The sheer size of converter transformers for ultra-high voltage DC (UHVDC) systems poses special challenges, not only for integration with other equipment, but even for design, testing, manufacturing, and transport to the site.

The names and even external shapes of many electrical components are often the same whether for small household appliances or gigantic industrial structures. For manufacturers, however, it is not simply a matter of scaling up the structure. Today, GE Vernova offers 800 kV converter transformers, but as David Wright, Senior Expert Engineer in the Power Transformer Division, explains, “Developing UHVDC systems meant practically starting from scratch in many cases. Initially there was little information available about the performance of insulation materials for converter transformers at the very high AC and DC voltages involved, so detailed testing on prototypes was needed to generate comprehensive design information."

Even before testing prototypes, bushings had to be developed. When you think of the complexity of grid equipment, power electronics probably come to mind, but even a seemingly simple mechanism like a bushing presents challenges. In a converter transformer, both AC and DC can be present, which adds to the complication. The voltage distribution of a bushing is determined by the capacitance between the foils for an AC voltage but by the resistance between the foils for DC. Special measures are needed to control voltage distribution when the bushing is in the transformer turret.

GE Vernova's Grid Solutions business - 800 kVDC bushing being tested in Graz, Austria

Getting it out of the door

There are physical constraints, too. Ratings have been increasing to achieve higher power transmission levels, but size cannot simply be increased proportionately. “The size and weight of the transformer is limited by shipping constraints,” says Wright.

The size and weight of the 800 kV converter transformer is limited by shipping constraints

“So if you want to get it out of the door and delivered to the site, insulation design has to improve the clearances and component structures to produce the most compact transformer possible but still maintain its thermal performance and respect IEC norms.” Trial and error would be too costly, so Wright and his colleagues took Grid Solutions’ SLIM finite element modeling package and added the capability to analyze the particular challenges of UHVDC bushings and transformers for studies of the core design, harmonics, hot spots, and dielectric components.

Finite Element Modelling

With the results suggesting that the proposed solutions were feasible, the next stage was building a prototype. The Technical University of Graz had a hall big enough for initial tests of the bushing, but to meet the requirements for test voltage supplies and very low levels of background partial discharge for a transformer weighing over 100 tonnes, Grid Solutions upgraded the facilities at its transformer factory in Wuhan, China, where the prototype was manufactured and tested successfully in 2011.

800 kV converter transformers being tested in Wuhan, China

One of the challenges Wright’s team had to overcome in moving from prototype to on-site installation was the fact that the internal, valve-side connections require large electrical clearances and can have a significant impact on the size of the transformer tank. To reduce the clearances, a system of preformed barriers with controlled oil duct spacing was developed and implemented for both 600 and 800 kV DC converter transformers. This allows the connections to remain inside the transformer tank while respecting the size limits for shipping the converter transformer.

Success justifies the investment

Converter transformers for Rio Madeira 600 kV DC bi-pole 2 were manufactured and tested in upgraded facilities in the GE Vernova's factories in Canoas (Brazil), Stafford (UK), and Wuhan (China). The 800 kV DC Champa-Kurukshetra transmission network in India involves the supply of 28 transformers from GE Vernova in the UK and Vadodara, India. Looking back, David Wright insists that “success required a major commitment of resources, expenditure and coordination of the many skills of the Power Transformer Product Line. Our approach enables us to offer transformers comprising the best available technology throughout, no matter which of our factories worldwide is producing them.”

To overcome the slow commercial uptake of hybrid HVDC circuit breakers, a long-range project is reconsidering gas-discharges tubes for use in HVDC circuit breakers. Stakes are high, since it may lead to a considerable decrease in cost, complexity and footprint of HVDC breakers and, moreover, with the opportunity to mount them in easy-to-install and maintain transportable containers…

Although the first DC circuit breaker concept was proposed in the 1970’s (using gas-discharge tubes at that time), it took around 40 years before the first economical, thus acceptable concept for a commercial use in a HVDC system^[1]  was developed: the ‘hybrid’ DC circuit breaker. Laboratory tested in 2013, it offered – at last – sufficiently low losses to be economic in a commercial HVDC system. “However, the commercial uptake of such hybrid breakers has been slow, mainly because of their relatively large cost, complexity and footprint” explains Colin C. Davidson, from GE Vernova's Grid Solutions business. “New developments using optimized gas-discharge tubes could completely change this picture”.
 
^[1] Grid operators increasingly use high Voltage Direct Current (HVDC) to carry high power over long distances, as direct current (DC) is superior to alternating current (AC) because it can transmit power without capacitive or inductive losses.

The basics of hybrid DC breakers

The first HVDC schemes indeed used mercury-arc valves, a type of gas-discharge tube for the conversion between AC and DC; these mercury-arc valves allowed to construct single switches offering voltage ratings of hundreds of kilovolts, a long operating life and a high robustness to faults. Afterwards, due to their high maintenance requirements, these mercury-arc valves were replaced by semi-conductor devices such as thyristors (for Line-Commutated Converter HVDC) and, later, IGBTs (for Voltage-Sourced Converter HVDC). Semiconductor devices were also proposed for all published variants of the ‘hybrid’ DC breaker concept until 2017.  Hybrid DC circuit breakers are built (see figure 1) with a mechanical switch (ultra-fast disconnector), low- and high-voltage semiconductor switches (PE1 and PE2) and a surge arrester which provides the reverse electro-magnetic force (EMF) needed to drive the fault current to zero, absorbing the inductive stored energy in doing so^[2].

Figure 1: Basic concept of a hybrid HVDC breaker

However, the component count in these hybrid concepts is very high (and expensive), due to the hundreds of semiconductor devices needed to withstand such high voltages. So “the advent of a single high-voltage switch capable of withstanding the entire terminal-to-terminal voltage of the DC breaker could be transformative” points out Davidson. Rather than searching for 100% innovative concepts, why not look back for the future?

^[2] The difficulty of the complete operation can be illustrated by comparing it to the successful catching, in a ‘blink of an eye’, of a 1-ton mass falling from a 450 m height.

A new generation of gas-discharge tubes

As a matter of fact, GE Vernova was one of the pioneers of HVDC starting with mercury arc valves, a type of gas-discharge tube, more than 50 years ago. The advantage of this technology was that the mercury cathode, being liquid, was self-restoring. This gave the valve a longer operating life than any gas discharge tubes using solid cathodes (such as thyratrons), and a robustness to faults that cannot be emulated by semiconductor-based switches. The company and its predecessors built both the first commercial thyristor-based HVDC scheme (Eel River, in 1972) and the last commercial mercury arc scheme rated at 150 kV dc and 1800 A, the largest such valves ever, both in Canada. So, what if there would be a chance to obtain the same advantage without the inconvenience?
 
They were in brief:

• the toxicity of the mercury cathode material,
• the long anode column needed to provide stable high-voltage operation
• the occurrence of “arc-backs” during which the valve temporarily and incorrectly conducted current in the reverse direction
• the high maintenance requirements arising from the vacuum pumps and other mechanical apparatus needed to maintain a vacuum on the – relatively large – tube assembly

“Experts of GE Vernova’s Global Research Center (GRC) thought that some old electrical concepts sometimes judged obsolete, could be given new life by steady improvements over the years in materials, components, processing, controls, and software, as it occurred in high power RF applications (microwave ovens, radio and TV transmission, radars) as well as in X-ray medical imaging” Davidson explains. This is all the truer since a new generation of gas-discharge tubes appeared, offering a much more compact solution than thyristor or mercury-arc based valves and – crucially – the ability to turn on but also to turn off current. An ideal first application for such a gas-discharge tube could be HVDC, to replace the complex and bulky high-voltage semiconductor system of the hybrid DC circuit breaker by a single gas-discharge tube.

Potential advantages are obvious. Single tubes can stand off and switch high voltages and for example, x-ray tubes operating at 600 kV can be purchased off-the-shelf. Tubes can carry potentially large currents, essentially in proportion to their active cross-sectional area, and they can switch quickly (the order of a microsecond), similar to thyristors.

Gas-plasma tubes

GRC selected gas-plasma tubes over vacuum tubes based on their lower forward voltage drop during operation. While HVDC converters were identified as a particular application for such tubes, they could particularly well function in frame of the DC breaker topology. With this in mind, GRC recently decided to launch a long-range project to investigate such tubes.

Several objectives have already been accomplished. Tube prototypes constructed at 40, 100, and 300 kV, provided knowledge of the necessary materials, engineering, and construction methods.  And unlike their mercury-arc predecessors, which required a long anode column with sophisticated grading electrodes to withstand high voltages, this new generation of tubes (Figure 2) is “remarkably compact, much smaller than traditional mercury arc or present-day thyristor valves”.

Figure 2: New generation gas-discharge tubes – cross section and principle of operation

Gas-discharge tubes in DC Circuit-Breakers

Various tests and a close examination of the plasma within the tubes during operation has revealed new, unexpected operational plasma states, some of which have lower forward voltage drops than previously expected, which can pay benefits in various applications. In HVDC hybrid breakers, Figure 1, the idea would be to substitute the auxiliary branch components (PE2) for a gas-discharge tube, keeping the main branch components (PE1 and the ultra-fast disconnector) essentially unchanged. “Moreover, since the DC circuit breaker operates infrequently, the operating life of the cathode material is not a concern, and the resulting DC circuit breaker could be much more compact than today’s solution, in a way that an outdoor, containerized, factory-tested solution could become feasible” reveals Davidson.

Splitting the dc breaker

Let’s take a ±320 kV VSC HVDC scheme with one breaker at each pole as an example. As the Transient Interruption Voltage (TIV) for a DC breaker—i.e. the peak voltage that the DC breaker should produce in order to force the current down to zero— is typically 150 percent of the nominal DC voltage, the breaker would require a TIV of 480 kV. This is fully achievable with a single gas discharge tube resulting in a very compact system. However, it is possible to divide the circuit breaker in smaller stages and to use it as current limiter. By using smaller stages as necessary for current interruption, the DC breaker can prevent the further rise of current due to remote (out of zone) faults, leaving the duty of interrupting the fault current to another DC breaker, further upstream.
 
Splitting the breaker presents two additional advantages:

• redundancy: if one modular DC breaker unit is unavailable, although the DC breaker may be unavailable for fault clearing, it can still be used for current limiting if required; and,
• simplification of mechanical configuration: the modular construction could simplify the mechanical configuration of the DC breaker and its housing.

Coming back to our example, the base of the 320-kV breaker’s structure are four identical modular sub-breakers, each of them with a nominal DC voltage of 80 kV and a Transient Interruption Voltage of 120 kV.

Outdoor mounting

One major limit of the commercial uptake of classic hybrid CBs is the (perceived) need for them to be located inside a large climate-controlled building similar to a valve hall, which precludes the possibility for DC breakers to be added as a retrofit on existing point-to-point HVDC schemes due to the lack of space.

Normally, HVDC converters are housed in special climate-controlled buildings because the high DC operating voltages cause particulate pollution to adhere to the insulating surfaces of the converter. In the case of a DC breaker, all components are normally operating at the same electrical potential – that of the DC line in which the breaker is inserted. It is therefore appropriate to enclose the DC breaker components in a conductive housing that is at DC line potential. The DC breaker components are therefore inside an equi-potential housing (in normal operation), and there is no tendency for these components to attract any atmospheric pollution. The enclosure therefore does not need onerous requirements for filtration or air-tightness.

As a result, a two sub-breaker scheme is obtained, each breaker rated at 80 kV nominal voltage (120 kV TIV) installed inside a midpoint-connected typical ISO 668 shipping container. The DC breaker components only see a transient voltage of up to 120 kV with respect to the container. As the air clearances at such a voltage are modest, it leaves enough room inside the container for the DC breaker equipment itself.

To make a complete 320 kV DC breaker, two such units are connected in series, each unit being mounted on an insulated pedestal (Figure 4). The DC breaker components are factory-assembled, tested and shipped to site inside the containers, with only the wall bushings, corona rings and support insulators being added on site.

Figure 3: A modular DC breaker unit consisting of two sub-breakers, each rated at 80 kV nominal voltage (120 kV TIV) inside a midpoint-connected container.

“Avoiding the need of a large climate-controlled building to house the breaker, just using a typical ISO 668 outdoor container could pave the way to the construction of DC grids,” concludes Davidson.

DC circuit breakers will be essential for the development of DC grids; however, the technology is in an intermediate state where the concepts have been proven up to mid TRLs but remain relatively large and potentially uneconomical. There is possible effective engineering, but full-scale product development is difficult to justify because of the limited commercial outputs. A gas-discharge tube-based hybrid DC breaker could potentially result in step-change as a more economically viable proposition with significant footprint and volume reduction compared what has been proposed so far.

The continued commitment to develop and evolve its products has enabled GE Vernova to maintain a strong position in a competitive HVDC market and provide an enhanced and flexible solution for new and replacement HVDC projects. The H450 HVDC thyristor-based valve is the latest such development, providing the ideal platform for future HVDC projects such as the Jeju Bipole 1 valve replacement in Korea.

HVDC converter station for Kepco's Jeju project installed on mainland

In Q1 2017, GE Vernova were successfully awarded an LCC HVDC Refurbishment Project in Korea. The project scope was for the replacement of the valves and controls of an existing 300 MW +/-180 kVdc Bi-Pole scheme. The scheme linked the mainland of Korea in Haenam to the island of Jeju. The key for GE Vernova to be able to undertake such a refurbishment project was having an LCC product portfolio flexible enough to provide an improved solution. Mark Donoghue, Principal Engineer at Grid Solutions, explains “This was critical for this type of scheme where there was a significant physical size and positioning constraint placed on the replacement valves due to the existing converter building which could not be modified”.

Haenam' HVDC converter substation building

H-Series valves evolving through the age

The solution was to use the latest development and evolution of the H400 series valves called the H450. This is the culmination of a number of major developments over the last fifty years. In order to understand where the H450 sits in the evolution of thyristor based HVDC valves, let’s look at the history of the GE Vernova valve family.  The first-generation oil-cooled outdoor thyristor valve was developed in the late 1960s with a pilot installation commissioned in 1971 using three parallel connected stacks of 37 mm 4 kV thyristors. This was followed in the early 1980s by the H200 series valves which were forced air-cooled, air-insulated indoor valves using 2 parallel 56 mm 4 kV thyristors per level. In the late 1980s, this was followed by the H300 series valves, the first water-cooled indoor floor mounted valve utilizing single 5.2 kV 100 mm thyristors per level. Finally, the latest H400 series suspended water-cooled indoor valve using single 8.5 kV thyristors with options for 100 mm or 125 mm thyristors per level was introduced in 2003.  This was developed further into the H420 in 2010, allowing for higher transmission voltages and the possibility to use 150 mm thyristors, and the latest evolution is the H450 introduced in 2017. This improvement of the LCC valve allows GE Vernova to be more competitive on the HVDC LCC market, by deploying the H450’s reduced physical size valve, without affecting electrical performance.

H400 valve module

H-series provides the core power converter technology in the Jeju Bipole 1 HVDC scheme

The key purpose of the Jeju Bipole 1 refurbishment project is to provide stable and economical power supply by a main equipment replacement and performance upgrade in order to meet an increase in continuous power demand. The existing equipment was originally installed by GE Vernova in 1994 and therefore around 25 years old. The original valves were based upon the 3^rd generation H300 thyristor valve.

“This valve is the core power converter technology for the traditional, and mature, LCC HVDC market. The present H400/H420 valve technology has been in use for about 15 years and was Grid Solutions’ first suspended valve design. The technology has been used for a variety of HVDC projects, including back–to–back and point–to–point projects, the latter including submarine cable and overhead lines (OHL) projects.  The valve has operated at DC voltages up to ± 800 kV on the Champa-Kurukshetra project in India and is able to accommodate 100 mm, 125 mm and 150 mm thyristor devices”, says Donoghue. 

Zoom on the H450 series valve

In common with all valves from the H300 series onwards, GE Vernova’s latest H450 valves use direct liquid cooling which enables a single-circuit system with either pure deionized water or a water/glycol mix, depending on ambient temperature conditions at site. The valves are air-insulated and suspended within a controlled environment. By suspension mounting the valves, the mechanical stresses are reduced, which is of particular importance for applications in seismic areas. However, in some cases, such as pre-existing structures with inadequate suspension facilities, the valve may be floor mounted by using ceramic or composite support insulators. The valves employ high power thyristors, together with associated gating, damping and grading circuits, arranged in 6- or 12-pulse converter groups. According to the application type, thyristors with different voltage ratings and diameters can be easily accommodated.

New H450 series, the need for a new valve module with same performance

In recent years, GE Vernova further evolved the H400 series valve with a "re-packaging" design of the existing H400 module and H400 valve arrangement. “The main scope of this development was the re-design of the module without affecting the electrical performance of the existing H400 design”, states Donoghue.  He adds, “Hence the same thyristor options, the same di/dt reactor and the same damping resistors have been reused on the new H450 module”.

The H450 development project followed the same New Product Introduction process as usual, with different technical gates from the conceptual designs to the industrialized product for the first H450 contract project in South Korea with KEPCO BP1 refurbishment scheme.

H450 valve hall used for Jeju HVDC Bipole 1 renovation project

But so much lighter and smaller! The new thyristor clamped assembly is key

The key part of the H450 development centered around what is called the Thyristor Clamped Assembly (TCA), an assembly that houses the thyristors and water cooled heatsinks. In the existing H400 series valves there were two separate but identical TCAs; however, as part of the H450 development these were combined into a single clamped assembly containing twice as many thyristor levels. The key components that make this possible are the filament wound glass reinforced plastic (GRP) banded straps used to provide the large clamping forces required by the modern-day power thyristors used in HVDC. Depending upon the size of the thyristors (diameter) the maximum clamping force can range from 90 kN for the 100 mm diameter devices up to 200 kN for the largest 150 mm diameter devices. A notable feature of the band design that was developed for the H450 was that only one design was needed, irrespective of the size of thyristor used, which was not the case for the original H400 series valves. Another key change within the TCA was the reduction of overall thickness of the thyristor heatsinks, enabling space saving compared to the original design. To ensure electrical continuity through the valve/TCA when we do not require a full complement of thyristors fitted into some of the modules, dummy thyristors are used. In a matter of fact, the total number of thyristors required for the project valve is not a multiple of 12 (the maximum that can be fitted in a TCA). That’s where an actual thyristor is replaced with a copper block, also called  dummy thyristor.

Thyristor clamped assembly (TCA) with thyristors (THY), dummy thyristors (Dummy THY) and heatsink (HSK)

The development of the single thyristor clamped assembly was the enabler to make significant reductions in the overall dimensions of the valve module; a key building block of an HVDC valve. A reduction in dimensions of some 38% and a reduction of weight of 20% were achieved, giving a significant flexibility in the valve arrangements and size of valve building. This size reduction was also key in the layout of the valves for the KEPCO valve replacement project.
 
On the valve structure stand-point, two significant improvements have been achieved.

1. The reduction of the electrical clearance around the valve thanks to the new shape of the corona shields (see sidebar article).
2. The addition of a second valve arrangement known as “in-line” where the modules are positioned end-to-end (see cover picture). By comparison, the traditional H400 series use a “square” arrangement where the modules are side-by-side.

Haenam HVDC LCC converter station In Korea: Valve hall

The new arrangement provides an opportunity for GE Vernova to improve the width of the valve hall using the in-line valve arrangement when other equipment, such as the converter transformer and busbars, dictate the length of the valve hall. By adding the choice of using an “in-line” arrangement or the existing “square” arrangement for either suspended or floor mounted options with two, four or eight valves per Multiple-Valve Unit, GE Vernova’s HVDC LCC product provides flexibility for the transmission operators.

Intensive type testing process according to IEC 60700-1

The design of a thyristor valve is a complex, multi-disciplinary process involving a range of engineering disciplines including power engineering, power electronics, analog electronics, semiconductor physics, heat transfer, fluid mechanics and mechanical and structural engineering. As there are no standards to follow for designing the HVDC value, GE Vernova relies on the vast experience and solid design practices gained through over 50 years in the HVDC industry.  Modern thyristor valves are relatively standardized, that is to say that the bulk of the real design work is carried out during the product development phase, such that applying the valves to a particular project is a relatively straightforward matter.  At its simplest, the work involved for a particular project may just involve adapting the number of series-connected thyristors according to the voltage rating requirements imposed by the overall system design. For the introduction of a new product and first project implementation this may not be so straightforward. We therefore sought to minimize manufacturing and testing risk by producing a batch of valve modules ahead of type testing.
 
While there are no specific standards for the design of HVDC valves, this is not the case for the testing of HVDC valves. IEC 60700-1: ‘Thyristor valves for high-voltage direct current (HVDC) power transmission – Part 1: Electrical testing’ defines the test program for the valve and covers two broad categories: dielectric tests and operational tests. The type tests form an important part of the design verification process as well as customer project requirements. In addition, the standard covers both production routine testing and sample testing.
 
When a new thyristor valve design has been produced or a previously tested valve design is modified, a program of type tests must be performed. Type testing of thyristor valves is complex, specialized and time-consuming. Some parts of it require extremely specific and expensive test circuits for which only a few serious players in HVDC can justify investment in. All thyristor valves are subjected to comprehensive routine testing in the factory. The purpose of this test program is to prove that the thyristor valves have been correctly assembled. It aims to identify wiring connections that have been incorrectly made, grading components that are out of tolerance, gate electronics that are malfunctioning, blockages in the cooling circuit, joints between the thyristor and heatsinks, etc.
 
For the KEPCO valve replacement project the valves needed to be floor mounted and sited within the converter building, essentially as in the original installation. The reduction in size, weight and increased flexibility of the H450 design made this possible. The figure below shows the valve arrangement. Each of the three structures are known as a quadri-valve (i.e. a structure comprises four valves) and forms the overall 12-pulse converter bridge and represents one pole end of the scheme.

H450 series valve arrangement

Highlights

• GE Vernova’s new H450 design is 38% smaller, and 20% lighter than the H400 product, despite using the same main components (thyristors, di/dt reactor, damping resistors.)
• The number of parts per module has been reduced by 25%. This allows the manufacturing line to be more efficient, quicker and more controlled.
• The maintainability at site is simplified, with easier access to the main components to replace during a planned maintenance outage.

Zoom on the corona shields

Design enhancement of the corona shields
 
Since the late 1970s, all commercial HVDC valves have been air-insulated; that is to say, the insulation between the valves and earth is achieved by using air instead of a higher-performance dielectric medium such as oil or SF₆. This is mainly because of the large physical size of the valves and the need to access the valve components at regular intervals to replace failed components.
 
As HVDC transmission voltages have increased sharply in the last decade (from 500 kV to 800 kV or even higher), the size of air clearances needed around the valves has also needed to increase, and since air clearances increase non-linearly with voltage, the air clearances around the valve are now having a dramatic effect on the size of the valve hall. The valve hall is a very large building with stringent requirements on air quality and there is therefore a considerable economic incentive to reduce its size.
 
An external profile as smooth as possible
 
For high voltages and large air clearances, the design of the corona shields at the top, bottom and sides of the valve is of paramount importance.  The aim of these corona shields is to make the external profile of the valve as “smooth” as possible, avoiding regions of high curvature which will lead to localized areas of high electric field and an increased risk of flashover.

The design of the predecessor H420 valve module was carried over from the earlier H400 valve and only the external corona shields were changed, leading to relatively limited shielding, and the need for long clearance distances.

Colin C. Davidson, Consulting Engineer at GE Vernova's Grid Solutions business, explains, “the H450 valve is a mechanical “re-packaging” of the H420 valve, using the same electrical components but in a better and more compact mechanical layout, considering the external corona shielding from the outset. The performance of the H450 valve has been verified by undertaking a series of “50% flashover voltage” tests (U₅₀ tests) which involve repeatedly applying switching impulses to the valve structure at gradually increasing voltages and for a range of different clearance distances”.  The H450 valve has been demonstrated to achieve dramatically smaller electrical clearance requirements than its predecessor, more than a 50% reduction for the so-called “inline” configuration at the highest voltages (pictured).

U₅₀ Test campaign

In a grid topology using HVDC circuit breakers able to provide fast clearance of a DC fault, two main contrasting, yet complementary, solutions appear possible. One would be to apply the same protection philosophy and principles used in AC systems. The second could be the “Open Grid” concept.

Adapting the AC grid protection philosophy

The principle, philosophy, and scheme for protection of HVDC systems can be inherited from AC systems. The greatest challenge is the need for a very short tripping time without losing selectivity, security and sensitivity.

Although HVDC grid protection is still in a development phase (to date no DC circuit breaker is in commercial use in the field), the protection principles of an AC system are still one option for application to an HVDC network. As Sankara Subramanian, head of the Innovation & Technology department at GE Vernova, explains: “A DC breaker that can provide fast clearance of the DC fault will play the key role for isolating the faulty line and devices in the HVDC system. In this context,” he adds, “the philosophy, principle and scheme for protection of HVDC grids may still follow those for AC systems.” However, as with AC system protection, the four requirements of a secure and reliable HVDC system (selectivity, speed, sensitivity and security), “are somewhat in contradiction with each other, and need to be balanced technologically and economically”.

Unit versus non-unit protection

Protection can be based on the information and measured voltage and/or current at one end (where the protection is installed) or at both ends. In the first case (one end), it is called a “non-unit protection”; in the second case (both ends), it is a “unit protection”.

In an AC system, the over-current protection and distance protection belong to the non-unit protection category, while the current differential/phase as well as directional comparison belongs to the unit protection category. “The advantage of the non-unit protection is that the communication links and devices are not required,” explains Subramanian. “This not only minimises costs but also means that the speed of the protection is not limited by the communication time delays. Meanwhile the reliability and security are not restrained by communication errors or failure.” On the other hand, non-unit protection has the disadvantage that it cannot provide absolute selectivity.  

Another important point is that transmission lines in an HVDC grid are normally longer than those in an AC system: the communication time delay is therefore longer, due to the distance involved. If the line length is longer than the limitation of direct communication, then inter-connection relaying to forward the information is required, which can cause an extra time delay (up to 100 ms for lines over 500 km). “Therefore, in future HVDC grids, the non-unit type is likely to play the dominant role of protection,” says Subramanian.

Protection algorithm: transient or steady-based?

Protection algorithms are generally formulated by the characteristic difference between internal and external faults. Algorithms based on the characteristic difference of transient voltage or current signals are called “transient-based protection”. Those based on the character of steady-state voltage and current signals are called “steady-state-based protection”.

Whereas in AC power grids most of the protection systems installed are steady-state based, in HVDC grids the steady-state signal is DC, so the Fourier-based protection does not work; the transient period is also much longer then in an AC system. Therefore, the preferred way of formulating an algorithm for HVDC grid protection of the conventional approach is by employing “transient-based protection”.

Sampling rate and time window

A very important element for transient-based protection is the time window, which will directly impact the speed of the protection algorithm. It determines the sampling rate such that, within the time window, there should be sufficient samples to detect the faults and determine whether they are internal or external faults. As per IEC 61869-9, the sampling rate for DC grid protection is 96 kHz (96 samples/ms). If we assume that the required time of the total fault clearance should be less than 5 ms, the window length should be less than 0.5 ms. Using the above sampling rate, the decision for an internal or external fault could be made by an algorithm in less than 0.5 ms, which could therefore meet the requirement of HVDC fault clearance.

New algorithms for HVDC grid protection

Based on the above analysis, the only difference between HVDC and AC grid protection schemes would be that non-unit protection would be the primary protection function. However, as for AC, there could be several protection schemes for HVDC grids: the primary protection may be a transient-based direction-over-current relay or a transient-based distance relay plus transient-based high-speed remote trip detection without relying on communications between the ends. The backup protection could be a transient current differential relay or a transient-based directional comparison unit protection or a transient-based distance unit protection plus an aided scheme (a scheme relying on communication between the ends). “All these protection philosophies are presently in the development stage, and there are already several patent applications in this field,” Subramanian concludes.

Open Grid: an alternative approach to HVDC grid protection

A major hurdle to overcome in the creation of a true HVDC grid, even when the HVDC circuit breaker technology is available, is the protection of the grid. “If the DC grid protection philosophy were to be based on that currently employed in AC grids, then the protection system would need to have the ability to detect the fault, to discriminate that the event is an HVDC grid fault and not an external disturbance, to locate where the fault is on the HVDC grid, and then to send a trip signal to the appropriate HVDC breakers,” explains Carl Barker, Chief Engineer at GE Vernova. “In the event that the fault is real, then this must take place while the DC current is rising towards a steady-state maximum value but before it has reached a level that is beyond the capability of the HVDC breakers to interrupt.

Fault interruption in a DC grid

Applying the same protection strategy as that used in AC systems – 20 ms detection and discrimination followed by tripping only the DC circuit breakers associated with the fault – results in the line voltages and currents displayed in the figure below (it is assumed that the DC circuit breaker operating time is 5 ms). “These results show that allowing the fault current to rise over a period of 20 ms and then only using the HVDC circuit breakers associated with the faulted line to clear, imposes a very high current interruption duty on those circuit breakers,” says Barker. “On the other hand, reducing the time for detection and discrimination before the HVDC circuit breakers are tripped reduces the time available for fault location.”

AC Methodology applied to DC grid protection

Reversing the protection sequence order

An alternative approach, referred to as “Open Grid”, is to reverse the protection sequence order. “This means allowing each HVDC circuit breaker to autonomously trip on detection of a fault without any delays associated with communications or discrimination logic, and then re-closing healthy circuits,” explains Barker. This strategy may offer practical advantages in terms of building a robust DC/AC grid: the HVDC circuit breaker opens at a much lower fault current, and the fault, as it propagates rapidly through the network, is “seen” by several breakers that will all open (except those more remote from the fault; they will not have time to “see” it and will therefore remain closed). “The energy requirements of the HVDC circuit breakers could thus be much lower – initial results indicate a reduction of some 95%.” This approach significantly reduces the duty on individual circuit breakers, facilitates their rapid opening and is complementary to ongoing HVDC circuit breaker development. Fault location would also be facilitated by this method.