Using Digital Instrument Transformers to Reduce Substation Design Costs

Authors: Dylan Stewart, GE, Canada, Allen Rose and Rich Hunt, GE, USA

The value in the intelligent digital substation is the ability to design and build substations faster and more efficiently, use fewer materials for the physical infrastructure of the substation, operate and maintain the substation (and substation equipment) at lower effort and cost, and unlock the value of data to improve the performance of the substation, substation equipment, and the power system itself.

The digital substation is, right now, moving from concept to reality, by expanding beyond simply replicating traditional SCADA concepts to using process bus. Process bus is nothing more than distributed I/O for protection and control, digitizing all the inputs to the protection and SCADA systems (currents, voltages, equipment status) at their source at the primary equipment.

The digital substation is essentially the virtualization of analog data. It is obvious that digitizing all analog information at the source has great value in the design and operation of protection, control and SCADA systems.

Once data is freely available and easy to access, it can be used to unlock the value the substation through better asset performance management and system operations. This leads to interesting questions: can we use the virtualization of data to improve the physical design of the substation and substation equipment? Can we go ever farther with substations in building more efficiently, using fewer materials, and maintaining them more cost effectively?

The obvious way that process bus improves physical design is by replacing the numerous copper cables required for measurements and equipment status with far fewer fiber optic cables: less design of cabling, less wiring, less installation effort. This can also permit the replacement of cable trench with simple conduit. 

A small air-insulated distribution substation may use 3,600m of copper cables across the switchyard for measurement, status, control, and power distribution, requiring investment in space and money for cable trench.

Process bus can reduce this to 400m of fiber optic cable, and 600m of copper cable for power distribution only. Beyond the savings due to less design of cabling, less wiring, and less installation effort, there is the physical savings afforded: a 0.1m diameter conduit as opposed to a 1m wide cable trench. In transmission substations, it is possible to realize even greater savings, reducing copper cables, and the facilities to support copper cables, by close to 90%. However, simply reducing copper cabling is not enough in terms of substation and equipment design. Virtualization must allow equipment to become smaller, lighter, more flexible, and more adaptable in terms of application, and easier to maintain and support.

One way to address design is to push digitizing data as close to the source as possible. Process bus starts at the terminal blocks of primary equipment, but the goal should be to remove the terminal blocks and be natively digital if possible. The place to start, because proven solutions exist, is with conventional wire-wound current transformers (CTs).

CTs are a physical device that add cost to every part of substation design. Their size and weight (which increase with accuracy class and voltage level), directly influence equipment design, equipment standards and specifications, substation layout, and substation footprint.

CTs influence operations, due to considerations around accuracy, linearity, and precision. They force engineers to mitigate the impacts of saturation and transients in both device algorithms and device settings. CTs even touch organization structure: they are primary, physical equipment that will therefore belong to the primary equipment group.

Process bus as a concept doesn’t directly address any of the issues of CTs. The present process bus design simply retains these issues. Process bus wires a merging unit, that digitizes signals, directly to the terminal blocks of the CT. Obviously, merging units are necessary because CTs are the installed base, and part of equipment specifications.

Merging units also match well with existing utility organization structures, as they maintain the boundaries between equipment, protection and control, and SCADA groups. However, non-conventional instrument transformers (NCITs) are a solution that directly addresses the limitations of conventional instrument transformers, improving substation design, substation performance, and operating and maintenance costs.

NCITs are any measurement transducer that uses a measurement technique other than the traditional CT, voltage transformer (VT), or capacitive voltage transformers (CVT). Examples include low-power CTs, Rogowski coils, fiber optic current sensors, and capacitive and resistive voltage dividers.

A digital instrument transformer (DIT) is an NCIT integrated with electronics to natively provide digital sampled values as an output.

NCITs use mature measurement techniques, that are well understood and trusted. Rogowski coils were first applied early in the 20th century. Fiber optic CTs have been used since the 1980s.

The limiting factor to their use has been the interface, as NCITs don’t produce the high energy 1A/5A/120V output that relays and meters expect. DITs remove this limit through adoption of common data formats under the IEC 61850 Standard.

One DIT, that replaces CTs with a fully digital substation, is the fiber optic current transducer (FOCT). FOCTs replace conventional terminal blocks with digital communications, and allow for smaller, more flexible substation equipment.

How FOCTs work

The design of a FOCT is very simple: it is a light source connected to a measurement fiber. The measurement fiber is wrapped around a conductor carrying current. The FOCT uses the Faraday Effect for the measurement: the magnetic field of the current flowing through the conductor impacts the speed of light in the measurement fiber. The measurement electronics simply measures this change in the speed of light. (Figure 1).

The common technique used for FOCTs is the interferometric technique, because it uses a relative measurement. A polarizer splits the light from the source into 2 separate waves with opposite polarization. The Faraday Effect speeds up one wave, and slows down the other. The current is proportional to the phase shift between the 2 waves. Because the interferometric technique is a relative measurement, this reduces impact of temperature and other environmental factors on the accuracy of the measurement.

The result is a measurement that is linear and accurate across the entire measurement range of the FOCT. This provides protection performance and metering accuracy with one measurement fiber, measuring from mA to kA. Depending on the design it is possible to measure primary currents greater than 200 kA, and provide 0.15% (revenue metering) accuracy.  (Figure 2).

The rest of the design of a FOCT is simply managing the physical installation, as in how to mount the measurement fiber around the conductor for reliability, safety, and ease of installation. It is necessary to design a standard, reliable, and safe method to install the measurement fiber around the conductor, to make the FOCT easy to manufacture, use, and order. And it is necessary to have a standard design of where to put the measurement / digital electronics, and provide the connection between these electronics and the measurement fiber in the yard. A well designed FOCT will be easy to specify, order, install, and maintain, with a design focus on reliability. (Figure 3).

FOCTs are better than CTs in terms of performance and application limits. The key detail is that the primary sensor is nothing more than a fiber optic cable. The size and weight of the cable are trivial compared to the equipment it will be mounted on. Since fiber optic cable is not a conductor, it is natively isolated.

The measurement fiber can therefore simply be wrapped around the conductor as long as good fiber management techniques are used. Due to the small size of the fiber, one sensor mounting assembly may contain multiple measurement fibers for redundant measurements, similar to a multi-core CT. A plus for applications is that the “turns ratio” that relates the output to the primary current is simply a multiplier setting that is entered into the configuration of the FOCT, and this multiplier can be set precisely for the application. FOCTs can also measure both AC and DC current, and can measure both simultaneously. This makes the FOCTs ideal for applications like detecting geomagnetically induced currents (GICs) in power transformers.

FOCTs do require electronics to perform the basic measurement. These electronics also act as a merging unit, as they publish digital samples representing the current. Since the connection from the measurement point is fiber optic, the electronics are easily installed in the control house.

Because a FOCT can measure fault currents while providing metering accuracy, a single fiber optic cable can replace up to 2 sets (one from a set of protection CTs and one from a set of metering CTs) of multi-conductor copper cables across the switchyard.

It is also important to understand that a FOCT simply measures current, and the same FOCT can measure current at any voltage, from 120V to 1100kV. The only voltage rating for FOCTs is actually a consideration for physical mounting and isolation. This allows the same FOCT to be used for any voltage class.

The best way to understand how FOCTs can improve the physical design of the substation is to look at actual, successful applications.

Dead Tank Circuit Breakers

FOCTs directly improve the application of dead tank circuit breakers in air-insulated substations (AIS) in terms of physical infrastructure and equipment design. There are a variety of methods to mount the FOCT to primary equipment. One method is a slipover bushing FOCT. The bushing mount FOCT simultaneously provides current measurements for protection and metering over a single fiber optic cable. Redundancy can be achieved either through a second slipover bushing mount, or by designing the slipover bushing mount with two measurement fibers. (Figure 4).

Using FOCTs with dead tank circuit breakers improves both equipment and substation design. The biggest benefit is in equipment design. The same FOCT can be specified on all circuit breakers, for any application or voltage class.

There is no risk of the FOCT being improperly sized in terms of accuracy or measurement multiplier. Circuit breaker bushings can be sized strictly for isolation, without having to provide for CTs, potentially leading to a smaller size bushing using less insulation oil. Four three-phase sets of FOCTs, each with its own bushing mount, will weigh less than 1 single bushing CT. As a whole, then, the circuit breaker gets lighter. This could result in lighter footings in some installations.

From a substation design perspective, the biggest benefit is in cabling. Six sets of multi-conductor copper cables (2 protection CT sets and 1 metering CT set on each side) are replaced by 4 individual fiber optic cables (2 FOCTs on each side), changing cable trench to conduit. In the case of the dead tank circuit breaker, going fully digital and removing the terminal blocks results in savings in equipment design, equipment specifications, and in the design of the substation.

Power Transformers

It is possible to extend the use of bushing mount FOCTs to power transformers as well. The same savings exist by going fully digital. The same FOCT can be used for all applications and voltage levels, with the same result in smaller bushings and fewer cables.

Beyond this, there is also a performance advantage. FOCT measurements are entirely linear, so saturation doesn’t exist. Therefore, FOCTs solve challenges caused by CT saturation during external faults, saturation due to high X/R ratios and fully offset faults, and saturation due to burden unbalance and high X/R ratios found at generator stepups.

In addition, FOCTs can be used for open-phase detection on delta-wye connected transformers.  They can also be used as part of a GIC detection / mitigation scheme due to their ability to simultaneously measure AC and DC currents. So going fully digital makes equipment design and specification better, while improving protection system performance.

Live Tank Circuit Breakers

Live tank circuit breakers used in AIS substations is where FOCTs have the largest benefit in regards to the digital substation.

A standard FOCT mount for outdoor use is a mechanical sensor head (to connect in series to the conductor) attached to a solid dielectric insulator column. Since this is a fiber optic cable mounted in a metal sensor head, a 245 kV rated FOCT weighs around 65 kg, almost all in the column structural components. (This is 700kg lighter than some traditional CT types.) This permits mounting FOCTs directly to the live tank circuit breakers. (Figure 7).

With live tank circuit breakers, going fully digital has advantage in both equipment specification and substation design. In terms of equipment design and specification, live tank circuit breakers can be ordered with FOCTS as a standard unit. Each voltage class of breaker will need a slightly different FOCT order code, as the sensor head and column must change with the voltage class. However, the measurement fiber and electronics are identical for all voltage classes.

The biggest benefit is in substation design. There is no need for a separate CT mounting footer, or for the space to maintain adequate clearance between the breaker and the CTs. For overhead line HV substations, this reduces the space needed for a circuit breaker with measurement by 15% to 25%. It is possible to design FOCT column mounts that contain 2 measurement fibers (or more) for redundancy. As with the slipover bushing FOCT, each measurement fiber replaces 1 or 2 multi-conductor copper cables.

Other Equipment Mounting

It is possible to use the same mounting concept on any equipment platform. Insulator column mount FOCTs, due to their small size and light weight can be mounted on bus structures, underground cable riser poles, air break switches, and similar devices. This allows easy, low cost addition of current measurements for underground cable fault detection, for in line current measurements – wherever a current measurement may be desirable.  (Figures 5 and 6).

Simple Retrofit with FOCTs

The dead-tank and live tank application examples show different methods of manufactured mountings of FOCTs for ease of installation, design, and specification. These models of FOCTs can be ordered as catalog numbers, just as with conventional CTs. Because of their small size and light weight, both mounting types may be easily used to retrofit existing equipment and substations. However, they both require outages to install, because they require opening a circuit to mount. There are also some applications where these mounting types don?t physically work due to size, space, and form factor, such as for zero-sequence CTs and split-phase bus.

Another mounting type for easy retrofit and difficult mounting locations reduces the FOCT to its most basic components: an armored fiber optic measurement cable and a cable management box. The management box is placed at the primary equipment, one end of the measurement cable is disconnected from the box, wrapped around the conductor, and reconnected to the box. This can be a completely tools-free installation, though proper safety procedures must be followed, and secure mounting methods for the fiber and cable management box should be employed. This type of mounting has been successfully used on circuit breaker and transformer bushings, on overhead lines for temporary measurements, and for revenue metering calibration in the field. This flexible form factor version of an FOCT is a simple way to add digital current measurements to existing substations and equipment.  (Figures 8 and 9).

Better Operations and Maintenance

FOCTs also better enable the digital substation in another way. Going fully digital also means self-monitoring of equipment and alarming on degradation or failure. A FOCT can monitor the entire electronics, including the light source and performance of the measurement path.

Alarms can be raised if the FOCT performance is degraded or deteriorating. FOCT lifespan can therefore be monitored and known, unlike conventional CTs.

Summary

By removing the terminal blocks for a full digital measurement, FOCTs fully enable the future digital substation. The small size and light weight of FOCTs and the use of fiber optic cables drives better substation design, using less materials, less space, and a better cost. Adding FOCTs to primary equipment results in more flexible equipment standards and specifications, as the same circuit breaker models can be applied anywhere on the system, using the same FOCT for all accuracy and voltage classes.

FOCTs can be mounted literally anywhere it is desirable to measure current, even as a retrofit solution, with no impact on substation design or cost. And FOCTs are flexible in application. Fault current measurement and metering accuracy in one fiber; accurate and linear across their entire measurement range. So going fully digital and removing the terminal blocks does provide an answer to the questions. Yes, FOCTs allow us to improve the design of the substation and substation equipment. And we can go farther than just replacing copper cabling to build more efficiently, user fewer materials, and operate more cost effectively

Biographies

Dylan Stewart is the Senior Commercial and Technology Manager for GE’s Digital Instrument Transformer product line.  He has global responsibility for Digital Instrument Transformer Sales, Marketing and Product Management.He has a Bachelor of Science in Computer Engineering with Distinction from the University of Alberta in 1998.  He started working on Digital Instrument Transformers in 2003 when he joined NxtPhase T&D as an Embedded Firmware Engineer writing software for optical current and voltage sensors.  In 2009, he joined Areva T&D’s Digital Instrument Transformer group.  As of 2015, the Digital Instrument Transformer group is part of GE’s Grid Solutions business. Dylan is active in developing guidelines and standards related to next generation substation equipment through his participation in IEEE PSRC, IEC TC38 and CIGRE A3.

Rich Hunt is a Senior Product Manager with GE Grid Solutions, focusing on digital substations. Rich has over 30 years’ experience in the electric power industry, including 15 years of experience with digital instrument transformers. He has an MSEE from Virginia Tech, is active in the IEEE Power System Relaying Committee, B5 Technical Committee, IEC TC57 Working Group 10, and is a registered Professional Engineer.

Allen H. Rose is the Engineering Manager of GE, in Phoenix, AZ and a GE technical expert.  He has worked for GE (2016-present), Alstom Grid Inc., Areva T&D Inc., and NxtPhase T&D Inc. (2001-2009). He worked at NIST in Boulder, CO from 1987 to 2001. He holds a Ph.D. and M.S. in physics from the University of Arkansas, Fayetteville, and a B.S. in physics from Abilene Christian University, Abilene, TX.  He completed an NRC Postdoctoral Fellow with the U.S. Army’s Ballistic Research Laboratory, Aberdeen Proving Ground, MD. Dr. Rose has published over 90 articles, 3 patents, received an R&D 100 award for the development of the optical fiber current sensor in 1991, and the NIST Bronze Medal for the development of a standard optical retarder in 1998.  He was an associate editor for Photonics Technology Letters from 2000-2008.  He is a senior member of the Optical Society of America, and a member of the IEEE Lasers & Electro-Optics Society.

BeijingSifang June 2016