IV DAS Pilot Project

4. Pilot Project Progress

4-1 Pilot Project Design and Make a Contract (2006. 11)

The participating scope of Pilot Project is a large range of distribution, communication and distribution materials and shall be proceeded with complexness. Therefore according to its complexness and circumstances KEPCO has proceeded Pilot Project with KDN, which is a subsidiary company of KEPCO taking a responsibility for installation and maintenance of DAS and participated in development of DAS.

(1) Contractor : KDN (Korea Electric Power Data Network)
(2) Business duration : 12months after starting (2006. 11. 03 – 2007. 11. 10)

4-2 Object Selection for Installation of Switches

Sayung S/S 4 feeders for installation object places were selected by PLN engineers with selection standard as following. Those are No. 2feeder of Sayung S/S No 1. M.Tr, No 5 and 6 of No.2 M.Tr, and No.9feeder of No.3 M.Tr, those location and kinds of switches for installation was decided by load characteristics and operation status of feeders.

Pilot project was conducted with new installation on the places having no load changing or with replacement of existing switches on the basis of the selected feeders. Following [Figure 4-3] is a single diagram for feeders marked with automated switches and automation object feeders. The installation status of the switches which is selected finally is as following [Table 4-1].

[Figure 4-3] Sayung S/S Single diagram

[Table 4-1] Installation Status of Automated switches

4-3 Measuring and Examination

4-3-1 Grasping Communication Status & Measuring (2006. 10 – 2006. 11)

To strength the operation efficiency of Indonesia DAS an optimum communication method shall be utilized to be suitable for local situation, so measuring for the communication method used in Indonesia commonly shall be preceded preferentially. Mc-40 was used in measuring for GSM and cell phone is used for CDMA.

4-3-2 Communication Method Examination & Conclusion (2006. 10 – 2006. 11)

There are two kinds of communication methods applying on Indonesia currently, one is GSM and the other is CDMA and as a result of sensitivity measuring for each communication method an optimum choice was conducted. Through a discussion and interviewing with technicians of TELKOMSEL and PLN communication engineers GSM having a high sensitivity and being used generally is selected as a main communication method for DAS.

4-3-3 Facility Technical Examination (2007. 1)

Facility survey and grasping if reinforcement is needed or not for the poles which automated switches would be installed was carried out in such points as taking a picture for poles, drawing a plan & input status, possibility for using the bands dimension, earthing condition and auxiliary power security. As a results of examination, installation of transformer and extension of low voltage line is needed for 8places, supplementation of different diameter of steel tubular pole is needed for 6places. These things were supplemented by PLN at March, 20th according to MOM.

4-4 Material Delivery and Transportation

4-4-1 Material Purchase and Inspection (2006. 12 – 2007. 01)

All materials to be installed in TDAS has been purchased by KDN of KEPCO subsidiary.

Main server, monitor and rack, etc. related with center control equipment was inspected by project manager with the naked eye (2006. 12). Recloser and Gas-insulated switches were inspected by Quality Inspection Center of KEPCO after the delivery (2007. 01).

4-4-2 Transportation and Customs Clearance (2007. 01 – 2007. 02)

Affairs related shipment and customs clearance for materials was carried out by KEPCO, 12 items (28boxes) including G/S and etc. were transported by ship to local. Duration for transportation took totally about 1month taking 2 weeks from the departure of Busan port (2007. 01.11) to Semarang (2007. 1.25) and 2 weeks for passing the customs. Also all shipped materials were applied without tariff according to a contract with Indonesia and after receiving all materials were kept in warehouse of PLN until installation.

4-5 Switch Construction

Automatic switches were directly constructed by PLN according to the division of the jobs agreed mutually, and KEPCO trained PLN supervisors and distribution electricians for the standard and procedures needed for construction to secure a operation efficiency. Main contents for training are composed of the installation method for switches, recloser and control box, and security method of auxiliary power, power and control cable connection, lighting arrestor installation and earthing method. Theory of these items explained with those pictures, so their understanding was high generally.

4-6 Communication Normality

4-6-1 Progressing Process with GSM Modem

1. November 2006 : Selecting the places for installation of switches and measuring for sensitivity of GSM/CDMA. GSM network shown more high quality than CDMA in sensitivity by measuring and initial investment cost for GSM network is more low because of its common use in Indonesia. Finally owing to above mentioned reasons GSM process is selected for DAS.
2. November 2006 – January 2007 : GSM modem was developed and produced with using mc-35i engine according to the result of meeting with SIEMENS KOREA network team, first test was conducted in China (January 20th, 2007) and succeeded in test between main server and FRTU (Using the CSD service).
3. February 2007 : In local test for voice call its connection was good but for data communication its success ratio was bad of 10%.
   • Analysis for reason
     1. CSD (Circuit switched data) capacity in Semarang was shortage in common testing with TELKOMSEL, so we discovered that its success ratio was low.
     2. Generally in GSM communication, VOICE capacity is the same as CSD but in Indonesia, differency between VOICE and CSD happened.
   • Solution
     1. We agreed on increasing the CSD capacity with TELKOMSEL head office (Need two weeks)
     2. We agreed on additional local test before 24, Mar. of Trial run
4. March 2007 : After revising software, second testing was in China, Success.
5. March 26th – April 4th, 2007 : After increasing CSD capacity, re-test was in Semarang
   • Poor success ratio. TELKOMSEL acknowledged to us that CSD capacity increasing is not possible any more so we changed into SMS, but we gave up it again because PLN dislike to accept SMS due to additional cost.
   • Communication method was changed into one having fixed IP (Using GPRS).
6. Early March, 2007 : Decision on communication configuration using GPRS.
   • To give main server and FRTU a fixed IP in wireless communication additional server (Radio Server) is necessary, therefore supporting fixed IP is not possible.
   • Revising the program needed and test was delayed.
   • After several meeting, final process was decided as this, through leased APN for only PLN [Terminal Modem having flexible IP-Wireless (Using GPRS) - APN - leased line - Main server having fixed IP].
7. April 19th, 2007 : Confirm a possibility of communication with the testing in TELKOMSEL head office.
8. April 23th, 2007 – 26th : Retest in Semarang branch office of TELKOMSEL.
9. May 1th, 2007 : Completion of installation of leased line in DCC.
10. May 2th – May 5th, 2007 : Test failure due to the APN network setting error of TELKOMSEL and registration failure of SIM card.
11. May 4th, 2007 : Test success after solving the reasons in clause (10) 4-6-2 Communication configuration.

4-6-2 Communication Configuration

[Figure 4-4] Communication network configuration

4-7 DB Construction (April 16th-20th, 2007)

We preferentially collected existing schematic diagram and single diagram for 9 feeders of Sayung S/S to be installed DAS and again constructed it by internal experts with the application program such as Nzed Editor and AgOLD. Re-constructed system network and single diagram was checked by the responsible person of PLN and then input them into Main server and simulator.

4-8 Main Server Installation (April 17th, 2007)

Main server was installed in DCC with UPS (5KVA) purchased in local supplier, Server, Simulator, Output equipment and Hub. Also the location for main server was changed by remodeling plan of PLN differently with original plan. UPS and battery were installed by a local supplier and the rest were installed by KDN.

In addition a leased line for communication was laid out by TELKOM with Main server.

4-9 Operation Test

4-9-1 1th Operation Test (May 4th-5th, 2007)

After succeeding communication test to GPRS, Operation Test were conducted to 15EA G/S and 3EA R/C installed on four feeders of Sayung S/S. Remote manipulation using Nzed program in DCC was followed with PLN engineers after a local manipulation. In Operation Test a Close/Open, Lock/Unlock, Current/Voltage measurement is carried out with Modem being supplied by DC24V and with interconnection to FRTU.

4-9-2 The 2nd Operation Test (June 15th-22nd, 2007)

The 2nd Operation Test was conducted on all 18 automated switches focusing on such devices as found to be unstable with poor signal quality in the 1st Operation Test. Key test activities were as follows:

1. Securing power for switches and reclosers, Measuring voltage (AC/DC) and signal quality
2. Installation of modem and antenna, Set up parameters for per device
3. Device Operation Test (remote/on-site) : Open/Close, Lock/Unlock, Battery test, Lamp test, Other measurements
4. Verifying communication program and correcting faults in programs

4-10 Operator and Manager Training

4-10-1 The 1st Training Session (May 1st-5th, 2007)

Theory education and operation practice for operator and Manager was carried out in DCC simultaneously with Trial run, this is for trial run after the successful Operation Test. 13 persons were participated in training, 9 persons of shift worker in DCC and 3persons of PLN SCADA manager. Contents for training are Nzed, AgOLD, Ag_works, DNP operation skill, FRTU manipulation and reading method, etc.

4-10-2 The 2nd Training Session (June 18th-20th, 2007)

1. Training Overview
   The 2nd operator and manager training session was performed additionally to help users improve skills in DAS manipulation and operation and operators manage and maintain DAS by enhancing their understanding of system and communication. In addition, unlike the 1st training session, the 2nd training session was attended by 2 operators and 13 users and consisted of not only pilot operation but also intensive courses on practices and theories (regarding communication and DAS) necessary for PLN to operate and maintain the system independently.
2. Training Results
   As the trainees who were young in general understood ideas fast and internalized them superbly, it is our judgment that they will be able to manipulate and operate the system without significant difficulties and it may be necessary to add more operators, arrange technology transfer program and provide opportunities for users to improve their manipulation skills by persistent exercise. Key training courses were as follows.
   1. Manual distribution : 5 manuals (15 copies in total) (Nzed, Nzed Editor, AgOLD, Simulator, Midhelper – including DNP)
   2. DAS operation manual course
   3. DAS program and device manipulation exercise using simulator
   4. Line fault diagnosis and response via fault indicator (FI)
   5. Communication status diagnosis & response to fix fault upon confirmation
   6. FRTU and modem manipulation and installation

4-11 DAS Operator Manipulation Capability Evaluation

In a meeting with the staffs of PLN Distribution Java (June 12) following the 2nd operator training session, a request for evaluating the DAS operation capability of the trainees was raised and the operation capability evaluation was conducted in DCC twice on June 28 and July 2 accordingly. Evaluation was conducted via written theory examination prepared by KEPCO and manipulation skill test using simulator with total score of 200 consisting of 100 scores for each evaluation category. Out of 13 operators in total, 10 applied for the evaluation. Evaluation description and result is as follows.

4-11-1 Key Evaluation Items

1. Basic operation knowledge of operators
2. Understanding of system operation program and parameter configuration capability
3. Line status diagnosis and response capability in case of fault
4. Device manipulation skills and familiarity with single line diagram

4-11-2 Evaluation Results

Most operators scored satisfactorily in theory examination with average score of 75 while some operators advanced in their age revealed a little more room for improvement. However, in manipulation skill test, the average score was 90, which indicated that most of the operators were well skilled in diagnosing faults and manipulating devices thanks to their experience in the existing SCADA system operation.

4-12 Trial run (July – October 2007)

Trial run was conducted for about 4 months since the Operation Test in June to allow managers and operators to develop operational capability and TDAS system to stabilize.

During the Trial run period, not only line data was acquired but also switch and recloser set values were determined based on acquired line data and resolutions to issues identified during the trial run period were prepared by analysis and review. Key observations made during the Trial run period were as below:

1. Communication status and success rate
2. Fault display accuracy, switch setting adjustment and fault description
3. Protection coordination confirmation

4-12-1 Communication Status and Success Rate

It is difficult to judge actual communication status simply with success ratio. However, as it was hard to measure communication status at all locations continuously, we confirmed communication status with success rate measured by polling of main server periodically ¹²⁾ at first. [Table 4-2] below shows fluctuation of communication success rate measured during the Trial run period.

¹²⁾ Polling : Periodic event transmission by modem upon request from main server. Polling interval is set to 1 hour in case of TDAS installed currently in Semarang, Indonesia.

[Table 4-2] Communication Success Rate during the Trial run Period

Following is the description of status analysis and resolution method for [Table 4-2] in the above.

1. As in [Table 4-2] in the above, communication status continued to deteriorate at most locations except for some from the operation test in June to August 20. The cause was found out to be the defective fiber optic cable installed in the fixed-line section (main server-Jakarta APN: owned by TELKOM, a state-owned telecom carrier) and the issue was resolved by replacing the cable. In addition, KEPCO also changed the reconnection try frequency setting of modem from 2 days to infinity to increase reconnection attempts as well.
2. However, communication status remained poor from late August to early September due to unstable communication status and the lack of understanding of communication network on the part of local telecom carrier. The success rate on Sep. 3 shown in [Table 4-2] above seemed to be satisfactory for a status indicator after resetting. However, the success rate was still unstable in comparison with the success rate on Sep. 11 and continued to fall up to significantly poor level at some locations. After conducting independent investigation and continuous discussion on communication network status with local telecom carriers (TELKOMSEL and TELKOM), we realized that there were complex causes and the major causes were found out to be the following.
   1. If data exchange between modem and main server does not occur for a certain period of time, wireless telecom carrier (TELKOMSEL) cuts the connection with APN network on its own.
   2. If the main server command fails when GPRS network (wireless section) is unstable, APN network disconnects automatically and communication fails in spite of reconnection attempt by modem.
   3. In case of loss of GPRS network connection, modem tries to reconnect to above application layer rather than trying to make PPP ¹³⁾ connection necessary for initial connection to GPRS network.
   4. Because of frequent maintenance of TELKOM network (fixed line network between Semarang-Jakarta) without advance notice, entire modems lose connectivity and fail to make reconnection.

¹³⁾ PPP : This is a protocol that is used for communication of the data in communication network, it has a benefit of multi-connection. Point-to-Point Protocol

As it is difficult for modem to maintain network connection, losing network connectivity frequently and not being able to understand communication network status because of the above reasons, KEPCO made the following arrangements to maintain communication network connection.

1. Modified modem program to generate event at regular interval (every 15 minutes) and notify APN of modem connection with main server maintain communication network connectivity continuously.
2. Modified modem program to monitor Internet and Tsel network status of local wireless network carrier (TELKOMSEL) all the time and restart modem independently and make reconnection from PPP if the data transmission network of wireless network carrier is in poor state (reconnection possible in 4 minutes at least).
3. Conducted communication quality investigation jointly with local wired/wireless network carriers and recommended as-is status investigation and communication quality improvement at vulnerable locations.

However, it is still premature to have any confidence in communication stability and observation must continue for a long time. However, if communication success rate still remains poor in spite of all such efforts, PLN must monitor the following items and perform management and monitoring activities to ensure stable communication status.

1. Confirm periodically and manage communication success rate and system operation status
2. Continue to manage and maintain on-site automation facilities (FRTU, modem and antenna status confirmation)
3. Identify poor communication quality locations by measuring signal quality periodically, develop communication quality improvement strategy with telecom carriers and continue to make improvement

4-12-2 Fault Indicator Accuracy & Fault Description

Faults break down into temporary and permanent faults in general and the Distribution Automation System (DAS) generates F.I. (Fault Indicator) for faults occurring in line to notify system operators.

In case of gas-insulated load breaker switch, automatic F.I and manual F.I are displayed against temporary and permanent faults respectively and the same F.I is generated for all types of faults in case of recloser. In addition, in case of automatic F.I, automatic resetting is performed upon fault recovery. However, in case of manual F.I, resetting must be performed manually. Therefore, recovery status must be checked before resetting even in case of line recovered automatically following momentary fault.

For F.I to be generated, following conditions must be satisfied.

1. Fault current detection and 3 phase no voltage experience
2. Fault current minimum pick-up current set value
3. No voltage : dead line or voltage below off level

F.I allows operators to identify fault section, fault current and phase with ease and will provide important input for distribution line fault analysis in the future by helping differentiate faults.

[Table 4-3-1] below shows faults found in lines where automated switches are installed in reference to the data accumulated by system during the trial run period.

[Fig. 4-5] also indicates F.I displayed against faults and follow-up actions on F.I.

[Table 4-3-1] Fault Occurrence Status per Switches

[Table 4-3-2] below shows the prediction of substation protective relay activation status per line which suffered fault in reference to F.I data during the trial run period.

Record on SYG.2 is absent but included on SYG.9 as PLN has switched the entire load of SYG.2 to SYG.9.

[Table 4-3-2] Substation Protective Relay Activation Status per Line

Numbers in ( ) represent incidents where substation circuit breaker was not activated in spite of fault occurrence as line recloser was locked out or re-closed.

[Table 4-3-3] below shows the breakdown of faults which occurred during the trial run period per F.I type.

[Table 4-3-3] F.I Breakdown per Type

4-12-3 Protection Coordination Confirmation

Activation set values of 3 reclosers are as in [Table 4-4] and accuracy and status of protection coordination between substation protective relay and recloser can be determined by status of activation triggered by actual fault. However, if substation protective relay is activated first or backup recloser is activated before downline recloser in spite of fault occurrence, protection coordination cannot be said to be accurate.

[Table 4-4] Recloser Set Values per Line

In case of ground in 143/T5-334, PU multiplier was determined in consideration of actual fault current (N phase average 700A as recorded in fault record on Sep. 11, 2007) and increased upward to ensure coordination with time-delay of downline recloser at grounding side.

[Figure 4-5] below shows the recloser activation status in relation to the faults which occurred during the trial run period and the result of activation status analysis is as follows.

[Figure 4-5-1] Initial Fault Detection

[Figure 4-5-1] in the above illustrates that automatic F.I was generated for 291/T1-24ZM switch because of temporary fault which occurred after terminal switch (291/T1-24ZM) and 2 reclosers on power side generated F.I after experiencing the fault with downline recloser (247/T1-24ZM) locked out after 2 instantaneous and as many time-delay activations in accordance with specified activation sequence. Of course, substation circuit breaker did not show any activation, which was possible as the fault section was separated by the lock out of downline recloser.

[Figure 4-5-2] Separation of Fault Section

[Figure 4-5-2] in the above depicts a status that 291/T1-24ZM switch was opened to separate normal section from fault section after fault occurrence and downline recloser (247/T1-24ZM) which had been locked out was closed again to supply load again from downline recloser to terminal switch (291/T1-24ZM).

[Figure 4-5-3] Recovery to Normal State

[Figure 4-5-3] in the above describes a status that all causes of faults are resolved and terminal switch (291/T1-24ZM)was closed to recover the line to its normal state.

[Figure 4-5] in the above indicates that the protection coordination between substation protective relay and recloser was performed correctly and the protection coordination between reclosers was also satisfactory. However, the reclosers experienced faults at the same time as the backup recloser was activated together at the 1st instantaneous activation of downside recloser. The activation of backup recloser, in spite of sequence coordination function, indicates that the protection coordination between reclosers was not perfect. However, it must be also recognized that such constraint was inevitable without increasing the pick-up current and modifying the T-C curve of substation protective relay.

If protection coordination is not performed well in addition or the following items are applicable, protection coordination must be reviewed and set values modified.

1. Substation protective relay is activated before recloser.
2. Downline recloser and backup recloser are locked out at the same time.
3. System on power side is changed.
4. System is changed significantly, involving permanent rather than temporary load switching.
5. Recloser location or protective relay is changed or a new recloser is added. However, maximum number of reclosers that can be installed per line is limited to 3.
6. Line type or length is changed.

4-13 TDAS Completion (2007. 10. 30)

TDAS completion ceremony was held in Semerang DCC with representatives from PLN, Bapenas, KOICA, KEPCO and other relevant organizations. The ceremony included not only presentation of the project overview by the project director of the pilot project but also system demonstration with simulator to provide an opportunity to experience the distribution automation system operation.