• Proceedings of the KIEE Conference
• /
• 2003.11a
• /
• pp.412-415
• /
• 2003

It is important to classily fault types, discriminate fault section and calculate the fault location by any detecting technique for combined transmission lines. This paper proposes the technique to classily the fault types and fault section using neuro-fuzzy systems. Neuro-fuzzy systems are composed of three parts to perform different works. First, neuro-fuzzy system for fault type classification is performed with approximation coefficient of currents obtained by wavelet transform. The second neuro-fuzzy system discriminates the fault section between overhead and underground with detail coefficients of voltage and current. The last neuro-fuzzy system calculates the fault location with impedance in this paper, neuro-furry system shows the excellent results for classification of fault types and discrimination of fault section.

• Proceedings of the KIEE Conference
• /
• 2003.07a
• /
• pp.534-536
• /
• 2003

It is important to classily fault types and discriminate fault section by any detecting technique for combined transmission lines. This paper proposes the technique to classify the fault types and fault section using neuro-fuzzy systems. Neuro-fuzzy systems are composed of two parts to perform different works. First, neuro-fuzzy system for fault type classification is performed with approximation coefficient of currents obtained by wavelet transform. Another neuro-fuzzy system discriminates the fault section between overhead and underground with detail coefficients of voltage and current. In this paper, neuro-fuzzy system shows the excellent results for classification of fault types and discrimination of fault section.

• Proceedings of the KIEE Conference
• /
• 1999.11b
• /
• pp.131-133
• /
• 1999

This paper proposed the technique of the fault-types classification using Adaptive Network Fuzzy Inference System in the distribution system. Fault and fault-like data in the linear RL load, arc furnace load and converter load were extracted by EMTP. These were characterized into 5 input variables and fuzzified automatically by learning. This technique was tested using another fault data unused learning.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.51 no.5
• /
• pp.209-216
• /
• 2002

Recently, there is greater demand for stable supply of electric power as higher level of our living. It becomes the important problem that the cause of fault in power system is found out in early stage, if once it occurs. In this respect, accurate classification of arcing faults in power systems is vitally important. This paper presents a new classification method for arcing faults in power system. To obtain data of various faults including high impedance fault(HIF) and low impedance fault(LIF), HIF model with the ZnO arrester is adopted and implemented within the overall transmission system model based on the electromagnetic transients program(EMTP). Results of phase plane trajectory if Clarke modal transformation using postfault current and voltage are utilized to classify types of arcing faults. The performance of the proposed method is tested on a typical 154 kV korean transmission system under various fault conditions. As can be seen from results, phase plane trajectory of postfault current should be combined with that of o component from Clarke modal transformation to give reliability of clear fault classification. Thus the proposed method can classify arcing faults including LIFs and HIFs accurately in power systems.

• Proceedings of the KIEE Conference
• /
• 2001.07a
• /
• pp.488-490
• /
• 2001

It is important to catch or classify fault types by any detecting technique for distribution protection. This paper proposes the technique to classify the fault types using wavelet transform in radial distribution line. Modeling of the radial distribution line is simulated using PSCAD/EMTDC and wavelet transform is performed in the Matlab program.

• Proceedings of the KIEE Conference
• /
• 1998.07g
• /
• pp.2272-2274
• /
• 1998

This paper presents recognition and classification of high impedance fault(HIF) patterns in the electrical power systems based on chaotic features. Chaotic features are obtained from two dimensional chaos attractors reconstructed from fault current waveform. The RBFN is trained with the two types of HIF data generated by the electromagnetic transient program and measured from actual faults. The RBFN successfully classifies normal and the three types of fault patterns based on the binary chaotic features.

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 2003.09a
• /
• pp.28-35
• /
• 2003

Quaternary faults found around the Ulsan Fault System can be divided into 4 types based on the fault outcrop features : Type I fault cuts basements and Quaternary deposits of which remain on both hangwall and footwall. Type II fault is developed only in Quaternary deposit. Type III fault has inclined unconformity after Quaternary faulting. Type IV fault is common type around the Ulsan fault system and has horizontal unconformity surface after cutting earlier Quaternary deposit. After erosion, later Quaternary deposit overlays on both old deposit and basement. The Ulsan Fault System consists of three segments at large scale from north to south based on the lineament rank and shape, Quaternary fault location, and slip rate. The segment boundaries are identified by the existence of the two intervals which show no lineaments and Quaternary faults. But, if detail fault parameters could be obtained and used in segmentation, it can be divided into more than three segments.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.4
• /
• pp.1614-1622
• /
• 2018

The research related to fault diagnosis in permanent magnet synchronous motors (PMSMs) has attracted considerable attention in recent years because various faults such as permanent magnet demagnetization and short-circuited turns can occur and result in unexpected failure of motor related system. Several conventional current and back electromotive force (BEMF) analysis techniques were proposed to detect certain faults in PMSMs; however, they generally deal with a single fault only. On the contrary, cases of multiple faults are common in PMSMs. We propose a fault diagnosis method for PMSMs with single and multiple combined faults. Our method uses three phase BEMF voltages based on the fast Fourier transform (FFT), support vector machine(SVM), and visualization tools for identifying fault types and severities in PMSMs. Principal component analysis (PCA) and t-distributed stochastic neighbor embedding (t-SNE) are used to visualize the high-dimensional data into two-dimensional space. Experimental results show good visualization performance and high classification accuracy to identify fault types and severities for single and multiple faults in PMSMs.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.9
• /
• pp.2927-2941
• /
• 2022

The primary task of machine fault diagnosis is to judge whether the current state is normal or damaged, so it is a typical binary classification problem with mutual exclusion. Mutually exclusive events and out-of-domain detection have one thing in common: there are two types of data and no intersection. We proposed a fusion model method to improve the accuracy of machine fault diagnosis, which is based on the mutual exclusivity of events and the commonality of out-of-distribution detection, and finally generalized to all binary classification problems. It is reported that the performance of a convolutional neural network (CNN) will decrease as the recognition type increases, so the variational auto-encoder (VAE) is used as the primary model. Two VAE models are used to train the machine's normal and fault sound data. Two reconstruction probabilities will be obtained during the test. The smaller value is transformed into a correction value of another value according to the mutually exclusive characteristics. Finally, the classification result is obtained according to the fusion algorithm. Filtering normal data features from fault data features is proposed, which shields the interference and makes the fault features more prominent. We confirm that good performance improvements have been achieved in the machine fault detection data set, and the results are better than most mainstream models.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.12
• /
• pp.1486-1491
• /
• 1999

This paper presents a method of recognizing high impedance fault(HIF) of electrical power systems and classifying fault patterns based on chaos attractors. Two dimensional chaos attractors are reconstructed from neutral point current waveforms. Reliable features for HIF pattern classification are obtained from the chaos attractors. Radial basis function network, trained with two types of HIF data generated by the electromagnetic transient program and measured form actual faults. The RBFN successfully classifies normal and the three types of fault patterns according to the features generated from the chaos attractors.