Forget your password?
Fields marked with an asterisk (*) are mandatory
DC networks and the challenges of protection
12/11/2014 - 12.01 pm
Protection of HVDC networks is a complex engineering challenge – and the subject of intense research into new concepts. Several potential, contrasting and collaborative, solutions are currently under study.
GETTY / ThinkStock
For technical and economic reasons, High Voltage Direct Current (HVDC) transmission is preferable to conventional AC transmission for the transition towards large power grids able to transmit renewable energy over long distances, that is, from the production zones to the consumption zones. However, this shift raises several fundamental challenges, notably in the field of DC grid protection and control, which calls for new fields of investigation. Alstom is today addressing the needs of future DC network protection through research undertaken jointly by its network protection experts and its HVDC development group.
Interrupting a DC current representsa principal technical challenge
How is a DC network different from an AC network, and what are the consequences relative to its protection strategy?
Carl Barker, Alstom DC Grids Chief Engineer: From the point of view of network protection, in case of a fault, interrupting a DC current represents a principal technical challenge as, in contrast to AC, there is no natural current zero providing an opportunity for current flow to be stopped. To interrupt fault currents, one method is to use specific high voltage DC circuit breakers; however, the energy driven by the fault has to be dissipated by the breaker itself, making it a much more complicated device than an AC breaker. Moreover, protecting a DC network requires the development of new fault detection systems that can operate many times faster than is required for an AC system; this is driven by the low inertia of a DC system where the impact of faults propagates rapidly across the network. Another issue to be considered is that, unlike an AC system supplied by synchronous machines with their inherent transient overload characteristic, an HVDC converter will very rapidly protect itself from the potentially large fault currents resulting from a DC network fault. This will then lead to the temporary loss of the whole DC network until the fault has been isolated (or cleared) if the converters are capable of “fault-blocking”. Otherwise, if, as is common today, the VSC converters are “non-fault blocking”, the converter AC side protection will operate, resulting in the shutdown of the DC network.
Typical DC grid
Several approaches appear possible In consequence, what are the main challenges to protecting HVDC networks?
Sankara Subramanian, Alstom Innovation and Technology Director: There are three major challenges in implementing HVDC network protection: speed, selectivity and time delay. First, high speed is required: the DC system circuit breaker should clear the fault current very rapidly (much faster than an AC system frequency cycle), due to the rapid increase in energy to be dissipated. The total tripping time for the DC protection is targeted to be less than 1 millisecond (including the time delay of hardware in the loop). The second challenge, selectivity, is to identify the faulted section. Unlike point-to-point HVDC, the DC grid is made up of several DC lines forming a DC network, where the protection is required to clear only the faulted line. The third challenge is due to the long length of a typical HVDC transmission line, which results in a relatively long communication time delay, which in turn would make current differential protection much slower than the speed that a DC grid requires.
Following on from these considerations, to use an HVDC breaker able to provide fast clearance of a DC fault, several approaches appear possible to the Alstom engineers. One is to adhere to the same protection philosophy and principles for DC networks as those used in AC systems, albeit adapted to their new purpose. A second approach, the “Open Grid” concept, is to consider circuit breakers autonomously and instantaneously tripping for a fault in the DC grid based only on local measurements and then selectively re-closing those not associated with the faulted section. Another consideration is the use of “fault-blocking” converters. These converters would minimise the use of HVDC circuit breakers but would still need new protection algorithms in order to locate a fault within a DC network, allowing it to be rapidly isolated before power transfer over the healthy part of the grid could be restored. These three contrasting solutions to the protection of future HVDC networks will be discussed in forthcoming articles.