Resilient MVDC and MVAC/MVDC Distribution Systems (21-02)

Principal Investigator: Dr. Rob Cuzner

This project falls under the Protection and Management thrust from the GRAPES needs document. It is built upon prior GRAPES projects that developed custom FPGA based real-time simulations using a NI FlexRIO system that is suitable for Controller Hardware-in-the-Loop (CHiL) simulation of MVDC systems and commercial platform Hardware-in-the-Loop (HiL) simulations of AC and DC microgrid fault behavior and protective relaying schemes. The project is informed heavily by experiences gained on a prior GRAPES project on development of medium voltage DC Circuit Breakers (DCCBs) and a realization that what is really needed is a system understanding of the impacts of DCCBs design on system fault response. It focuses on the distribution networks within the GRAPES challenge project frameworks. Year One (Y1) will introduce and simulate a reference mixed medium voltage AC and DC (MVAC/MVDC) grid architecture that will enable the characterization of faults in MVDC and MVDC systems. This reference system will also be used to characterize the behavior MVAC distribution systems having a high level of Power Electronic Converter (PEC) interfacing Distributed Energy Resources (DERs). Fault characterization will be used to determine fault interruption time and current requirements for DC Circuit Breakers (DCCBs) in MVDC systems and to design adaptive protective schemes to ensure resilient and dependent power and energy delivery. Application of DCCBs to most likely MVDC systems that will be market drivers. These include MVDC Collectors and Corridors, and MVDC-fed multi-EV fast charging stations, factories and buildings. Voltage ranges between 1kV and 36kV will be considered. The mixed MVAC and MVDC system will be simulated with particular focus on fault behaviors of MVDC Corridors, multiple MVDC Corridors forming a meshed grid, localized MVDC buses at 1kV-1.5kV with direct-connected hybrid energy storage and responses to Line-to-Ground (LG) faults within a multiple EV fast charging system. The impacts of Line-to-Line (LL), LG and Multiple Line-to-Ground (MLG) faults on MVAC and MVDC systems connected by non-isolated Modular Multi-Level Converters (MMCs) and other multi-level Voltage Source Converters (VSCs) will be studied and how LL, LG and MLG in the MVAC systems affect the MVDC system and vice-versa. A HiL-CHiL system will be developed using combined NI FlexRIO and Typhoon platforms. Offline, accelerated time and real-time simulations will be performed using these platforms. The novelty of this approach is to utilize a highly detailed system-wide simulation of PEC-based distribution systems in order to discover the behaviors associated with emerging MVDC and MVAC/MVDC electrical grid structures and MVAC structures with high PEC-based DER penetration. The focus is on the very short time-frame transients that affect the system resiliency. The detailed system simulations include full-switching PECs and their controls, a range of DERs with adequate detail for simulation of short-time events, specific component locations with associated cable lengths and impedances, state machine-based protection scheme and device implementations. The unique outcomes of this research will be to discover the interactions between PECs under a wide range of MVAC/MVDC system configurations and fault scenarios. These results will help to define the requirements for DCCB implementations and provide a foundation for design of protective schemes and derivation of protective control hardware requirements. Year Two (Y2) will focus on a resilient distribution design process and focuses on the responses of interconnected PECs and DCCBs with real hardware digital controls. The CHiL part of the system will replicate practical implementations of digital-based protective relaying systems, including data monitoring and the implementation of adaptive protective schemes. Using classification and regression technics from machine learning algorithms, informed by previously collected and classified data and detection of configuration states, reconstructed from data exchanged between adjacent DCCBs and PECs in the system.


Posted on

December 22, 2020

Submit a Comment