EFFECTIVE FAULT ISOLATION METHODS TO IMPROVE 33kV NETWORK RELIABILITY A dissertation submitted to the Department of Electrical Engineering, University of Moratuwa in partial fulfillment of the requirements for the Degree of Master of Science S.M.M.S. WEERATUNA A IVERSITYS " ^ • • - • . ^ m . m i n i u r t( Wf Supervised by: Prof. H.Y.R. Perera ^ s n / / 6 2 A - 3 o r _ _ _ _ _ ^ Department of Electrical Engineering University of Moratuwa, Sri Lanka January 2009 University of Moratuwa 92963 T H 9*963 DECLARATION The work submitted in this dissertation is the result of my own investigation, except where otherwise stated. It has not already been accepted for any degree, and is also not being concurrently submitted for any other degree. S.M.M.S. Weeratunga I endorse the declaration by the candidate. Prof. H.Y.R. Perera i CONTENTS Page No. Declaration i Abstract iv-v Acknowledgement vi List of Tables vii List of Figures vi'i- 1. Introduction 1 1.1 Background 1 1.2 Reliability of33kV network of C.E.B 1 1.3 Motivation 2 2. Problem statement 3 2.1 Identification of the Problem 3 2.2 Objectives of the Study 3 2.3 Importance of the Study 3 3. Reliability analysis of 33kV network 4 3.1 Recloser event analysis 5 3.1.1 Re view of events 11 3.2 Type of faults 11 3.2.1 Transient fault 11 3.2.2 Permanent fault 3.3 Outage reduction method 13 3.3.1 Reduce faults in the network 13 3.3.2 Installation of fault indicators for remote indication 13 3.3.3 Remote operation of reclosers 14 3.3.4 Using line sectionalizing devices 14 3.4 Development model for reliability evaluation 15 4. An effective protection scheme reduce outage rate 19 4.1 Protection scheme of C.E.B 33kV network 19 4.1.1 Downstream protection scheme 21 4.2 Design new protection scheme 26 4.2.1 Design algorithm 26 4.2.2 Further mprovement 34 ii 4.3 Sample design 37 4.3.1 Calculation of shortcircut current 37 4.3.2 Calculation of eartfault current 39 4.3.3 Selection of recloser attempts to lockout 42 4.3.4 Selection of protection setting 43 4.3.5 Selection fuses for spurs at Kalumale feeder 51 4.4 Computer based model for new protection scheme 54 4.5 Advantages using new protection scheme 55 5. Conclusion and Recommendation 57 5.1 Improvement of reliability 57 5.2 Recommendation 58 5.2.1 Using event log of recloser 58 5.2.2 Reduce outage in the network 58 5.2.3 Isolating faulty line from the feeder 59 References 60 Appendix I, II & 111 iii ABSTRACT Improving reliability in power system is very important to the utilities as well as to the country as it is an important attribute of the supply quality and increases the end user satisfaction. The aim of this study is to investigate the fault identification and isolation techniques to improve reliability. This study focuses on the following; 1. Reliability status of 33kV network of C.E.B. 2. Type of faults in 33kV network and outage reduction methods. 3. Develop a mathematical model for reliability evaluation. 4. Analyze protection scheme of C.E.B 33kV network. 5. Design new protection scheme to reduce outages. The analyzed reliability indices of C.E.B are far below compared to internationally accepted levels. In this study reliability status of 33kV network is analyzed using event log data of reclosers and breakdown reports of Consumer Service Centers. The author is working at Region -04, C.E.B and two switching gantries of Region-04 are selected to analyze events one in wet zone and other in coastal zone. It can be seen that outage time and fault frequency are high during both monsoon periods, 86% of faults are line to earth, 67% of faults are of transient nature and the repair time in case of overcurrent fault is high due to poor workmanship in line connections. With this study predominant causes of faults in the network can be identified by using event log of recloser. The several methods for outage reduction are identified, installation of fault indicators for remote indication and remote operation of reclosers, using line sectionalize devices. iv The protection scheme of 33kV network is analyzed to find techniques to reduce outages. The downstream protective devices (fuse on Spurs) beyond the recloser do not distinguish permanent and transient faults. It can be seen nuisance fuse blowing in spurs as well as nuisance recloser lockout affecting all customers in feeder. A new protection scheme is designed to overcome the above problems. The developed method isolates unhealthy spurs from healthy sections, improves reliability and reduces maintenance cost and extent of unserved energy. ACKNOWLEDGEMENT First, I pay my sincere gratitude to Professor Ranjit Perera who encouraged and guided me to conduct this investigation and on perpetration of final dissertation. I also thank to Eng. J. Karunanayake who gave the valuable instructions during the study and valuable advice for perpetration of final dissertation. I would like to take this opportunity to extend my sincere thanks to Mr.U.K.W. Silva, Deputy General Manager (PHM-R4), Mr.L.TJ. Fernando, Deputy General Manger (Planning & Devlolopment-R4), Mr.S. Bogahawatta, Project Manager (Lightning Hambantota Project) Mr.S.T.S. Shantha (System Planning Engineer-Southern Province), Mr.K. Perera (System Planning Engineer-Rl), Mr.K. Wimalendra (Area Engineer-Kalutara), Mr.V.Ediriweera (Area Engineer-Matara), Mr.R. Thilakarathna (Electrical Superintendent - Agalawatta CSC), Mr.W.M. Premarathna (Electrical Superintendent-Dickella CSC) of Ceylon Electricity Board who gave their co- operation to conduct my investigation work successfully. It is a great pleasure to remember the kind co-operation extended by the colleagues in the post graduate programme, friends and specially my wife who helped me to continue the studies from start to end. vi List of tables Table number Description Table 1.1 Reliability statistics of 33kV C.E.B network Table 1.2 Reliability indices of 33kV C.E.B network Table 3.1 Rainfall data at Kithulgoda area Table 3.2 Rainfall data at Dickwella area Table 3.3 The occurrences of faults between lines and lines to earth Table 3.4 Average repair time of feeders Table 3.5 Recloser attempts of feeders Table 3.6 Types of faults at Kithulgoda Gantry Table 3.7 Types of faults at Dickwella Gantry Table 3.8 Assumed failure rates of feeder Table 3.9 Assumed consumers of feeder Table 3.10 Annual outage time when spurs connected directly to sections Table 3.11 Annual outage time when spurs connected with fuses and manual isolation Table 3.12 Annual outage time when spurs connected with fuses and sectionalizers Table 4.1 Connected fuse links at Kalumale feeder Table 4.2 Multiplying factor (k) for recloser curves Table 4.3 Connected fuse links in Kalumale feeder with new protection scheme Table 4.4 Reliability indices of Kalumale feeder Table 4.5 Annual requirement of fuse types of C.E.B vii List of figures Figure number Description Figure 3.1 Feeding arrangement of 33kV switching Gantry Figure 3.2 Outage period of feeders at Kithulgoda Gantry Figure 3.3 Reclosing of feeders at Kithulgoda Gantry Figure 3.4 Earthfault of feeder at Kihulgoda Gantry Figure 3.5 Overcurrent of feeders at Kithulgoda Gantry Figure 3.6 Outage period of feeders at Dickwella Gantry Figure 3.7 Reclosing of feeders at Dickwella Gantry Figure 3.8 Earthfault of feeder at Dickwella Gantry Figure 3.9 Overcurrent of feeders at Dickwella Gantry Figure 3.10 Spur lines connected directly to sections Figure 3.11 Spur lines connected across fuse with sections Figure 4.1 Protection scheme between GSS CB and Recloser Figure 4.2 Protection curves between GSS CB and Recloser Figure 4.3 Protection curves when spur line is protected by 10A fuse Figure 4.4 Protection curves when spur line is protected by 15A fuse Figure 4.5 Protection curves when spur line is protected by 20A fuse Figure 4.6 Protection curves when spur line is protected by 40A fuse Figure 4.7 Network diagram ' Figure 4.8 Algorithm for new protection scheme design Figure 4.9 Recloser to fuse miscoordination classification Figure 4.10 Recloser to fuse coordination curves viii Figure number Description Figure 4.11 Relay to fuse coordination curves Figure 4.12 Fuse to fuse miscoordination classification Figure 4.13 Fuse to fuse coordination of curves Figure 4.14 Relay to relay miscoordination classification Figure 4.15 Relay to relay coordination curves Figure 4.16 Sectionalizer device Figure 4.17 Sectionalizers in 33kV network Figure 4.18 Single line diagram of Kaumale feeder Figure 4.19 Overcurrent settings of protective devices with 50A fuse Figure 4.20 Earthfault settings of protective devices with 50A fuse Figure 4.21 Coordination of 50A (T type) fuse and 20A (K type) fuse Figure 4.22 The model for 33kV feeder to analyze protection coordination IX Chapter 1 Introduction 1.1. Background Electricity has become a fundamental commodity of modern life. Hence the quality of distribution of electricity has gained utmost concern. Distribution system consists of 33kV, l lkV networks and domestic service lines. Major portion of these networks is overhead lines and hence vulnerable to many a system fault. Faults in the distribution network lead to supply failure. Therefore, prevention of faults in the network is a primary concern in improving reliability of power supply. 1.2. Reliability status of 33kV network of C.E.B The reliability statistics shown in Table 1.1 give an idea of present status and how the network behaved during the past five years [14]. It is difficult to find data for 10 years and available data are considered in this study. The data comprise only the reported breakdowns and actual values are slightly higher than the reported values. Item 2003 2004 2005 2006 2007 No. of 33kV breakdowns 26345 26978 27678 29123 31200 Outage time due to breakdowns (hour) 54352 56789 58743 59765 61234 Table 1.1 - Reliability statistics of 33kV network The internationally accepted reliability indices of SAIDI, SAIFI and ASAI can be used to get a clear idea regarding the performance. SAIDI = System Average Interruption Duration Index (Minutes/year per connected consumer) SAIFI = System Average Interruption Frequency Index (Interruptions/year per connected consumer) ASAI = Average System Availability Index Page 1 of 60 Z (Outage duration) x (No.of customers affected) Total customers c.rr^r £ No.of customers interrupted • SAIFI = Total customers . _ . r Customer hours of service availability ASAI = Customer hours of service demand The reliability indices of 33kV network of whole country for 2003 to 2007 year are shown in Table 1.2. The indices are calculated without incorporating planned interruptions, the breakdowns which were not reported and the source side outage. Item 2003 2004 2005 2006 2007 Average SAIDI(min/yr) 1023 1167 1244 1456 1567 SAIFI (int/yr) 7.1 7.5 7.8 8.2 8.5 ASAI 0.99805 0.99778 0.99763 0.99723 0.99702 Table 1.2 - Reliability indices The reliability targets of the United States of America are as follows. These indices of C.E.B are far below. SAIDI = 95.9 min/yr SAIFI =1.18 int/yr ASAI = 0.99983 1.3. Motivation The outcome of this project will develop an effective fault isolating methodology and sets guide lines for proper planning of maintenance schedules. The understanding of reliability concepts and its proper application would lead to improve the consumer satisfaction. Also the reliability improvement will have a definite positive effect on industries which in turn improve economy of the country. Page 2 of 60 Chapter 2 Problem statement 2.1. Identification of the problem The analysis of reliability figures mentioned in Chapter-01 clearly shows the present condition of overhead network of C.E.B and annual outages without incorporating the planned interruptions and the breakdowns which were not reported. The poor condition of the network is highlighted in public media and the criticism leveled against C.E.B. 2.2. Objective of the study All 33kV switching gantries are equipped with reclosers and the only purpose they serve today is just to respond to faults. Although these modern reclosers are equipped with data storing facilities (event log), this facility is not used at present in the C.E.B network. This study explores the possibilities of using these data to; Improve reliability of network by reducing outage time in the network. 2.3. Importance of the study The average operation and maintenance cost of distribution sector of C.E.B is about LKR 1,800 million in year 2007 and the cost of unserved energy is about LKR 520,000 million in year 2007. The reliability status of 33kV network shows present condition of the system. Therefore it is essential to investigate the reliability status of the network and see how to improve it. This study will help to find a feasible solution for the above problems and the results obtained through this study could be used to develop proper maintenance schedules. It also identifies fault isolating methodologies. Page 3 of 60 Chapter 3 Reliability analysis of 33kV network 3.1. Recloser events analysis Modern reclosers are equipped with data storing facility (Event log) and this facility could be used to evaluate reliability of network and identify faults in the network. When the status of the control electronics or switchgear changes, events are generated which are recorded in an Event Log for display to the operator. More than 3000 events are recorded in recloser control unit. A typical event log of recloser data is as follows (This particular event list shows activation of earthfault). Date & Time 18/05/2008 10:29:22.39 18/05/2008 10:09:04.36 18/05/2008 10:08:59.41 18/05/2008 10:08:59.41 18/05/2008 10:08:59.41 18/05/2008 10:08:59.31 18/05/2008 10:08:59.31 18/05/2008 10:08:59.31 18/05/2008 10:08:59.31 18/05/2008 10:08:59.19 18/05/2008 10:08:57.24 18/05/2008 10:08:57.24 18/05/2008 10:08:57.19 18/05/2008 10:08:57.19 18/05/2008 10:08:57.19 18/05/2008 10:08:57.18 18/05/2008 10:08:56.81 18/05/2008 10:08:56.37 18/05/2008 10:08:56.37 18/05/2008 10:08:56.31 18/05/2008 10:08:56.31 18/05/2008 10:08:56.31 Event Panel Close Req Load Supply OFF E Max 492 Amp C Max 534 Amp Lockout Prot Trip 3 Earth Prot Trip Prot Group A Active Pickup Automatic Reclose E Max 499 Amp C Max 513 Amp Prot Trip 2 Earth Prot Trip Prot Group A Active Pickup Automatic Reclose E Max 498 Amp C Max 500 Amp Prot Trip 1 Earth Prot Trip Prot Group A Active Page 4 of 60 3.1.1. Review of events A switching gantry is fed by grid substation and 33kV reclosers are fitted to 33kV feeders. A typical switching arrangement is shown in Figure 3.1. Two Switching Gantries are selected for analysis of events. One is Kithulgoda gantry, WPS-1 Province -Region-04 close to wet zone and other is Dickwella gantry, Southern Province-Region -04 close to coastal zone of Ceylon Electricity Board. One year (2007) data are reviewed. The permanent faults (Lockouts) and temporary faults (reclosing) are identified. Then above faults are analyzed as to check whether they are Over Currents (OC) or Earth Faults (EF). 1. Wet zone-Kithulgoda Gantry, which is situated in Kalutara Area and fed by Mathugama Grid, has three outgoing feeders called Pellawatta, Sirikadura, Kalumale. 2. Coastal Zone-Dickwella Gantry, which is situated in Matara Area and fed by Matara Grid Substation, has two outgoing 33kV feeders called Dandeniya and Thangalle. 33kV Bus at GSS Feeder CB Switching Gantry Lynx line r v Isolator 33kV Bus X X RecioserX X 1 I * * Feeders T " Figure 3.1 - Feeding arrangement of 33kV Switching Gantry Page 5 of 60 Figure 3.2 shows bar chart presentation of outage period of each feeder at Kithulgoda Gantry and Table 3.1 represents rainfall data at Kithulgoda area for comparison [10]. Kalmale Pellawatta • Pellawatta q Sirikadura • Kalmale Figure 3.2- Outage period of feeders at Kithulgoda Gantry Figure 3.3 shows bar chart presentation of reclosing of each feeder at Kithulgoda Gantry. Kalumale Pellawatta I Pellawatta l Sirikadura D Kalumale Month 10 11 12 Figure 3.3- Reclosing of feeders at Kithulgoda Gantry Page 6 of 60 Figure 3.4 shows bar chart presentation of earthfaults of each feeder at Kithulgoda Gantry. Figure 3.4- Eathfault of feeders at Kithulgoda Gantry Figure 3.5 shows bar chart presentation of overcurrent of each feeder at Kithulgoda Gantry. • Pelawatta a Sirkadura • Kalumale Figure 3.5- Overcurrent of feeders at Kithulgoda Gantry Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Rain fall in mm 196 38 91 445 814 425 245 368 434 700 264 205 Table 3.1 - Rainfall data at Kithulgoda Gantry Page 7 of 60 Figure 3.6 shows bar chart presentation of outage period of each feeder at Dickwella Gantry and Table 3.2 represents rainfall data at Dickwella area for comparison [10]. 45.00 40.00 -C 35.00 ® 30.00 "a 25.00 » 20.00 © jf 15.00 O 10.00 • Dandeniya • Thangalla 7 8 Month 9 10 0 Dandeniya 11 1 2 Figure 3.6- Outage period of feeders at Dickwella Gantry Figure 3.7 shows bar chart presentation of reclosing of each feeder at Dickwella Gantrty. 3 4 5 6 Month ] 9 10 11 Dandeniya 12 I Dandeniya Thangalla Figure 3.7- Reclosing of feeders at Dickwella Gantry Page 8 of 60 Figure 3.8 shows bar chart presentation of eartfault of each feeder at Dickwella Gantry. • Dandeniya • Thangalla Figure 3.8- Earthfault of feeders at Dickwella Gantry Figure 3.9 shows bar chart presentation of overcurrent of each feeder at Dickwella Gantry. 8 7 Dandeniya I Dandeniya I Thangalla Figure 3.9- Overcurrent of feeders at Dickwella Gantry Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Rain fall in mm 21 23 26 153 67 111 108 456 569 370 79 36 Table 3.2 - Rainfall data at Dickwella area Page 9 of 60 * i^'stOS^^f , , . . , ... — - * t ' " t It can be seen that outage time and faults are high for two gantries during both monsoon periods. The occurrence of faults between lines (L) and line to earth (E) can be categorized as shown in Table 3.3. It can be seen most of the line to earth faults are caused by way leaves. The percentage of line to earth faults is 86%. LE LLE LLLE LL LLL Pellawatta 168 23 6 14 9 Sirikadura 45 5 2 7 4 Kalumale 96 10 4 5 3 Dandeniya 106 7 3 40 0 Thangalla 146 23 8 13 5 Table 3.3 - The occurrence of faults between lines and lines to earth The average repair time for overcurrent and earthfault are given in Table 3.4. More time is required for repairing overcurrent faults. Overcurrent fault causes to flow high current through conductors, thus overheating it and damaging the connection of poor workmanship. This eventually opens up the line. Gantry name Feeder Name Average repair time for OC fault(hour) Average repair time for EF(hour^ Dickwella Thangalla 3.20 2.74 Dandeniya 5.28 3.29 Kithulgoda Pellawatta 4.45 1.27 Sirikadura 5.23 4.81 Kalumale 2.08 2.17 Table 3.4 - Average repair time of feeders 9 2 9 G 3 Page 10 of 60 3.2. Type of faults Maximum service reliability is achieved when the distribution system is designed and operated to minimize the effect of any fault that may occur. Two types of faults are encountered on the above overhead network and can be categorized as transient and permanent. 3.2.1. Transient faults A transient fault is one that does not require corrective action to remove the fault from the system. If the arc can be cleared quickly, before burning into a permanent fault, the cause of the fault is gone. Since no equipment damage has occurred, the circuit can be re-energized immediately and supply reinstated. A transient fault would result from occurrences such as lightning, an arc caused by an animal or tree branch that then falls clear or the wind momentary blowing two conductors closer. The analyzed recloser attempts in each feeder are shown in Table 3.5. It can be seen that 67% of faults are in transient nature. The internationally proved level of transient faults in distribution system is between 60-80% [13]. 1st Attempt 2nd Attempt 3rd Attempt Lockout Pellawatta 170 17 33 33 Sirikadura 41 7 15 15 Kalumale 94 3 21 21 Dandeniya 110 10 36 36 Thangalla 149 9 37 37 Table 3.5 - Reloser attempts of feeders 3.2.2. Permanent faults A permanent fault is one in which permanent damage has resulted from the cause of the fault. A permanent fault usually requires some form of repair before power can be stored (Examples; a broken insulator, a broken conductor or an automobile knocking a pole down). It also includes faults that are initially transient in nature but result in permanent damage to the system. In such a situation, the line must be de-energized, a line crew brought to the site, and repairs made. The type of permanent faults occurred in the selected Gantries during 2007 year were grouped by comparing breakdown reports of Consumer Service Centre and the Page 11 of 60 correspondence event log of reclosers. The failure records obtained from the breakdown register is validated after the discussion with the Electrical Superintend and his work gang. The type of faults of Kithulgoda Gantry is shown in Table 3.6 and that of Dikwella Gantry in Table 3.7. Pellawatta Sirikadura Kalumale Total Jumper broken 5 3 3 11 Tree branches fallen 17 6 7 30 Tree fallen 2 1 2 5 Lightning 0 0 0 0 Animals 0 0 0 0 Vehicles 0 1 0 1 Intermittent 8 2 7 17 Bird 0 0 0 0 Earth broke 0 0 0 0 Conductor broken 1 2 0 3 Wind 0 0 0 0 Insulators 0 1 2 3 Table 3.6- Type of faults at Kithulgoda Gantry. Thangalla Dandeniya Total Jumper broken 5 3 8 Tree branches fallen 13 19 32 Tree fallen 2 1 3 Lightning 0 0 0 Animals 0 0 0 Vehicles 0 0 0 Intermittent 6 7 13 Bird 0 0 0 Earth broke 2 1 3 Conductor broken 4 3 7 Wind 0 0 0 Insulators 5 2 7 Table 3.7- Type of faults at Dickwella Gantry. Page 12 of 60 It can be seen from above tables tree branches fallen to line is common to all feeders and most of them occur in monsoon periods. More insulators have damaged in Tangalla Feeder, which is near to coastal zone. The spread of salt is mainly responsible for this type of faults. The non identified faults are categorized as intermittent faults. They are normally caused by way leaves or wind. 3.3. Outage reduction methods There are several methods to reduce outage time and discussed in next section. 3.3.1 Reduce faults in the network The two rainy seasons and fast growing thick vegetation surrounding medium voltage lines are responsible for many faults due to fall of trees and distance tree branches during windy and rainy days. These faults can be eliminated by carrying out way leaves programs before and after rainy seasons. Adequate clearance shall be maintained. The other suitable solution to minimize the effect will be the introduction of partially insulated Aluminum conductors for medium voltage line construction. The problem of salt formation over the line insulators and flashover gives enormous problems to system reliability. The insulation paint can be applied over the insulators or composite type insulators can be used. Re-stringing of conductors and adjustment of clearance of T-ofif should be carried out to overcome faults by way leaves and wind. Compression type H connectors should be introduced for jumper connection at tension points or angle points and spur points to overcome opening during heavy fault current. The compression type midspan joints should be used to connect damaged conductors. 3.3.2 Installation of remote fault indicators The power supply of healthy section can be reinstated with the least possible delay after isolating the faulty section. Fault indicators can be used to detect faulty section. Its fault indication can be used to send remote signal to control centre. Then it can be Page 13 of 60 given fast instruction to isolate faulty section. Then fault location of difficult areas will become easy and the outage time of such areas can be minimized. Recommended location of fault indicators; i. Easy to access line points for immediate monitoring ii. Next to the spur line iii. After switch (Air Break Switch or Load Break Switch) iv. Before and after the points that is difficult to reach such as forest or mountain. 3.3.3 Remote operation of reclosers Modern Reclosers used in C.E.B can be remote operated by control centre. All the fault events can be reviewed. Therefore operation time of recloser can be minimized by remote operation. Then outage time of network can be reduced. It can be seen more intermittent faults occur in two Gantries (Table 3.6 and 3.7). The recloser can be operated from control centre without physically attending to the site. Hence it reduces outage time of network and also traveling and labour cost. 3.3.4 Using line sectionalize devices For faults on the main feeder line, a line sectionalizing device (recloser or sectioalizer) can be used to divide the feeder into smaller line segments. All taps should have a protective device (fuses for small taps, a recloser or sectionalizer for large taps) where they connect to the main feeder. Even on very small taps, a fuse should be used. The justification is that although this type of tap fuse does not protect the tap; it protects the remainder of the distribution feeder from a fault on the tap. The extent of the outage can be minimized by limiting the size and length of the affected line. The shorter line segment minimizes the number of customers affected and minimizes the time required to patrol the line and locate the fault. Page 14 of 60 3.4. Development of a model for reliability evaluation A feeder can be represented by a single line diagram as follows. a) Spur lines (lateral) connected to feeder directly is shown in Figure 3.10. b) Spur lines protected by fuse link and sections separated by isolators are shown in Figure 3.11. Sec -1 Sec- 2 Sec- 3 Sec-4 Spur-A Spur-B Spur-C Spur- D Figure 3.10 Spur lines connected directly to sections Sec-1 Sec- 2 Sec-3 Sec-4 Isolator Fuse link Spur-A Spur-B Spur-C Spur-D Figure 3.11 - Spur lines connected across fuse with sections Page 15 of 60 The assumed failure rates of sections and spurs are as shown in Table 3.8. Item Length (Km) No. of failures/year (A) Repair time-r (hour) Section - 1 2 0.2 3 Section - 2 3 0.3 3 Section - 3 3 0.3 3 Section - 4 2 0.2 3 Spur - A 2 0.4 2 Spur - B 3 0.6 2 Spur - C 1 0.2 2 Spur - D 2 0.4 2 Table 3.8- Assumed failure rates of feeder The assumed customers are available is shown in Table 3.9. Load point No. of customers A 1000 B 2000 C 4000 D 3000 Table 3.9 - Assumed customers of feeder Effect of Spur lines (Lateral) connected to feeder directly The annual outage time (U) of section -1 = 0.2x3 = 0.6 hours The annual outage time of other sections and spur point also can be calculated as above and results are shown in Table 3.10. The total failures ()*) = Z 1 = 2 . 6 % Average Annual outage time (Us) = ]T A,rt = 6.2 hours i The customer orientated indices could be calculated using above formula and data, and the results are given bellow. SAIFI =2.6 interruption / customer yr SAIDI = 6.2 hours / customer yr ASAI =0.999292 Page 16 of 60 Customer A Customer B Customer C Customer D I r U X r U 1 r U 1 r Item (f/y) (h) (h) (f/y) (h) (h) (f/y) (h) (h) (f/y) (h) U(h) Section -1 0.2 3 0.6 0.2 3 0.6 0.2 3 0.6 0.2 3 0.6 Section - 2 0.3 3 0.9 0.3 3 0.9 0.3 3 0.9 0.3 3 0.9 Section - 3 0.3 3 0.9 0.3 3 0.9 0.3 3 0.9 0.3 3 0.9 Section - 4 0.2 3 0.6 0.2 3 0.6 0.2 3 0.6 0.2 3 0.6 Spur-A 0.4 2 0.8 0.4 2 0.8 0.4 2 0.8 0.4 2 0.8 Spur-B 0.6 2 1.2 0.6 2 1.2 0.6 2 1.2 0.6 2 1.2 Spur-C 0.2 2 0.4 0.2 2 0.4 0.2 2 0.4 0.2 2 0.4 Spur-D 0.4 2 0.8 0.4 2 0.8 0.4 2 0.8 0.4 2 0.8 Total 2.6 2.4 6.2 2.6 2.4 6.2 2.6 2.4 6.2 2.6 2.4 6.2 Table 3.10 - Annual outage time when spur connected directly to section Effect of Spur lines protected by fuse link The effects of protection failures and load transfer circuits are not considered in this evaluation due to non-availability of data. It is assumed feeder sections are isolated manually and average time for manual isolation (i) is 1 hour. When the fault is in feeder section it is sectionalized manually as far as it is not in the mainstream of the fault current. When faults occur in the spurs (lateral) the fuse links installed at the beginning operates and isolates the spur from the main feeder. This happens for all the spur points. The annual outage time of sections and spur points calculated are shown in Table 3.11. The improved reliability indices are evaluated bellow. SAIFI =1.4 interruption / customer yr SAIDI = 3.2 hours / customer yr ASAI =0.999635 Page 17 of 60 Customer A Customer B Customer C Customer D Item X fly (h) i(h) U (h) X ffy (h) i(h) U (h) X fly (h) i(h) U (h) X % (h) i(h) U (h) Section - 1 0.2 3 0.6 0.2 3 0.6 0.2 3 0.6 0.2 3 0.6 Section - 2 0.3 1 0.3 0.3 3 0.9 0.3 3 0.9 0.3 3 0.9 Section - 3 0.3 1 0.3 0.3 1 0.3 0.3 3 0.9 0.3 3 0.9 Section - 4 0.2 1 0.2 0.2 1 0.2 0.2 1 0.2 0.2 3 0.6 Spur-A 0.4 2 0.8 Spur-B 0.6 2 1.2 Spur-C 0.2 2 0.4 Spur-D 0.4 2 0.8 Total 1.4 1.6 2.2 1.6 2 3.2 1.2 2.5 3 1.4 2.7 3.8 Table 3.11 -Annual outage time when spur equipped with fuses and manual isolation The isolating time of isolator can be assumed as zero if isolator is replaced with sectionalizer. The annual outage time of sections and spur points calculated are shown in Table 3.12. Then improved reliability indices are evaluated below; SAIFI = 1.4 interruption / customer yr SAIDI = 2.9 hours / customer yr ASAI =0.999664 Customer A Customer B Customer C Customer D Item X f / y r(h) U(h) XVy r00 U(h) X f / y r(h) U(h) X f l y r(b) U(h) Section - 1 0.2 3 0.6 0.2 3 0.6 0.2 3 0.6 0.2 3 0.6 Section - 2 0.3 0.3 3 0.9 0.3 3 0.9 0.3 3 0.9 Section - 3 0.3 0.3 0.3 3 0.9 0.3 3 0.9 Section - 4 0.2 0.2 0.2 0.2 3 0.6 Spur-A 0.4 2 0.8 Spur-B 0.6 2 1.2 Spur-C 0.2 2 0.4 Spur-D 0.4 2 0.8 Total 1.4 1 1.4 1.6 1.7 2.7 1.2 2.3 2.8 1.4 2.7 3.8 Table 3.12 - Annual outage time when spur equipped with fuses and sectionalizers It can be seen from above analysis that reliability improvements can be achieved by proposed replacement of protection devices in spurs and sections. Page 18 of 60 g L . . SfL: VF-iriTY OF •r.r-"'V Jf Chapter 4 An effective protection scheme to reduce outage rate 4.1. Protection scheme of C.E.B 33kV network As a case study protection scheme developed from Mathugama gantry to Kithulgoda Gantry shown in Figure 4.1 is selected for analysis. The overcurrent and earthfault settings of circuit breakers at Mathugama GSS and reclosers at Kithulgoda gantry are shown in Figure 4.1. The protection curves drawn by SynerGEE Electric Program are shown in Figure 4.2 [4]. This protection scheme was issued by Distribution Planning Branch-C.E.B and is common for all gantries in Medium Voltage Network of C.E.B [2]. CT: 400: 1 o / c PMS=1 (400A) TMS=0.1 Inst=8 Curve: SI E/F PMS=0.1 40A) TMS=0.1 Inst=8 Curve: SI Phase Trip =200A (O/C) Time Mult. =0.1 Inst. Mult. = 8 Curve: VI Earth Trip = 20 A (E/F) Time Mult. = 0.1 Inst Mult = 8 Curve: VI Figure 4.1-Protection scheme between GSS CB and Gantry recloser Page 19 of 60 Current in Amperes Figure 4.2-Protection curves between GSS CB and Gantry Recloser Page 20 of 60 4.1.1. Downstream protection scheme The downstream protection devices of Kalumale feeder of Kithulgoda gantry is selected for analysis. Feeder arrangement of Kalumale feeder is attached in Appendix -1. There are 7 Spur points numbered 1, 2, 3 etc. The connected fuse links for Transformers and Spurs are listed in Table 4.1 [11]. Spur Point Sin No Transformer- Capacity (kVA) Recommended fuse link connected to Transformer (A) Present Connected Fuse link in spurs 1 A12 250 8 40 A71 160 6 Al l 160 6 A90 1000 20 2 A50 100 3 15 A60 100 3 A80 100 3 A68 100 3 3 A59 100 3 15 A13 160 6 A73 100 3 4 A16 100 3 20 A46 160 6 A17 250 8 A 28 100 3 5 A19 250 8 15 A20 160 6 6 A64 160 6 10 7 A24 100 3 10 A44 100 3 Table 4.1 Connected fuse links at Kalumale feeder The spur lines are protected by 10A, 15A, 20A and 40A fuse links. A 'K' type (fast type - low speed ratio) fuses are used by C.E.B. The downstream protection scheme Page 21 of 60 is analyzed by drawing fuse curves with upper stream protection scheme. These curves are drawn by SynerGEE Electric Program. — H - l_j — 1 1 -]— 1 1 10A Fuse A! — —H— — H ^ - — | I I — ——H _ L _ L i _ L L J — ——r~ -H 4 | T l — ' — -H — i H - M - -J A -40A (CB) — 0 _LI I'G 1 1 \\ EF-20A(RBC) I \ \ - • 1 1 I * v c = = ———f : : X = — r H — — 1 \ v 4 t f k \ — -H i m f m Y)N\ c H — OC-IO 0A(REC)| ftj; 1— i | M f c r — u ~ r n - H - W(A o a k i*mi -320 OCt (X L ffi luw.l — i — 1 L _ _ _LL \m m XX Vv. W Current in Amperes Figure 4.3 -Protection curves when a spur line is protected by 10A fuse The protection scenario of a spur line protected by 10A fuse is shown in Figure 4.3. In this scheme all transient and permanent faults between 70A to 160A cause to operate fuse link. But the desired operation is to clear transient faults by the recloser and Page 22 of 60 permanent faults by the fuse. Therefore, existing protection scheme does not serve the purpose by isolating the faulty section only, but trips the spur line affecting all the consumers on that spur unnecessarily. T j f f — — —i- +1 w —J— f "4-4- —r -I— 15AFus f —LI _ 1 - L L j y "tv l" : H i •j £ i\ _L_L g = = " n '—bX + = p V p i EF-40A (CB) — t - — i j j . - nr^MA era 1 EF-20 -U \(R£C ) \ \ s \ — —L 1 J — 1 —J—L \ k - - K — — — — - — — 4 - - \ — [ • 4 - i _ L N oc -20QA m\ — 4 — — ^ — 3 —L i | —L - 4 - 7 — mma -320 OA 6 6 L I6QJ H32C — — p r A S k. Current in Amperes Figure 4.4 - Protection curves when a spur line is protected by 15A fuse The protection scenario of a spur line protected by 15A fuse is shown in Figure 4.4. In this scheme all transient and permanent faults below 1600A cause to operate the fuse link. If a permanent earthfault (below 160A) occurs in spur line it causes to lockout recloser but not the spur line leading to supply outage to all customers in Feeder. Fuse is not actually an earthfault detection device, but at the time of the earthfault the earthfault current is equal to one of phase current in Feeder. It occurs 90% of the time during an earthfault situation as proven by the event log records of recloser, the equivalent of earth fault current and phase current at the time of earth fault is Page 23 of 60 deducted in Chapter 3.1. The desired operation is by recloser for clear transient faults and by fuse for permanent faults. Therefore existing protection scheme does not serve the purpose by isolating the faulty section. hHEtEF t+fEEEE — 1 - n -H j~\ —H — —U- — M 20AF use i U U E f^TT M 1 1 R~TTT TTj — h- I r 4 ~TW1 —[4- 1 MI WiA -f-v VT'J. nna T m _j_i_j =cf _L_L ——™t H—t~r+ ~Mr4 tH — - fef III 1 M J 1 t T — — — — v EF-40A (CB) K —nr OC-MfCB-l EF-20A(REC) V s= ^ A \ : n n i jii 1 T T 1 N — • -U-k x — 4-H— f-i ,.... f v l V AA —V oc- -44- MA(REC)| — v 4 -k •+H i - — — U T T I — H—— -320 OA V V • 1 1 1 16GA — 32( — A 10 tB I30C Current n Amperes Figure 4.5 - Protection curves when a spur line is protected by 20A fuse The above behavior is applicable also for spur line protected by 20A (see Figure 4.5) and 40A (see Figure 4.6) Page 24 of 60 K:: — — — — — = = - — — — — — — EE J 1 = = i OAFac ' tui N — - — — - — — — — = -FFJfim | | f | oc 1 1 1 1, I P ) t # 4 _Li -LL- — — — : Et - ...... |—T~T Til—r" j \ 1 — — \ i MTI yyu \ ;r 1 f N jo t : — —H — b m fVn—^ H[ bd — - • 1 1 I MM 'Mmr f v - K v •f- 1 jl. E • r |j]S IKXJI. SJ \ ——M "r f jj \ It;; 1_ i= n LJ :=f J $ i i r —i- j-f-j-j F HO A ••kit ••ft'. IHI Wfe — — i j p r-n 44 llv'vl 1 31A i \~M " 1 •T " • • - r i l LEO HIT 1L UJ Current in Amoeres Figure 4.6 - Protection curves when a spur line is protected by 40A fuse It is understood from the above analysis that the down stream protective devices beyond the recloser do not distinguish permanent and transient faults in the spurs of the network. It can be seen that nuisance fuse blowings in spurs as well as nuisance recloser lockouts occur affecting all customers in the Feeder. * ' \ Therefore there are possibilities of improving reliability of network by considering - downstream protective devices and minimizing unnecessary fuse blowings. Page 25 of 60 4.2. Design new protection scheme With proper selection of protection devices (fast tripping and fast reclosing) transient faults can be cleared without a sustained outage and permanent faulty sections can be isolated. Following three operations are available [2]; I. The main protection should clear a temporary fault before fuse ruptures. For a permanent fault the fuse should rupture and disconnect the faulty section. II. Let the fuse blow even for temporary faults. III. Let the fuse not blow at all. The criteria ' 1' is selected for design of the new protection scheme. 4.2.1. Design algorithm The algorithm for design of the new protection scheme is shown in Figure 4.8. The design algorithm comprises following steps. 1. The applied algorithm network is shown in Figure 4.7. Collection of following network data; • Fault levels at GSS • Line impedances • Transformer data 2. Collection of the functions and data of circuit breakers at GSS, reclosers at Gantries, fuses at spurs and transformers. 3. Calculate fault level in network. 4. Select the devices to check coordination by considering continuous current ratings of equipment. Page 26 of 60 Figure 4.7 - Network diagram 5. If the coordination between devices is success, new protection scheme could be adopted. 6. If there is a miscoordination between protective devices, It should be classified recloser to fuse, fuse to fuse and relay to relay as follows; Page 27 of 60 Figure 4.8- Algorithm for the new protection scheme design Page 28 of 60 Recloser - fuse miscoordination classification Figure - 4.9 Recloser to fuse miscoordination classification The procedure to co-ordinate a reloser and a fuse is shown in Figure 4.9, when the fuse is at the load side, the following rules is applied [6]: • The minimum melting time of the fuse must be greater than the fast curve of the recloser times the multiplying factor (k), given in Table 4.2. The recloser to fuse coordination curves are shown in Figure 4.10. Page 29 of 60 • The maximum clearing time of the fuse must be smaller than the delay curve of the recloser without any multiplying factor. If the delay curve of recloser and fuse maximum clearing curve coincide, it causes to trip recloser as well as blow fuse. It can be overcome by giving grading margin between delay curve of the recloser and the fuse as follows [13]. t'=0At + 0.\5Seconds Where t = nominal operating time of the fuse t'= grading margin to be allowed between the relay and the fuse The relay to fuse coordination is shown in Figure 4.11 Figure 4.10 - Recloser to fuse coordination curves Page 30 of 60 Current in Amperes Figure 4.11 - Relay to fuse coordination curves • The recloser should have at least two or more delayed operations to prevent loss of service in case the recloser trips when the fuse operates. Better co-ordination between a recloser and a fuse is obtained by setting the recloser to give two instantaneous operations followed by two timed operations. In general, the first opening of the recloser will clear 80% of the temporary faults, and the second will clear a further 10%. The load fuses are set to operate before the third opening, clearing permanent faults. The co-ordination obtained using one instantaneous operations followed by three timed operations is less effective. Reclosing time in cycle Multipliers for one fast operation two fast operation 25-30 1.25 1.80 60 1.25 1.35 90 1.25 1.35 120 1.25 1.35 Table 4.2- Multiplying factor (k) for recloser curve Page 31 of 60 Fuse to fuse miscoordination classification Figure 4.12 - Fuse to fuse miscoordination classification The fuse to fuse coordination criteria is shown in Figure 4.12. The essential criterion when using fuses is that the maximum clearance time for a main fuse should not exceed 75% of the minimum melting time of the back-upfiise. (Tl<0.75xT2) [13]. The fuse to fuse coordination of curves is shown in Figure 4.13. Simple 'RULE OF THUMB' for grading Choose IFA~ 2>2184A Now it should be checked whether the new protection settings of GSS CB is matched with GSS T/F CB. GSS T/F Protection settings: Current transformer ratio = 600/5A PSM = 5A TMS = 0.325 Curve = Standard Inverse Operating time of standard inverse curve Maximum short circuit current at GSS Bus Operating time of GSS Feeder CB Operating time of T/F GSS CB Time margin of at maximum short circuit current Time margin is above 0.35 S.Therefore existing protection setting of GSS TR CB is enough. The overcurrent setting of protective devices with 50A fuse is shown in Figure 4.19. 0A4TMS r j \ 0 02 — - 1 .Is) 10000A 13.5x0.20 10000 1 . 400 , 0.11 Sec 0.14x0.325 (10000Y02 , 600 J 0.78 S 0.78-0.11=0.67S Page 45 of 60 I | I I I I ITT IIZZLIIILIuIIUI n Mycint-R«kr-1A I yyi "vnr m \ \ m i i Wi T e c 0 |jh M LI i i i i i m ™ n i III III TVM I \ I "111 m\ | \i mi m V W I TT I ' I I I 1 KMi i ! li l " •t— , \ \ c s m w r Curat h A p e s Figure 4.19 - Overcurrent settings of protective devices with 50A fuse Page 46 of 60 Then earthfault setting can be found. The selected Earthfault setting of recloser = 60A (Unbalance allowance is taken as 30% from overcurrent setting [6]) The maximum earthfault current of Kalumale Feeder =808A Therefore Instantaneous current setting =60x7= 420A (Instantaneous setting = 50% of the maximum short-circuit current at the point of connection of the relay or between six to ten times of earthfault setting) The opening time of the recloser at above 420A (earthfault current)=0.04 S The minimum times to prevent hot the fuse at maximum earthfault current =1.8x0.04 (Select maximum multiplying factor as 1.8 from Table 4.2) = 0.072 S Earth fault current at Spur Point' V = 715A (Calculated as earlier-4.2km from Switching Gantry) The minimum melting time of the 50A fuse at maximum fault Current (715A) = 0.34S The fuse minimum melting time (0.34S) is above the recloser times the multiplying factor. (Time with multiplying factor = 0.072S) Therefore fuse will be protected by preheating. The selected fast tripping curve of earthfault recloser = IEC Ext Inv TMS of fast tripping Curve = 0.05 (Above curve and TMS is selected to operate the recloser at minimum time and to the match Fuse Curve.) The recloser operating time at the maximum earthfault fault current at Spur point -1 =0.49S (Time is selected by drawing the fuse curve) Therefore the operating time of the recloser earthfault delay curve should be above 0.49S and can be selected as 0.50S. Select delay operating time of recloser as extremely inverse to better coordination of fuse „ .. .. - . , . 80TMS Operating time of extremely inverse = — — - 1 \Is) Page 47 of 60 "715 Therefore suitable TMS value for recoser delay curve = 0.50 80 = 0.88 Therefore TMS value can be selected for recloser delay curve above 0.88. Above curves are drawn in SynerGEE Electric Program and TMS value of delay curve of reloser is selected as 2.0 in order to overcome minimized the intersect delay curve of recloser and the 50A fuse. (The maximum TMS setting for the recloser is Then protection settings of GSS Feeder Circuit breaker should be found because the present setting does not coordinate with the recloser. Maximum eartfault current at Gantry Operating time of recloser at this short circuit current at Gantry The usually better margin for relay to relay Operating time of the GSS CB a fault current of 776A The selected earthfault current setting of GSS Feeder CB (Unbalance allowance is taken as 25% from overcurrent setting) The selected curve of GSS CB is Extremely Inverse TMS of GSS CB Therefore TMS value can be selected for GSS CB delay curve is selected as l.u by drawing above curves in SynerGEE Electric program. The maximum earthfault current at Kalumale Feeder = 808A Suitable instantaneous value to prevent relay operation for outside faults = 808x 12=910A The selected instantaneous value of GSS Feeder CB = 100x 10=1000A>970A Now it should be checked whether the new protection settings of GSS CB is matched with GSS T/F CB. Page 48 of 60 808A 80x2.0 808 \2 60 . 0.89 S 0.35 S 0.89+0.35 1.24 S 100A 1 = 1.24- 808 100 80 = 1.0 GSS T/F Protection settings: Current transformer ratio = 600/5A Earth fault settings PSM = 0.65A TMS = 0.325 Curve = Standard Inverse Maximum earthfault current at GSS Bus = 1890A Operating time of GSS Feeder CB = 0.04S Operating time of T/F GSS CB 0.14x0.325 1890 \ 0.02 - 1 Time margin at maximum short circuit current . 78 . = 0.69 S = 0.69-0.04 =0.65S Time margin is above 0.35S. The margin is sufficient, but it can be seen curves of GSS CB and GSS Transformer intersect at lower current values Earth fault setting of GSS CB is 100A and GSS Transformer CB is 78A. Therefore new protection setting should be applied to GSS Transformer CB. As earlier calculations and by drawing curves in SynerGEE Electric program following settings is selected. Earth fault settings of GSS Transformer CB = 150A Curve = IEC Very inv TMS • =1.0 The stand by earth fault setting can also be set of the same values. The earth fault settngs of protective devices with 50A fuse is shown in Figure 4.20. Page 49 of 60 I I I I i i i rv/rrr T | ; irT T T T T rf M 11 _ I l l l T | [ l | X I " Ddzv curv e - Recloser - 60A I I I I I I I I i M111 1 Ml nyvu 1 wm TO \ \ i t r XX U« \ IVTT TTT IT HAAl \ | \ i | Wt M \ l\ m k N \ GSSUCB-li: m i \ \ \\ m i \ j \ | K r T ¥ \ \ INJ i U l i X i \ \ Fast cun-e - Recb'ser - 60 A: IVUj I i \1\T vnl I k * \ x x XL 7 I wm > 11 I T n w I I I I I I II I II' I l\ i r TTTTT J3_ I I I M I I | \ | I I I ! I\ I II I I Fuse - 50A \VYYTH m l v m W xxxxxsT 130C IE Figure 4.20 - Earthfault settings of protective devices with 50A fuse Page 50 of 60 4.3.5. Selection fuses for spurs at Kalumale Feeder. The selected fuse for spur point' 1' is 50A. It is the maximum short circuit current and also the earthfault current point of spurs. The short circuit current and the earthfault currents of all other points are below maximum short-circuit current and the earth fault current at spur point '1'. Therefore 5OA fuse can be selected for other points also. The connected fuse links for Transformers and Spurs are listed in Table 4.3. Spur Point Sin No Transformer- Capacity (kVA) Recommended fuse link connected to Transformer (A) Present Connected Fuse Capacity Selected Fuse Capacity (A) 1 A12 250 8 40 50 A71 160 6 Al l 160 6 A90 1000 20 2 A50 100 3 15 50 A60 100 3 A80 100 3 A68 100 3 3 A59 100 3 15 50 A13 160 6 A73 100 3 4 A16 100 3 20 50 A46 160 '6 A17 250 8 A28 100 3 5 A19 250 8 15 50 A20 160 6 6 A64 160 6 10 50 7 A24 100 3 10 50 A44 100 3 Table 4.3 Connected fuse links in Kalumale feeder with new protection scheme Page 51 of 60 There are fuses at spurs for transformer protection. It should be checked whether the 50A fuse is properly coordinated with them. Following fuse coordination should be checked. 1. 50A (T type) fuse to 20A (K type) fuse 2. 50 A (T type) fuse to 8A(K type) fuse 3. 50 A (T type) fuse to 6A (K type) fuse 4. 50 A (T type) fuse to 3A (K type) fuse The fuse coordination between 50A (T type) fuse to 20A (K type) fuse should be matched because 50A fuse is 2.5 times 20A fuse. The 20A fuse is protected by lOOOkVA Transformer and situated in Spur Point-1 with a distance from point-1 of 1.1km. The total clearing time of maximum short circuit current (1439A) of 20A (K type) fuse is 0.016S and the minimum melting time of that current for the 50A (T type) fuse is 0.080S. Therefore, the maximum clearance time for the main fuse does not exceed 75% of the minimum melting time of the back-up fuse. The Figure 4.21 shows coordination between 50A and 20A fuses. Other fuse coordination should also be matched similarly. Page 52 of 60 ffl IXC Current in Amperes Figure 4.21 - Coordination of 50A (T type) fuse and 20A (K type) fuse Page 53 of 60 4.4. Computer based model for new protection scheme Normally, coordination between protections devices are done manually. Existing practice is based upon set of rules that require human interface. This may lead to errors and incorrect operation. The model for any 33kV feeder developed by SynerGEE Electric program is shown in Figure 4.22. In this model source impedance, line length and impedances could be set. Then suitable values for protective devices could be found by changing settings of devices and checking the achieved coordination. The algorithm given in section 4.2 could be tested with the aid of network computer model. Feeder 33kV L r n fine Racoon line GSS TF CB GSS Feeder CB Xode-1 Section-1 Figure 4.22 - The model for 33kV Feeder to analyze protection coordination The above model is used to verily coordination between protective devices of Kalumale Feeder as a case study. The summarized fault report and the coordination results between protection devices of Kalumale Feeder are given in Appendix II & III. Page 54 of 60 4.5. Advantages using new protection scheme. i. The new reliability indices for Kalumale Feeder if above protection scheme is adopted are shown in Table 4.4. These values are evaluated based on the model developed in Chapter 3.4. With old protection scheme With new protection scheme(Insert 50A Fuse for Spurs) With new protection scheme (Insert 50A Fuse for Spurs & Sectionalizers for Isolators) SAIFI(int/yr) 5.6 3.5 3.5 SAIDI(h/yr) 67.5 39.3 35.6 ASAI 0.992295 0.995514 0.995936 Table 4.4 - Reliability indices of Kalumale Feeder It can be seen that the reliability indices are significantly improved with new protection scheme. ii. Only one type of fuse can be selected for all spurs. This reduces human errors, and also the number of different types of fuses to be stocked. The annual requirement of fuses used with old protection scheme in C.E.B is shown in Table 4.5 [15]. The old protection scheme causes nuisance fuse blowing. It saves 67% (Transient faults) of fuse requirement with new protection scheme and the saving financial is estimated to be Rs. 14,173,850.00. Fuse type Unit price (Rs) Quantity Cost (Rs) 10A 121.00 45000 5,445,000.00 15A 161.00 30000 4,830,000.00 20A 163.00 35000 5,705,000.00 40A 207.00 25000 5,175,000.00 Total cost 21,155,000.00 Table 4.5 - Annual requirement of fuse types of C.E.B Page 55 of 60 The traveling time to find the fault is reduced because only faulty line is isolated by new protection scheme. (Consumers at faulty section will inform about supply outage to Consumer Service Centre). When we consider Kalumale Feeder with old protection scheme, the average fault finding distance is about 20km and average fault repairing time is 2hours. Page 56 of 60 Chapter 5 Conclusion & Recommendation 5.1. Improvement of reliability The objective of this study is to improve reliability of 33kV network of C.E.B using effective fault isolating methodologies. The results of the study show that reliability of 33kV network is very poor compared to international level. The reliability of 33kV network is directly related to the faults in the network. Most of the faults in the network are during the two rainy periods of the year. There is no proper protection scheme to isolate the faulty sections of the network. As the faulty section cannot be identified at the very beginning of the incident, the repairing time and labour involvement is high. The situation is more severe in rural areas. 5.2. Recommendation In order to overcome this problem and to increase the reliability of 33kV network of C.E.B some recommendations are given below: 5.2.1. Using event log of recloser It is recommended to analyze event log of reclosers. Following benefits can be obtained: a. Predominant causes of faults in the network can be identified. The maintenance schedules such as way leaves program can be better planned. b. The faults can be categorized as overcurrent and earthfaults. More repair time is required to repair overcurrent faults caused by loose jumper connections and damaged conductors in the line. c. The momentary interruption of duration (below 'l'min) can be analyzed. If the momentary interruption is higher for a certain period, it can be predicted that tree branches are close to line. Then immediate way leaves clearing program can be arranged after line inspection. Page 57 of 60 5.2.2. Reduce outages in the network Faults in the network highly affect the reliability of the network. Faults are higher in the period of two rainy seasons. These faults can be eliminated by carrying out way leaves programs before and after rainy seasons. Predefined clearance should be strictly maintained. Re-stringing of conductors and adjustment of clearance of T-off should be carried out to overcome faults by way leaves and wind. Compression type of H connectors for jumper connections should be introduced at tension points, angle points and spur points to overcome open circuit under heavy fault current. The compression type midspan joints should be used to connect damaged conductors. The outage time of the network can be reduced by a considerable amount by installing fault indicators because it reduces fault finding time and gives a remote indication to the control centre. Recommended location of fault indicators are given below; i. Easy to access line points for immediate monitoring ii. Next to the spur line iii. After switch (Air Break Switch or Load Break Switch) iv. Before and after the points that is difficult to reach such as forest or mountain. Reclosers used in C.E.B can be operated by remotely. This function is still not initiated by C.E.B. The outage time can be reduced by using remote operation. Thereby it improves reliability of the network. A line sectionalizing device shall be used to divide the feeder in to smaller line segments. All taps should have protective devices such as reclosers or sectionalizers for large taps and fuses for all taps. Fuses shall be used even for small taps. Page 58 of 60 The outage time can be minimized by limiting the size and length of the affected line. Therefore the locating time of the fault can be minimized. 5.2.3. Isolating faulty line from the feeder. The recloser to fuse protection coordination should be designed to clear transient faults by the recloser and the permanent faults by fuse. The faulty section can be isolated more effectively with this protection scheme. The nuisance fuse blows can be reduced with this protection scheme. Thereby a stock of fuses for spurs can be reduced by 67%. It is recommended to use fuses of same rating for all spurs. The sectionalizers can be used for spurs and sections when rising fault level in the network because it has no time-current characteristics. This does not require changing of GSS protection setting. A single phase sectionalizer can be used for spurs in rural areas. A three phase sectionalizers can be used for industrial consumers and heavy load current lines. Page 59 of 60 References [1] Ceylon Electricity Board - Network Improvement unit - North Western Province," Overhead Electricity Network of Ceylon Electricity Board and its performance and improvements" CEB, September 2003 [2] Ceylon Electricity Board - Distribution Planning Branch, "Protection of Electricity Distribution Network", CEB, April 2002 [3] Ceylon Electricity Board - Commercial and Corporate Branch - Distribution Region - 3 , "Statistical Digest - 2007 " CEB, February 2008 [4] Advantica Stoner, "Users Guide", SynerGEE Electric 3.5. [5] NU - LEC Industries. A Schneider Electric Company, "Technical Manual of Pole Mounted Enclosure N12/ N24/ N36." [6] J.M. Gers and E.J. Holmes, "Protection of Electricity Distribution Network", The Institution of Electrical Engineers - United Kingdom, 1998, pp 65-123 [7] J.H. Naylor, "Power System Protection, 1 Principles and Components- second edition", The Electricity Council- United Kingdom, 1981, pp 53-246 [8] Cooper Power Systems - "Electricity Distribution System Protection-third edition", 1990, pp 82-143 [9] Ceylon Electricity Board, "Breakdown reports of Agalawatta, and Dickwella CSCs from January 2007 to December 2007" CEB, 2007 [10] Weather Department - Sri Lanka, "Rain data reports of Kithulgoda and Dickwella area from January 2007 to December 2007", 2007 [11] Ceylon Electricity Board - Distribution Planning and Development Branch - Region 4, "Selection of MV Fuses", CEB, July 2006 [12] D. Stone, "Switchgear Manual"- ninth edition", Asea Brown Boveri Pocket Book - Federal Republic of Germany, 1992, pp 88-93 [13] Protective Relay Application Guide - GEC Alsthom T and D, 2000, pp 129- 157 [ 14] Provincial Breakdown Reports, CEB, 2003 to 2007 [15] Annual Procurement Reports, CEB, 2007 Page 60 of 60 Appendix II Fault Report for Feeder at Kalumale Feeder Section Fit Phase Dist Phs Cuml. Impedanc 9 Nominal Symmetrical Amps Id Loc Conductor kM Cfg R1 X1 R0 xo kV Min G Max G L-L L-L-G 3 Ph | Feeder Kalumale Feeder ABCN 0.118 1.899 -0.000 26.460 33.0 461 1889 8672 8688 10014 33kV Bus 33kV Bus 1 CU 0.0 ABCN 0.118 1.899 0.000 26.460 33.0 461 1889 8670 8686 10012 Mathugama GSSto Kithulgoda Gant Kithulgoda Gantry 1431ACSR 22.9 ABCN 4.172 9.571 7.443 49.908 33.0 375 807 1580 1600 1825 Kithulgoda Gantry to Node - 1 Node - 1 1/0 ACSR 27.1 ABCN 5.982 11.146 9.874 54.579 33.0 354 715 1304 1323 1506 Spur-1 Spur-1 1/0ACSR 28.1 ABCN 6.413 11.521 10.453 55.692 33.0 350 696 1251 1270 1445 Section-1 Section-1 1/0 ACSR 40.0 ABCN 11.542 15.983 17.343 68.928 33.0 302 526 837 854 966 Device Device Device Fault Values Interrupt Percent Interrupt Dwn Stream Min. LG Dwn Stream Min. 3Ph Name Category Type LG LL LLG 3Ph Rating LG LL LLG 3Ph Section Amps Section Amps | Matugama GSS TF OG CB Brkr GEC, MCGG 1889 8672 8688 10014 25000 8 35 35 40 33kV Bus 461.0 33kV Bus 10011.6 Matugama GSS Feeder CB Brkr GEC. MCGG 1889 8670 8686 10012 25000 8 35 35 40 Mathugama GSSto Kithulgoda Gant 375.2 Mathugama GSSto Kithulgoda Gant 1824.9 Recloser - Kaiumle Feeder Reel Nu-Lec, N series 807 1580 1600 1825 16000 5 10 10 11 Section-1 301.6 Section-1 966.4 Fuse Spur - 1 Fuse Chance, T 715 1304 1323 1506 — — — — — Spur-1 349.6 Spur-1 1445.0 Appendix III Check Coordination at Kalumale Feeder 0 Project: Licensee: SAI In house (Carlisle) Analysis : Check Coordination Run: 06:12PM on January 06, 2009 I SynerGEE Electric 3.5 (Oct 212002) Stoner Associates, Inc. © 2001 Feeder Number Checks Failed Checks Passed Checks Kalumale Feeder 39 0 39 i i Category Margin | Relay Protecting Fuse (pet) 100.0 % Relay Protecting Fuse (sec) 0.3 Sec Fuse Protecting Relay (sec) 0.3 Sec Fuse Protecting Fuse (pet) 75.0 % Recloser Protecting Fuse (pet) 90.0 % Fuse Protecting Recloser (pet) 95.0 % Relay Protecting Relay (pet) 75.0 % Relay Protecting Relay (sec) 0.3 Sec Recloser Protecting Transformer (pet) 75.0 % Relay Protecting Transformer (pet) 75.0 % Fuse Protecting Transformer (pet) 75.0 % Recloser Protecting Recloser (pet) 75.0 % Recloser Protecting Relay (pet) 75.0 % Relay Protecting Recloser (pet) 70.0 % Fuse - Transformer Inrush (pet) 75.0 % Recloser - Transformer Inrush (pet) 75.0 % Relay - Transformer Inrush (pet) 75.0 % Fuse Minimum Fault Margin (pet) 220.0 % Fuse Load Current Margin (pet) 90.0% Recloser Protecting Recloser (sec) 0.2 Sec Check Coordination Page 1 of 4 Protecting Device (A) Protected Device (B) Protecting Phase Amps Protecting Ground Amps | Pair (Downstream) (Upstream) Min Max Load Shift Min Max Shift 1 * Type Name Sect Type Name Sect Fit At Fit Amps B -> A Fit At Fit B->A | Feeder Kalumale Feeder 1 Brkr Matugama GSSTF OG CB 33kV Bus — — — 756 33kV Bus 10014 0 — 756 33kV Bus 1889 — 2 Brkr Matugama GSS Feeder CB Mathugama GSSto Kithulgoda Gantry — — — 323 Mathugama GSSto Kithulgoda Gantry 10012 0 — 323 Mathugama GSSto Kithulgoda Gantry 1889 — 3 Reel Recloser - Kalumle Feeder Kithulgoda Gantry to Node - 1 — — — 210 Section-1 1825 0 210 Section-1 807 — 4 Fuse Fuse Spur - 1 Spur-1 — — — 278 Spur-1 1506 0 — 278 Spur-1 715 — 5 Brkr Matugama GSS Feeder CB Mathugama GSSto Kithulgoda Gantry Brkr Matugama GSSTF OG CB 33kV Bus 323 Mathugama GSSto Kithulgoda Gantry 10012 0 None 323 Mathugama GSSto Kithulgoda Gantry 1889 None 1 ! 6 Reel Recloser - Kalumle Feeder Kithulgoda Gantry to Node - 1 Brkr Matugama GSS Feeder CB Mathugama GSSto Kithulgoda Gantry 210 Section-1 1825 0 None 210 Section-1 807 None 7 Reel Recloser - Kalumle Feeder Kithulgoda Gantry to Node - 1 Brkr Matugama GSSTF OG CB 33kV Bus 210 Section-1 1825 0 None 210 Section-1 807 None 8 Fuse Fuse Spur - 1 Spur-1 Reel Recloser - Kalumle Feeder Kithulgoda Gantry to Node - 1 278 Spur-1 1506 0 None 278 Spur-1 715 None 9 Fuse Fuse Spur - 1 Spur-1 Brier Matugama GSS Feeder CB Mathugama GSSto Kithulgoda Gantry 278 Spur-1 1506 0 None 278 Spur-1 715 None 10 Fuse Fuse Spur - 1 Spur-1 Brkr Matugama GSSTF OG CB 33kV Bus 278 Spur-1 1506 0 None 278 Spur-1 715 None Fuse Checks 4 Pass Fuse Spur -1 10-32 223% 278 A - Minimum phase Suit of 278A shout: eceed 223% of t s e 53 amp rating 4 Pass Fuse Spur - 1 10-33 223% 278 A - Mirimun ground faut of 278A should exsee: 223% of t s e 53 amp rating. 4 Pass Fuse Spur - 1 10-05 - _ Fuse has valid mm-nrelt (34581) and max-dear (34583) curves Check Coordination Page 2 of 4 Fuse Protecting Recloser Checks. 8 Pass Fi.se Spur -11 Recloser -Kaiumle Feecer 12-01 95% 278- 1535 A Recloser fast phase curve |K Factor « 1.8) should lie below fuse min met! curve 8 Pass Fuse Spur -1 / Recloser -Kaiumle Feeder 12-02 95% 278-715A Recloser fast ground curve (K Factor = 1.8) should lie bek>w fuse min. melt curve. 8 Pass Fuse Spur -1 / Recloser -Kaiumle Feeder 12-03 95% 278- 1555 A Recloser slaw phase curve should lie above the i s e max clear curve 8 Pass Fuse Spur -1 / Recloser -Kaiumle Feeder 12-04 95% 278-715A Recloser slow ground curve should lie above the l;se max. clear curve. Fuse Protecting Relay Checks. 13 Pass Fuse Spur -1 / Matugama GSS T FOG C8 13-01 0.30 278- 1533 A The tise max clear curve should lie below the relay phase cuve. 9 Pass Fi.se Spur -1 / Matugama GSS Feeder CS 13-01 0.33 sec 278- 15D5A The I s e max clear curve should lie below the relay phase cuve. Recloser Checks. 3 Pass Redoser - Kaiumle Feeder 20-01 - - 1825A phase fault current should not exceed recloser "3333 amp interrupt rating 3 Pass Redoser - Kalunle Feecer 20-02 - - 837A ground faut current should not exceed redoser! 3333 amp interrupt rating 3 Pass Recloser - Kalunle Feeder 20-33 - - Redoser should pickup minimum phase feult current of210A 3 Pass Recloser - Kalunle Feecer 23-34 - - Redoser should pickup minimum ground aUt current of210A. 3 Pass Redoser - Kalunle Feecer 23-35 - - Redoser Sst ground cuve should lie beta'.1 its siow curve. 3 Pass Redoser - Kalunle Feeder 23-05 - - Redoser &st phase curve should lie below is slow curve. 3 P3SS Recloser - Kalunle Feeder 23-35 - - Redoser should have at least one fast or ere time-delayground operation 3 Pass Recloser - Kalunle Feecer 23-33 - - Redoser should ha ve at least one fast or ere time-delay phase operation. 3 Pass Redoser - Kalunle Feeder 2307 - 210-1825 A Redoser ast and slow phase operation cuves shout extend past 210A minimum &ult level. 1 Pass Redoser - Kalunle Feeder 23-38 - 210-837 A Redoser Sst and sloe.1 ground operator curves shout extend past 210A minimum feult level 3 Pass Redoser - Kalunle Feeder 23-39 - 210-1825A Redoser ast and slow phase operation cuves shout pickup beta'. 1825 maximum Suit amps 3 3 Pass Redoser - Kaiumle Feecer 23-10 - 210-837 A Redoser ast and stow Ground operation curves shout pickup beta.' 837 maximum ault amps. Pass Recloser - Kalunle Feecer 23-12 - 210-837 A Maxmum &ult cuirent of 837A should not excee: either of the reckser ground operator curves 3 Pass Recloser - Kalunle Feecer 23-13 - - Redoser should have phase or ground curves Recloser Protecting Relay Checks. 7 Pass Redoser - Kalunle Feeser 1.* atugama GSS TF OGCS 23-31 75% 210- 1825A Redoser phase td. curve should lie below relay phase curve | lo t Pass Redoser - Kalunle Feecer Matugama GSS Feeder CS 23-31 75% 210-1825A Redoser phase td. curve should lie below relay phase curve 7 Pass Redoser - Kalunle Feeder Matugama GSS TF OG CS 23-32 75% 210-837 A Redoser ground t.d. curve should lie beta': relay ground curve S Pass Redoser - Kalunle Feecer Matugama GSS Feeder CS 23-32 75% 210-837 A Redoser ground t.d. curve should lie below relay ground curve. Relay Checks. 2 Pass Matugama GSS Feeder CB 33-01 - - 13312A phase &ult current should not exceed breaker s 25330 amp interrupt rating Check Coordination Page 3 of 4 Relay Checks. 2 Pass M atugarra GSS Feeder CS 33-01 - — 1X12A phase Suit current should not exseec beakers 25333 amp irterrupt rating 1 Pass M atugarra GSS TF OG CS 33-31 - - 10014A phase feult current should not eroeec beaker's 25333 amp irterrupt rating 2 Pass Manama GSS Feeder CS 33-32 - - 1SS5A ground fault current should not exsee: beaker's 25333 amp irterrupt rating 1 Pass Matugama GSS TF OG CS 33-32 - - 'SS9A ground fault currert should not exsee: beaker s 25333 amp interrupt rating 1 Pass M atugarra GSS TF OG CS 33-34 - - Rsl3y Ground curve (pickup = 153) should pchL'p minimum Suit current of 756A. 2 Pass U atugarra GSS Feeder CS 33-34 - - Relay Sound curve (pickup = 103) should pickup minimum Suit current of 323A. 1 Pass M atugarra GSS TF OG CB 33-39 - - Beaker should have phase or ground elay curves. 2 Pass Matugama GSS Feeder CS 33-39 - - Beaker should have phase or ground elay curves. Relay Protecting Relay Checks 5 Pass I.'atugarra GSS Fee:er CS / l."anj;arra GSS TF OG CS 33-31 75% 323- 13312 A Protecting phase relay curve should lie below protected relayphase curve. 5 Pass 1.'atugarra GSS Fee:er CB ' M atugarra GSS TF OG CB 33-02 75% 323- 18S9A Protecting grouvd relay cirve should lie below protected relay ground curve 5 Pass Matugama GSS Fee:er CB / M atugama GSS TF OG CB 33-33 0 36 sec 323- 13312 A Protecting phase relay curve should lie below protected relayphase curve. 5 Pass 1." atugarra GSS Fee:er CS / M aoigama GSS TF OG CB 33-34 0.36 se: 323- 1S89A Protecting groird relay ct/ve should lie below protected relayground curve Page 4 of 4