Using this approach an actuator and a sensor fault in the pitch system and a sensor fault in the drive train system are successfully detected. Results are compared with Combined Observer and Kalman Filter (COK) approach (Chen et al. 2011) used for wind turbine fault detection with this approach requiring less detection time thus providing a more useful solution to the wind industry.

Single-parameter fault identification through information entropy analysis at the startup-transient current in induction motors

Opportune diagnosis of rotating machines contributes to avoid expensive reparations and unscheduled shutting downs, and prevents incomings losses. Most of the developed techniques for induction motor condition monitoring fall in one of three classifications: the detection of a single fault by analyzing one or multiple parameters; the detection of different faults by combining multiple parameters and processing techniques; and expert systems that combine several computing-intensive techniques to analyze different electrical and mechanical parameters in order to detect multiple faults. Recent works have been oriented to provide computationally effective diagnostic tools for condition monitoring, which able to discriminate different faults by analyzing a minimum set of parameters. This work presents a methodology for induction motor condition monitoring by analyzing a single parameter. The analysis combines the reconstruction of a single wavelet-packet node with information entropy to obtain one parameter, which allows the detection of different faults quantitatively by analyzing the startup-transient current signal from the induction motor. Experimental results show that the proposed methodology allows the detection of a healthy motor, a motor with one broken bar, a motor with unbalanced mechanical load, and a motor with a faulty bearing in a quantitative way; with a certainty of more than 99.7%.

Accurate diagnosis of induction machine faults using optimal time–frequency representations

This paper presents a new diagnosis method of induction motor faults based on time–frequency classification of the current waveforms. This method is composed of two sequential processes: a feature extraction and a rule decision. In the process of feature extraction, the time–frequency representation (TFR) has been designed for maximizing the separability between classes representing different faults. The diagnosis is realised in two levels; the first one allows the detection of different faults—bearing fault, stator fault and rotor fault. The second one refines this detection by the determination of severity degree of faults, which are already identified on the previous level. The diagnosis is independent of the level of load. This method is validated on a 5.5 kW induction motor test bench.