The Modular Multilevel Converter and Fault Current Management in Medium Voltage DC System of an Electric Ship

• Other Authors: Sun, Ke (authoraut)
• Format: Others
• Language: English; English
• Published: Florida State University
• Subjects: Electrical engineering
• Online Access: http://purl.flvc.org/fsu/fd/FSU_2016SP_Sun_fsu_0071N_13261

The Modular Multilevel Converter (MMC) is a potential candidate for power conversion in a Medium Voltage DC System (MVDC) based electric ship. One of the major advantages of utilizing MMC in an MVDC environment is the capability of limiting DC side fault current and fast re-start process because re-energizing of the MMC cells is not necessary. The MMC cells have various configurations e.g. half-bridge, full-bridge. The full-bridge MMC is more suitable for the MVDC system and fault current handling. However, the modeling, control, coordination in a multi-MMC system, and fault handling of a full-bridge MMC based MVDC system is still not fully investigated and understood. This thesis focused on the key issues of the full-bridge MMC controls and modeling in an MVDC environment and the fault current limiting using multiple MMCs. The fundamental characteristics of the MMC topology are also discussed. Followed by the single MMC control design, the MMC control scheme for the MVDC system is designed to adapt to the capability of having a fast and controllable DC voltage and current. To decrease the complexity of the MMC circuit, two simple averaged models of MMC are proposed. To verify the accuracy of the averaged models, the simulation results are compared with the results from Controller Hardware in The Loop (CHIL). The results of the comparison show that the proposed two types of averaged models predict the steady state values with very a good accuracy. For studying the behavior of a multi-MMC based MVDC system under DC side fault scenarios, an MVDC test system is proposed in this work. For comparison purposes, the real-time system model and off-line model are developed respectively. The off-line MMC model uses the individual IGBT component from the MATLAB/Simulink/SimPowerSystem software package whereas the real-time model is built using the library provided by OPAL-RT. The multi-cell circuit which has many nodes is simplified as a two-node voltage source and an equivalent resistance in series connection by applying the Th\'evenin equivalent. This thesis also discusses the challenges of determining the sampling time and how to group the MVDC system component models so that it is able to run in a multi-core real-time simulator. Besides the modeling of the MVDC system components (e.g. the MMCs, the loads), a fault current limiting strategy is also proposed in this work. This thesis put forward an operation mode for the multi-MMC system in a way that only one MMC is allowed to run in voltage controlled mode and the other MMCs are required to run in power controlled mode. By employing this operation mode, the fault current can be limited in the case of a DC side fault scenario. And no operation mode switching is needed as this operation mode also works for normal operation. The proposed fault current limiting strategy also contains the sequence of the converter actions. Five simulation cases are designed to test the proposed fault handling strategy. The simulation results show that the peak fault current is related to the operation conditions e.g. the pre-fault load current carried by MMCs, and the MMC control has some effect on mitigating the peak fault current. The proposed fault current limiting strategy is able to limit the fault current to a certain level in an MVDC system made up of single MMC, two MMCs and four MMCs with different loads conditions. === A Thesis submitted to the Department of Electrical and Computer Engineering in partial fulfillment of the requirements for the degree of Master of Science. === Spring Semester 2016. === April 13, 2016. === Electric ship, Fault current limiting, Medium Voltage DC (MVDC), Modular Multilevel Converter (MMC) === Includes bibliographical references. === Omar Faruque, Professor Co-Directing Thesis; Chris S. Edrington, Professor Co-Directing Thesis; Simon Foo, Committee Member.