You are on page 1of 39

Work In Progress

Eastern Interconnection Phasor Project


Part 1: Targeted Applications: Raw Data Utilization
E I D

Mas sena

E I D

E I D N iagara obinson R oad R

E I D E I D

MA R C Y 765 MA R C Y 345

E I D

FR A S E R

E I D

O range, A E P

E I D

C allaw ay, A meren

E I D

R us h Is land, A meren
E I D

K anaw ha R iver

E I D

R ockport, A E P

E I D E I D

P aradise Fossil P lant

S haw nee Fos sil P lant E I D Mars hall


E I D

J ack sons Ferry, 765 k V


E I D

Montgomery
E I D

E I D E I D

C umberland Foss il P lant


E I D E I D E I D

S ullivan, 500 kV

W eak ley

W ilson, TN
E I D

DIEavidso n H ook , TN D P in Johns onv ille Fos sil P lant Maury

E I D

E I D V olunteer B ull R un Fossil R oane

E I D E I D

E I D Jac kson, TN H ayw ood, TN S W S TA

E I D

S helby C ordova
E I D R ock S pring Metering B ellefonte N uc lear

E I D E I D

Freeport
E I D

M adison
E I D

E I D

P leas ant H ill, MS U nion

E I D E I D

Trico S teel Trinity

E I D

S heridan, A R

E I D

E I D

Limes tone, A L
E I D

B ow en

E I D

E l D orado, A R

E I D E I D I D E

Low ndes , MS W est P oint

Fa rley

E I D

Miller

E I D

Frank lin, MS

E I D

H artburg, TX
E I D

W aterford

Standards and Performance Task Team


Prepared by A. P. Meliopoulos, CERTS George Cokkinides, CERTS Georgia Institute of Technology, Atlanta, Georgia Bruce Fardanesh, NYPA, Chairman of S&PTT Henry Huang, CERTS Matthew Ford, SEL Fahrudin Mekic, ABB Contributors/Task Team Members Ullattil Manmandhan, ABB Ray Hayes, AEP Matt Donnelly, CERTS

Standards and Performance Task Team


Scope: The scope of the Standards and Performance Task Team includes coordinating and acting as liaison to standards efforts, determining consistent and satisfactory performance of synchronized measurement devices, and insuring the security of the data in accordance with best practices and the terms of the data sharing agreements. The task force has identified potential applications of phasor data. Each application places specific requirements on hardware and firmware of GPS-synchronized equipment. The following four classes of applications have been identified ordered in increasing hardware/software feature requirements: 1. Streaming data to displays (or utilization of raw data) 2. Data filtering via state estimation methods 3. Real time dynamic analysis of system performance 4. Relaying applications with GPS-synchronized data The goal of the task force is to define hardware and firmware requirements for each application class. Another goal is to determine testing procedures for evaluating existing GPS-synchronized equipment as well as future products. The goals of the task force will be achieved with two documents. The first document (present) is focused on determining the requirements for the first application class. The second document will cover the other application classes.

Table of Contents
Definitions ____________________________________________________________ 4 1. Introduction _________________________________________________________ 8 2. Data Accuracy ______________________________________________________ 12 3. Reliability of Data Delivery ____________________________________________ 14
Data Reliability of GPS-synchronized Equipment _______________________________ 14 Data Access _______________________________________________________________ 14 Communications Standards _________________________________________________ 15 Communication Layers _____________________________________________________ 15 Physical Communications Layer _____________________________________________ 15

4. Suggested Minimum Requirements _____________________________________ 17


GPS-synchronized Equipment _______________________________________________ 17 Phasor Data Concentrators (PDC) ___________________________________________ 18 Databases ________________________________________________________________ 19 Communication Network ___________________________________________________ 19

5. Characterization of GPS-Synchronized Measurement Devices________________ 20 6. Characterization of Instrumentation Channels ____________________________ 22 7. EMC Issues ________________________________________________________ 23 8. References _________________________________________________________ 24 Appendix A: GPS-Synchronized Measurement Devices _______________________ 27 Appendix B: Instrumentation Channel Characterization ______________________ 28
B.1 CT Steady State Response _______________________________________________ 28 B.2 PT Steady State Response________________________________________________ 28 B.3 CCVT Steady State Response_____________________________________________ 29

Appendix C: SuperCalibrator ____________________________________________ 33 Appendix D: Testing of GPS-synchronized Equipment ________________________ 36

Definitions
This section provides some useful definitions pertinent to GPS-synchronized devices, communication protocols and communications media. DFR Digital Fault Recorder DDR Dynamic Disturbance Recorder SER Sequence of Events Recorder PMU Phasor Measurement Unit. A device that samples analog voltage and current data in synchronism with a GPS-clock. The samples are used to compute the corresponding phasors. Phasors are computed based on an absolute time reference (UTC), typically derived from a built in GPS receiver. PDC Phasor Data Concentrator. A logical unit that collects phasor data, and discrete event data from PMUs and possibly from other PDCs, and transmits data to other applications. PDCs may buffer data for a short time period but do not store the data. Relay An electromechanical or electronic device applied to the purpose of power apparatus protection. A relay typically monitors voltages and currents associated with a certain power system device and may trip appropriate breakers when a potentially damaging condition is detected. IED Intelligent Electronic Device. A general term indicating a multipurpose electronic device typically associated with substation control and protection. UTC Coordinated Universal Time (initials order based on French). UTC represents the time-of-day at the Earth's prime meridian (0 longitude). IRIG-B Time transmission formats developed by the Inter-Range Instrumentation Group (IRIG). The most common version is IRIG-B, which transmits day of year, hour, minute, and second once per second, over a 1 kHz carrier signal. GOES Geostationary Operational Environmental Satellites. Operated by the National Oceanic and Atmospheric Agency (NOAA). Two GOES satellites broadcast a time code referenced to UTC. Clocks based on this transmission are accurate to 100 microseconds. GPS Global Positioning System. A satellite based system for providing position and time. The accuracy of GPS based clocks can be better than 1 microsecond. pps Pulse-Per-Second. A signal consisting of a train of square pulses occurring at a frequency of 1 Hz, with the rising edge synchronized with UTC seconds. This signal is typically generated by GPS receivers.

kpps One thousand pulses per second. A signal consisting of a train of square pulses occurring at a frequency of 1 kHz, with the rising edge of synchronized with UTC milliseconds. This signal is typically generated by GPS receivers. Sampling Rate The number of samples (measurements) per second taken by an analog to digital converter system. Navigation The mode in which GPS receiver has locked onto signals from three or more satellites thus providing accurate time, as well as position. COMTRADE-file format COMTRADE file format is a standardized ASCII text or binary file (2 formats), originally designed for Digital Fault Recorders. It can be used to transfer locally recorded values from a PMU over to the central data storage. COMTRADE ASCII format is not efficient for long-term data storage but could be used for event file retrieval. PhasorFile A binary storage format that is used by PDC for long-term storage of SynchroPhasor data. Currently, this format is not standardized, and may be left in such a state as long as stored data is made available in an industry standard format (i.e. COMTRADE).

Communications Related Terms


Unicast UDP transmission from one host to another (source/destination). When talking about a link on a network, typically a unicast link is inferred. Broadcast Data transmission from one host to many. The destinations will be all computers on a network, for example, all the computers in the office building. With broadcast, every computer on the network must be trusted. Multicast Data transmission from one host to many. Data is transmitted to a group IP address. Any member of the group can access the address to receive the data. Anybody can then join in this multicast group, and when a server sends to the group, everyone in the group will receive the data. The advantage is that it is very simple to set up groups. WAN Wide Area Network, stretching across large geographical distances. Latencys can be very large, up to many seconds. LAN Local Area Network, within a small building or office. Very low latency between endpoints on the network, often less then a couple ms. VLAN A simulated LAN that is spread across a LAN, but uses special IP addresses so that it appears a physically separate LAN.

Lossless data compression Data stored in a lossless format can be retrieved exactly as it was created. Lossy data compression Data that is stored in a lossy format will have degraded accuracy when retrieved. However, more data can be stored in the same amount of space. Lossy data compression should be applied with caution and is not expected to play major role in synchrophasor data collection.

Communication Protocols Terms


IEEE 1344 A highly efficient protocol for real time SynchroPhasor data. Typically data is streamed in this format over UDP/IP or across a serial link. BPA/PDCStream A variant of IEEE 1344, widely used by the BPA PDC and HMI software on the West Coast. IEEE C37.118 Related to IEEE 1344, but adds much needed capability. This protocol and its associated standard are intended to replace IEEE 1344 and the BPA/PDCStream protocols. Typically data is streamed in this format over UDP/IP or across a serial link. OPC DA (Open Process Control Data Access) OPC was created for industrial automation, for use within a factory, for example. It is designed to share simple data between computers running only Microsoft Windows. There are 3 revisions that are commonly used. Different revisions are generally not compatible. This protocol is useful for simple data sharing between computers in a small LAN, but has serious security and performance issues when deployed across a WAN. OPC uses TCP/IP the underlying link. OPC HAD (Open Process Control Historical Data Access) an offshoot of OPC DA which allows a client to request stored data. This is a separate protocol, and different servers/clients must be developed. OPC AE (Open Process Control Alarms and Events) an offshoot of OPC DA, which allows clients to be notified on alarm conditions. As with OPC HAD, this is a separate protocol, and different servers/clients must be developed. OPC XML DA (Open Process Control XML Data Access) An OPC DA protocol designed for use across a WAN. This protocol uses the standard Web Services structure, using SOAP and XML. This protocol is simple to work with and will allow PMU devices that dont run Windows as an operating system, to be an OPC server for providing data to any client. The OPC Foundation is creating a Unified Architecture using the XML-based structure as the foundation for future development. TCP/IP TCP/IP is a low-level protocol for use mainly on Ethernet or related networks. Most of the higher-level protocols use TCP/IP to transport the data. TCP/IP provides a highly reliable connection over unreliable networks, using checksums, congestion 6

control, and automatic resending of bad or missing data. TCP/IP requires time to handshake new connections and will block if missing data is being resent. UDP/IP UDP/IP is a low-level protocol that is typically unreliable. However it provides low-latency communication across Ethernet or related networks. UDP/IP does not provide any error-control or resending of missing or bad data. The Application will need to check data for correctness. UDP/IP however, does not require time for handshaking and will not block, making it ideal for real-time data communications. HTTP HTTP is a protocol made popular by the Internet and web pages. Web pages are transmitted using HTTP. It has also become the mechanism for the Web Services Paradigm using SOAP and XML. HTTP uses TCP/IP as the underlying protocol. FTP FTP is the file transfer protocol. It is a simple protocol where a client can connect and request a file to be downloaded. A separate data connection is automatically created where the data is then transferred across the network while the command connection becomes unavailable. FTP is commonly used to get recorded data from devices.

1. Introduction
Digital data acquisition equipment with GPS synchronization adds another dimension to the utilization and application of the data. The technology is young and as such the performance of similar equipment from different manufacturers varies. Yet, for the smooth development of applications using this data in a multi-vendor environment, it is necessary to develop standards that will accommodate the rapid development and deployment of such applications. The purpose of this document is to define the desirable performance characteristics of GPS- synchronized devices. It is very common to refer to these devices as Phasor Measurement Units (or PMU) because of the fact that the first application of the GPS-synchronized measurements was to provide the phasor of the positive sequence components. Today, it is not appropriate to use this term. Indeed these devices are simply data acquisition units with the capability to time tag the data with GPS time precision, i.e. better than one microsecond. The applications of utilizing this data go beyond the initial objective of computing the phasor of the positive sequence component. It is important to recognize that the GPS-synchronized measurement devices are normally part of the overall data acquisition and utilization system. Figure 1.1 illustrates the basic components of the overall system as it may appear in a present application. The equipment comprising the system is organized in three layers: (a) the measurement layer, consisting of instrument transformers, instrumentation cables, burdens, signal conditioning circuits and GPS-synchronized equipment (PMUs), (b) a data collection layer, consisting of equipment that receive the data, align the data and transmit them to the various application layers, and (c) application layers, which utilize the phasor data to generate displays, view events, archive data, perform state estimation, dynamic analysis and other tasks. Our objective is to characterize the performance of the overall system. There are two major issues associated with this system: (a) what is the accuracy of the collected data? and (b) what is the reliability of the availability of the data for various applications. The accuracy of the data is determined at the measurement layer. Specifically observe that the measurement layer contains analog and digital systems. Once in digital form, there is no further accuracy deterioration. Therefore, accuracy issues are confined on the analog system, i.e. the instrumentation subsystem. The reliability of data availability depends on the communication channels and software. Both of these questions are addressed in this document. It is important to recognize that the requirements for each one of the posed questions depend on the end application or applications for this system. This is the reason why several application layers have been defined. The first application layer is to utilization of the system for streaming data (and all display, visualization and animation of the data) as well as event capturing and visualization.

It is also important to recognize that the system beyond the data acquisition unit (PMU) is a digital system and as such it is flexible. This means that in many cases it is possible to enhance the performance of the system with additional software development. Such possibilities are identified in this document. It has been mentioned that the question of data accuracy is confined within the measurement layer. The objective of this document is to identify the performance criteria of the measurement layer, identify the sources of errors and characterize the overall performance of the various components of the measurement layer as well as the overall measurement layer for specific applications. The GPS-synchronized equipment may be part of an information network so that the information can easily become available to (a) users and (b) application programs. It is important to note that GPS-synchronized measurement technology and in particular communications and protocols are in a flux as advances provide better products and better approaches. Therefore it is important to recognize that maximum flexibility should be maintained to adapt to new products and approaches as they become available.

Measurement Layer
Phase Conductor Current Transformer Potential Transformer Burden PMU Vendor C Optional Analog Signal Conditioning PMU Unit Vendor A

Data Collection Layer


Data Concentrator

Data Concentrator

Instrumentation Cables

PMU Vendor B Burden

Data Proprietary Concentrator Protocol

WAMS Application/Energy Management Layer


Comtrade '91 PC37.118/OPC

VPN/Internet
Database Data Archiving
1 ade '9 Comtr
OP C

3 PC

7.1

18

Planning Model Validation Real-Time Data Firewall Global Trigger System Reader SCADA State Estimator

Figure 1.1: Typical Instrumentation Setup Figure 1.1 illustrates the basic components of the overall system as it may appear in a present application. Note that the PMUs may be communicating to Data Concentrators via a standard protocol. A PMU may also directly communicate with the VPN/Internet infrastructure. There is also the possibility that a MPU may communicated with a data concentrator via a proprietary protocol. The last approach is not recommended. As a matter of fact, the recommended architecture for the Phase 1 EIPP is illustrated in Figure 1.2.

10

Measurement Layer
Phase Conductor Current Transformer Potential Transformer Burden PMU Vendor C Optional Analog Signal Conditioning PMU Unit Vendor A

Data Collection Layer


Data Concentrator

Data Concentrator

Instrumentation Cables

PMU Vendor B Burden

Data Concentrator

WAMS Application/Energy Management Layer


Comtrade '91 PC37.118/OPC

VPN/Internet
Database Data Archiving
1 ade '9 Comtr
OP C

3 PC

7.1

18

Planning Model Validation Real-Time Data Firewall Global Trigger System Reader SCADA State Estimator

Figure 1.2: Recommended Instrumentation Setup Figure 1.2 illustrates the basic components of the recommended overall system for phase 1 of the EIPP system. Note that the PMUs should communicate with the Data Concentrators via the recommended protocol to provide streaming data. A PMU may also directly communicate with the VPN/Internet infrastructure via standard protocols.

11

2. Data Accuracy
GPS-synchronized equipment has the capability to provide a data acquisition system with the following precision: 1. Time tagging with precision better than 1 microsecond (or equivalently 0.02 degrees of phase at 60 Hz). 2. Magnitude precision of 0.1% or better. This precision cannot be achieved for the overall system in any practical application, i.e. in the substation environment. In addition, depending on the implementation approach and equipment used, the accuracy of the collected data and the reliability of the data availability may differ. Typical GPS synchronized equipment (PMUs) are very accurate devices. However, the inputs to this equipment are scaled down voltages and current via instrument transformers, control cables, attenuators, etc. We collectively refer to it as the instrumentation channel. The instrumentation channel components are typically less accurate. Specifically, potential and current instrument transformers may introduce magnitude and phase errors that can be magnitudes of order higher than the typical PMU accuracy. Although, high accuracy laboratory grade instrument transformers are available, their application in substation environment is practically and economically infeasible. For the first layer of applications (i.e. data streaming and event capturing) the typical accuracy of the overall system may be acceptable. Specifically for this layer of applications, it will be assumed that the requirements are: (a) recording of phasor quantities (positive sequence or segregated phase), (b) time precision of several microseconds and (b) magnitude precision of 1%. For the phase 1 of the EIPP system we recommend adaptation of the IEEE Standard C37.118 that specifies that the total vector error should not exceed 1%. It is noted that for applications beyond the layer 1, a total vector error of 1% may not be acceptable. For these applications the data accuracy can be improved with two distinct approaches: (a) improving the hardware or (b) by software-based correction methods. The first approach is a practical impossibility. The second approach is described in Appendix C. Specifically, it is proposed to use a super-calibration approach based on state estimation. The proposed architecture for implementing the super-calibrator is illustrated in Figure 2.1. Specifically, Figure 2.1 illustrates an alternative organization of the overall system, which facilitates the super-calibrator implementation. It includes a Phasor Data Concentrator (PDC) at the measurement layer. The best placement of this PDC is at the substation. It is also proposed that the PDC receives data from all GPS-synchronized equipment (PMUs) and other IEDs existing in the substation. The super-calibrator PDC will apply error correction on the collected phasor data and align the data before 12

streaming the data to the system data collection layer. The data error correction and data alignment will be based on a state estimation approach using mathematical models of all the substation equipment.
Measurement Layer
Phase Conductor v(t) Potential Transformer i1(t) i2(t) i(t) Current Transformer Relay Vendor C
Data Processing

IED Vendor D

Data Collection Layer


PDC

Attenuator Burden PMU Vendor A

Substation PDC

PDC

Instrumentation Cables v1(t) v2(t)

Attenuator

Anti-Aliasing Filters PMU Vendor C Proprietary Protocol

PDC

Burden

WAMS Application/Energy Management Layer


Comtrade PC37.118/OPC

VPN/Internet
Database Data Archiving
ade Comtr
OP C

3 PC

11 7.

Planning Model Validation Real-Time Data Firewall Global Trigger System Reader SCADA State Estimator

Figure 2.1: Alternative Approach to Instrumentation Setup Figure 2.1 illustrates the basic components of the recommended overall system for phase 2 of the EIPP system. Note that it is recommended that a data concentrator be placed at each substation (substation data concentrator) to handle the data from all PMUs, IEDs, relays, etc. The substation data concentrator could communicate with other data concentrators. The substation data concentrator may also directly communicate with the VPN/Internet infrastructure via standard protocols. The advantages of the proposed phase 2 EIPP system are: (a) the substation data concentrator can provide proper filtering of the data via the super-calibrator (see Appendix C), and (b) the substation data concentrator can fuse data into a minimum set of unique data, i.e. bus voltages, circuit currents, etc. but eliminating the redundant data. Best approach is via the functions of the supercalibrator. This fusing of the data will drastically improve the throughput of the system.

13

3. Reliability of Data Delivery


The reliability of data delivery depends on two requirements: (a) ability of the GPSsynchronized equipment to output correct data continuously for example, some equipment may interrupt phasor computations for a short time, and (b) ability of the communication layers to deliver data on-time. Standards for item (a) above do not exist. Standards for item (b) exist in various forms and in general are very complex. Discussion of these issues is provided below. For the purpose of setting a goal regarding reliability of data delivery, the following requirements will be placed for layer 1 applications: 1. Correct data streaming from GPS-synchronized equipment will not be interrupted for more than 0.5 seconds and no more than once in 10 minutes. 2. The reliability of the communication layer should be better than 0.999 with the ability to locally store data and avoid data loss. This section presents discussion and evaluation of various options that affect reliability of data delivery.

Data Reliability of GPS-synchronized Equipment


Present GPS-synchronized measurement equipment has different methods of data generation priorities. For example dedicated GPS-synchronized equipment may have independent channel for each measurement and computation of phasors frequency and rate of change of frequency that is uninterrupted from other calculations. Other equipment may use multiplexing and the priority of the phasor, frequency and rate of frequency change computational algorithms may be affected by other functions. If the priority of such computations is suspended, the accuracy of the data will be also suspended.

Data Access
Present GPS-synchronized measurement equipment has different methods of data access and different levels of data availability. For example one manufacturer outputs only the phasors without access to the data of all three phases or point-on-waveform data. Some devices have RS232 ports, other Ethernet, etc. Some devices are interfaced to data concentrators at the substation and the data concentrator is charged with communications. With the EIPP scheme, each GPS synchronized measurement device should communicate with the PDC and each PDC should communicate with other PDCs. In other words there are several levels of communications.

14

Communications Standards
Within WECC, much work has been performed that resulted in standards for communicating and transferring data. The C37.118 IEEE Standard defines the data formats for PMU to PDC communications. WECC has also defined a standard protocol for PDC to PDC communications. These standards are usable today and should be retained. There is also activity in defining communications protocols for more general applications. The IEC Technical Committee 57 Working Group 03 (TC57 WG03) has the responsibility of developing a standard for telecontrol, teleprotection and associated telecommunications for electric power systems. The group has created the IEC 60870-5, a group of five utility-specific protocol standards. The IEC standard is being accepted by all major manufacturers of intelligent equipment for utility applications. It is important to note that there is a way to merge the C37.118 standard within the IEC 60870. The issue of network and protocols is a very complex issue and a moving target. There is a great deal of activity that has resulted in several interoperability projects and demonstrations. On the other hand BPA/PNNL has invested a great deal of effort in developing communications/network/protocols that are close to some standards but not interoperable with the new IEC standard. The issue of standardization should be extensively discussed and decisions taking early in the game. One way to approach it is to develop filters/converters at the PDC level that will make the EIPP interoperable with other IEDs. It is important to recognize that the deployment of relays with GPSsynchronization is presently occurring and it will proliferate in the future. If we tap on this resource, it will dramatically boost the EIPP project. As the industry is moving in this direction, the two major issues will be: (a) interoperability (the relay manufacturers are addressing this issue) and (b) throughput.

Communication Layers
Data communications are usually organized as multilayer entities. The lowest layer upon which all others rest is the physical layer. A multitude of layers exits above the physical layers, where various protocols are implemented. Detailed analysis and evaluation of each communication layer is beyond the scope of this document. However, it is useful to touch on the requirements of the bottom layer the physical layer.

Physical Communications Layer


There are several options for implementing the physical communications layer of the data transfer:

15

Serial asynchronous communication channels (RS-232, 9.6 to 64 kbps) 10/100 Mbps Ethernet network interface SONET

The RS-232 standard is relatively simple and has been effectively used for many years in transferring data from PMUs to data concentrators. The main drawback of the RS-232 based communications is limited bandwidth. Table 3.1 illustrates the bandwidth requirements of PMU data transfers as a function of number of channels, and data format. The communications channel bandwidth is expressed in kilo-bits per second (kbps). The table assumes a rate of 60 phasors per second for each channel. The data from each sampling instant are organized into packets which include the phasor values as well as additional information such as PMU identification and status codes as defined in the IEEE C37.118 Standard. Table 3.1: Required Communications Bandwidth for PMU Data Number of Phasors 2 4 6 12 24 100 Required Bandwidth (16 Bit Integer Format) 15.3 kbps 19.2 kbps 23.0 kbps 34.6 kbps 57.6 kbps 203.5 kbps Required Bandwidth (32 Bit Floating Point Format) 21.1 kbps 28.8 kbps 36.5 kbps 59.5 kbps 105.6 kbps 397.4 kbps

Comparing the required bandwidth to that of a standard 56kbps RS232 port it is concluded that RS232 is usable for up to 24 channels for 16 bit integer phasor representation, or 12 channels for 32 bit floating point phasor representation. Thus, for a higher number of channels a different communications hardware layer must be selected. For example, it is estimated that a 10Mbps Ethernet will support up to 1500 phasors originating from 150 different PMUs at a rate of 30 samples/sec, while a 100Mbps Ethernet will support 15,000 phasors originating from 1500 PMUs. It is expected that due to bandwidth requirements, data links beyond the PMU to local PDC will require faster communication hardware layer such as 10/100Mbs Ethernet. It is important to note that the Ethernet network to be used for real-time phasor data communications must be free of data transfer latencies. It is possible to deploy such networks with proper equipment and protocol selection. The details of this process are beyond the scope of this document and are best handled by IT staff.

16

4. Suggested Minimum Requirements


This section provides the minimum requirements for the various components of the phasor measurement network. It is based on the discussions and analysis of the issues presented in the previous sections. The suggested minimum requirements are deemed adequate for the layer 1 applications of this system, namely streaming of phasor data and event capturing.

GPS-synchronized Equipment
Table 4.1 summarizes the minimum requirements for phasor measurement units. Table 4.1: Phasor Measurement Equipment Suggested Minimum Requirements Specification
Synchronization Mechanism Measured Quantities Frequency Response

Suggested Minimum Requirement


Data synchronized with UTC time from a GPS receiver with accuracy of 1 microsecond or better. Voltage phasor, Current phasors, Frequency (f), rate of frequency change (df/dt) At least 5 Hz . -40 dB at frequencies above the sampling Nyquist frequency, -60 dB at power system harmonic frequencies Minimum of 1440 samples per second 12 Bits Support IEEE 1344, upgrade to IEEE C37.118, Default: 30 samples per second IEEE1344, upgrade to PC37.118 in 2005 if available Over-Current, Under/Over-Voltage, Under/Over-Frequency, Under/Over df/dt, status change New Standard under development (WG Report: File naming convention for time sequence data) COMTRADE Phasors, Optionally waveforms, Individual phase phasors 10 cycles of pre-trigger data, 1 second total duration 30 seconds of pre-trigger data, 120 seconds total duration 1000 events 14 days Ethernet (Optional leased line, dial-up, microwave) UDP, (Optional TCP/IP)

Raw Data Sampling Rate A/D Conversion Resolution Phasor Data Rate Streaming Data Format Event Triggering Naming of Event Files Event Data Formats Event Data Types Event Data Record Lengths (Waveform Data) Event Data Record Lengths (Phasor Data) Event Data Retention Continuous Data Retention (Phasor Data) Data Link Protocol

17

Phasor Data Concentrators (PDC)


A Phasor Data Concentrator is a logical unit that collects phasor data, and discrete event data from PMUs and other PDCs, and transmits data to other applications. PDCs should have storage capability to buffer data for a reasonable time to allow data alignment and other vital tasks. Thus, a PDC is capable of receiving, aligning, storing and transmitting GPS-synchronized data. Table 4.2 below summarizes the minimum requirements for a phasor data concentrator. The minimum requirements enable applications of streaming phasor data and event capturing. Table 4.2: Phasor Data Concentrator Suggested Minimum Requirements Specification
Input Data Format Output Data Format

Suggested Minimum Requirement


IEEE1344, upgrade to PC37.118 in 2005 if available Optional: COMTRADE, OPC COMTRADE. IEEE 1344. Upgrade to PC37.118 in 2005 if available. (Optional: PDC Stream, PDCxchng) Adopt BPA standard* It should support IEEE1344 and PC37.118 (future). Default value: 30 samples per second User Defined Configuration 32 Days

Data Alignment Output Data Rate Streaming Channels Continuous Data Retention

* We recommend adaptation of the BPA standard described in reference [19]. It is important to note that it is possible that a PDC may receive data from PMUs from different manufacturers. Aligning data from different manufacturer PMUs may be a complex task that requires knowledge of the characteristics of each unit. For application level one, the alignment of data is done on the basis of the time tag that each PMU data has. This may result in misalignments of several microseconds. For streaming data applications and event capturing applications, this misalignment is not critical. For other applications it may be critical.

18

Databases
A database resides locally with the PDC to read record store and manage phasor data. Table 4.3 below summarizes the minimum requirements for a phasor database. The PDC software should be flexible to accommodate user selections. Table 4.3: Phasor Database Suggested Minimum Requirements Specification
Data Recording Organization Data Retention Data Compression Input Data Format

Suggested Minimum Requirement


Circular Buffer (new data overwrites oldest data) Length of buffer should be user selected Archive important data Lossless Compression OPC, PDCstream, COMTRADE. IEEE 1344. Upgrade to PC37.118 in 2005 if available Optional: (Optional PDCxchng, Phasorfile, XML) OPC. Optional: IEEE 1344, PC37.118, PDCstream, COMTRADE, PDCxchng, Phasorfile, XML Yes

Output Data Format Continuous Recording

Communication Network
The network comprises the physical communication paths for phasor data exchange. Table 4.4 summarizes the minimum requirements for a network. Table 4.4: Communication Network Suggested Minimum Requirements Specification
PMU to PDC PDC-PDC DB-DB Bandwidth PMU to PDC Security Latency Protocols

Suggested Minimum Requirement


Ethernet (Optional dial-up, microwave) Leased Line Internet (Optional NERCNet) 36800 baud (Optional 10/100 T-Base) VPN/Internet 20-100 milliseconds See section 3

19

5. Characterization of GPS-Synchronized Measurement Devices


Equipment for synchronized measurements from various vendors may have different designs and therefore different ways of data acquisition and processing and different precision characteristics. As an example, in a recent characterization of devices from two different manufacturers we have found a difference of one degree at 60 Hz (this is equivalent to 46 microseconds error). This error can not be ignored. When all PMUs come from the same manufacturer, the systematic errors are irrelevant. However, in a multi-vendor environment (which is the case presently, and as more manufacturers will start offering GPS synchronization) this issue must be addressed. It is important to note that PC37.118 provides instructions for compliance verification. As a minimum PMU should be PC37.118 compliant. The objective of this section is to identify the issues related to data acquisition, data processing and data output of GPS-synchronized devices. The discussion in this section will be used as a blueprint for testing GPS-synchronized equipment. The following performance issues will be addressed: Data Types Phasor (positive sequence) Phasor (segregated phases) Time samples I/O Characteristics Reporting rates Computational algorithms Streaming data capability (throughput capability) Supported formats Time tagging Time tagging precision (true synchronization/multiplexing) Precision characteristics Frequency error Magnitude error Phase error Total Vector Error Transfer function Output time latency Disturbance recording capability

20

Trigger options Data length capability Throughput capability Phasor computation priorities Determine whether phasor computations may be disrupted for any reason This section describes the known approaches and the characteristics of each approach relative to the above items. Appendix A provides a list of US products and their major specification. The list in Appendix A will increase in the near future as more manufacturers planning the release of new products with GPS-synchronization. In addition, many foreigner manufacturers are already offering GPS-synchronized equipment. In view of the multiplicity of manufacturers, it is necessary and recommended that GPSsynchronized equipment be tested to quantify their performance. Testing results are available in the literature and from the manufacturer. It is also recommended that a standard testing procedure be developed for characterizing this equipment. Appendix E provides a proposal for this standard testing.

21

6. Characterization of Instrumentation Channels


High voltage instrumentation channels introduce errors to phasor measurements. The level of error is dependent upon the type of instrument transformers, control cable type and length and protection circuitry at the input of the A/D converters. Depending on the instrumentation channel, the characterization of these errors may be possible. Reference [35] provides some additional information. In most cases these errors can be accounted for and corrected via software. Two approaches are very promising: (a) model the instrumentation channel and provide model based correction algorithms, and (b) use state estimation methods to correct the error. A combination of the two will be ideal. This issue is very important to the EIPP. Appendix B provides examples of instrumentation channel characterization and the effects on the overall precision of the GPS synchronized measurements. We recommend that further work be contacted to develop methodologies for characterizing the instrumentation channel errors and algorithms to correct for these errors. This work should be coordinated with the work on remote calibration. It should be recognized that GPS-synchronized equipment may be also connected to existing instrumentation in substations that may be for other purposes, i.e. metering. Many times the instrument transformers are connected in an arrangement that generates a phase shift, for example delta connection. The resulting phase shift must be accounted for.

22

7. EMC Issues
Synchronized measurement equipment is highly sensitive electronic systems that must perform satisfactory under adverse conditions in the harsh electromagnetic environment of a substation. There are two issues: (a) what should the withstand capability of the equipment be and (b) how they should be installed (shielding, grounding and bonding). There is a plethora of work in this area already. We recommend the following. Equipment should be designed to withstand 3 kV of transient input voltages. Equipment should be properly grounded. Equipment should have a common mode rejection of -60 db minimum.

23

8. References
1. IEEE Std 1344-1995, IEEE Standard for Synchrophasors for Power Systems, 1995. 2. C37.118, IEEE Standard for Synchrophasors for Power Systems, 2004 Draft. 3. IEEE Std C37.2 1991, IEEE Standard Electrical Power System Device Function Numbers and Contact Designations. 4. IEEE Std 1379-1997, IEEE Trial-Use Recommended Practice for Data Communications Between Intelligent Electronic Devices and Remote Terminal Units in a Substation. 5. IEEE C37.111-1999, IEEE Standard Common Format for Transient Data Exchange (COMTRADE) for Power Systems, June 1999. 6. IRIG Standard 200-98, IRIG Serial Time Code Formats, May 1998, Telecommunications Group, Range Commanders Council, US Army White Sands Missile Range, NM. 7. K. E. Martin, J. F. Hauer, W. A. Mittelstadt and A. Ellis, WECC Disturbance/Performance Monitor Equipment: Proposed Standards for WECC Certification and Reimbursement, March 17, 2004 8. IEEE-PES-PSRC, Working Group H-8, K. E. Martin, Chairman, IEEE Standard for Synchrophasors for Power Systems, IEEE Transactions on Power Delivery, Vol 13, N0. 1, pp 73-77, January 1998. 9. IEEE C37.111-1999, IEEE Standard Common Format for Transient Data Exchange (COMTRADE) for Power Systems, June 1999. 10. IRIG Standard 200-98, IRIG Serial Time Code Formats, May 1998 11. IEC 60870, Telecontrol Equipment and Systems, Transmission Protocols. 12. A. P. Meliopoulos, F. Zhang, S. Zelingher, G. Stillmam, G. J. Cokkinides, L. Coffeen, R. Burnett, J. McBride, 'Transmission Level Instrument Transformers and Transient Event Recorders Characterization for Harmonic Measurements,' IEEE Transactions on Power Delivery, Vol 8, No. 3, pp 1507-1517, July 1993. 13. A. P. Sakis Meliopoulos, F. Zhang, and S. Zelingher, 'Power System Harmonic State Estimation,' IEEE Transactions on Power Systems, Vol 9, No. 3, pp 17011709, July 1994. 14. A. Arifian, M. Ibrahim, S. Meliopoulos, and S. Zelingher, Optic Technology Monitors HV Bus, Transmission and Distribution, Vol. 49, No. 5, pp. 62-68, May 1997. 15. B. Fardanesh, S. Zelingher, A. P. Sakis Meliopoulos, G. Cokkinides and Jim Ingleson, Multifunctional Synchronized Measurement Network, IEEE Computer Applications in Power, Volume 11, Number 1, pp 26-30, January 1998. 16. T. K. Hamrita, B. S. Heck and A. P. Sakis Meliopoulos, On-Line Correction of Errors Introduced By Instrument Transformers In Transmission-Level Power

24

Waveform Steady-State Measurements, IEEE Transactions on Power Delivery, Vol. 15, No. 4, pp 1116-1120, October 2000. 17. A. P. Sakis Meliopoulos and George J. Cokkinides, A Virtual Environment for Protective Relaying Evaluation and Testing, IEEE Transactions of Power Systems, Vol. 19, No. 1, pp. 104-111, February, 2004. 18. Evaluating Dynamic Performance of Phasor Measurement Units: Experience in the Western Power System, J. F. Hauer, Ken Martin, and Harry Lee. Interim Report of the WECC Disturbance Monitoring Work Group, partial draft of April 28, 2004. 19. Synchronized System Measurement Networks in North America: Operating Process and System Formats Based Upon BPA's Phasor Data Concentrator, K. E. Martin. WAMS Working Note, June 1, 2004. 20. Integrated Monitor Facilities for the Western Power System: The WECC WAMS in 2003, J. F. Hauer, W. A. Mittelstadt, K. E. Martin, and J. W. Burns. Interim report of the WECC Disturbance Monitoring Work Group, June 25, 2003. (Available at http://www.wecc.biz/committees/JGC/DMWG/documents/). 21. "Performance of 'WAMS East' in Providing Dynamic Information for the North East Blackout of August 14, 2003", J. F. Hauer, Navin Bhatt, Kirit Shah, and Sharma Kolluri. IEEE/PES Panel on Major Grid Blackouts of 2003 in North America and Europe, IEEE PES General Meeting, Denver, CO, June 6-12, 2004. 22. "Dynamic Signatures Recorded on the "WAMS East" Monitor Backbone: August 14, 2003 vs. Other Days", J. F. Hauer, Navin Bhatt, Rajesh Pudhota, Sujit Mandal, and Michael Ingram. Presented by J. F. Hauer to the IEEE/PES IEEE Work Group on Power System Dynamics Measurements, IEEE PES General Meeting, Denver, CO, June 7, 2004. 23. Juancarlo Depablos, Virgilio Centeno, Arun G. Phadke, and Michael Ingram, Comparative Testing of Synchronized Phasor Measurement Units, Paper presented at the IEEE/PES General Meeting, Denver, CO, June 2004. 24. S. Zelingher, G.I. Stillmann, A. P. Sakis Meliopoulos, Transmission System Harmonic Measurement System: A Feasibility Study," Proceedings of the Fourth International Conference on Harmonics in Power Systems (ICHPS IV), pp. 436444, Budapest, Hungary. October 1990. 25. A. P. Sakis Meliopoulos, F. Zhang, and S. Zelingher, "Hardware and Software Requirements for a Transmission System Harmonic Measurement System," Proceedings of the Fifth International Conference on Harmonics in Power Systems (ICHPS V), pp. 330-338, Atlanta, GA. September 1992. 26. A. P. Meliopoulos, F. Zhang, S. Zelingher, G. Stillmam, G. J. Cokkinides, L. Coffeen, R. Burnett, J. McBride, 'Transmission Level Instrument Transformers and Transient Event Recorders Characterization for Harmonic Measurements,' IEEE Transactions on Power Delivery, Vol 8, No. 3, pp 1507-1517, July 1993. 27. Real-time applications task team, EIPP goals and objectives 28. EIPP evaluating dynamic performance of phasor measurement units, March 2004 29. EIPP Standards and Performance Task Team preliminary report, August 2004 30. Marzio Pozzuoli, Ethernet in Substation Automation Applications Issues and Requirements, www.ruggedcom.com

25

31. Cisco Systems, QOS Quality Of Service, web published document, http://www.cisco.com/en/US/tech/tk543/tech_topology_and_network_serv_and_p rotocol_suite_home.html, 32. IEEE Power Systems Relaying Committee, Application Considerations of UCA 2 for Substation Ethernet Local Area Network Communication for Protection and Control, http://www.pes-psrc.org/h/DRAFT%235(7-16-04)H6%20PAPER.zip, WEB published paper. 33. IEC 61850 consists of the several parts.
34. A. P. Sakis Meliopoulos and G. J. Cokkinides, Visualization and Animation of Instrumentation Channel Effects on DFR Data Accuracy, Proceedings of the 2002 Georgia Tech Fault and Disturbance Analysis Conference, Atlanta, Georgia, April 29-30, 2002.

26

Appendix A: GPS-Synchronized Measurement Devices


This Appendix lists the presently available GPS-synchronized measurement devices for power system applications. The list will always be a moving target since many data acquisition system manufacturers are adding GPS synchronization capability. Presently the US based manufacturers of this equipment are: MACRODYNE 1620, ABB 521, SEL 421, Arbiter 1133A A checklist for the major specifications of any GPS synchronized device is given in Table A-1. Table A-1. Specifications of US Based GPS-Synchronized Measurement Systems Device ID A/D Converter Technology A/D Converter Word Length Sampling Rate (samples per second per channel) Automatic Calibration GPS Time Tagging Simultaneous Sampling GPS synchronized sampling Voltage Input Isolation Current Input Isolation Anti-Aliasing Filters Communications Support of IEEE Standard 1344 Support of IEEE C37.118

27

Appendix B: Instrumentation Channel Characterization


This Appendix provides characterization of errors resulting from instrumentation channels. The instrumentation channel may be current (CT based) or voltage (PT based or CCVT based).

B.1 CT Steady State Response


The conventional CT steady state response is very accurate. The steady state response can be extracted from the frequency response of the device. Figure B.1 provides a typical frequency response of a CT. Note that the response is flat in the frequency range of interest. It is important to note that errors may be present due to inaccurate determination of the transformation ratio. These errors are typically small.
1.25 Magnitude 1.00 Phase 0.75 0 1 2 3 Frequency kHz 4 5 6 -45.0 0.0 45.0 Phase (Degrees)

Normalized Magnitude

Figure B.1: Typical 600 V Metering Class CT Frequency Response

B.2 PT Steady State Response


Wound type PTs are in general less accurate than CTs. Again the steady state response can be obtained from the frequency response of the device. Figure B.2 provides a typical frequency response of a wound type PT. Note that the response is flat in a small frequency range around the nominal frequency. Our work has shown that the higher the transformation ratio of the PT the higher the errors will be.

28

0.0012 0.0008 0.0004 0.0000 0 500 1000 1500 2000 2500 3000
Frequency (Hz)

Magnitude
Phase (Deg)

90.00 45.00 0.00 -45.00 -90.00 -135.00 0 500 1000 1500 2000 2500 3000
Frequency (Hz)

Figure B.2: 200kV/115V Potential Transformer Frequency Response

B.3 CCVT Steady State Response


By appropriate selection of the circuit components a CCVT can be designed to generate an output voltage with any desirable transformation ratio and most importantly with zero phase shift between input and output voltage waveforms. In this section we examine the possible deviations from this ideal behavior due to various causes by means of a parametric analysis, namely: Power Frequency Drift Circuit component parameter Drift Burden Impedance

The parametric analysis was performed using the CCVT equivalent circuit model illustrated in Figure B.3. The model parameters are given in Table B.1:

29

CP C1
Vin A

LC

RC RF
Cable Model

C2 LD LF

RB Vout

CF

Figure B.3: CCVT Equivalent Circuit Table B.1: CCVT Equivalent Circuit Parameters Parameter Description CCVT Capacitance Class Input Voltage Output Voltage Upper Capacitor Size Lower Capacitor Size Drain Inductor Compensating Reactor Inductance Compensating Reactor Resistance Burden Resistance Ferroresonance Suppression Damping Resistor Ferroresonance Suppression Circuit Inductor Ferroresonance Suppression Circuit Capacitor Cable Type Cable Length Transformer Power Rating Transformer Voltage Rating Leakage Reactance Parasitic Capacitance Schematic Reference Value Normal 288 kV 120 V 1.407 nF 99.9 nF 2.65 mH 68.74 H 3000 Ohms 200 Ohms 70 Ohms 0.398 H 17.7 uF RG-8 100 Feet 300 VA 4kV/120V 3% 500 pF

C1 C2 LD LC RC RB RF LF CF

CP

Figure B.4 shows the results of a frequency scan. Note that over the frequency range of 0 to 500 Hz the response varies substantially both in magnitude and phase. Near 60 Hz (55 to 65 Hz) the response magnitude is practically constant but the phase varies at the rate of 0.25 degrees per Hz.

30

Table B.2 shows the results of a parametric analysis with respect to Burden resistance and instrumentation cable length. Note that the system is tuned for zero phase error for a short instrumentation cable and with a 200 Ohm Burden. Table B.3 shows the results of a parametric analysis with respect to CCVT component parameter inaccuracies. Specifically the varied parameters were the compensating reactor inductance and the capacitive divider capacitance. Table B.2: Phase Error (in Degrees) Versus Burden Resistance and Cable Length Cable Length (feet) 1000 2000 -0.155 -0.096 -0.063 -0.047 -0.036 -0.365 -0.213 -0.127 -0.080 -0.052

Burden Resistance 50 Ohms 100 Ohms 200 Ohms 400 Ohms 1000 Ohms

10 0.077 0.026 0.000 -0.013 -0.022

Table B.3: Phase Error (in Degrees) Versus Capacitance and Inductance Inductance Error (%) 1% 5% -0.066 -0.132 -0.396 -0.331 -0.397 -0.661

Capacitance Error (%) 0% -1% -5%

0% 0.000 -0.066 -0.330

31

Figure B.4: CCVT Computed Frequency Response over 10-600 Hz

32

Appendix C: SuperCalibrator
Background: GPS-synchronized equipment receives inputs from instrument transformers, control cables, attenuators, etc. Many utilities will use CCVTs for instrument transformers. While GPS-synchronized equipment are more accurate devices than typical utility SCADA and other monitoring equipment, the overall precision of GPS-synchronized measurements is limited by the instrumentation channel. For example, while GPS-synchronized equipment has the capability of measuring the phase angle to a precision of 0.01 degrees, the instrumentation channel may introduce a phase error that may be ten times larger. In particular, CCVTs introduce an error to the measurement that is appreciable at the power frequency as compared to the precision of PMU equipment and very large with respect to PMU equipment at other frequencies. Depending on the application, this performance may be unacceptable. Conceptually, the overall precision issue can be resolved with sophisticated calibration methods. This approach is quite expensive and faces difficult technical problems. It is extremely difficult to calibrate instrument transformers and the overall instrumentation channel in the field. Laboratory calibration of instrument transformers is possible but a very expensive proposition if all instrument transformers need to be calibrated. In the early 90's I directed a research project in which we developed calibration procedures for selected NYPAs high voltage instrument transformers [1]. From the practical point of view, this approach is an economic impossibility. We propose a viable and practical approach to correct for these errors. The approach is based on an estimation process at the substation level for correcting for these errors. Specifically, we propose to develop a methodology and a software product that will perform as a supercalibrator. This product may reside at the PDC level, and it will operate on the streaming data. The process is expected to be fast and therefore it will introduce only minor time latencies. The procedure will maintain the data format including the time tags of the data. A brief description of the methodology follows. A detailed model of the substation (from which PMU data are originating) will be constructed. This model will be an integrated model, i.e. it will include the three phase model of the substation, the model of the instrumentation channels that feed inputs to the PMUs and the model of the PMUs. As the data stream, each set of data at a specific time tag will be processed via a general state estimation process. The procedure will provide the best estimate of the data as well as performance metrics of the estimation process. The most important metric will be the expected value of the error of the estimates. The software will continuously display these metrics and it will also store the computed metrics. The best estimate of the data will be used to regenerate the streaming data flow.

33

It is important to note that the proposed super-calibrator is also a tool for remote calibration. Since these equipment are digital and since the super-calibrator will determine what the reading of each device should be, a calibration factor can be inserted into each channel of the GPS-synchronized equipment. This very simple method is also very effective.

34

Measurement Layer
Phase Conductor v(t) Potential Transformer i1(t) i2(t) i(t) Current Transformer

IED Vendor D

Relay Vendor C
Data Processing

Attenuator Burden

PMU Vendor A

Substation PDC

Instrumentation Cables v1(t) v2(t)

Attenuator

Anti-Aliasing Filters PMU Vendor C

Proprietary Protocol

Burden

Appendix D: Testing of GPS-synchronized Equipment


This appendix defines testing procedures for GPS-synchronized equipment. It defines the functions/accuracy/data reliability/noise immunity etc. and test procedures to quantify the performance of the GPS-synchronized equipment to all these metrics. It provides information on the type of testing that is appropriate for PMUs. The objective of this testing is to characterize the performance of PMUs in a system wide environment. Additional testing that is not included in this discussion is: (a) environmental testing of the devices and (b) seismic testing. The test system is illustrated in Figure D.1. It is a PC based testing procedure. The PC generates the tests signals injected into the GPS-synchronized equipment under tests and also collect the data from the GPS-synchronized equipment as well as data from high precision GPS-synchronized instrumentation. The details of the hardware selection are to be determined. The PC software includes two major modules: (a) an integrated high fidelity power system simulator, (WinIGS-T) which can generate test waveforms for a variety of power system operating scenarios (steady state, faults, switching transients, harmonics etc). The simulator includes full 3-phase models of major power system components (generators, transmission lines transformers etc) including explicit representation of grounds. The models are physically based and represent phase to phase asymmetries, frequency dependent phenomena, grounding effects etc. and (b) a waveform analysis and comparison software package (XFM). This software has the capability to compute even the smallest time/phase variations between two waveforms or the smallest magnitude differences. This capability is necessary for the purposes of this project. The program XFM is fully developed. The program WinIGS-T is also a developed program however this program requires continuous development for the purpose of increasing the modeling capabilities of the program. It is important to note that the same approach can be achieved by using the PC to play back COMTRADE files into the relays. This approach will severely limit the testing capabilities of the system. Another alternate approach is to use commercially available simulators, such as the EMTDC. This approach will be costly and again it will be limited by the size of the hardware. The proposed approach is much more flexible, its implementation requires only a PC and the high precision GPS-synchronized instrumentation.

digital

Relay PC D/A
analog

PMU

digital

Devices Under Test


High Precision GPS synchronized Data Acquisition System

Test Instrumentation
Figure D.1: Test Setup for GPS-synchronized Equipment The waveforms generated by the simulator are converted to analog form via a D/A converter, and amplified to the standard voltage and current levels normally interfaced with standard digital relay and PMU inputs. There are commercially available D/A converter hardware. The digital streaming data generated by the PMU or relay under test are transferred to the same PC for accuracy evaluation. The accuracy of the phasor data is performed by comparison to waveforms simultaneously captured by a reference data acquisition system. The reference data acquisition system will be a high precision system with simultaneous sampling capability and GPS synchronization. The reference data acquisition system hardware will be selected on the basis of cost and requirements for this testing. The PMU data and reference data acquisition system data comparison is performed with the aid of a second PC Software component (Program Xfm). This program has been developed for the purpose of waveform and phasor data analysis (the first version of this program was developed for NYPA for exactly this purpose to compare waveforms. Later it was further improved with work for ENTERGY. Recently this program has been upgraded to commercial program running under any Microsoft Windows system). It accepts data in many custom/proprietary formats, including standard COMTRADE files. The software has extensive data analysis, plotting and visualization capabilities. Analysis examples include harmonic analysis, frequency estimation, fault distance 37

estimation, transformer thermal modeling etc. A notable feature that is particularly useful for the present application is the waveform calculator. This feature allows any operations among any number of the input waveforms to be easily defined by the user in the style of a RPN (Reverse Polish Notation) calculator. The proposed test setup will provide accurate performance evaluation of PMUs and Digital Relays. Specifically, it will measure both magnitude and phase accuracy, under various power system operating conditions, such as faults, harmonic distortion, frequency swings, etc. The proposed system will allow testing various GPS-synchronized equipment under various conditions. For example in case of relay equipment with GPS-synchronization, the proposed testing system it will be able to generate a plethora of disturbance data going into the relay and at the same time will characterize the phasor data coming out of the relay. This type of testing will reveal whether the phasor computations are less accurate during this period, how long it takes after the disturbance has been removed for complete restoration of phasor computations, etc. We do not know of any system that has this capability presently. It is expected that once this system is developed, it will be available to other organization that may wish to set-up their own test facility. The proposed system will address the following issues regarding GPS-synchronized equipment. Data Types Phasor (positive sequence) Phasor (segregated phases) Time samples I/O Characteristics Reporting rates Computational algorithms Streaming data capability (throughput capability) Supported formats Time tagging Time tagging precision (true synchronization/multiplexing) Precision characteristics Frequency error Magnitude error Phase error Transfer function Output time latency Disturbance recording capability 38

Trigger options Data length capability Throughput capability Phasor computation priorities determine whether phasor computations may be disrupted for any reason

39