A Numerical Protection Relay Solution: Application Report
A Numerical Protection Relay Solution: Application Report
ABSTRACT
Numerical Protection Relays (NPRs) are critical elements in any power distribution network. Generally,
there are several different types of NPRs. Each type, however, shares a similar architecture, thus enabling
designers to build an entire system solution that is based on a relatively small number of flexible
components. This application note demonstrates how an NPR can be implemented using TI's extensive
portfolio of semiconductor devices.
Contents
1 Introduction .................................................................................................................. 1
2 Speed Requirements ....................................................................................................... 2
3 Accuracy Requirements .................................................................................................... 3
4 Simultaneous Sampling Requirements .................................................................................. 4
5 Component Selection ....................................................................................................... 4
6 Conclusion ................................................................................................................... 6
7 References ................................................................................................................... 6
List of Figures
1 Typical Substation in Electrical Energy Distribution Network ......................................................... 2
2 Example IDMTL Curve: Long-Time, Extremely Inverse Delay ........................................................ 3
3 Simultaneous Sampling Requirement of Two Input Signals in an NPR ............................................. 4
4 Typical Multi-Channel Overcurrent Protection Relay ................................................................... 4
5 Proposed NPR Solution .................................................................................................... 6
1 Introduction
Electrical energy generated by various electrical power stations is gathered by a central electrical grid and
then redistributed to various users based on real-time supply and demand characteristics. Have you ever
wondered what would happen when one or more of these power stations stopped producing electrical
energy, or when some of the loads suddenly demanded excessive power from the grid? As a result of the
high energy levels involved (powers measuring in MW), this imbalance generated by such sudden events
could lead to catastrophic failures.
Some of these situations are unavoidable and beyond human control. Consequently, the central electrical
grid must run a real-time algorithm to detect such fault conditions and react quickly in order to minimize
adverse impact. This monitoring function is typically managed by make-or-break contacts called
switch-gears or relays. These relays are, in turn, controlled by a smart controlling unit that continually
monitors the grid parameters (such as voltages, currents, temperature, and so forth) and switches the
appropriate relays in case fault conditions occur. Most of the data processing happens in the digital
domain; thus, these relays are often called Numerical Protection Relays, or NPRs.
Let us first review where these NPRs are used and the specifications governing the design of these
components. Figure 1 shows a typical substation in an electrical distribution system.
400 kV/200 kV
Transmission Transmission Remote
Network Substations Control
110-kV Network
Power Plants
Remote Control
Interface
Protection
Relay
Distribution
Substation
20-kV Overhead
20-kV Cable Power Line
Network Distribution
Secondary Substation Transformer
(Distribution Transformer) 230 V/400 V
230 V/400 V
2 Speed Requirements
Based on the end application and respective user geography, there are various standards such as ANSI
C37.90, IEC255-4, IEC60255-3, IAC, etc. that govern the response time of the relay to the fault conditions
that may occur. These response characteristics are generally represented by performance graphs called
Inverse Definite Minimum Time Lag (IDMTL) curves. IDMTL is the measure of an NPR response time to
the amount of fault/error.
Figure 2 illustrates one example IDMTL curve. The associated formula is given in Equation 1.
A
+B
C
tRESPONSE = DMF ·
( IMeasured ( -1
ISet
Where:
• tRESPONSE = Response time
• DMF = Delay Multiplier Factor
• IMeasured = Measured current
• ISet = Set current
• A, B, C = IDMTL Parameters (1)
Parameter
Delay Type
A B C
LTI Long-time inverse 0.086 0.185 0.02
LTVI Long-time very inverse 28.55 0.712 2
LTEI Long-time extremely inverse 64.07 0.250 2
MI Moderately inverse 0.0515 0.1140 0.02
VI Very inverse 19.61 0.491 2
EI Extremely inverse 28.2 0.1217 2
STI Short-time inverse 0.16758 0.11858 0.02
STEI Short-time extremely inverse 1.281 0.005 2
From the equation given in Equation 1, it is clear that any NPR which can support a lower Delay
Multiplication Factor (DMF) for a given curve will have a faster response time to the fault signal, and thus
will achieve a higher product rating. In order to obtain this lower DMF and correlating faster response time,
all components in the control loop must be as fast as possible without compromising accuracy.
3 Accuracy Requirements
In an NPR, the dynamic range of the input signal is often very high. For example, a typical overcurrent
protection relay will be expected to monitor currents starting from few amperes (A) all the way up to its trip
point setting, which could be in kilo-amperes (kA). It is also expected that the relay maintain its
measurement accuracy across the entire input range. Depending on the critical nature of the application,
this accuracy requirement can be anywhere from 1% to 0.05%; generally, however, the higher the
accuracy, the better the relay. To achieve higher accuracy, then, the analog front end (AFE) of the system
should have high resolution (~10-bit for 1%, ~16-bit for 0.05%), high linearity, high stability, and low noise,
even at higher sampling speeds.
Successive-approximation register analog-to-digital converters (SAR ADCs) offer zero latency and higher
linearity at higher sampling rates, compared to delta-sigma (ΔΣ) ADCs, and are therefore preferred in this
type of application. Low-noise, low-drift op amps are used in the input stage to limit noise and drift-related
errors.
Internal
Voltage Signal Impedance Postprocessing Logic
Grid
V Tripping
z=
I & Display
Sample & Analog-to-
Antialiasing Hold Digital Windowing Preprocessing DFT
Harmonic Current
Current Signal
5 Component Selection
An NPR must be fast and accurate. From the perspective of electronics design, this requirement
translates to a very high-performance AFE backed up by fast, feature-rich, digital back-end processing. In
this section, we will try to determine the exact system requirements from the individual blocks.
Based on the respective responses to certain types of fault conditions, NPRs are generally classified as
Overcurrent Protection Relays, Overvoltage Protection Relays, Earth Fault Protection Relays, Feeder
Protection Relays, or Distant Protection Relays, as well as other types. It is important to understand that
these different types of relays actually have very similar hardware configurations. The differentiation
between the various types is based on the firmware programming. For illustration purposes, let us take an
example of an Overcurrent Protection Relay.
Figure 4 shows a block diagram of a typical Overcurrent Protection Relay.
Relay Drivers
Optional
CT
Multiple Multi-channel,
Channels High-resolution.
Controller/
Simultaneous
DSP
CT sampling
ADC
Instrumentation ADC Driver
I
Amplifier Amplifier
Power Supply
Interface
Reference Voltage
Devices
5.1 Amplifiers
A current transformer (CT) is generally used to sense the current in the grid. This sensed current is then
converted to a differential voltage signal using the load resistor. It is preferable to use the smallest
possible load resistor on the secondary side of the CT in order to reduce the reflected load seen on its
primary side. Consequently, the voltage signal on the secondary side will often be as low as a few
millivolts, but it can be riding on a much higher common-mode voltage. Therefore, the component used to
amplify this signal should be able to provide high gain, but should also support high common-mode
voltages with high CMRR. Texas Instruments’ INA128 is ideal for this application.
One INA128 will be required for every CT in the system.
Based on the ADC used in the application, an extra amplifier stage may be required between the output of
the instrumentation amplifier and the data converter in order to meet the settling time parameter of the
ADC. This amplifier should have sufficient bandwidth, low drift, and low-noise performance. In order to
maximize the usable range of the converter, this amplifier should be able to support the necessary output
voltage range. Texas Instruments' OPAx277 series is very well suited for this type of application.
5.2 ADCs
The various measurements discussed earlier with the desired accuracy and speed can be summarized in
this way:
• Multi-channel, simultaneous sampling ADCs are preferred in NPRs
• Faster ADCs will allow averaging of the output codes to improve overall noise performance
• A higher resolution ADC is preferred so that quantization errors are minimal. (Typically, 12- to 16-bit
ADCs are used in NPRs.)
• ADCs that support a high input range are preferred. This parameter helps to increase the dynamic
range of the input signal, thereby improving the signal-to-noise ratio (SNR).
Based on these requirements, Texas Instruments’ ADS8556/7/8 family is best suited for this application.
5.3 Reference
The noise performance of the data converter can be only as good as its input voltage and its reference
voltage. Thus, the reference voltage used by the ADC should be accurate and stable across temperature.
Texas Instruments’ ADS8556/7/8 ADCs have an on-chip reference voltage generator. The output of this
reference can be trimmed around the center value, according to the system requirements. Once set, the
reference voltage does not drift more than ±10ppm/°C across the complete temperature range. Generally,
this degree of drift is sufficient in most applications. If higher stability is desired, the ADC can be used in
external reference mode and Texas Instruments' REF50xx series of low-noise, very low-drift (3ppm/°C
max), precision voltage references should be used to provide this reference voltage. Note that no external
reference buffer is required while using the REF50xx.
5.4 Processor
As seen in the previous example (refer to Figure 3), in order to calculate various parameters such as
active power, impedance, harmonics, and so forth, an FFT calculation is required based on the captured
data. Consequently, a DSP engine is preferred in these types of applications. In particular, floating-point
DSPs have a distinct performance advantage over fixed-point DSPs in certain applications.
In addition to the required processing power, various timers and connectivity functions in the controller are
essential in order to actuate the relays and to communicate with other units.
Because this application calls for DSP-like processing and microcontroller-like peripherals, Texas
Instruments’ C2000 family of Digital Signal Controllers is ideal for these applications.
6 Conclusion
Numerical Protection Relays are critical elements in any power distribution subsystem. In order to avoid
catastrophic failures, these relays should employ high-speed and high-accuracy electronics. A
comprehensive hardware solution has been proposed (as shown in Figure 5) by using Texas Instruments’
vast portfolio of analog solutions and embedded microcontrollers.
ULN2003
Optional
CT
I INA128 OPA277
Multiple
Channels ADS8556/
ADS8557/ TMS320F28335
CT ADS8558
I INA128 OPA277
Power Supply
ISO7221/
REF5025 ISO3082/
TRS232
7 References
1. Power System Testing Solutions: www.omicron.at
Products Applications
Amplifiers amplifier.ti.com Audio www.ti.com/audio
Data Converters dataconverter.ti.com Automotive www.ti.com/automotive
DLP® Products www.dlp.com Communications and www.ti.com/communications
Telecom
DSP dsp.ti.com Computers and www.ti.com/computers
Peripherals
Clocks and Timers www.ti.com/clocks Consumer Electronics www.ti.com/consumer-apps
Interface interface.ti.com Energy www.ti.com/energy
Logic logic.ti.com Industrial www.ti.com/industrial
Power Mgmt power.ti.com Medical www.ti.com/medical
Microcontrollers microcontroller.ti.com Security www.ti.com/security
RFID www.ti-rfid.com Space, Avionics & www.ti.com/space-avionics-defense
Defense
RF/IF and ZigBee® Solutions www.ti.com/lprf Video and Imaging www.ti.com/video
Wireless www.ti.com/wireless-apps
Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265
Copyright © 2010, Texas Instruments Incorporated