이 논문에서는 신호 모델에 기반하여 유도전동기의 고장 검출 및 고장 진단을 위한 새로운 시스템을 제안한다. 산업현장에 적용하는 기존의 제품들은 신호가 문턱치를 넘어면 고장을 검출하는 단순한 알고리듬을 가지고 있어 고장의 유형이나 고장을 예측하는데 문제가 있다. 이 논문에서는 이러한 문제들을 해결하기 위한 시스템을 제안한다. 이 시스템은 고장 검출 과정과 고장 진단 과정으로 구성되며, 고장 검출 과정은 기계 신호음들이 웨이블렛 필터뱅크를 통과한 후 웨이블렛 계수들의 분산과 상관도를 분석하여 고장을 검출한다. 고장 진단 과정은 패턴분류기술을 적용하여 고장의 유형을 진단하게 된다. 대표적인 유도전동기 고장 유형들로서는 불평형, 미스얼라이먼트, 그리고 베어링 루스 등이 있으며, 이러한 유형들은 제안하는 시스템에서 분석되고 진단을 받게 된다. 제안하는 시스템에 적용한 결과 상관도를 이용한 방법은 78 %, 분산을 이용한 방법은 95 % 이상의 고장진단율을 보이는 우수한 결과를 나타내었다.
이 논문에서는 신호 모델에 기반하여 유도전동기의 고장 검출 및 고장 진단을 위한 새로운 시스템을 제안한다. 산업현장에 적용하는 기존의 제품들은 신호가 문턱치를 넘어면 고장을 검출하는 단순한 알고리듬을 가지고 있어 고장의 유형이나 고장을 예측하는데 문제가 있다. 이 논문에서는 이러한 문제들을 해결하기 위한 시스템을 제안한다. 이 시스템은 고장 검출 과정과 고장 진단 과정으로 구성되며, 고장 검출 과정은 기계 신호음들이 웨이블렛 필터뱅크를 통과한 후 웨이블렛 계수들의 분산과 상관도를 분석하여 고장을 검출한다. 고장 진단 과정은 패턴분류기술을 적용하여 고장의 유형을 진단하게 된다. 대표적인 유도전동기 고장 유형들로서는 불평형, 미스얼라이먼트, 그리고 베어링 루스 등이 있으며, 이러한 유형들은 제안하는 시스템에서 분석되고 진단을 받게 된다. 제안하는 시스템에 적용한 결과 상관도를 이용한 방법은 78 %, 분산을 이용한 방법은 95 % 이상의 고장진단율을 보이는 우수한 결과를 나타내었다.
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. ...
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.
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.
* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.
문제 정의
In this approach, we proposed an approach to use the wavelet and cross-correlation techniques for fault detection and fault diagnosis processes. The basic idea of this approach is to find the most important frequency bands for each reference signal in each faulty category. Then, the comparison between the trained signal and the reference signal is checked just in the most important frequency band.
가설 설정
The thresholds for each the wavelet bands are assigned as shown in Table 5. We just consider the important bands and neglect the other bands since only the important bands provide the essential information for identifying the faults. The thresholds are dependent on the engineering expert’s experience.
In this experiment we set the thresholds using our feel on the sound and vision on frequency spectrum. Finally, the total classification results are presented in Table 6. Among nine signals in the BF category, five signals are not correctly classified that is a big trouble in this approach. The overall classification accuracy reaches more than 78 percent.
제안 방법
Dictionary : For each category, a standard signal is selected as a reference signal that is closest to the other sample signals in the same category. After using the wavelet transform to decompose the reference signals from all categories into frequency bands(suppose N bands created), each frequency band are saved to the dictionary as the frequency feature vectors, which are used to make cross-correlations with trained signals. Therefore, in a dictionary for a faulty category, there are N frequency feature vectors existing.
The thresholds are dependent on the engineering expert’s experience. In this experiment we set the thresholds using our feel on the sound and vision on frequency spectrum. Finally, the total classification results are presented in Table 6.
By analyzing the features of the sounds to generate the faulty symptoms in the fault detection process, we predict the motor condition in the fault diagnosis process. In this experiment, sounds from typical faults of the induction motor are collected categorized into faulty categories. Through the faulty symptoms, any sound signals generated from the motor are classified to the corresponding faulty category.
The methods for selecting the reference feature vector are various such as the vector that has the minimum Euclidean distance with other feature vectors in the category. In this paper, we determine the reference feature vector that contains collection of variance values of entire wavelet bands with each variance value of the reference feature vector is the mean of all variance values from sample signals in that category.
대상 데이터
For each case, a one-second duration segment of a sound signal is stored as a separate signal. In this experiment, there are 42 sound signals used for training with nine signals for BF, nine for NOR, eight for LO, seven for UN and nine for MIS. This set of sound signal is referred to the testing database.
성능/효과
The final decision for classification is that an arbitrary trained signal is referred to the equivalent signal category if it is the most similar to the reference feature vector of this category as well as the Euclidean distance must be smaller than an assigned threshold. The threshold may be allocated to a scalar value depending on faulty conditions(normally, assigned by the expert’s experience).
참고문헌 (23)
Cohen, L., 1995, “Time-Frequency Analysis,” Prentice Hall
Eren, L., Teotrakool, K. and Devaney, M., 2007, “Bearing Fault Detection via Wavelet Packet Decomposition with Spectral Post Processing,” Instrumentation and Measurement Technology Conference - IMTC 2007,Warsaw, Poland
Gertler, J. J., 1998, Fault Detection and Diagnosis in Engineering Systems, Marcel Dekker, Inc, America
Norton, M. P. and Karczub, D. G., 2003, “Fundamentals of Noise and Vibration Analysis for Engineers,” Cambridge University Press, United Kingdom
Vetterli, M. and Kovacevic, J., 1995, “Wavelet and Subband Coding,” Prentice Hall
Akansu, A. N. and Haddad, R. A., 1992, “Multiresolution Signal Decomposition,” Academic Press
Chung, B.-H. and Shin, D.-C., 2004, “A Development of the Algorithm to Detect the Fault of the Induction Motor Using Motor Current Signature Analysis,” Transactions of the Korean Society for Noise and Vibration Engineering, Vol. 14, No. 8, pp. 675~683
McInerny, S. A. and Dai, Y., 2003, “Basic Vibration Signal Processing for Bearing Fault Detection,” IEEE transaction on Education, Vol. 46, pp. 149~156
Isermann, R., 2006, “Fault Diagnosis System : An Introduction from Fault Detection to Fault Tolerance,” Springer, Germany
Isermann, R., 1994, “Integration of Fault Detection and Diagnosis Methods,” In Proceeding of IFAC Symposium on fault detection, Supervision and Safety for Technical Processes, Espoo, Finland, pp. 587~609
Isermann, R., 1997, “Supervision, Fault Detection and Diagnosis Methods- an Introduction,” Control Engineering Practice ？ CEP. Vol. 5, No. 5, pp. 639~652
Chung, B.-H. and Shin, D.-C., 2002, “A Study on Detection of Broken Rotor Bars in Induction Motors Using Current Signature Analysis,” Transactions of the Korean Society for Noise and Vibration Engineering, Vol. 12, No. 4, pp. 287~293
Widodo, A. and Yang, B.-S., 2006, “Intelligent Fault Diagnosis of Induction Motor Using Support Vector Machines,” Proceedings of the KSNVE Annual Autumn Conference 2006, KSNVE06A-11-03
Tran, V. T., Yang, B.-S. and Oh, M.-S., 2006, “Fault Diagnosis of Induction Motors Using Decision Trees,” Proceedings of the KSNVE Annual Autumn Conference 2006, KSNVE06A-11-04
Calis, H., 1998, “Current-based Detection of Mechanical Faults in Induction Motors,” PhD thesis, University of Sussex
Chow, T. W. S., 1996, “Condition Monitoring of Electrical Machines Using Third-order Spectrum Analysis,” IEEE IAS Annual Meeting. Vol. 1, pp. 679~686
Schoen, R., 1994, “On-line Current-based Condition Monitoring of Three-phase Induction Machines,” PhD thesis at Georgia Institute of Technology
Boyle, C., et al, 1994, “Online Current Monitoring to Detect Misalignment and Dynamic Eccentricity in Three-phase Induction Motor Drives,” In: Proceedings of the 29th Universities Power Engineering Conference 1, Galway, Ireland, pp. 5~8
Cameron, J. R., 1987, “Vibration and Current Monitoring for Online Detection of Airgap Eccentricity in Induction Motors. PhD Thesis at Robert Gordon Institute of Technology
Arkan, M., Calis, A. H. and Tagluk, A. M. E, 2005, 'Bearing and Misalignment Fault Detection in Induction Motors by Using the Space Vector Angular Fluctuation Signal,' Electrical Engineering. Vol. 87, pp. 197~206
McInerny, S. A. and Dai, Y., 2003, “Basic Vibration Signal Processing for Bearing Fault Detection,” IEEE Transaction on Education, Vol. 46, No. 1, pp. 149~156
Yang, B.-S., Kim, K. J. and Han, T., 2004, “Fault Diagnosis of Induction Motors Using Data Fusion of Vibration and Current Signals,” Transactions of the Korean Society for Noise and Vibration Engineering, Vol. 14, No. 11, pp. 1091~1100
※ AI-Helper는 부적절한 답변을 할 수 있습니다.