, pp.113-117 http://dx.doi.org/10.14257/astl.2016.141.23 Partial Disconnected Cable Fault Detection Using Improved SSTDR Ga-Ram Han 1, Jeong-Chay Jeon 1, Jae-Jin Kim 1 and Myeong-Il Choi 1 1 Korea Electrical Safety Corporation #12 Ogong-ro, Iseo-myeon, Wanju-gun, Jeollabuk-do, 55365, Rep. of Korea Abstract. To prevent electric shock accidents caused by electric leakage, it is important to detect cable insulation faults in advance. This paper shows the results that an improved SSTDR is more effective in detecting partial disconnected cable fault than the conventional method. The existing SSTDR has difficulties to determine accurate partial fault location due to signal attenuation. The improved SSTDR was validated through comparison with existing methods in partial cable fault. Keywords: Cable fault location, Partial disconnection, SSTDR, Fault detection 1 Introduction Power cables are susceptible to many faults such as insulation damage, open fault, or short circuit due to inappropriate installation and other various physical, electrical, or environmental factors, which could lead to electrical fires. According to the electrical accident statistics published by the Korea Electrical Safety Corporation, 20% of electric facility accidents are caused by cables, and most electrical fires are related to cables [1]. Many methods have been proposed to diagnose cable faults and detect their locations such as partial discharge measurement. Among them, reflectometry has been most commonly used for fault detection, in which a specific pulse such as radar is injected into a cable, thereby measuring a reflected signal produced due to the mismatch of characteristic impedance from fault location. (TDR, TFDR, STDR, SSTDR, etc.) SSTDR (Spread Spectrum Time Domain Reflectometry) have been studied to minimize measurement error and facilitate easy fault detection [2]. But, in the case of partial disconnection, fault detection is more difficult to find correct location than open- and short-circuit fault due to small change of characteristic impedance. This paper demonstrates an improved SSTDR for Partial Disconnection Fault(PDF) detection. Improved SSTDR proposed in [3] is consisted of two steps: peak value of the correlation coefficient of the reference signal is detected using timefrequency correlation analysis, and then a peak value of the correlation coefficient of ISSN: 2287-1233 ASTL Copyright 2016 SERSC
the reflected signal is detected after removing the reference signal to solve the problem of the inaccurate PDF detection due to signal attenuation. 2 The PDF Detection Using Improved SSTDR The existing SSTDR analyzes signals in the time correlation function and it has difficulties to determine the accurate PDF location due to side lobe and reflected signal attenuation. The improved method analyzes signals in the time-frequency correlation to solve the problem. That s why it has enhanced performance in determine the PDF point with main lobe despite signal attenuation. The timefrequency correlation analysis in the SSTDR improved employs the Wigner Ville Distribution (WVD) to analyze the reference and reflected signals in the timefrequency domain [4]. Also, PDF detecting error increases when the reflected signal overlapped with the injected signal or the reflected signal is too small to be analyzed in contrast to the injected. In order to solve such weaknesses, it is possible to make the reflected signal stand out by removing the influence of the reference signal from the measured signal. According to the reference [3], the improved method searches the maximum value 1 C ( sr ) location of the reference signal from the time-frequency correlation function of reference and the reflected signals to find a the reference signal is removed from the measured signal to make e ( t) r( t) s( t 1), and the peak value of the correlation function of the reflected C signal is found via the second time-frequency correlation function se ( ) of and. Finally, a time difference d 1 2 between the peak values is calculated to obtain the distance to the PDF location. s(t) s(t) 2 s(t) location as shown in Fig. 1. Then, r(t) e(t) Fig. 1. Diagram of the improved SSTDR 114 Copyright 2016 SERSC
3 Experimental Results To validate the performance of the improved SSTDR, an experiment was conducted as shown in Fig. 2. An F-CV2C6SQ cable was used for the experimental target cable because it has been most widely used in low-voltage power systems. The SSTDR experimental setup in Fig. 2 consists of a control unit, an arbitrary waveform generator, a digital oscilloscope and T connector. To automatically control the Arbitrary Waveform Generator (NI PXI 5422, 16bits, 200 MS/s) that generates a signal injected into a cable and digital oscilloscope (NI PXIe-5162, 10bits, 5 GS/s, 1.5 GHz) that acquires a signal reflected from the PDF point, the NI LabVIEW program was developed and MATLAB was used to analyze correlations between reference and measured signals. In the experiment, the reference and measurement signals were injected and measured through the T connector and RG58 cables. Fig. 2. Experimental setup for the improved SSTDR In this experiment, in order to simulate the PDF at 70m, 130m on a 200m cable, one of 2 cores was cut step by step (a core consists 7line). The PDF location D using a time difference and VOP (Velocity Of Propagation) which is a variable that represents the propagation velocity of the electromagnetic wave in a corresponding cable is given by (1) t r Where is detection time of reflection signal, and t s is start time of reference signal.( VOP of 1.905 108m/s) As shown in Fig. 3., the existing SSTDR using the time correlation cannot determine the PDF location because of the weak reflected signal due to attenuation. Copyright 2016 SERSC 115
Fig. 3. Measurement for PDF with the conventional SSTDR The improved SSTDR has correlation value that can detect the PDF location more effectively than the conventional method through time-frequency correlation analysis as shown in Fig. 4(a). However, in some cases a larger correlation value was occured in the reference signal portion and the detecting error occurs in which the PDF location is measured at 38.1m. In oder to solve this problem, it is possible to accurately detect the PDF as 128.7m as shown Fig. 4(b) by removing the reference signal(error in 1%). (a) (b) Fig. 4. Measurement for PDF with the improved SSTDR 4 Conclusions Even if the PDF (due to various physical or electrical factors) is occurred, the traditional methods of detection is difficult to find fault points because of small changes of the impedance. 116 Copyright 2016 SERSC
The performance of the improved SSTDR, which used time-frequency correlation analysis and remove the reference signal, was evaluated through the PDF detection experiments using 200m (PDF at 70m, 130m) low-voltage power cables. Against the existing SSTDR, the ability to detect the cable fault and the accuracy of finding the PDF point are increase. The improved SSTDR is expected to apply for resolving the difficulties of the PDF detection due to various noises and signal attenuation. Acknowledgments. This study was supported by "2013 Dual Use Technology Program". References 1. Korea Electrical Safety Corporation: A Statistical Analysis on the Electrical Accident (2013) 2. Chirag R. Sharma, Cynthia Furse and Reid R. Harrison: Low-Power STDR CMOS Sensor for Location Faults in Aging Aricraft Wiring: IEEE Sensors Journal, Vol. 7, No. 1, pp. 43 50 (2007) 3. Jeong-Chay Jeon, Jae-Jin Kim, Myeong-Il Choi: Detection and Location of Cable Fault Using Improved SSTDR: KIEE The Transactions of the Korean Institute of Electrical Engineers, Vol. 65, No. 9, pp. 1583 1589, (2016) 4. Y. J. Shin, E. J. Powers, T. S. Choe, C. Y. Hong, E. S. Song, J. G. Yook and J. B. Park: Application of Time -Frequency Domain Reflectometry for Detection and Localization of a Fault on a Coaxial Cable: IEEE Trans. on Instrumentation and Measurement, Vol. 54, No. 6, pp. 2493--2500(2005) Copyright 2016 SERSC 117