Targeting a climate-neutral maritime sector drives the adoption of the all-electric ship (AES). While AESs can utilize both ac and dc shipboard power systems (SPS), a dc system offers advantages in efficiency, power density, and source synchronization. However, the enhanced network complexity of dc grids combined with the high penetration of power electronic devices and harsh environmental conditions can compromise the system's reliability. Therefore, this article provides an overview of the reliability aspect of dc-SPSs, addressing the power system design, adequacy assessment, and reliability improvement. First, the performance tradeoffs associated with the SPS design are examined, revealing how changes in the power system topology and dc bus structure impact the vessel's reliability along with other performance parameters. Second, a hierarchical reliability model framework is proposed for the adequacy assessment of dc-SPSs, considering the reliability from the component level up to the system level. To determine the system-level reliability, multiple probabilistic methods, including simulation and analytical models, are compared using a propulsion subsystem example. Finally, an overview of the reliability improvement strategies is provided, addressing methods at the system, device, and component level. These three topics combined aim to provide guidance in the design of future reliable dc-SPSs.

@en

High-power flexible dc links employ modular multilevel converters (MMC) for compact active power redirection in medium and high voltage grids. During contingencies, such converters may need to provide an enhanced active power capacity to avoid overload in vulnerable grid locations. This paper achieves this target by using the capacitor voltage ripple margin of the MMC submodules (SM) to enhance the dc voltage beyond the rated value. This voltage enhancement enables the enhanced active power capacity of the MMC while maintaining rated electro-thermal stresses on the components. Moreover, dynamically varying the dc side voltage reduces the MMC's circulating current, improving its operating efficiency. Because the average capacitor voltage is controlled to remain constant, the overall stresses and harmonic performance of the enhanced MMC remain the same as in the base case. In this paper, the analytical expressions for the voltage and power enhancement limits are derived, revealing a dependence on the grid-injected reactive power. Furthermore, a controller is designed to achieve stable operation during transient conditions when the power enhancement is carried out. Finally, the enhancement concept is validated using simulations and experiments with a down-scaled laboratory MMC prototype.

@en

In DC shipboard power systems (DC-SPS), the enhanced network complexity and high penetration of power electronic devices make the system level reliability a critical design aspect. This paper proposes a stochastic framework for the reliability assessment of DC-SPSs based on a three-stage Monte Carlo (MC) simulation, including component failure sampling, active fault propagation, and reliability index calculation. The proposed MC framework is verified for a simplified meshed DC grid through comparison with an analytical method. Later, the advantages of the MC method are demonstrated for a dynamic positioning vessel equipped with a ring-type DC power system architecture. The results quantify the impact of redundancy on the reliability of a DC-SPS, show the spread in the subsystem repair times, and reveal the system's availability during both the initialization and steady-state. Combined, the simulation results reveal the strengths and weaknesses of the designed grid, guiding the focus for future reliability enhancements.

@en

The role of power electronics in advancing electrification and sustainability is pivotal. The Modular Multilevel Converter (MMC) is a leading candidate for connecting offshore wind farms to the power grid. However, one of the primary concerns with MMC is its reliability, primarily due to the high number of components, with semiconductors and capacitors being the main sources of failures. This study examines how the modularity and redundancy of the MMC affect its reliability and, consequently, its impact on power system adequacy. Our findings reveal a substantial influence of MMC's modularity and redundancy on power system adequacy. A high level of modularity with no redundancy leads to the worst-case scenario. On the other hand, lower modularity combined with higher redundancy results in the best scenario for power system adequacy. However, it's important to note that lower modularity and higher redundancy come with increased capital costs of MMC, representing a trade-off between reliability and affordability that we explore in this paper.

@en

When operating a modular multilevel converter (MMC), a margin appears between the arm voltage and sum capacitor voltage corresponding to the power dependent ripple. This margin can be used to enhance the DC link voltage and increase the transfer capacity of an MMC-based distribution link while keeping the submodule (SM) stresses fixed. Consequently, this dynamic enhancement in transfer capacity can be achieved with the same submodule switch and capacitor voltage ratings. Using an arm-level averaged simulation model of a 10MW MMC-based MVDC link, the enhancement concept is verified and shown to be beneficial to a practical link application. Besides, a dependency is discovered between the enhancement limit and the grid-injected reactive power, which defines the basis of the proposed control for dynamic enhanced operation.@en

Reliable Power Electronic Systems (PES) are vital for enabling energy transition technologies of the future. Power hardware-in-the-Loop (PHIL) test bed can be used to validate such systems cost-effectively and time-efficiently. In general, the Real Time Digital Twin (RTDT) is a virtual representation of the PES and its operating environment that mimics its behavior in real-time to provide adequate flexibility to the test bed. The workflow of alternating between the prototype and twin, for instance, overcomes the dilemma of needing 100 % details (due to fast dynamics), but optimization during design choices requires cheap flexibility. In this paper, some use cases in applications of RTDT-based PHIL test bed such as fault tolerant converters, power electronic interface for green technologies, survivable all-electric ships, mission profile-based reliability testing, protection of multi terminal dc systems and reconfigurable hybrid ac-dc links is discussed. Furthermore, the co-simulation potential of real-time platforms is briefly described.@en