Cover Page

VSC-FACTS-HVDC

Analysis, Modelling and Simulation in Power Grids

Professor Dr Enrique Acha

Laboratory of Electrical Energy Engineering
Tampere University
Tampere, Finland

 

Dr Pedro Roncero-Sánchez

Department of Electronics
Electrical Engineering and Control Systems
University of Castilla-La Mancha, Spain

 

Dr Antonio de la Villa Jaén

Department of Electrical Engineering
University of Seville, Spain

 

Dr LuisM. Castro

Faculty of Engineering
National University of Mexico (UNAM)
Mexico City, Mexico

 

Dr Behzad Kazemtabrizi

School of Engineering
Durham University, UK

 

Wiley Logo

Preface

Electrical power transmission using high voltage direct current (HVDC) is a well‐established practice. There is common agreement that the world's first commercial HVDC link was the Gotland link, built in 1954, designed to carry undersea power from the east coast of Sweden to the Island of Gotland, some 90 km away. The original design was rated at 20 MW, 100 kV and used mercury‐arc valve converters. Its power and voltage ratings were increased in 1970 to 30 MW and 150 kV, respectively. Solid‐state electronic valves were used for the first time in the upgrade, with the new type of valve, termed silicon‐controlled rectifier (SCR) or thyristor, being connected in series with the mercury‐arc valves. This kind of HVDC link, and ancillary technology, has been magnificently described in earlier treatises by Adamson and Hingorani, Kimbark, Uhlmann and Arrillaga.

By the turn of the second millennium, there had been 56 HVDC links of various topologies and capacities built around the world: 22 in North and South America, 14 in Europe, 2 in Africa and 18 in Australasia. They ranged from the small, 25 MW, Corsica tapping of the Sardinia–Italy HVDC link to the large, 6300 MW HVDC link, a part of the awesome Itaipu hydro‐electric development on the Brazil–Paraguay border. At the time, three other large capacity HVDC links were at the planning/construction stage in China, to transport hydro‐electric power from the Three Gorges to the east and southeast of the country, each spanning distances of around 1000 km, rated at 3000 MW – the Three Gorges is a gigantic hydro resource in central China, with an estimated power capacity of 22 MW. Heretofore all the HVDC links in the world had employed either mercury‐arc rectifier or thyristor bridges and phase control to enable the rectification/inversion process. These converters are said to be line commutated and when applied to HVDC transmission are termed LCC‐HVDC converters. The LCC‐HVDC technology has continued its upward trend and five other high‐power, high‐voltage, long‐distance DC links have been built in China since 2010. The most recent LCC‐HVDC in operation was commissioned in 2012; it is the Jinping‐Sunan link in East China, rated at 7200 MW and ±800 kV, spanning a distance of 2100 km.

In LCC‐HVDC systems the current is unidirectional, flowing from the rectifier to the inverter stations. Such a fundamental physical constraint in thyristor‐based converters limits applicability to the following HVDC system topologies: point‐to‐point, back‐to‐back and radial, multi‐terminal links. In this context, the conventional, or classical, HVDC transmission technology is not a meshed grid maker; rather, its role has been to interconnect AC systems where an AC interconnection is deemed too expensive or technically infeasible.

However, one has to bear in mind that nowadays, in many situations, robust AC interconnections may be achieved more economically using one or more of the options afforded by the Flexible Alternating Current Transmission Systems (FACTS) technology, an array of power electronics‐based equipment and control methods which became commercially available in around 1990. It is widely acknowledged that N.G. Hingorani and L. Gyugyi stand out prominently as the intellectual driving force behind the development of the FACTS technology.

The main aim of the FACTS technology is to enable almost instantaneous control of the nodal voltages and power flows in the vicinity of where the FACTS equipment has been installed. We should not forget that power flows over an AC line can be manipulated very effectively by controlling the line impedance, or the phase angles, or the voltages, or a combination of these parameters up to the thermal rating of the equipment. A key element of the FACTS technology is the so‐called static compensator (STATCOM), which, in the parlance of a power electronics engineer, is a voltage source converter (VSC) and serves the purpose of injecting/absorbing reactive power to enable tight voltage magnitude regulation at its point of connection with the AC power grid. The advent of the STATCOM in the mid‐1990s was made possible by the development of power semiconductor valves with forced turn‐off capabilities, like the gate turn‐off (GTO) first and the insulated gate bi‐polar transistor (IGBT) soon afterwards. GTOs are like thyristors, which can be turned on by a positive gate pulse when the anode–cathode voltage is positive, and, unlike thyristors, can be turned off by a negative gate pulse. This turn‐off feature led to new circuit concepts and methods such as self‐commutated, pulse‐width‐modulated, soft‐switching, voltage‐driven and multi‐level converters. These circuits may be made to operate at higher internal switching frequencies than the fundamental level, at several hundreds of hertz, which, in turn, reduces low‐order harmonics and allows operation at unity and leading power factors. This contrasts sharply with what can be achieved with the normal thyristors.

Advances in the design of the power GTO and its applications in Japan and the USA continued apace by virtue of strategic collaborative R&D projects funded by utilities, manufacturers and governments. In Japan there was a target to develop 300 MW GTO converters for back‐to‐back HVDC interconnections, while in the USA a 100 MVAR GTO‐STATCOM was commissioned in 1996 for the Tennessee Valley Authority. Meanwhile, similar efforts were conducted in Europe in the design of the power IGBT. It is reported that on 10 March 1997, power was first transmitted between Hellsjön and Grängesberg in central Sweden using an HVDC link employing IGBT converters driven by pulse‐width‐modulation (PWM) control. The link is 10 km long, rated at 3 MW, 10 kV and is used to test new components for HVDC.

In spite of the great many technical advantages and operational flexibility of the VSC compared with the thyristor bridge, the GTO‐based converters did not make inroads into HVDC applications because of the much higher power losses and cost of GTOs compared with thyristors. A further reason is that the ratings of GTOs are low compared with those of thyristors. All this conspired to make VSC‐HVDC installations expensive. The impasse was broken with the use of IGBT valves, which exhibit lower switching losses than GTO valves, and decreasing manufacturing costs. Three years after the commissioning of the Hellsjön‐Grängesberg, four other VSC‐HVDC links had been commissioned in very distant parts of the world: a 50 MVA DC link in the emblematic Island of Gotland to evacuate wind power, an 8 MVA DC link in West Denmark to link an offshore wind farm, the 180 MVA Directlink or Terranora project in Australia for power export from New South Wales into Southern Queensland, and a 36 MVA DC link for system interconnection on the Mexican–Texan border. The undersea Estlink 1, linking the Estonian and Finnish power grids, was commissioned in 2006, rated at 350 MW and using VSC stations. Intriguingly, the Estlink 2, rated at 650 MW and commissioned in 2014, uses the classical thyristor‐converter technology.

It should be noted that all the VSC stations used in HVDC projects until 2010 had been of the so‐called two‐ and three‐level power converters. In around 2008, a new breed of VSCs was introduced into the market, the modular multilevel converters (MMCs), which switch at low frequencies, yield minimum harmonic production and have power losses just above those of the classical thyristor‐based HVDC converters. Equally important is the fact that it has been possible to increase the capacity of VSC‐HVDC links using MMC, by a very considerable margin, say 1000 MW per circuit, such as in the INELFE DC link between Baixas, France, and Santa Llogaia, Spain. Two identical circuits make up for a transmission capacity of 2000 MW. The link was commissioned at the end of 2013. Note that this application comes into the realm of bulk power transmission and is already eating into the niche area of classical thyristor‐based HVDC technology, namely asynchronous bulk power transmission, an area until recently thought to be unassailable. The Trans Bay Cable link was the first MMC VSC‐HVDC, commissioned in 2010, transmitting up to 400 MW of power from Pittsburg in the East Bay to Potrero Hill in the centre of San Francisco, California.

Furthermore, there are new application areas in which VSC‐HVDC transmission does not seem to have a competitor in sight – the connection of wind sites lying more than 70 km away from the shore is one of the most obvious applications, but there are a few others. For instance, the connection of microgrids with insufficient local generation and little or no inertia (inertia‐less power grids), the electricity supply of oil and gas rigs in deep waters, the infeed of densely populated urban centres with power grids already experiencing high short‐circuit ratios. Moreover, the unassailable characteristic of the HVDC transmission using the VSC technology is that it is a natural enabler of meshed DC power grids, with such a high level of operational flexibility, reliability and efficiency that one day may surpass that of the meshed AC power grids. To get to this point, though, further technological breakthroughs are still awaited in the ancillary areas of DC circuit breaker technology and high‐temperature superconductor cables and circuit breakers, as well as more affordable VSCs.