Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Machine Learning-Based Protection and Fault Identification of 100% Inverter-Based Microgrids

Milad Beikbabaei, Michael Lindemann, Mohammad Heidari Kapourchali, Ali Mehrizi-Sani

TL;DR

The paper addresses protection challenges in microgrids consisting entirely of inverter-based resources, where low fault currents and bidirectional flows complicate conventional schemes. It proposes a lightweight decision-tree–based protection method that relies solely on local RMS measurements (I, V, P, Q) to simultaneously detect faults and classify fault types, with training data generated from PSCAD/EMTDC simulations. The approach is validated on a 4-inverter, 4-bus microgrid and demonstrates fault detection times under 5 ms and accurate fault-type identification across seven fault scenarios, though performance degrades for untrained high-impedance faults. The work offers a communication-free, real-time protection solution suitable for practical deployment on digital relays and informs future extensions to larger networks and varied configurations.

Abstract

100% inverter-based renewable units are becoming more prevalent, introducing new challenges in the protection of microgrids that incorporate these resources. This is particularly due to low fault currents and bidirectional flows. Previous work has studied the protection of microgrids with high penetration of inverter-interfaced distributed generators; however, very few have studied the protection of a 100% inverter-based microgrid. This work proposes machine learning (ML)-based protection solutions using local electrical measurements that consider implementation challenges and effectively combine short-circuit fault detection and type identification. A decision tree method is used to analyze a wide range of fault scenarios. PSCAD/EMTDC simulation environment is used to create a dataset for training and testing the proposed method. The effectiveness of the proposed methods is examined under seven distinct fault types, each featuring varying fault resistance, in a 100% inverter-based microgrid consisting of four inverters.

Machine Learning-Based Protection and Fault Identification of 100% Inverter-Based Microgrids

TL;DR

The paper addresses protection challenges in microgrids consisting entirely of inverter-based resources, where low fault currents and bidirectional flows complicate conventional schemes. It proposes a lightweight decision-tree–based protection method that relies solely on local RMS measurements (I, V, P, Q) to simultaneously detect faults and classify fault types, with training data generated from PSCAD/EMTDC simulations. The approach is validated on a 4-inverter, 4-bus microgrid and demonstrates fault detection times under 5 ms and accurate fault-type identification across seven fault scenarios, though performance degrades for untrained high-impedance faults. The work offers a communication-free, real-time protection solution suitable for practical deployment on digital relays and informs future extensions to larger networks and varied configurations.

Abstract

100% inverter-based renewable units are becoming more prevalent, introducing new challenges in the protection of microgrids that incorporate these resources. This is particularly due to low fault currents and bidirectional flows. Previous work has studied the protection of microgrids with high penetration of inverter-interfaced distributed generators; however, very few have studied the protection of a 100% inverter-based microgrid. This work proposes machine learning (ML)-based protection solutions using local electrical measurements that consider implementation challenges and effectively combine short-circuit fault detection and type identification. A decision tree method is used to analyze a wide range of fault scenarios. PSCAD/EMTDC simulation environment is used to create a dataset for training and testing the proposed method. The effectiveness of the proposed methods is examined under seven distinct fault types, each featuring varying fault resistance, in a 100% inverter-based microgrid consisting of four inverters.
Paper Structure (16 sections, 1 equation, 6 figures, 2 tables)

This paper contains 16 sections, 1 equation, 6 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Test System
  3. Basics of Inverter Control
  4. Grid-Following Mode
  5. Grid-Forming Mode
  6. 100% Inverter-Based Microgrid
  7. Methodology
  8. Algorithm Selection
  9. Decision Tree Basics
  10. Dataset Preparation and Feature Selection
  11. Training and Hyperparamters Selection
  12. Simulation results and Evaluation
  13. Case 1: $AG$ Fault on Bus 1.
  14. Case 2: $ACG$ Fault on Bus 2.
  15. Case 3: $ABCG$ Fault on Bus 3.
  16. ...and 1 more sections

Figures (6)

  • Figure 1: Decoupled control scheme of a grid-following inverter for $d$- and $q$-axes.
  • Figure 2: Droop control scheme of a grid-forming inverter for: (a) $P-f$, (b) $Q-V$.
  • Figure 3: Single-line diagram of the microgrid.
  • Figure 4: $AG$ fault located at bus 1: (a) fault detection and (b) fault type.
  • Figure 5: $ACG$ fault located at bus 2: (a) fault detection and (b) fault type.
  • ...and 1 more figures