Recruiting & HR

A new unsynchronized fault-location technique for three-terminal lines

Description
A new unsynchronized fault-location technique for three-terminal lines
Published
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Unsynchronised fault-location technique forthree-terminal lines ISSN 1751-8687Received on 16th October 2014Revised on 9th June 2015Accepted on 24th June 2015doi:10.1049/iet-gtd.2015.0062www.ietdl.org Behnam Mahamedi  1  ✉ , Majid Sanaye-Pasand  2  , Sadegh Azizi  2  , Jian Guo Zhu  1 1 School of Electrical, Mechanical and Mechatronic Systems, University of Technology, PO Box 123, Broadway, Sydney, NSW 2007,Australia  2  School of Electrical and Computer Engineering, University of Tehran, North Kargar Avenue, Tehran, Islamic Republic of Iran  ✉  E-mail: behnam.mahaamedi@gmail.com  Abstract:  This study describes a new fault-location technique using negative-sequence voltage for three-terminal lines.The ratios between the negative-sequence voltage magnitudes measured at each terminal are utilised to firstdetermine the faulted section and then to estimate the exact fault location within the section. Since the current data isnot deployed, the influence of inherent errors of current transformers can be avoided. The proposed method canaccurately locate the unbalanced faults, that is, single-phase-to-ground, double-phase-to-ground, and phase-to-phasefaults, regardless of the fault resistance and pre-fault conditions and without any need to identify the fault type. Themethod requires only the negative-sequence reactance behind each terminal which can be estimated by the short-circuit analysis with an acceptable accuracy. Reliability and practicality of the proposed method make it an attractiveoption to include in numerical protective relays. Simulation experiments with different fault cases reveal the capabilityof the proposed method. 1 Introduction Three-terminal lines may be a reasonable option for supplyingdemands for various economical and technical reasons [1]. Insome cases, it is economical to serve a growing load with a tapped line at least for a limited period of time [2]. This con fi gurationmay last until the load growth justi fi es installing a new substation.A three-terminal line could even be the only feasible solution dueto the dif  fi culties in obtaining a right of way for new lines.However, the protection of such lines is not as simple as that of the conventional two-terminal ones. They usually undergo a seriesof problems owing to the intermediate infeed/outfeed from thethird terminal, difference in line length to the tee point, and dissimilarity of equivalent source impedance behind the lineterminals. In this sense, fault location on three-terminal lines becomes an issue due to the possibility of multiple sources feedinga fault.When a short-circuit fault occurs on a three-terminal line, it takesmuch time for maintenance crews to locate the exact fault point,since the faulted section of the line should be distinguished at  fi rst. This may not only increase the maintenance cost, but alsoreduce the system reliability [3]. On the other hand, most of thereported research works in the fault-location (FL) area have dealt with problems on two-terminal lines [4 – 6]. Therefore, aneffective algorithm is needed for fault location onthree-terminal lines.The proposed method in [7] used the synchronised voltage and current data at two terminals to locate a fault on three-terminallines. It is assumed that the fault impedance is purely resistive and the fault type is known. A technique based on synchronised datawas suggested in [8] requiring three-phase current and voltage phasors at all line terminals. Although the proposed techniqueconsiders fault location during dynamic conditions of power systems, it involves an iterative solving process, which may result in a divergent solution though.Reference [9] utilised negative-sequence circuits to develop an FLtechnique for two- and three-terminal lines. This method requires thenegative-sequence components of both voltage and current. In [10],de Pereira and Zanetta described a technique which needs the fault type and faulted section to be known at   fi rst though it can be used for multi-tapped lines. A method has been proposed in [11] usingsuperimposed positive-sequence voltages, which is sensitive to themeasurement errors such that it may fail to identify faulted sectiondue to the errors.Fault locators exploiting current data are all prone to large errorsdue to the magnetic saturation of current transformers (CTs).Moreover, fault locators based on current phasors are very slowand sometimes fail to accurately locate faults if a decaying DCremoval module has not been embedded in the relay algorithm.To overcome the aforementioned problem, transient-based methods can be an alternative. A transient-based FL method for three-terminal transmission lines using the wavelet transform and neural network is presented in [12]. In [13], another  transient-based FL algorithm based on WT is presented, whichcan locate faults using data of one terminal out of three terminals.In [14], Ahmadimanesh and Shahrtash presented a different method of FL on three-terminal lines using time-to-timetransform. Despite satisfactory results achieved by transient-based methods, the required high sampling frequency is a mainchallenging point which requires appropriate solution. Limited  band of instrument transformers is another obstacle against usingthese algorithms.This paper presents a new FL method for three-terminal linesusing only the negative-sequence voltage magnitudes. These dataalong with the equivalent source impedance behind the lineterminals, which can be provided by load dispatch centres [15],are used to identify reliably the faulted section and further to pinpoint the fault location. Since only magnitudes are required,there is no need for synchronised measurements. The accuracytogether with practicality of the proposed method makes it areliable option to implement in the protective relays as an FLfunction for three-terminal lines.The rest of this paper is organised as follows. Section 2 explainsthe basic principles, and presents the proposed method as well asthe numerical implementation of the FL algorithm. In Section 3, anumber of simulation studies are conducted on a test system to IET Generation, Transmission & Distribution Research Article  IET Gener. Transm. Distrib. , 2015, Vol. 9, Iss. 15, pp. 2099 – 21072099 & The Institution of Engineering and Technology 2015  investigate the effectiveness and accuracy of the proposed method.Conclusions are drawn in Section 4. 2 Proposed method The majority of short-circuit faults, occurring on power lines areunbalanced. This  fi gure can reach as far as 98% based on theexperiences observed in different utilities across the world [16].In addition, three-phase balanced faults are experienced duringsystem repairs when the operators have not removed systemground wires before re-energising the system. As such, thelocation of three-phase faults is known already. Thus, the proposed method aims at locating unbalanced faults onthree-terminal lines, namely, single-phase-to-ground, double- phase-to-ground, and phase-to-phase faults. In this section, the basic principles of the proposed FL method are explained and then the algorithm for locating a fault on a three-terminalline is elaborated. 2.1 Basic principles and derivations  Consider a common application of three-terminal lines shown inFig. 1, where the tapped line ends up to a transformer feeding aload [2]. It is to be noted that in presence of any source at terminal3, all derivations are still valid.The extra links between line terminals are not always present,especially in power networks that are not highly interconnected [17]. Moreover, in the lower-voltage levels (voltage <230 kV)where three-terminal lines are mostly used, those links betweendifferent terminals can be safely ignored as their equal impedancesare much more than the line impedances. In other words, it isexpected that the extra links are weak enough to be neglected in aT-shape line.When an unbalanced fault occurs in Section 1 (line  L 1 ) at distance m 1  (per-unit) from relay  R 1 , the negative-sequence circuit can then be shown in Fig. 2. The reasons why the negative-sequence circuit has been chosen over zero-sequence one for developing the proposed method have been described in [18] for the interested readers.Referring to Fig. 2, one can derive the following equations: V  2  R 1 =  V  2  F   ·  Z  2 S  1  Z  2 S  1 +  m 1  Z  2  L 1 (1) V  2  R 2 =  V  2  F   ·  Z  2 2  Z  2 3   Z  2 2  Z  2 3   +  (1  −  m 1 )  Z  2  L 1 ·  Z  2 S  2  Z  2 S  2 +  Z  2  L 2 (2) V  2  R 3 =  V  2  F   ·  Z  2 2  Z  2 3   Z  2 2  Z  2 3   +  (1  −  m 1 )  Z  2  L 1 ·  Z  2 S  3  Z  2 S  3 +  Z  2  L 3 (3)where  Z  2 2  =  Z  2 S  2 +  Z  2  L 2 (4)  Z  2 3  =  Z  2 S  3 +  Z  2  L 3 (5)  Z  2 2  Z  2 3   =  Z  2 S  2 +  Z  2  L 2    ·  Z  2 S  3 +  Z  2  L 3    Z  2 S  2 +  Z  2  L 2 +  Z  2 S  3 +  Z  2  L 3 =  R 23  +  j  X  23  (6)and   V  2  Ri  is the negative-sequence voltage measured at relay locationin Section  i .Dividing (1) by (2) results in V  2  R 1 V  2  R 2 =  Z  2 S  1  Z  2 S  2 ·  Z  2 S  2 +  Z  2  L 2  Z  2 2  Z  2 3  ·  Z  2 2  Z  2 3   +  1  −  m 1    Z  2  L 1  Z  2 S  1 +  m 1  Z  2  L 1 (7)Rearranging (7) yields V  2  R 1 V  2  R 2 ·  Z  2 S  2  Z  2 S  1 ·  Z  2 2  Z  2 3   Z  2 S  2 +  Z  2  L 2 =  Z  2 2  Z  2 3   +  1  −  m 1    Z  2  L 1  Z  2 S  1 +  m 1  Z  2  L 1 (8)Assuming  A 12  = V  2  R 1  V  2  R 2  ·  Z  2 S  2   Z  2 S  1  ·  Z  2 2  Z  2 3   Z  2 S  2 +  Z  2  L 2  =  Z  2 2  Z  2 3   +  1  −  m 1    Z  2  L 1   Z  2 S  1 +  m 1  Z  2  L 1  (9)one obtains m 1  =− b 12  +   b 122 −  4 a 12 c 12   2 a 12 (10)where a 12  =  A 122 −  1    ·  R 2  L 1 2 +  X  2  L 1 2    (11) b 12  =  2  A 212  ·  R 2 S  1 ·  R 2  L 1  +  X  2 S  1 ·  X  2  L 1   +  2  R 2 2  L 1  +  X  2 2  L 1  +  R 23  ·  R 2  L 1  +  X  23  ·  X  2  L 1    (12) c 12  =  A 212  ·  R 2 2 S  1 +  X  2 2 S  1   −  R 223  +  X  223  +  R 2 2  L 1 +  X  2 2  L 1 +  2  R 23  ·  R 2  L 1 +  2  X  23  ·  X  2  L 1   (13) Fig. 1  Three-terminal line protected by protective relays, R 1  , R 2  , and R 3 Fig. 2  Negative-sequence circuit for an unbalanced fault in Section 1 at distance m 1  from relay R 1 IET Gener. Transm. Distrib. , 2015, Vol. 9, Iss. 15, pp. 2099 – 21072100  & The Institution of Engineering and Technology 2015  In (11) – (13),  R 2 L1  and   X  2 L1  represent for the negative-sequenceresistance and reactance of the line, respectively. The similar de fi nition holds for resistive and reactive parts of source impedances.Mathematically, to the second-order equation derived from (9),two solutions exist, but   m 1  in (10) is the physically acceptableone, whereas the other is not acceptable since it is out of the rangeof [0, 1]. The fault location in Section 1 (line  L 1 ) which isobtained by using  A 12  is denoted by  m 1.12  henceforth.On the other hand, dividing (1) by (3) leads to V  2  R 1 V  2  R 3 =  Z  2 S  1  Z  2 S  3 ·  Z  2 S  3 +  Z  2  L 3  Z  2 2  Z  2 3  ·  Z  2 2  Z  2 3   +  (1  −  m 1 )  Z  2  L 1  Z  2 S  1 +  m 1  Z  2  L 1 (14)Assuming  A 13  = V  2  R 1  V  2  R 3  ·  Z  2 S  3   Z  2 S  1  ·  Z  2 2  Z  2 3   Z  2 S  3 +  Z  2  L 3  =  Z  2 2  Z  2 3   +  (1  −  m 1 )  Z  2  L 1   Z  2 S  1 +  m 1  Z  2  L 1  (15)one obtains m 1  =− b 13  +   b 213  −  4 a 13 c 13   2 a 13 (16)where a 13  =  A 213  −  1    ·  R 2 2  L 1 +  X  2 2  L 1    (17) b 13  =  2  A 213  ·  R 2 S  1 ·  R 2  L 1 +  X  2 S  1 ·  X  2  L 1   +  2  R 2 2  L 1 +  X  2 2  L 1 +  R 23  ·  R 2  L 1 +  X  23  ·  X  2  L 1    (18) c 13  =  A 213  ·  R 2 2 S  1 +  X  2 2 S  1   −  R 223  +  X  223  +  R 2 2  L 1 +  X  2 2  L 1 +  2  R 23  ·  R 2  L 1 +  2  X  23  ·  X  2  L 1   (19)The fault location in Section 1 obtained by using  A 13  is known as m 1.13  hereafter.Both  m 1.12  and   m 1.13  pinpoint the fault location in Section 1. If anunbalanced fault occurs in Section 1,  m 1.12  and   m 1.13  would be equaland reveal the fault location. Thus,  m 1.12  and   m 1.13  can be considered as indicators for fault location in Section 1. When both indicators fallin the range of [0, 1], the fault is located in Section 1 and their valuesgive out the exact fault distance from relay  R 1 . Now, assume an unbalanced fault to occur in Section 2 at distance m 2  (per unit (pu)) from relay  R 2 . In this case, the respectivenegative-sequence circuit is shown in Fig. 3.On the basis of Fig. 3 and using the same procedure pursued for locating faults in Section 1, the fault location in Section 2 can beobtained as m 2  =− b 21  +   b 221  −  4 a 21 c 21   2 a 21 (20)where a 21  =  A 221  −  1    ·  R 2 2  L 2 +  X  2 2  L 2    (21) b 21  =  2  A 221  ·  R 2 S  2 ·  R 2  L 2 +  X  2 S  2 ·  X  2  L 2   +  2  R 2 2  L 2 +  X  2 2  L 2 +  R 13  ·  R 2  L 2 +  X  13  ·  X  2  L 2   (22) c 21  =  A 221  ·  R 2 2 S  2 +  X  2 2 S  2   −  R 213  +  X  213  +  R 2 2  L 2 +  X  2 2  L 2 +  2  R 13  ·  R 2  L 2 +  2  X  13  ·  X  2  L 2   (23)and   A 21  = V  2  R 2  V  2  R 1  ·  Z  2 S  1   Z  2 S  2  ·  Z  2 1  Z  2 3   Z  2 S  1 +  Z  2  L 1  (24)In (24)  Z  2 1  =  Z  2 S  1 +  Z  2  L 1 (25)  Z  2 1  Z  2 3   =  Z  2 S  1 +  Z  2  L 1    ·  Z2 S  3 +  Z2  L 3    Z  2 S  1 +  Z  2  L 1 +  Z  2 S  3 +  Z  2  L 3 =  R 13  +  j  X  13  (26)The fault location in Section 2 which is obtained by using  A 21  isknown as  m 2.21  henceforth.As in case of Section 1, fault location in Section 2 can be obtained in another way as m 2  =− b 23  +   b 232 −  4 a 23 c 23   2 a 23 (27)where a 23  =  A 223  −  1    ·  R 2 2  L 2 +  X  2 2  L 2    (28) b 23  =  2  A 223  ·  R 2 S  2 ·  R 2  L 2 +  X  2 S  2 ·  X  2  L 2   +  2  R 2 2  L 2 +  X  2 2  L 2 +  R 13  ·  R 2  L 2 +  X  13  ·  X  2  L 2   (29) c 23  =  A 223  ·  R 2 2 S  2 +  X  2 2 S  2   −  R 213  +  X  213  +  R 2 2  L 2 +  X  2 2  L 2 +  2  R 13  ·  R 2  L 2 +  2  X  13  ·  X  2  L 2   (30)  A 23  = V  2  R 2  V  2  R 3  ·  Z  2 S  3   Z  2 S  2  ·  Z  2 1  Z  2 3   Z  2 S  3 +  Z  2  L 3  (31) Fig. 3  Negative-sequence circuit for an unbalanced fault in Section 2 at distance m 2  from relay R 2 IET Gener. Transm. Distrib. , 2015, Vol. 9, Iss. 15, pp. 2099 – 21072101 & The Institution of Engineering and Technology 2015  The fault location in Section 2 which is obtained by using  A 23  isknown as  m 2.23  from now on.It can then be concluded when an unbalanced fault occurs inSection 2,  m 2.21  and   m 2.23  would be equal and reveal the fault location. Thus,  m 2.21  and   m 2.23  can be considered as indicators for fault location in Section 2. When both indicators fall in the rangeof [0, 1], the fault is located in Section 2 and their values indicatethe exact location from relay  R 2 .At last, consider an unbalanced fault occurring in Section 3 at distance  m 3  (pu) from relay  R 3 . In this case, the negative-sequencecircuit is shown in Fig. 4.On the basis of Fig. 4 and following the same analysis carried out for Section 1, the fault location in Section 3 can be obtained by m 3  =− b 31  +   b 231  −  4 a 31 c 31   2 a 31 (32)where a 31  =  A 231  −  1    ·  R 2 2  L 3 +  X  2 2  L 3    (33) b 31  =  2  A 231  ·  R 2 S  3 ·  R 2  L 3 +  X  2 S  3 ·  X  2  L 3   +  2  R 2 2  L 3 +  X  2 2  L 3 +  R 12  ·  R 2  L 3 +  X  12  ·  X  2  L 3   (34) c 31  =  A 231  ·  R 2 2 S  3 +  X  2 2 S  3   −  R 212  +  X  212  +  R 2 2  L 3 +  X  2 2  L 3 +  2  R 12  ·  R 2  L 3 +  2  X  12  ·  X  2  L 3   (35)and   A 31  = V  2  R 3  V  2  R 1  ·  Z  2 S  1   Z  2 S  3  ·  Z  2 1  Z  2 2   Z  2 S  1 +  Z  2  L 1  (36)In (36),  Z  2 1 ||  Z  2 2  is given by  Z  2 1  Z  2 2   =  Z  2 S  1 +  Z  2  L 1    ·  Z  2 S  2 +  Z  2  L 2    Z  2 S  1 +  Z  2  L 1 +  Z  2 S  2 +  Z  2  L 2 =  R 12  +  j  X  12  (37)As a result, fault location in Section 3 is determined by using  A 31  and denoted by  m 3.31  henceforth.On the other hand, a fault can also be located in Section 3 by m 3  =− b 32  +   b 232  −  4 a 32 c 32   2 a 32 (38)where a 32  =  A 232  −  1    ·  R 2 2  L 3 +  X  2 2  L 3    (39) b 32  =  2  A 232  ·  R 2 S  3 ·  R 2  L 3 +  X  2 S  3 ·  X  2  L 3   +  2  R 2 2  L 3 +  X  2 2  L 3 +  R 12  ·  R 2  L 3 +  X  12  ·  X  2  L 3   (40) c 32  =  A 232  ·  R 2 2 S  3 +  X  2 2 S  3   −  R 212  +  X  212  +  R 2 2  L 3 +  X  2 2  L 3 +  2  R 12  ·  R 2  L 3 +  2  X  12  ·  X  2  L 3   (41)  A 32  = V  2  R 3  V  2  R 2  ·  Z  2 S  2   Z  2 S  3  ·  Z  2 1  Z  2 2   Z  2 S  2 +  Z  2  L 2  (42)The fault location in Section 3 which is obtained by using  A 32  isknown as  m 3.32  from now on.As similar for Sections 1 and 2,  m 3.31  and   m 3.32  indicate the fault location in Section 3 and can be utilised as indicators for fault location in Section 3. When both indicators fall in the range of [0, 1], the fault is located in Section 3 and their values yield upthe exact fault location from relay  R 3 . 2.2 Notification on indicators  As already explained, the two indicators of a section will be of thesame value and fall in the range of [0, 1] when a fault occurs inthat section. However, to complete the proposed FL technique, thevalue of the indicators of other two sections should be examined as well. To do this, assume a fault to occur in Section 1. It isdesired to obtain the variations of   m 2.21 ,  m 2.23  when fault locationin Section 1 varies along line  L 1 .It is claimed that   m 2.21  will be >1 and   m 2.23  will remain 1 for anyfault located in Section 1. To prove the  fi rst claim, we need to solve m 2 . 21  = − b 21  +   b 221  −  4 a 21 c 21   2 a 21 . 1 (43)which results in a 21  +  b 21  +  c 21 , 0 (44)that is  A 21 ,  Z  2 1  Z  2 3   Z  2 S  2 +  Z  2  L 2  (45)or   A 21  = V  2  R 2  V  2  R 1  ·  Z  2 S  1   Z  2 S  2  ·  Z  2 1  Z  2 3   Z  2 S  1 +  Z  2  L 1  ,  Z  2 1  Z  2 3   Z  2 S  2 +  Z  2  L 2  (46)Rearranging (46) yields V  2  R 2  V  2  R 1  ,  Z  2 S  2   Z  2 S  1  ·  Z  2 S  1 +  Z  2  L 1   Z  2 S  2 +  Z  2  L 2  (47) Fig. 4  Negative-sequence circuit for an unbalanced fault in Section 3 at distance m 3  from relay R 3 IET Gener. Transm. Distrib. , 2015, Vol. 9, Iss. 15, pp. 2099 – 21072102  & The Institution of Engineering and Technology 2015  On the other hand, when a fault occurs in Section 1, we have V  2  R 2  V  2  R 1  =  Z  2 S  2   Z  2 S  1  ·  Z  2 2  Z  2 3   Z  2 2  Z  2 3  +  m 1  Z  2  L 1  ·  Z  2 S  1 +  m 1  Z  2  L 1   Z  2 S  2 +  Z  2  L 2  (48)Comparing (48) with the right side of (47), one can conclude that | V  2  R 2 |/| V  2  R 1 | will meet (43) for a fault occurring in Section 1.To prove the second claim, it is required that  m 2 . 23  =− b 23  +   b 223  −  4 a 23 c 23   2 a 23 =  1 (49)or  a 23  +  b 23  +  c 23  =  0 (50)resulting in  A 23  =  Z  2 1  Z  2 3   Z  2 S  2 +  Z  2  L 2  (51)Examining  A 23 , one can readily  fi nd that (51) is true when a fault occurs in Section 1.The same justi fi cation holds for   m 3.31  and   m 3.32 , that is,  m 3.31  will be >1 and   m 3.32  will be equal to 1 for any fault located inSection 1. Similarly, when an unbalanced fault occurs in Section 2, m 1.12  and   m 3.32  will be >1, whereas  m 1.13  and   m 3.31  will remain 1for any fault located in Section 2. The corresponding features arealso true for an unbalanced fault occurring in Section 3. 2.3 FL procedure  Fig. 5 elaborates the proposed FL algorithm using a  fl owchart diagram. Once a fault is detected on the three-terminal line byrelays action, the FL procedure will be triggered. On the basis of the negative-sequence voltage measured at each end,  A 12 ,  A 13 ,  A 21 ,  A 23 ,  A 31 , and   A 32  are calculated. Next, the correspondingindicators for each section are obtained using the formulas alreadyderived. If   m 1.12  and   m 1.13  are both in the range of [0, 1], Section 1is identi fi ed as faulty. Accordingly, both  m 1.12  and   m 1.13  are used to estimate the FL. If any of   m 1.12  and   m 1.13  is out of range [0, 1],the fault is in either Section 2 or 3. By checking the indicators of Sections 2 and 3 in the same manner, the faulted section and FLcan be determined.When the initial transients induced by faults, especially close-infaults, are signi fi cantly superimposed to the voltage signals, theindicators of no section may meet the condition. Therefore,re-calculation after one power cycle is suggested to override themisleading results caused by the transient signals. Moreover, it would be reasonable to average the two indicators,  m 1.12  and  m 1.13 , in order to locate the fault with higher accuracy. 3 Simulation experiments To evaluate the effectiveness and the accuracy of the proposed method, a three-terminal 132 kV line illustrated in Fig. 1 wasdeveloped in MATLAB/SIMULINK. The lengths of Section 1(  L 1 ) and that of Section 2 (  L 2 ) are assumed 100 and 50 km,respectively. The tapped line is 20 km long feeding a 132/20 kVtransformer. Other system parameters are provided in Appendix.The generated post-fault voltage signals are fed into the proposed algorithm as inputs. A large number of fault cases presented inTable 1 are studied to scrutinise the accuracy and reliability of the proposed method. Fault cases are considered at different locationswith different fault types and fault resistances. The fault inceptiontime is set to  t   =0.1 s for all cases. Moreover, thenegative-sequence voltage magnitude is calculated by using thediscrete Fourier transform with a 1 kHz sampling frequency. 3.1 Reliability and accuracy evaluation  Fig. 6 shows the indicators of Section 1 when faults occur in thissection as per Case 1. As shown, the indicators fall in the range of [0, 1] and track the actual FL. Note that in practice,  m 1.12  and  m 1.13  may be quite close to each other, but not precisely the same because of the measurement errors in addition to those raised byignoring line capacitances. On the other hand, one indicator of Sections 2 and 3 is out of the range as shown in Figs. 7 and  8, respectively. This indicates that the faulted section has beenreliably identi fi ed. Fig. 9 shows the estimation errors obtained by m 1.12 ,  m 1.13 , and the average of them. The estimation errors are Fig. 5  Proposed FL technique in a logical pattern Table 1  Different fault cases considered for evaluation studiesCase Fault type Fault resistance,  Ω  Actual FL, km1 Ph-G 10 Section 1: [5 … 95] every 1 km2 Ph-G 1003 Ph-Ph 5 Section 2: [2.5 … 47.5] every 1 km4 Ph-Ph 205 Ph-Ph-G 10 Section 3: [1 … 19] every 1 km6 Ph-Ph-G 100 IET Gener. Transm. Distrib. , 2015, Vol. 9, Iss. 15, pp. 2099 – 21072103 & The Institution of Engineering and Technology 2015
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks