Skip to main content
To KTH's start page To KTH's start page

Multi-terminal HVDC protections based on transient line modeling

Time: Fri 2022-10-28 10.00

Location: H1, Teknikringen 33, Stockholm

Video link:

Language: English

Subject area: Electrical Engineering

Doctoral student: Niclas Johannesson , Elkraftteknik

Opponent: Professor Jef Berteen, KU Leuven

Supervisor: Staffan Norrga, Elkraftteknik; Hans-Peter Nee, Elkraftteknik

Export to calendar

QC 20221006


High voltage direct current (HVDC) is considered one of the critical technologies required for the power system to enable the transition toward renewables. With an increasing geographical density of HVDC converters, there is a potential for optimization by connecting more than two converters into a shared DC transmission system, thus forming a multi-terminal HVDC (MTDC) system.

Larger MTDC systems are expected to require HVDC circuit breakers, thereby allowing disconnection of system subsections in case of faults rather than a complete shutdown of all converters. Thus, the protection system in MTDC systems with DC breakers differs from a conventional point-to-point system, as differentiation between DC faults is required to ensure that only the minimum subsection of the system is disconnected in the event of a fault. The main topic of this thesis is to achieve reliable detection of DC line faults (i.e., underground/submarine cables or overhead lines) in MTDC systems.

In this thesis, two different methods are proposed. The first is based solely on locally obtained measurements, thus requiring a reactor at the opposite end to provide a boundary of the protection zone. The method extracts the incident traveling wave using time-domain modeling techniques to represent the frequency-dependent characteristic admittance, thereby making it independent of line terminal reflections. Differentiation between internal and external faults is achieved by determining the steepness of the incident wave-front.

The second method, traveling wave differential protection, requires telecommunication between the two ends of a line, thereby not requiring a reactor to differentiate between internal and external faults. The method is based on a differential calculation of traveling waves obtained from voltages and currents at both ends and the frequency-dependent representations of the characteristic admittance and propagation function. Compared with other telecommunication-based methods, it is found that the method can operate faster because of the included wave propagation time in the differential calculation. 

The traveling wave differential protection relies on the transmission line parameters to accurately calculate and compare the traveling waves at both ends of a line. Thus, parameter errors will result in a false non-zero differential current during external disturbances, potentially causing false operation and reducing reliability. Therefore, the method's sensitivity was evaluated in a cable application using a procedure to automatically generate cable models with parameter variations and perform a transient simulation of an external fault. It was found that the propagation time used for synchronizing the waves in the differential calculation was the most critical parameter. Therefore, a method was developed to minimize any time-shift errors that otherwise would result in a false differential current.