A Comparison of Machine Learning Methods to Identify Broken Bar Failures in Induction Motors Using Statistical Moments

Induction motors are reported as the horse power in industries. Due to its importance, researchers have been studied how to predict its faults in order to improve reliability. Condition health monitoring plays an important role in this field, since it is possible to predict failures by analyzing its operational data. This paper proposes the usage of vibration signals, combined with Higher-Order Statistics (HOS) and machine learning methods to detect broken bars in a squirrel-cage three-phase induction motor. The Multi-Layer Perceptron and Optimum-Path Forest have presented as promising approaches for faults classifications in an induction motor.

A novel electrocardiogram feature extraction approach for cardiac arrhythmia classification

The use of energy sources boosts technological development over time, but the continuous and unrelenting usage of fossil fuel energy has resulted in harmful impacts to the ecosystem. Then came an alert in the scientific community, generating discourses and research in the field of renewable energy, for example, wind, solar and biofuels. Among various fields of study in wind power, this work is concerned in the study of electric generators, specifically in the detection of faults, using magnetic axial flux acquired from a squirrel cage induction generator (SCIG). We compare the performance of classifiers Bayes, Support Vector Machines, Optimum Path Forest and K- Nearest Neighbor in three different feature extraction methods, High Order Statistics (HOS), Fourier Transform and Structural Co-occurrence Matrix (SCM). We were able to correctly identify 99.4% of the Normal conditions using a Bayesian classifier alongside with as well identify incipient short-circuits with 95.6% of hit rate. SVM had a lowers false positive, 0.6% and 2% in HOS and SCM, respectively, while identifying 98% of the Normal conditions, so is more suitable for binary classification between fault and non-fault short-circuits.

The detection of broken rotor bars in induction motors is studied. The hypothesis on which detection is based is that the apparent rotor resistance of an induction motor will increase when a rotor bar breaks. Here, the apparent rotor resistance is that in the balanced steady-state single-phase electrical model of an induction motor. To detect broken rotor bars, measurements of stator voltage, stator current, stator excitation frequency, and rotor velocity are taken over a small range of velocity. These measurements are processed by a near least square error estimator to produce estimated motor states and parameters. In particular, rotor resistance is estimated and compared with its nominal value to detect broken rotor bars. As part of this estimation process, it is necessary to compensate for the thermal variation in rotor resistance. The broken rotor bar detector is evaluated experimentally using one stator and three rotors, one with a broken bar, from identical 3 hp induction motors. The estimated rotor resistance is clearly greater for the rotor with the broken bar. >

Cutting analysis is a important and crucial task task to detect and prevent problems during the petroleum well drilling process. Several studies have been developed for drilling inspection, but none of them takes care about analysing the generated cutting at the vibrating shale shakers. Here we proposed a system to analyse the cutting's concentration at the vibrating shale shakers, which can indicate problems during the petroleum well drilling process, such that the collapse of the well borehole walls. Cutting's images are acquired and sent to the data analysis module, which has as the main goal to extract features and to classify frames according to one of three previously classes of cutting's volume. A collection of supervised classifiers were applied in order to allow comparisons about their accuracy and efficiency. We used the Optimum-Path Forest (OPF), Artificial Neural Network using Multilayer Perceptrons (ANN-MLP), Support Vector Machines (SVM) and a Bayesian Classifier (BC) for this task. The first one outperformed all the remaining classifiers. Recall that we are also the first to introduce the OPF classifier in this field of knowledge. Very good results show the robustness of the proposed system, which can be also integrated with other commonly system (Mud-Logging) in order to improve the last one's efficiency.

The reduction in the electromagnetic vibration and noise in three-phase squirrel-cage induction motors (IMs) has become very important from an environmental standpoint. Although electromagnetic vibration in IMs has been studied for several years, the relationship between the electromagnetic force and the radial distributions of the resulting electromagnetic vibration and noise has not yet been analyzed in sufficient detail. Our results clearly show that the radial distributions of the dominant electromagnetic vibration and noise components caused by an IM under load conditions match the mode shape of the dominant electromagnetic force producing these components.

A methodology for automatic incipient broken rotor bar detection in induction motors (IMs) is presented. Sparse representations of signals are applied as a diagnosis technique. The novelty of this technique is that it can analyze the frequency spectra from vibration signals even when the differences among signals are small. This representation allows decomposing or reconstructing signals through a trained dictionary that has learned the features of one specific group/class. The main feature of this paper is the use of over-complete dictionaries trained from sets of signals with faults to be detected. In this way, trained dictionaries perform the decomposition of signals using the orthogonal matching pursuit (OMP) algorithm. The decomposition is evaluated and classified by error-based criteria and a majority decision classifier, allowing the detection of early damage, ranging from 1 mm to one broken bar. The detection is performed by the decomposition of vibration signals from three axes (x, y, and z) of IMs under three load conditions (unloaded, half loaded, and three-fourths loaded) and different levels of damage (healthy or 0 mm, 1-9 mm, and one broken bar). These signals are processed by the Fourier transform and the spectrum obtained is evaluated by the OMP algorithm. Finally, the retrieved information is evaluated and the diagnosis is given. All algorithms are developed in MATLAB software and the detection accuracy is higher than 90% for damages as small as 1 mm.

In this paper, the dynamic strain measurement of a four-pole induction motor stator operating at 75% and 100% of the rated load is used to detect a broken rotor bar. Four fiber Bragg grating (FBG) strain sensors are installed inside the motor, between two stator teeth. Electromagnetic immunity and decreased size of the FBGs make them suitable for this application. When a broken bar is present, some characteristic frequencies in the stator strain spectrum appear. The values of these frequencies are proportional to the motor slip. Instrumentation allows the diagnosis of the broken bar in two different regions of the stator strain frequency spectrum: first, near the rotational frequency, and second, close to twice as much the supply frequency. Two different supply conditions are presented, feeding the motor from an electrical grid and by using a variable-frequency driver. The experimental results indicate that the broken rotor bar can be detected in all supply situations by using the two regions of the stator strain spectrum. By using the method presented in this paper, the broken rotor bar can be detected, even with a harmonic distortion in voltage. This is an advantage over the classic motor current signature analysis method.

Emotion recognition plays a pivotal role in advancing Human-Computer Interaction (HCI) by enabling systems to understand and respond to human emotional states. Among various modalities, Speech Emotion Recognition (SER) stands out as a non-invasive, cost-effective, and temporally efficient approach for detecting emotions. This study leverages the RAVDESS dataset, which provides high-quality audio recordings for emotion classification. The proposed methodology involves preprocessing audio signals to remove noise, extracting temporal and spectral features in both time and frequency domains, and implementing machine learning models for multi-class emotion classification. The study evaluates and compares the performance of several classification models, including Random Forest (RF), Multilayer Perceptron (MLP), Support Vector Machine (SVM), Convolutional Neural Networks (CNN), and Decision Trees (DT). Experimental results demonstrate promising accuracy levels, highlighting the potential of these machine learning techniques in SER. This research contributes to the integration of SER within Brain-Computer Interfaces (BCI) and other emotion-aware applications, paving the way for enhanced interactive systems. Keywords – Speech Emotion Recognition -Machine Learning - Random Forest -Multilayer Perceptron - Support Vector Machine - Convolutional Neural Networks - Decision Tree - RAVDESS Dataset

This study compares single and hybrid soft computing models for estimating daily solar radiation flux for two scenarios. Scenario I developed single soft computing models, including multilayer perceptron (MLP), support vector machines (SVM), adaptive neuro-fuzzy inference system (ANFIS), and multivariate adaptive regression splines (MARS), for estimating daily solar radiation flux at two stations from the USA and South Korea. The MLP model was used to evaluate the effect of factors controlling daily solar radiation flux. Using different combinations of controlling factors as input, the MLP and SVM models, based on evaluation measures, were found to be superior to the ANFIS and MARS models at Big Bend station, USA. In addition, the MLP, SVM, and MARS models performed better than did the ANFIS model at Incheon station, South Korea. Scenario II combined the discrete wavelet transform (DWT) and single soft computing models (e.g., MLP and SVM) for improved performance using 4-input combination. The wavelet-based MLP (WMLP) and SVM (WSVM) models were superior to other single soft computing models (MLP, SVM, ANFIS, and MARS) at two stations. Taylor diagrams, violin plots and point density plots were also utilized to examine the similarity between the observed and estimated solar radiation flux values. Results showed that scenarios I and II can be alternatives for estimating daily solar radiation flux based on different meteorological information, such as peninsular and continental conditions.

The paper deals with the investigation of the effects of three-phase voltage dips on the induction motors. The experimental set-up comprises a dip generator, a three-phase analyzer and a control system. As model, a three-phase squirrel cage induction motor of 1.5 kW / 0.4 kV is tested with balanced three-phase individual voltage dips of different magnitudes and durations, and of different rectangular and non-rectangular forms. A comparative analysis is done regarding the wave form and shape of rms three-phase currents and voltage dips curves. The decreasing of current and power of induction motor was correlated with voltage dip magnitudes. The experimental determinations of the effects of balanced three-phase voltage dips are compared with characteristics of induction motor operating in normal conditions.

This paper presents a new controller of the photovoltaic (PV) pumping system based on a multicell voltage inverter feeding three-phase squirrel cage induction motor (IM). The system consists of a centrifugal pump connected to IM is drived by multicell voltage inverter. To improve the efficiency of the IM motor and minimize the number of required solar panels, a delta of 220V connection is established (instead of 380V). Two control loops are used in the overall controller design. A DC/DC control loop based on robust loop shaping control method associated with a robust PI approach scalar control [for the IM] to generate variable frequency PWM signal. Simulation results are presented to show the effectiveness of the proposed approach.

Automatic inspection of petroleum well drilling has became paramount in the last years, mainly because of the crucial importance of saving time and operations during the drilling process in order to avoid some problems, such as the collapse of the well borehole walls. In this paper, we extended another work by proposing a fast petroleum well drilling monitoring through a modified version of the Optimum-Path Forest classifier. Given that the cutting’s volume at the vibrating shale shaker can provide several information about drilling, we used computer vision techniques to extract texture informations from cutting images acquired by a digital camera. A collection of supervised classifiers were applied in order to allow comparisons about their accuracy and effciency. We used the OptimumPath Forest (OPF), EOPF (Efficient OPF), Artificial Neural Network using Multilayer Perceptrons (ANN-MLP) Support Vector Machines (SVM), and a Bayesian Classifier (BC) to assess the robustness of our proposed schema for petroleum well drilling monitoring through cutting image analysis.

In order to avoid lower efficiency, poor power factor and lower torque density, the airgap length of a three-phase squirrel cage induction motor (SCIM)should not be designed very large. On other hand, the distortion of airgap flux density distribution can be minimized by enlarging the airgap length. This has informed machine designers to choose very carefully the airgap length while designing a three-phase SCIM. This paper deals with the influence that the airgap length has on the performance of a three-phase SCIM having a capacitive auxiliary three-phase stator winding. The three-phase auxiliary winding is only magnetically coupled to the stator main winding. A conventional 5.5 kW, 50-Hz, and 4-pole three-phase SCIM is modified to accommodate both main and auxiliary windings in the same stator slots, while maintaining the rated power to 5.5 kW. The results obtained from practical measurements evidenced that for a large airgap it is possible to obtain high efficiency and good power factor with the presence of a capacitive auxiliary stator winding. The high efficiency is mainly due the decrease in iron and stator copper losses. The iron losses decrease with increase in airgap length for the SCIM with or without a capacitive auxiliary winding, while the stator copper loss is obtained by reactively exciting the three-phase auxiliary stator winding.

Recently, induction motors (IMs) have been used as power sources for industrial machines. In general, middle and small size IMs have skewed rotor-bar slots using die cast to reduce abnormal starting torque and electromagnetic vibration and noise. Then, the current in the direction of the rotation (inter-bar current) adversely affects the starting characteristics and decreases the motor efficiency of the IM. Hence, we have studied the effects of the inter-bar current on the characteristics of IM, which has a rotor with skewed slot, a rotor with a skewed slot and short-circuited by two aluminum rings, or a post-heated rotor. In this study, we examine the effects of the inter-bar current on the IM characteristics based on experimental results obtained from several test motors.