A Transient Current Based Wavelet-Fuzzy Approach for the Protection of Six-Terminal Transmission System

Transcription

1 Abstract International Journal of Exploration in Science and Technology A Transient Current Based Wavelet-Fuzzy Approach for the Protection of Six-Terminal Transmission System J.Uday Bhaskar 1, G.Ravi Kumar 2 and S.S.Tulasi Ram 3 1 Associate Professor, EEE Department, DMSSVH College of Engineering, Machilipatnam udayadisar@gmail.com 2 Professor, EEE Department, Kakatiya Institute of Technology and Sciences, Warangal goli.ravikumar@yahoo.com 3 Professor, EEE Department, University College of Engineering, JNTU, Hyderabad. ramsankara@gmail.com In order to supply power to communities in the remote rural and densely populated areas, multi terminal transmission lines provide an inviting solution at minimum cost and the difficulties in obtaining the Right-of-Way, constructing a new sub- station have also led to the development of multi terminal transmission systems. Introducing a tap point in the line complicates the protection and also these transmission systems are more vulnerable to faults which are required to be cleared immediately for minimizing the power system disturbances and for efficient transmission. This paper projects the systematic approach for fault detection, classification, faulty terminal identification and fault location estimation of a six-terminal transmission system based on frequency domain approach employing Wavelet multi resolution analysis with the variations in fault inception angle, fault distance by extracting the fault indices of transient current signals of all the three phases at all terminals. These fault indices were then compared with threshold value to detect and classify the faults and also identify the faulty terminal. The fault indices were used as inputs to the Fuzzy Inference System (FIS) for approximate fault location estimation from the respective terminals. Bior 1.5Wavelet has been chosen as mother wavelet as it has given the accurate results when compared with other members of wavelet family. An extensive series of simulations were carried out in MATLAB environment with 110 Km, 400KV six- terminal transmission system to verify the accuracy of the proposed algorithm which can be used as effective tool for real time digital relaying purpose as it has taken less than half cycle to detect the faults. Keywords: Multi terminal transmission, wavelet, threshold value, fault indices, fault inception angle, Fuzzy inference system 1. INTRODUCTION The protection of six terminal transmission systems is more difficult and complicated, as they experience problems such as the intermediate in feed of the currents from the other terminals or an out feed to the terminals, variations in section lengths, source impedances and superimposing of currents which require the system to be protected under fault conditions. The accurate fault detection, classification and location estimation are vitally important for efficient transmission as the faults cause interruption of power flow. Quick detection of faults helps in faster maintenance and restoration of supply which results in improved economy and power supply reliability. Wavelet transform is an effective tool in analyzing transient current signals associated with faults both in frequency and time domain. A.H.Osman et al considered two terminal transmission system and used wavelet transform for detection of faults with analysis of voltage and current signals [1]. Feng Liang et al found the effectiveness of relaying operation with application of wavelet transforms in distance protection of transmission lines[2]. Ebha Koley et al considered six phase transmission line and classified the faults using wavelets and located the fault distance using ANN[3]. M.J.Reddy et al considered two terminal 100

2 transmission line to detect and classify the faults using wavelet fuzzy approach[4].quanyuan jiang et al utilized phasor measurement units to locate the multi terminal transmission lines[5].r.k.aggarwal et al proposed protection of teed feeders by utilising high frequency travelling wave information contained in the post-fault voltage and current signals [6]. P.G.Mclarenn et al considered travelling wave techniques for the protection of teed circuits[7] Bo, Z.Q proposed non-communication protection technique for transmission lines[8].bhalija,b et al considered high resistance faults in the protection of two terminal parallel transmission lines[9].brahma S.M et al considered synchronized voltage measurements for fault location on transmission lines[10]. Lyonette et al considered directional comparison technique for protection of teed transmission circuits[11]. Bhalija,B et al projected differential protection scheme for tapped transmission lines[12] T.Nagasawa et al developed a new fault location algorithm for multi terminal two parallel transmission lines by using current differentials at terminals to reduce multiterminal lines to a two-terminal line[13]. T.Funabashi et al considered parallel double- circuit multi terminal transmission lines.[14] Prarthana Warlyani et al considered ANN for fault classification and faulty section identification in Teed transmission circuits by using voltage and current signals of each section of teed circuit to detect and classify L-L-G faults[15].adel A. lbaset et al used ANFIS for transmission line protection[16].g.ravikumar et al considered neuro wavelet approach for multi terminal transmission line protecton[17] S.Chandrasekhar et al considered multi terminal line protection connected with microgrid[18]. There must be some innovative methods to be developed for multi terminal transmission line protection. In this paper, Wavelet Multi-Resolution Analysis was used for detection and classification of faults and faulty terminal identification on a six-terminal transmission system. Fuzzy Inference System has been applied for fault location estimation. Detail D1 coefficients of current signals at all the six terminals were used to detect and classify the faults. The current signals were analyzed taking into consideration that sum of the current coefficients at all the six terminals. The faulty terminal identification was done by considering the fault indices of the faulty phase at all the terminals and the fault index of the faulty terminal is maximum when compared with the fault indices for the same phase at the remaining terminals which was a very novel feature in terminal identification. The entire work has been carried out in MATLAB Simulink environment by considering Bior 1.5 as mother wavelet which was found to be superior in detecting the faults when compared with the other members of the wavelet family, and series of extensive simulations were carried out with variations in fault inception angle, fault impedance, and distance with the positional change of fault along the path of the terminals from one-two, two-three, three-four, four-five, five-six, six-one and the results obtained were found to be accurate and effective. 2. WAVELET ANALYSIS Wavelet transform is a mathematical tool applied for feature extraction in signal analysis. The basic idea of wavelet transform is that it allows only changes in time extension. A wavelet with an amplitude that begins at zero, increases, and then decreases back to zero. Wavelet converts a continuous time signal into different scale of components and data into frequency- components, and analyzes each component with a resolution matched to its scale. By fine tuning the band of analysis, the high frequency components and low frequency components are detected accurately which is the major advantage of the wavelet transform. Multi resolution analyzes the signal at different frequencies with different resolutions The wavelet transform can be analyzed in both time domain and the frequency domain. Wavelet Transform is defined as a sequence of a function {h(n)}(low pass filter) and {g(n)} (high pass filter). The scaling function ⱷ(t) and wavelet Ψ(t) are defined by the following equations. t 2 h n 2t n t 2 g n 2t n, 101

3 Where g(n)=(-1)n h(1-n). A sequence of {h(n)} defines a Wavelet Transform. Wavelets allow the decomposition of a signal into various different levels of resolution which gives a better signal characterization and a more reliable discrimination. The multi resolution of a signal analyses to decompose into an approximation and a detail. The approximation is further decomposed into another approximation and detail and the process is repeated and these successive decompositions are called levels. This feature extraction property of Wavelets Transforms is exploited in the area of protection of transmission line to detect and classify the faults on various components. There are many types of wavelets like Daubachies,Haar, Symlet, Bior etc. The mother wavelet selection was based on the type of application and after extensive work with all wavelets, Bior-1.5 was found to be effective and has been selected as mother wavelet with the 1st level decomposition of the signal has been considered for the analysis of fault classification and terminal identification since it satisfies the characteristic relationships for all types of faults for protection purpose. 3. SYSTEM CONSIDERED FOR FAULT DETECTION AND CLASSIFICATION The single line diagram of the system considered for the protection scheme is given below which is incorporated with six-terminal transmission system of 110-km, 400 kv transmission lines whose line parameters are R 0= Ω/km,R 1 =0.02Ω/km,L 0 =3.5Mh/km, L 1 =0.94mH/km,C 0 =0.0083μf/km., C 1 =0.012μf/km. A sampling frequency of 16KHZ was chosen to capture the high frequency content of current signals. The system has been modeled in MATLAB Simulink environment. The network was simulated for various fault situations. Series of extensive simulations were carried out for L-G,L-L, L- L-G, L-L-L faults occurring at different locations along the paths of Terminals 1-2,2-3,3-4,4-5,5-6,6-1. To evaluate the performance of the proposed scheme for each type of fault, at a particular location, the fault inception angle was varied from 20 0 to The fault inception angle has a considerable effect on the current samples of all the phases and also on the outputs of post fault signals obtained from wavelet transform. The waves are periodic and it is sufficient to consider the fault inception angle in the range from 20 0 to for all types of faults. Influence of fault resistance was also considered with value of 5 ohms. Synchronized sampling of three phase currents at all terminals was carried out and the detail D1 coefficients were used for detection and classification of the type of fault. The performance of the scheme in detection and classification of the faults was evaluated. In all the cases studied, the scheme was able to detect the faults.. The simulations show that the fault inception angle has a considerable effect on the phase current samples and therefore on Wavelet Transform output of post-fault signals. The impedance measurement involves both the voltage and current signals to be analyzed and takes more time to measure faulty section impedance thereby the relay operation. The faulty phases will be identified with the analysis of only current signals and fault clearing time will be less than half cycle The protection scheme involves the comparison of all phases with threshold value(th), where the faulty phase(s) indices have higher value compared with the threshold value and the healthy phase indices are less than the threshold value and therefore the digital values of the fault indices play a significant role in the protection of six- terminal transmission system. 102

4 Fig.1 Single line model of the proposed six-terminal transmission system Fig.2 Simulink model of the proposed six-terminal transmission system 103

5 Fig.3. Three phase current waveforms at Terminal-1 for A-G fault at T-1 Fig.4. Three phase current waveforms at all six terminals for A-G fault at T-1 3.1: Variation of Fault Indices with Fault Inception Angle(FIA) at a constant Distance 50km and 90km from Terminal for A-G Fault Table-1: Variation of fault indices of all the three phases with fault inception angle FIA 50km 90km Th (Degrees) Ia Ib Ic Ia Ib Ic Fig.5 Variation of fault indices of all the three phases with fault inception angle for A-G fault occuring at distance of 50km and 90km from Terminal-1 104

6 3.1.2 for A-B Fault Table-2: Variation of fault indices of all the three phases with fault inception angle FIA(Degrees) 50km 90km Th Ia Ib Ic Ia Ib Ic Fig.6. Variation of fault indices of all the three phases with fault inception angle for A-B fault occuring at distance of 50km from Terminal For A-B-G Fault Table-3: Variation of fault indices of all the three phases with fault inception angle FIA(Degrees) 50km 90km Th Ia Ib Ic Ia Ib Ic Fig.7. Variation of fault indices of all the three phases with fault inception angle for A-B-G fault occuring at distance of 50km and 90km from Terminal-1 105

7 3.1.4 For A-B-C Fault Table-4: Variation of fault indices of all the three phases with fault inception angle at a constant distance FIA(Degees) 50km 90km Th Ia Ib Ic Ia Ib Ic Fig.8. Variation of fault indices of all the three phases with fault inception angle for A-B-C fault occuring at distance of 50km and 90km from Terminal Variation of Fault Indices with Distance at Fault Inception Angles and from Terminal For A-G Fault Table-5: Variation of fault indices of all the three phases with Distance. Distance,km Th Ia Ib Ic Ia Ib Ic Fig.9. Variation of fault indices of all the three phases with distance for A-G fault occuring at Fault Inception Angles and from Terminal

8 3.2.2 For A-B Fault Table-6: Variation of fault indices of all the three phases with Distance Th Distance,km Ia Ib Ic Ia Ib Ic Fig.10. Variation of fault indices of all the three phases with distance for A-B fault occuring at Fault Inception Angles and For A-B-G Fault Table-7: Variation of fault indices of all the three phases with Distance. Distance,km Th Ia Ib Ic Ia Ib Ic Fig.11. Variation of fault indices of all the three phases with distance for A-B-G fault occuring at Fault Inception Angle and

9 3.2.4 For A-B-C Fault Table-8: Variation of fault indices of all the three phases with Distance. Distance,km Th Ia Ib Ic Ia Ib Ic Fig.12. Variation of fault indices of all the three phases with distance for A-B-C fault occuring at Fault Inception Angle and FAULTY TERMINAL IDENTIFICATION 4.1 Variation of fault indices of three Phase currents at all terminals with fault inception angle at a distance of 50km from Terminal A-G fault at Terminal-1 Table-9: Variation of fault indices of all the three phases with fault inception angle FIA(Degrees) Ia at T1 Ia at T2 Ia at T3 Ia at T4 Ia at T5 Ia at T

10 Fig.13 Variation of fault indices of Phase-A at all terminals with fault inception angle for A-G fault occuring at distance of 50km from Terminal A-B fault at Terminal-1 Table-10 Variation of fault indices of PhasesA and B at all terminals with fault inception angle for A-B fault occuring at distance of 50km from Terminal-1 FIA(Degrees) Ia-b-g act T1 Ia-b-g at T2 Ia-b-g at T3 Ia-b-g at T4 Ia-b-g at T5 Ia-b-g at T Fig.14 Variation of fault indices of Phases A and B at all terminals with fault inception angle for A-B fault occuring at distance of 50km from Terminal-1 109

11 4.1.3 A-B-G fault at Terminal-1 International Journal of Exploration in Science and Technology Table-11 Variation of fault indices of Phases A and B at all terminals with fault inception angle for A -B-G fault occuring at a distance of 50km from Terminal-1 FIA(Degrees) Iab at T1 Iab at T2 Iab at T3 Iab at T4 Iab at T5 Iab at T Fig.15 Variation of fault indices of Phases A and B at all terminals with fault inception angle for A -B-G fault occuring at distance of 50km from Terminal A-B-C fault at Terminal-1 Table-12 Variation of fault indices of Phases A,B and C at all terminals with fault inception angle for A -B-C fault occuring at a distance of 50km from Terminal-1 Distance,km Ia T1 Ia at T2 Ia at T3 Ia at T4 Ia at T5 Ia at T Fig.16 Variation of fault indices of Phases A,B and C at all terminals with fault inception angle for A-B-C fault occuring at distance of 50km from Terminal-1 110

12 4.2. Variation of fault indices with distance at a constant fault inception angle of A-G Fault Table-13 Variation of fault indices of Phase A with distance at a constant fault inception angle of for A-G Fault at Terminal-1 FIA(Degrees) Ia-b-c act T1 Ia-b-c at T2 Ia-b-c at T3 Ia-b-c at T4 Ia-b-c at T5 Ia-b-c at T Fig.17 Variation of fault indices of Phase A with distance at a constant fault inception angle of for A-G Fault at Terminal A-B Fault Table-14 Variation of fault indices of Phases A and B with distance at a constant fault inception angle of for A-B Fault at Terminal-1 Distance,km Iab T1 Iab at T2 Iab at T3 Iab at T4 Iab at T5 Iab at T

13 Fig.18 Variation of fault indices of Phases A and B with distance at a constant fault inception angle of for A-B Fault at Terminal A-B-G Fault Table-15 Variation of fault indices of Phases A and B with distance at a constant fault inception angle of for A-B-G Fault at Terminal-1 Distance,km Iab T1 Iab at T2 Iab at T3 Iab at T4 Iab at T5 Iab at T Fig.19 Variation of fault indices of Phases A and B with distance at a constant fault inception angle of for A-B-G Fault at Terminal A-B-C Fault Table-16 Variation of fault indices of Phases A and B with distance at a constant fault inception angle of for A-B-C Fault at Terminal-1 Distance,km Iab T1 Iab at T2 Iab at T3 Iab at T4 Iab at T5 Iab at T

14 Fig.20 Variation of fault indices of Phases A,B and C with distance at a constant fault inception angle of for A-B-C Fault at Terminal-1 Figures 5-8 indicate the variation of fault indices of three phase currents with fault inception angle at a distance of 50 km from Terminal-1 for A-G A-B, A-B-G, A-B-C faults respectively, figures 9-12 indicate the variation of fault indices with distance at a fault inception angle for the same type of faults from the same Terminal. In both the cases, it was observed that the fault indices of faulty phase(s) was(were) large as compared with that of healthy phase(s) by comparing the fault index with a Threshold value (Th) which was taken as 400,which clearly show that the healthy phases lie below the threshold value and faulty phases lie above the threshold value, thus the faulty phase(s) was(were) determined. After the faulty phase(s) identification, the faulty terminal has been identified by considering the fault indices of that particular phase at all the six terminals and comparing all the index values and the index value of that particular phase(s) at the faulty terminal was(were) higher than the index values of the same phase(s) at the remaining terminals. Figures indicate the variation of fault index of phase A current at all the terminals for A-G fault, fault indices of phases A and B for A-B and A-B-G faults, fault indices of phases A,B,C for A-B-C fault with fault inception angle at a distance of 50km from Terminal-1, it was observed that the fault indices of that particular phase(s) at faulty terminal is(are) higher as compared with that of other terminals. Figures indicate the variation of fault index of current of phase A at all the terminals for A-G fault, fault indices of phases A and B for A-B and A-B-G faults, fault indices of phases A,B,C for A-B-C fault with distance at a fault inception angle of 100 0, it was observed that the fault indices of that particular phase(s) at faulty terminal is(are) higher as compared with that of other terminals. Thus the faulty phase(s) was(were) determined by comparing the fault index with a threshold value (Th) and the faulty terminal was identified by comparing the fault index of the same phase(s) at all the terminals and the faulty terminal fault index will be higher than the other terminals for the same phase which identifies the faulty terminal. The proposed scheme was found to be effective in detecting all types of faults at any distance along the length of the terminal and at any fault inception angle in the half cycle period. 5. ESTIMATION OF FAULT LOCATION USING FUZZY INFERENCE SYSTEM After the detection and classification of the faults and identification of faulty terminal, the location of the fault has been estimated with Fuzzy logic. For this purpose, the three phase current signals at the corresponding terminal have been decomposed with Bior1.5 mother wavelet and the fault indices obtained were the inputs to the Fuzzy inference system. The standard Fuzzy membership function taken was triangular and the output function taken was distance(d). Each input variable is quantized into the linguistic variables such as Very 113

15 Low(VL), Low(L), High(H), Very High(VH) for the universe of discourse spanning from 0 to 1 and for the output variable, it is the length of the transmission system divided into four zones D1,D2,D3,D4.The inputs were combined together based on the expert opinion and the possible rules were framed out and the output was defuzzified to get the crisp value of D. Simulations were carried out considering variations in fault location and fault inception angle for different types of faults. Figure 21: Fuzzy Inference System for fault location estimation Figure 21: Input variable (fault indices) for fault location from terminal-1 Figure 22: Output variable for location estimation of distance Table 17: Fuzzy based fault location estimation analysis from Terminal-1 of the transmission system Fault type A-G A-B A-B-G A-B-C Actual distance from terminal-1 (KM) Fuzzy distance, (KM) Transmission system Error distance % error CONCLUSION Depending upon the type of fault the conventional distance relay is likely to over reach or under reach and can be rectified by wavelet based multi-resolution analysis approach that is applied for effective fault detection, classification and faulty terminal identification in multi-terminal transmission lines. The above algorithm has been implemented for all types of faults with variations in fault inception angle, 114

16 fault impedance and fault distance at all terminals. The results indicate the accuracy in fault detection, classification and faulty terminal identification, and fault location estimation. The effectiveness of the scheme has to be tested when the transmission lines are compensated by FACTS devices. The proposed protection scheme was found to be fast, reliable and accurate for various types of faults on six terminal transmission lines at different locations and with various inception angles. REFERENCES (Journal Article) [1] A. H. Osman and O. P. Malik, Transmission line distance protection based on wavelet transform, IEEE Trans. Power Del., vol. 19, no. 2, pp , Apr [2] Feng Liang, B.Jeyasurya, Transmission Line Distance Protection Using Wavelet Transform Algorithm. IEEE Trans. on Power Delivery, Vol.19, No.2 April 2004, pp [3] Ebha Koley, Khushaboo Verma and Subhojit Ghosh, An improved fault detection classification and location scheme based on wavelet transform and artificial neural network for six phase transmission line using single end data only SpringerPlus (2015) 4:551, DOI /s SCI Index Journal. [4] M. J. Reddy and D. K. Mohanta, A wavelet-fuzzy combined approach for classification and location of transmission line faults, International Journal of Electrical Power and Energy Systems, vol. 29, no. 9, pp , [5] Quanyuan jiang, BO Wang, Xingpeng li An Efficient PMU based fault location technique for multi transmission lines, IEEE Transactions on Power Delivery, Vol.29, NO.04 August [6].R.K.Aggarwal,D.V.Coury,A.T.Johns and A.Kalam A practical approach to accurate fault location on extra high voltage teed feeders. IEEE transactions on power delivery, vol.8,july 1993, pp [7].P.G.Mclarenn, S.Rajendra Travelling wave technique applied to the protection of teed circuits: Principle of travelling wave technique.ieee transactions on power apparatus and systems, vol. PAS- 104,No.12 December 1985,pp [8].Bo, Z.Q A new non-communication protection technique for transmission lines, IEEE Trans. Power Deliv., 1998, 13(4), pp [9]. Bhalija,B.,and Maheswari,R.P, High resistance faults on two terminal parallel transmission line; analysis, simulation studies, and an adaptive distance relaying scheme IEEE Trans. Power Deliv., 2007, 22, (2), pp [10].Brahma, S.M and Girgis, A..A. Fault location on a transmission line using synchronized voltage measurements, IEEE Trans.Power Deliv.,2009,(4) [11].Lyonette,D.R.M.,Bo,Z.Q.,Weller, G., and Jiang,G A new directional comparison technique for the protection of teed transmission circuits. Power Eng. Soc.Winter Meeting, IEEE., January 2000, vol.3, pp [12] Bhalija,B., and Maheswari, R.P New differential protection scheme for tapped transmission line. IET Gener.Transm. Distrib., 2008, 2,(2),pp

17 [13].T.Nagasawa,M.Abe,N.Otsuzuki,.Emura,Y.Jikihara,and M.Takeuchi, Development of a new fault location algorithm for multi terminal two parallel transmission lines, IEEE Trans.Power Deliv., vol-7. 3, pp , July [14].T.Funabashi, H.Otoguro, Y.Mizuma, L.Dube, and A.Ametani, Digital fault location algorithm for parallel double- circuit multi terminal transmission lines., IEEE Trans.Power Deliv., vol.15. 2, pp April-2000 [15].Prarthana Warlyani, Anamika Jain,A.S.Thoke, R.N.Patel Fault classification and faulty section identification in Teed transmission circuits using ANN. International Journal of Computer and Electrical Engineering, Vol.3,No.6 Dec-2012 [16] Adel A. Elbaset Takashi Hiyama, Fault Detection and Classification in Transmission Lines Using ANFIS IEEJ Trans. IA, Vol.129, No.7, 2009 [17] G.Ravikumar, M.Vaidehi, R. Kameswara Rao Neuro-wavelet approach for transient current based multi terminal transmission system protection scheme Journal of Signal Processing Systems Vol. 1, Issue 1, Dec 2015, [18]S.Chandra Shekar, G.Ravi Kumar, SVNL Lalitha Wavelet Based Multi-Terminal Transmission Line Protection with MicroGrid WSEAS Transactions on Power Systems, Volume 11, 2016,pp Authors profile J.Uday Bhaskar J.Uday Bhaskar received B.E from GITAM, Andhra University, Visakhapatnam. M.Tech from JNTU Kakinada. Presently pursuing Ph.D from JNTU Kakinada. Research areas include Multiterminal Transmission line protection, FACTS Devices. G.Ravi Kumar G.Ravi Kumar received B.Tech, M.tech and Ph.D from JNTU college of Engineering, Kakinada. Presently working as professor in the Department of Electrical and Electronics Engineering in KITS, Warangal. His areas of interest include FACTS Devices, Power System Security. S.S Tulasi Ram S.S Tulasi Ram received B.Tech, M.tech and Ph.D from JNTU college of Engineering, Kakinada. Presently working as professor in the Department of Electrical and Electronics Engineering in College of Engineering, JNTUH, Hyderabad. His areas of interest include High Voltage Engineering, Power System Analysis and Control. 116