• Transactions of the Korean Society of Automotive Engineers
• /
• v.25 no.3
• /
• pp.273-286
• /
• 2017

We develop an inspection method for bearing faults with a rapid change in the rotation speed and present indexes for the pass/fail inspection. At the end of line, impulse noises generated by the operation of machines and conveyors may distort the inspection results. In this paper, we present robust inspection indexes for bearing faults under impulse noises, by taking into account fault signals having pulse train. Using logistic regression, we optimize the pass/fail criterion for each index and evaluate the performance of the inspection indexes based on the total error rate.

• Journal of the Korean Society for Precision Engineering
• /
• v.25 no.3
• /
• pp.83-92
• /
• 2008

This paper develops the theory for a fault-tolerant, permanent magnet biased, homopolar magnetic bearing. If some of the coils or power amplifiers suddenly fail, the remaining coil currents change via a novel distribution matrix such that the same magnetic forces are maintained before and after failure. Lagrange multiplier optimization with equality constraints is utilized to calculate the optimal distribution matrix that maximizes the load capacity of the failed bearing. Some numerical examples of distribution matrices are provided to illustrate the theory. Simulations show that very much the same dynamic responses (orbits or displacements) are maintained throughout failure events (up to any combination of 3 coils failed for the 6 pole magnetic bearing) while currents and fluxes change significantly. The overall load capacity of the bearing actuator is reduced as coils fail. The same magnetic forces are then preserved up to the load capacity of the failed.

• Journal of KSNVE
• /
• v.10 no.6
• /
• pp.985-990
• /
• 2000

The signals that can be obtained from rotating machines often convey the information of machine. For example, if the machine under investigation has faults, then these signals often have pulse signals, embedded in noise. Therefore the ability to detect the fault signal in noise is major concern of fault diagnosis of rotating machine, In this paper, minimum variance cepstrum (MV cepstrum) . which can easily detect impulse in noise, has been applied to detect the type of faults of ball bearing system. To test the performance of this technique. various experiments have been performed for ball bearing elements that have man made faults. Results show that minimum variance cepstrum can easily detect the periodicity due to faults and also shows the pattern of excitation by the faults.

• 제어로봇시스템학회:학술대회논문집
• /
• 1989.10a
• /
• pp.445-450
• /
• 1989

In this study, a variety of method to diagnose a fault of rotatory machine is suggested. Apprehending the physical meaning of each techniques, computer simulation is performed. The result from this computer simulation and the signal of the faulted ball bearing is studied from all its aspect. It is found that this conditioning monitor system is effective.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2001.05a
• /
• pp.603-608
• /
• 2001

In order to distinguish fault electric motors automatically in real time. an intelligent diagnosis technique may be required. This paper presents an automatic fault detection system for electric motors by using their acoustic noises. Time signals of each candidate motor were measured in an anechoic chamber for further analysis. Spectral analysis was first carried out and they showed that two typical types of fault motors could be successfully distinguished in the frequency domain; bearing faults and scratches. Unlike the trend of normal motors that shows only a single dominant peak at around 2000 ㎐, several peaks are bunched together in bearing fault motors. On the other hand, large frequency noises at around 6500 ㎐ are newly arisen in scratchy fault motors. However, the processing time for spectral analysis was rather long for a real time application in production lines. Thus, a number of band-pass filters were used in the time domain instead for a real time application. Before applying filters, the bands of filters were set from the information of spectral analysis. By applying a set of band-pass filters, the RMS values of each filtered signal were calculated, and thus the normal and damaged motors could be successfully distinguished.

• Smart Structures and Systems
• /
• v.33 no.4
• /
• pp.253-262
• /
• 2024

Current-based mechanical fault diagnosis is more convenient and low cost since additional sensors are not required. However, it is still challenging to achieve this goal due to the weak fault information in current signals. In this paper, a dual-loss convolutional neural network (DLCNN) is proposed to implement the intelligent bearing fault diagnosis via current signals. First, a novel similarity loss (SimL) function is developed, which is expected to maximize the intra-class similarity and minimize the inter-class similarity in the model optimization operation. In the loss function, a weight parameter is further introduced to achieve a balance and leverage the performance of SimL function. Second, the DLCNN model is constructed using the presented SimL and the cross-entropy loss. Finally, the two-phase current signals are fused and then fed into the DLCNN to provide more fault information. The proposed DLCNN is tested by experiment data, and the results confirm that the DLCNN achieves higher accuracy compared to the conventional CNN. Meanwhile, the feature visualization presents that the samples of different classes are separated well.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.11
• /
• pp.17-24
• /
• 2014

This paper proposes a fault detection method for low-speed rolling element bearings of an induction motor using acoustic emission signals and histogram modeling. The proposed method performs envelop modeling of the histogram of normalized fault signals. It then extracts and selects significant features of each fault using partial autocorrelation coefficients and distance evaluation technique, respectively. Finally, using the extracted features as inputs, the support vector regression (SVR) classifies bearing's inner, outer, and roller faults. To obtain optimal classification performance, we evaluate the proposed method with varying an adjustable parameter of the Gaussian radial basis function of SVR from 0.01 to 1.0 and the number of features from 2 to 150. Experimental results show that the proposed fault identification method using 0.64-0.65 of the adjustable parameter and 75 features achieves 91% in classification performance and outperforms conventional fault diagnosis methods as well.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.10a
• /
• pp.819-820
• /
• 2016

Electrical power distribution is consists of high voltage, low voltage and motor control center(MCC). Motor control centers involves turning the motor on and off, it is configured electronic over current relay to detect a motor overcurrent flows. Existing electronic over current relay detects electrical fault such as overcurrent, undercurrent, phase sequence, negative sequence current, current unbalance and earth fault. However, it is difficult to detect mechanical fault such as locked rotor, motor stator and rotor and bearing fault. In this paper, we propose a condition monitoring and fault diagnosis system for electrical and mechanical fault detection of rotating machinery. The proposed system is designed with signal input and control part, system interface part and data acquisition board for condition monitoring and fault diagnosis, it was possible to detect electrical fault and mechanical fault through measurement and control of insulation resistance, locked rotor, MC counter and bearing temperature.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2009.04a
• /
• pp.681-688
• /
• 2009

This paper deals with a bearing faults localization technique based on holographic approach by visualizing sound radiated from the faults. The main idea stems from the phenomenon that bearing faults in a moving vehicle generate impulsive sound. To visualize fault signal from the moving vehicle, we can use the moving frame acoustic holography [H.-S. Kwon and Y.-H. Kim, "Moving frame technique for planar acoustic holography," J. Acoust. Soc. Am. 103(4), 1734-1741, 1998]. However, it is not easy to localize faults only by applying the method. This is because the microphone array measures noise (for example, noise from other parts of the vehicle and the wind noise) as well as the fault signal while the vehicle passes by the array. To reduce the effect of noise, we propose two ideas which utilize the characteristics of fault signal. The first one is to average holograms for several frequencies to reduce the random noise. The second one is to apply the partial field decomposition algorithm [K.-U. Nam, Y.-H. Kim, "A partial field decomposition algorithm and its examples for near-field acoustic holography," J. of Acoust. Soc. Am. 116(1), 172-185, 2004] to the moving source, which can separate the fault signal and noise. Basic theory of those methods is introduced and how they can be applied to localize bearing faults is demonstrated. Experimental results via a miniature vehicle showed how well the proposed method finds out the location of source in practice.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.19 no.7
• /
• pp.726-735
• /
• 2009

In this paper, we propose an approach to signal model-based fault detection and diagnosis system for induction motors. The current fault detection techniques used in the industry are limit checking techniques, which are simple but cannot predict the types of faults and the initiation of the faults. The system consists of two consecutive processes: fault detection process and fault diagnosis process. In the fault detection process, the system extracts the significant features from sound signals using combination of variance, cross-correlation and wavelet. Consequently, the pattern classification technique is applied to the fault diagnosis process to recognize the system faults based on faulty symptoms. The sounds generated from different kinds of typical motor's faults such as motor unbalance, bearing misalignment and bearing loose are examined. We propose two approaches for fault detection and diagnosis system that are waveletand-variance-based and wavelet-and-crosscorrelation-based approaches. The results of our experiment show more than 95 and 78 percent accuracy for fault classification, respectively.