Search results

Vehicle fault diagnosis is a key factor in ensuring the safe and efficient operation of the railway system. Due to the numerous vehicle categories and different fault mechanisms, there is an unbalanced fault category problem. Most of the current methods to solve this problem have complex algorithm structures, low efficiency and require prior knowledge. This study aims to propose a new method which has a simple structure and does not require any prior knowledge to achieve a fast diagnosis of unbalanced vehicle faults.

This study proposes a novel K-means with feature learning based on the feature learning K-means-improved cluster-centers selection (FKM-ICS) method, which includes the ICS and the FKM. Specifically, this study defines cluster centers approximation to select the initialized cluster centers in the ICS. This study uses improved term frequency-inverse document frequency to measure and adjust the feature word weights in each cluster, retaining the top τ feature words with the highest weight in each cluster and perform the clustering process again in the FKM. With the FKM-ICS method, clustering performance for unbalanced vehicle fault diagnosis can be significantly enhanced.

This study finds that the FKM-ICS can achieve a fast diagnosis of vehicle faults on the vehicle fault text (VFT) data set from a railway station in the 2017 (VFT) data set. The experimental results on VFT indicate the proposed method in this paper, outperforms several state-of-the-art methods.

This is the first effort to address the vehicle fault diagnostic problem and the proposed method performs effectively and efficiently. The ICS enables the FKM-ICS method to exclude the effect of outliers, solves the disadvantages of the fault text data contained a certain amount of noisy data, which effectively enhanced the method stability. The FKM enhances the distribution of feature words that discriminate between different fault categories and reduces the number of feature words to make the FKM-ICS method faster and better cluster for unbalanced vehicle fault diagnostic.

The purpose of this paper is to develop an appropriate approach for detecting unbalanced fault in rotating machines using KNN and SVM classifiers.

To fulfil this goal, a fault diagnosis approach based on signal processing, feature extraction and fault classification, was used. Vibration signals were acquired from a designed experimental system with three conditions, namely, no load, balanced load and unbalanced load. FFT technique was applied to transform the vibration signals from time-domain into frequency-domain. In total, 29 feature parameters were extracted from FFT amplitude of the signals. SVM and KNN were employed to classify the three different conditions. The performances of the two classifiers were obtained under different values of their parameter.

The experimental results show the potential application of SVM for machine fault diagnosis.

The results demonstrate that the proposed approach can be used effectively for detecting unbalanced condition in rotating machines.

In this paper, an intelligent approach for unbalanced fault detection was proposed based on supervised learning method. Also, a performance comparison was made between KNN and SVM in fault classification. In addition, this approach gave a high level of classification accuracy. The proposed intelligent approach can be used for other mechanical faults.

The purpose is using generative adversarial network (GAN) to solve the problem of sample augmentation in the case of imbalanced bearing fault data sets and improving residual network is used to improve the diagnostic accuracy of the bearing fault intelligent diagnosis model in the environment of high signal noise.

A bearing vibration data generation model based on conditional GAN (CGAN) framework is proposed. The method generates data based on the adversarial mechanism of GANs and uses a small number of real samples to generate data, thereby effectively expanding imbalanced data sets. Combined with the data augmentation method based on CGAN, a fault diagnosis model of rolling bearing under the condition of data imbalance based on CGAN and improved residual network with attention mechanism is proposed.

The method proposed in this paper is verified by the western reserve data set and the truck bearing test bench data set, proving that the CGAN-based data generation method can form a high-quality augmented data set, while the CGAN-based and improved residual with attention mechanism. The diagnostic model of the network has better diagnostic accuracy under low signal-to-noise ratio samples.

A bearing vibration data generation model based on CGAN framework is proposed. The method generates data based on the adversarial mechanism of GAN and uses a small number of real samples to generate data, thereby effectively expanding imbalanced data sets. Combined with the data augmentation method based on CGAN, a fault diagnosis model of rolling bearing under the condition of data imbalance based on CGAN and improved residual network with attention mechanism is proposed.

Electric motors are present in most industries today, being the main source of power. Thus, detection of faults is very important to rise reliability, reduce the production cost, improving uptime and safety. Vibration analysis for condition-based maintenance is a mature technique in view of these objectives.

This paper shows a methodology to analyze the vibration signal of electric rotating motors and diagnosis the health of the motor using time and frequency domain responses. The analysis lies in the fact that all rotating motor has a stable vibration pattern on health conditions. If the motor becomes faulty, the vibration pattern gets changed.

Results showed that through the vibration analysis using the frequency domain response it is possible to detect and classify the motors in several induced operation conditions: healthy, unbalanced, mechanical looseness, misalignment, bent shaft, broken bar and bearing fault condition.

The proposed methodology is verified through a real experimental setup.

The individual machine learning methods used for fault detection and classification have accuracy performance at a certain level. A combined learning model composed of different base classifiers rather than an individual machine learning model is introduced to ensure diversity. In this way, this study aims to improve the generalization capability of fault detection and classification scheme.

This study presents a probabilistic weighted voting model (PWVM) with multiple learning models for fault detection and classification. The working principle of this study’s proposed model relies on weight selection and per-class possibilities corresponding to predictions of base classifiers. Moreover, it can improve the power of the prediction model and cope with imbalanced class distribution through validation metrics and F-score.

The performance of the proposed PWVM was better than the performance of the individual machine learning methods. Besides, the proposed voting model’s performance was compared with different voting mechanisms involving weighted and unweighted voting models. It can be seen from the results that the presented model is superior to voting mechanisms. The performance results revealed PWVM has a powerful predictive model even in noisy conditions. This study determines the optimal model from among voting models with the prioritization method on data sets partitioned different ratios. The obtained results with statistical analysis verified the validity of the proposed model. Besides, the comparative results from different benchmark data sets verified the effectiveness and robustness of this study’s proposed model.

The contribution of this study is that PWVM is an ensemble model with outstanding generalization capability. To the best of the authors’ knowledge, no study has been performed using a PWVM composed of multiple classifiers to detect no-faulted/faulted cases and classify faulted phases.

The purpose of this paper is to investigate a control strategy to fulfill low-voltage ride through (LVRT) requirements in wind energy conversion system (WECS).

This paper considers an active front-end converter of a grid connected WECS working under grid fault conditions. Two strategies based on symmetrical components are studied and proposed: the first one considers control only for positive sequence control (PSC); the second one considered a dual controller for positive and negative sequence controller (PNSC). The performance of each strategy is studied on LVRT requirements fulfillment.

This paper shows presents a control strategy based on symmetrical component to keep the operation of grid-connected WECS under unsymmetrical grid fault conditions.

This work is being applied to a 2 kVA laboratory prototype. The lab prototype emulates a grid connected WECS.

This paper validate the PNSC strategy to LVRT requirements fulfillment by experimental results obtained for a 2 kVA laboratory prototype. PNSC strategy allows constant active power delivery through grid-voltage dips. In addition, the proposed strategy is able to grid-voltage support by injection of reactive power. Additional features are incorporated to PNSC: sequence separation method using delay signal cancellation and grid frequency identification using phase locked loop.

The purpose of this paper is to present an analytical calculation for estimating the leakages field distribution in surface-mounted permanent magnet synchronous motors (SMPMSMs) according to a sub-domain field model for eccentricity fault detection.

The magnetic field domain is classified into four sub-domains of PMs, air gap, stator core and outer region. In the proposed method, the governing equations taking the rotor eccentricity effect into account per region and the interface boundary conditions between sub-domains are formulated using the regular perturbation technique, Taylor series and Fourier series expansion. Maxwell's equations are solved in different regions in the polar coordinate system regarding the boundary conditions.

The radial and tangential components of electromagnetic field distribution in all sub-domains of one SMPMSM are obtained using the proposed method analytically. Finite element analysis is used to validate the results of the proposed method; the results indicated that the analytical model matches the finite-element prediction up to 30% eccentricity, except for some peak values that depend on the harmonic order value. The results of this paper demonstrated that in the event of eccentricity, an asymmetric magnetic field is generated in the outer region of the machine. Although its amplitude is small, it can be an indicator for detecting eccentricity faults from the outside environment of the machine.

The formulas presented in this paper can be applied as a new technique for detecting eccentricity faults in these motors from the outside environment.

One of the major problems faced by industry is vibrations in rotating parts. Vibration is a to-and-fro movement of rotating mechanical parts and has many detrimental effects on machinery. It is obvious that no movement can be achieved without consumption of energy. All the energy consumed in vibration of mechanical parts is useless. Unbalance is one of the most common reasons for vibrations. This paper aims to experimentally evaluate the effect of unbalance in a shaft–rotor system on power consumption. An experimental setup consisting of a shaft and a rotor mounted on antifriction bearing was built-up. The shaft was driven through a flexible coupling, by a variable speed DC motor. The shaft–rotor system was rotated at different speeds and electrical power consumed by the system was measured at specific speeds varying from 1,200 to 2400 rpm. The rotor was balanced to grade G6.3 at 1,200 rpm. The power consumption by shaft in balanced condition was taken as baseline data for the further work. The rotor was then made unbalanced by adding different masses at 60 mm radius, and power consumption was recorded again at the same speeds. It was observed that average power loss due to unbalance is of 0.11watt/gm.mm unbalance. This can amount to 2.75 kw if there is unbalance of 50 gm at a radius of 500 mm. This work is meant to emphasis on the fact that the power consumption can be reduced if the vibrations can be kept under control.

The experimental setup consisting of a rotor–shaft system was fabricated. The shaft was supported on two anti-friction bearings. The shaft is driven by a 0.25 HP DC motor. The speed of the motor can be varied by a speed controlling device. A digital ammeter and voltmeter are connected to measure the input current and voltage to the system. The rotor was rotated at different speeds after two-plane balancing and the parameters like voltage, current drawn, rms velocity (average of drive and non-drive side bearing) and displacement at 1× frequency were recorded. The base line data for the balanced shaft–rotor system were recorded for further use.

Power consumption increases with increase in unbalance at each of the speeds. Total power consumed at resonant frequency is high. The average power consumed “W/gm.mm” increases at higher speed due to increased damping force of lubricant in bearings combined with the effect of resonance. Average power consumed due to unbalance is about 0.11 W/gm.mm unbalance. It is important to reduce the vibration to save power which can be effectively achieved by balancing the rotating parts in the machinery.

The experimentation is done on a small rotor. When the same work is done on real situations where the rotors are heavy, we may expect some differences in the actual effect of unbalance on the power consumption.

The experimental work have a huge application in industry in condition monitoring. The power may tend to increase not only because of the unbalance but also due to other reasons of vibrations like misalignment, loose foundation, poor bearing conditions, etc. The power loss may increase due to any other reasons mentioned above. The degree of power saving due to steps taken for reducing vibration will depend on the existing vibration levels.

The work highlights the effect of power loss due to vibrations. Even (1 per cent) small amount of power saved can save millions of dollars in industry, as there are many rotating parts which run 24 × 7. The emphasis is on condition-based monitoring which will help in power saving beyond the conventional advantages of condition monitoring.

The experimentation clearly quantifies power loss in absolute form that is the power loss is expressed per gm.mm of unbalance and not as the percentage of electrical or mechanical power, input or output. The percentage values may be misleading some times, as SMALL percentage of large values is also LARGE and hence should be taken into consideration.

Efficient calculations of the transient behaviour after disturbances of large-scale power systems are complex because of, among other things, the non-linearity and the stiffness of the overall state equation system (SES). Because of the rising amount of flexible transmission system elements, there is an increasing need for reduced order models with a negligible loss of accuracy. With the Extended Nodal Approach and the application of the singular perturbation method, it is possible to reduce the order of the SES adapted to the respective setting of the desired tasks and accuracy requirements.

Based on a differential-algebraic equation for the electric power system which is formulated with the Extended Nodal Approach, the automatic decomposition into reduced order models is shown in this paper. The paper investigates the effects of different coordinate systems for an automatic order reduction with the singular perturbation method, as well as a comparison of results calculated with the full and reduced order models.

The eigenvalues of the full system are approximated sufficiently by the three subsystems. A simulation example demonstrates the good agreement between the reduced order models and the full model independent of the choice of the coordinate system. The decomposed subsystems in rotating coordinates have benefits as compared to those in static coordinates.

The paper presents a systematic decomposition based only on a differential-algebraic equation system of the electric power system into three subsystems.

The purpose of this paper is to study variation regularization with a positive sequence extraction-normalized least mean square (VRP-NLMS) algorithm for frequency estimation in a three-phase electrical distribution system. A simulation test is provided to validate the performance and convergence rate of the proposed estimation algorithm.

Least mean square (LMS) algorithms for frequency estimation encounter problems when voltage contains unbalance, sags and harmonic distortion. The convergence rate of the LMS algorithm is sensitive to the adjustment of the step-size parameter used in the update equation. This paper proposes VRP-NLMS algorithm for frequency estimation in a power system. Regularization parameter is variable in the NLMS algorithm to adjust step-size parameter. Delayed signal cancellation (DSC) operator suppresses harmonics and negative sequence component of the voltage vector in a two-phase Î ± β plane. The DSC part is placed in front of the NLMS algorithm as a pre-filter and a positive sequence of the grid voltage is extracted.

By adapting of the step-size parameter, speed and accuracy of the LMS algorithm are improved. The DSC operator is augmented to the NLMS algorithm for more improvement of the performance of this adaptive filter. Simulation results validate that the proposed VRP-NLMS algorithm has a less misalignment of performance with more convergence rate.

This paper is a theoretical support to simulated system performance.