• The microgrid consists of two synchronous generators, one photovoltaic
(PV) module, and one wind farm.
• The IEC microgrid is modelled in RSCAD.
• The RTDS allows realistic and accurate analysis of the
dynamics of the power system.
• Physical power, control and protection devices can be
interfaced to the RTDS for hard-ware in the loop simulations.
Abstract
• This project develops a data-mining based intelligent
protection scheme for fault detection and classification in a
microgrid with multiple distributed generation interfaces.
• The protection scheme retrieves one cycle post-fault
current signal samples of each phase from fault inception at
bus ends of feeders to derive some effective features.
• The retrieved current samples are pre-processed using S-
transform and TT-transform to obtain statistical features,
such as energy, mean, standard deviation (STD), entropy,
median absolute deviation (MAD) which are further used
to build a machine-learning-based model.
• The model will be validated on the unseen data set for fault
detection and classification in the microgrid.
• The proposed relaying scheme will be developed on a real
time digital simulator (RTDS) platform which is integrated
with Matlab.
• Using extensive test results, the performance of the
proposed intelligent relaying scheme will be evaluated for
microgrids with wide variations in operating parameters
.
IEC Microgrid
An Intelligent Protection Scheme for Microgrids
using Data-mining-based Models
O. A. Gashteroodkhani, S. Aznavi, H. Livani, M. Majidi, M. Etezadi Amoli
Electrical and Biomedical Engineering Department, University of Nevada, Reno
Diesel Generating System in RSCAD
Conclusions and Future Work
Microgrid Under Study
System Implementation in Real Time Digital Simulator
PV System in RSCAD
DFIG Wind Generator System in RSCAD
Simulation Results
• The IEC microgrid is simulated in RSCAD.
• The performance of the microgrid has been evaluated by simulating different
types of faults.
• Using the simulated microgrid, fault data will be generated by different types
of fault conditions including fault resistance, fault locations, fault inception
angles, and various levels of DG penetration within different topologies of
microgrid (radial and mesh).
• Also, machine learning models such as deep belief network (DBN) will be
applied for developing the protection scheme.
Current signal at bus 3
in grid connected mode
Current signal at bus 3
in islanded mode
Fault current at bus 2 for a three
phase fault on bus 2
in grid connected mode
Fault current at bus 6 for a three
phase fault on bus 2
in grid connected mode
Acknowledgement
This material is based upon work supported by the National Science Foundation
under Grant No. IIA-1301726.
A three phase to ground fault with duration of 0.05 and Rf=50.

Mg protection 04022018

  • 1.
    • The microgridconsists of two synchronous generators, one photovoltaic (PV) module, and one wind farm. • The IEC microgrid is modelled in RSCAD. • The RTDS allows realistic and accurate analysis of the dynamics of the power system. • Physical power, control and protection devices can be interfaced to the RTDS for hard-ware in the loop simulations. Abstract • This project develops a data-mining based intelligent protection scheme for fault detection and classification in a microgrid with multiple distributed generation interfaces. • The protection scheme retrieves one cycle post-fault current signal samples of each phase from fault inception at bus ends of feeders to derive some effective features. • The retrieved current samples are pre-processed using S- transform and TT-transform to obtain statistical features, such as energy, mean, standard deviation (STD), entropy, median absolute deviation (MAD) which are further used to build a machine-learning-based model. • The model will be validated on the unseen data set for fault detection and classification in the microgrid. • The proposed relaying scheme will be developed on a real time digital simulator (RTDS) platform which is integrated with Matlab. • Using extensive test results, the performance of the proposed intelligent relaying scheme will be evaluated for microgrids with wide variations in operating parameters . IEC Microgrid An Intelligent Protection Scheme for Microgrids using Data-mining-based Models O. A. Gashteroodkhani, S. Aznavi, H. Livani, M. Majidi, M. Etezadi Amoli Electrical and Biomedical Engineering Department, University of Nevada, Reno Diesel Generating System in RSCAD Conclusions and Future Work Microgrid Under Study System Implementation in Real Time Digital Simulator PV System in RSCAD DFIG Wind Generator System in RSCAD Simulation Results • The IEC microgrid is simulated in RSCAD. • The performance of the microgrid has been evaluated by simulating different types of faults. • Using the simulated microgrid, fault data will be generated by different types of fault conditions including fault resistance, fault locations, fault inception angles, and various levels of DG penetration within different topologies of microgrid (radial and mesh). • Also, machine learning models such as deep belief network (DBN) will be applied for developing the protection scheme. Current signal at bus 3 in grid connected mode Current signal at bus 3 in islanded mode Fault current at bus 2 for a three phase fault on bus 2 in grid connected mode Fault current at bus 6 for a three phase fault on bus 2 in grid connected mode Acknowledgement This material is based upon work supported by the National Science Foundation under Grant No. IIA-1301726. A three phase to ground fault with duration of 0.05 and Rf=50.