Enhancing robustness of DC microgrid protection during weather intermittency and source outage for improved resilience and system integrity. (28th November 2021)

• Record Type: Journal Article
• Authors: Prasad Tiwari, Shankarshan; Koley, Ebha; Manohar, Murli; Ghosh, Subhojit; Mohanta, Dusmanta Kumar; Bansal, Ramesh C.
• Is Part Of: International transactions on electrical energy systems. Volume 31:Number 12(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-11-28
• Subjects: DC microgrid -- distributed energy resources -- distributed power generation -- photovoltaic system -- power grid -- power system protection -- weather intermittency; Electric power -- Periodicals; Electric power systems -- Periodicals; Electrical engineering -- Periodicals; 621.3
• DOI: 10.1002/2050-7038.13243
• Languages: English
• ISSNs: 2050-7038

Abstract

Abstract: Despite the advantages pertaining to efficiency, stability, and ease of integration with renewable energy sources, the wider acceptance of DC microgrids has been hindered by the complexity of the protection task. In addition to the absence of zero crossings, the weather‐dependent fluctuations in the operation of renewable sources and the high probability of distributed energy resources (DERs) outage further complicate the fault detection and classification task. This paper proposes a reliable DC microgrid protection approach with improved robustness against weather intermittency and possible source outages. The protection framework involves identifying the network topology from local information followed by executing a set of convolution neural networks (CNNs) to perform the protection tasks. The proposed CNN‐based scheme reduces the complexity and the associated computational cost of feature extraction for performing the intended protection tasks. The novelty of the scheme pertains to the inclusion of the stochastic model for representing weather variation and a mechanism for adapting the relaying mechanism during changes in the network structure. The same allows for avoiding relay malfunction during weather intermittency and source outage, thereby improving grid resilience and maintaining system integrity. The DC microgrid model has been simulated using MATLAB and Simulink. The proposed scheme has been extensively validated for various operating scenarios involving variations in wind speed, solar irradiance, fault parameters, and network configuration. Improvement toward reduction in memory requirement is still a challenge for the hardware implementation of CNN, which can be considered as a potential area for future research. Abstract : A DC microgrid protection scheme is proposed with improved robustness against weather intermittency and DER outages. The protection framework based on the stochastic modeling of weather variables involves online identification of network topology and execution of a set of convolution neural networks to perform fault detection, classification, and location estimation.