Fully digital substation
Developments in communication have done much to realize the digital substation, but to realize a full digital substation it is necessary to have everything in digital form. Whilst much substation protection, control, and automation technology has always been digital (trip signals, interlocking signals, etc.), the principal power system inputs of voltage and current have traditionally been presented in analogue form.
It has been traditional to present scaled versions of power system currents and voltages to measuring devices, protective relays etc. Scaled versions can easily be produced using conventional electrical transformers, although capacitor dividers may be additionally employed for transforming very high voltages.
Conventional transformers based on iron cores introduce measurement errors. Due to the wide dynamic range of current signals on power systems, current transformers for protection need large cores to avoid saturation under fault conditions.
Due to the nature of the magnetic core material, however, these large cores produce significant errors at nominal current, which renders them impractical for metering purposes.
Therefore metering-class transformers need to be introduced resulting in increasing costs.
Conventional wired 1A/5A current transformers (CT) circuits have thermal overload constraints, and pose increasing burdens on the core as cross-site wire run lengths increase.
This can degrade protection performance, potentially leading to the need to duplicate CTs. Conventional voltage transformer (VT) circuits may experience ferro-resonance phenomena, with thermal overstress resulting. Capacitor voltage transformers (CVT) can produce high frequency interference signals.
Techniques that do not require the iron core of conventional transducers can overcome the limitations. The solutions use different sensor technologies such as optical and Rogowski coils. In practical implementations, the techniques require sophisticated solutions employing digital signal processors and microprocessors in numerical products.
This presentation of analogue power system quantities in the form of standardized digital communication signals is the final element in realizing the digital substation.
Solutions providing signal transformation based on technology other than wound transformers are often referred to as non-conventional instrument transformers (NCIT), and the devices that provide the standardized digital communication equivalents of the power system signals are referred to as
NCIT technologies may be based on optical techniques, or Rogowski coils, and overcome the limitation of iron-cored transformers by delivering:
- Single devices providing measurement class accuracy with dynamic ranges also capable of faithfully reproducing fault currents.
- Reliable, repeatable accuracy.
- High measurement bandwidth for rated frequency, harmonics, and sub-harmonics.
- Low electrical stress insulation – no premature ageing, moisture ingress, or risk of explosion.
An NCIT device based on Rogowski coil technology is shown in Figure 3 below.
Some pilot applications using NCIT (Non Conventional Instrument Transformers) have been implemented in France and UK on actual 245 and 420 kV GIS. These field trial installations have confirmed the performances of these modern sensors, as well as the robustness of a comprehensive Protection and Metering system governed by the former applications of the IEC 61850.
Making use of Protection relays from different vendors, these pilots also proved the perspectives of interoperability, absolutely mandatory for the End users.
Despite the maturity of these technologies, the limitation of their utilization is a reality and some reasons can be reminded to explain that.
Three ways of reasons can be mentioned and many works are undertaken to overcome the issues in order to start a large deployment in the coming years:
- Technology acceptance,
- Standardization of the interface,
- Testing methods.
The current trends address a large number of different technologies and applications of the sensors like, for instance, Rogowski or optical-type current transformers, electronic or magnetic-type core for metering, capacitive effect voltage transformers of different technologies, together with demanding high specification in term of:
“Reliability, Availability and Maintenance” performance criteria.
The emergence of Non Conventional Instrument Transformers (NCIT) in the field of current and voltage measurements has been driven by the need for improved performances.
- Accuracy over a large metering range: NCIT are manufactured in series with a spread linked to dimension tolerances, which is corrected during the calibration phase using parameters held in an electronic memory;
- Non-saturation of magnetic circuits on extended metering ranges. NCIT are characterized by good linearity, both in the native state or after correction;
- No need for the considerable power occasionally required from the secondary units of conventional CT and VT;
- More compact, while allowing new metering points for a more selective protection scheme;
- Communication solutions for providing data to local or remote systems that belong to the power producer and the T&D network operator;
- New operating requirements in relation to the inter-operability / inter-changeability of components of the chain;
- Cabling simplification: in deed conventional instrument transformers are equipped with multiple secondary units and cabling is extensive with significant cross-sections. These parameters also result in current and voltage transformers that differ from one station to another.
We limit ourselves here to a reminder of the more common NCIT technologies in high voltage substations, which are suited to GIS substations.
Merging units present signals such as power system voltages and currents to IEDs within the substation, in the form of numerical values adhering to standardized definitions.
The use of NCIT sensors has made it possible for raw measurement information to be fed into so-called Merging Units for further distribution. It is these merging units that are one of the main contributors to the digital substation, and these will be discussed in the next paragraphs.
The merging unit (MU) is the interface device which takes as its inputs connections from the instrument transformer sensors, and performs signal processing to generate and distribute output sampled value streams.
As well as connecting with NCIT technology such as optical sensors and Rogowski coils, merging units can also be provided in conjunction with conventional transformer technology such that numerical equivalents of system current and voltage can be provided over communication buses.
Using merging units with conventional current and voltage transformers allows the different life-cycle expectations of primary and secondary plant to de de-coupled as per Figure 5.
Key features of merging units are:
- They can support signal processing for all transformer types including conventional and NCIT.
- They provide accurately time-stamped sampled analogue values.
- They deliver Ethernet multicast transmission of sampled analogue values via a process bus.
- They can support Ethernet connection to the station bus.
- They feature watchdog self monitoring of the NCIT sensors as well as the MU itself.
The connections to the process bus and station bus are in accordance with the IEC61850 standard that is introduced later in this chapter and is shown in Figure 6.
Each merging unit module offers signal processing to provide sampled values of phase currents (Ia, Ib, Ic), phase voltages (Va, Vb, Vc), plus residual current and residual voltage. The sampled value frames are multicast via Ethernet, using a fibre optic, or copper Ethernet connection. The outputs of the merging units need to be accurately time stamped.
This requires the merging units to be accurately time-synchronised, and this is discussed in the next section.
The accurate time synchronisation needed by the merging units is realised in the same way as for synchronising phasor measurement units (PMUs).
Synchronisation can be achieved thanks to the global positioning satellite system. Synchronising signals may either be delivered over fibre-optic links in the form of one-pulse-persecond (1pps) signals or over Ethernet according to IEEE1588.
In 1956 the American Inter Range Instrumentation Group standardized the different time code formats. These were published in the IRIG Document 104-60. This was revised in 1970 to IRIG Document 104-70, and published later as IRIG Standard 200-70.
The name of an IRIG code format consists of a single letter plus 3 subsequent digits. Each letter or digit reflects an attribute of the corresponding IRIG code.
The following tables contain the meanings of the suffixes and descriptions of the abbreviations used.
Table 1 – Serial time code formats
Table 2 – Suffix Descriptions
There are many subsets of the IRIG-B format. IRIG-B is the standard for time synchronization using 100 PPS. It was this flavour that was embraced by the utility industry to provide real-time information exchange between substations.
For IED time synchronisation, IRIG-B12x is typically used for modulated signals and IRIG-B00x for demodulated signals.
The IRIG-B time code signal is a sequence of one second time frames. Each frame is split up into ten 100ms slots as follows:
- Time-slot 1: Seconds
- Time-slot 2: Minutes
- Time-slot 3: Hours
- Time-slot 4: Days
- Time-slot 5 and 6: Control functions
- Time-slots 7 to 10: Straight binary time of day
Each frame starts with a position reference and a position identifier. Each time-slot is further separated by an 8mS position identifier.
A typical 1 second time frame is illustrated in Figure 7. If the control function or SBS time-slots are not used, the bits defined within those fields are set as a string of zeroes.
The 74-bit time code contains 30 bits of BCD time-of-year information in days, hours, minutes and seconds, 17 bits of SBS, 9 bits for year information and 18 bits for control functions.
An index marker occurs between the decimal digits in each sub-word to provide separation for visual resolution.
The SBS word begins at index count 80 and is between position identifiers P8 and P0 with a position identifier bit, P9 between the 9th and 10th SBS coded bits. The SBS time code recycles each 24-hour period. The eighteen control bits occur between position identifiers P6 and P8 with a position identifier every 10 bits.
The frame rate is 1.0 second with resolutions of 10ms (dc level shift) and 1ms (modulated 1 kHz carrier).
With IEDs interconnected via Ethernet on the station bus, and with time synchronized merging units multicasting time stamped analogue sampled values via the process bus, the building blocks for the digital substation are all in place.
The need for a common understanding between source and destination has already been highlighted.
The introduction of standard protocols such as Modbus, DNP3 and IEC60870-5-103 go some way to opening the doors to common understanding, but it was the introduction of IEC61850 that brought true possibilities for interoperability and ‘plug-and-play’ opportunities for substation automation.
The NCIT devices, developed from several different physical principles, installed as part of the GIS pilot applications benefit from very promising experience feedback in a context that is favorable to the emergence of this technology.
A large number of industrial installations, linked to the growth requirements of emerging countries, where voltage is spread mainly between 110 and 525kV, will be equipped with this technology in the near future.
Knowledge of the environment, associated with experience feedback and knowledge of the standards and the technology, make it possible to propose NCIT equipment suited to requirements.
- Network Protection and Automation Guide by Alstom
- Gas-Insulated Switchgear Type 8VN1 blue GIS up to 145 kV, 40 kA, 3150 A by Siemens
- DIGITAL SUBSTATION – Tests of Process Bus with GIS Non Conventional Instrument Transformers by D. CHATREFOU, J.L. RAYON and C. LINDNER