Abstract
The growth in demand for low-cost bulk power transmission over long distances, coupled with the integration of renewable energy sources, has led to increased adoption of high voltage direct current (HVDC) technologies, with increasing interest in multi-terminal direct current (MTDC) systems. These systems enable more flexibility in regional power exchange, increased redundancy, and reduced operational and investment costs. Offshore wind energy is anticipated to play a significant role in the expansion of the renewable energy sector and drive the need for multi-terminal HVDC grids.
Nevertheless, the protection of the interconnected DC lines within the MTDC grid is more complicated than the protection of conventional AC power grids due to the need to detect and clear faults within two to five milliseconds to avoid the collapse of the converter dc voltages. In addition, due to the absence of natural zero-crossings in DC fault currents, DC circuit breakers (DCCBs) are essential for quick and selective isolation of DC faults to avoid shutting down the entire DC system.
Protection algorithms utilized in MTDC systems use either local or add communication based measurements. Local measurement-based algorithms, which examine the shape of the response of measured local voltage and current after the fault, are quick and don't necessitate communication from other points in the network. Communication-based algorithms, which use information from both ends of the DC line to make more precise tripping decisions, require secure, redundant channels for reliable high-speed reliable communication. However, such algorithms experience communication delays, which increase with the length of the protected line. These delays may pose problems for multi-terminal systems which require fault isolation within a few milliseconds. Therefore, communication-based protection algorithms are often used as backup protection.
The main goal of this thesis is to develop a comprehensive protection scheme for HVDC transmission systems, based on a combination of single-ended and communication-aided methods, with a speed of response suitable for multi-terminal HVDC grids. The proposed protection system consists of three protection schemes, namely: a local rate of change of the current scheme implemented as the primary protection scheme, a novel differential rate of change of the current scheme implemented as a secondary protection scheme, and a conventional line current differential scheme used as a backup protection scheme. The combined primary and secondary protection schemes ensure fast fault detection and trip commands to DCCBs, with the capability of detecting relatively high resistance faults. The backup protection scheme is implemented to provide coverage for detecting faults with higher fault resistances. The primary protection is a single-ended protection scheme whereas the secondary and backup protection schemes are communication-aided protection schemes. The communication-aided schemes do not required time synchronized measurements, improving response time. In addition, despite their ability for fast detection, these protection schemes are also capable of fast DC fault type identification.
The operation of the proposed protection system is validated and analyzed through simulations of a four-terminal MTDC grid by creating different fault scenarios. The performance of the proposed approach is compared with existing algorithms. The ability of the approach to achieve fast detection, selective protection, and high sensitivity to high-resistance faults is verified.