Science Score: 54.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
    Found CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
  • Academic publication links
    Links to: arxiv.org
  • Committers with academic emails
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (13.3%) to scientific vocabulary
Last synced: 6 months ago · JSON representation ·

Repository

Basic Info
  • Host: GitHub
  • Owner: openEDI
  • License: bsd-3-clause
  • Language: Python
  • Default Branch: main
  • Size: 1.56 MB
Statistics
  • Stars: 0
  • Watchers: 2
  • Forks: 1
  • Open Issues: 0
  • Releases: 0
Created about 2 years ago · Last pushed over 1 year ago
Metadata Files
Readme License Citation

README.md

Project Title

A Data-Driven Approach for High-Impedance Fault Localization in Distribution Systems

Appendix

https://arxiv.org/abs/2311.15168

NREL Software Record

SWR-24-13

Usage

```javascript

Pick some high impedance fault data for classification

filepaths = ['IEEE123PVL1C1B5HIF.csv', 'IEEE123PVL1C1B34HIF.csv', 'IEEE123PV_L1C1B45HIF.csv'] ```

The high impedance fault (HIF) data for the IEEE 123-bus system can be downloaded from here.

The user can select the HIF data from a few locations for the classification. The algorithm first processes the data and apply piecewise linear approximation to the voltage-current trajectory. Following that, the simplified function features are collected as inputs and SVM is utilized for the HIF identification.

When SVM is used for the HIF identification task, the learning output is the fault location label (e.g., bus number), and we propose utilizing the features from the approximation functions as the input. Specifically, for the piecewise linear approximation, the SVM input can be constructed as:

math x_{\mathcal{L}} = \{s_1, s_2, s_3\} which consists of slope rates for all segments in the piecewise linear function.

Since quadratic functions contain more information than linear functions, SVM inputs under quadratic approximation will be lifted to a higher-dimensional space:

math x_{\mathcal{Q}} = \{(m_1,n_1),(m_2,n_2),(m_3,n_3)\}

which include both quadratic and linear coefficients for each piece of the function.

Examples

The performance of the piecewise approximation and SVM algorithm for HIF identification tasks can be validated using the IEEE 123-node test feeder. The test feeder is simulated in EMTP-ATP (ElectroMagnetic Transients Program, version Alternative Transients Program) with 14 rooftop photovoltaic (PV) units incorporated. In this model, the loading condition ({0.4, 1.0} per unit) and PV capacity ({0.4, 0.6, 0.8, 1.0} per unit) are varied for different steady-state scenarios. The HIF simulations span across 66 locations in total, with 63 internal nodes as well as 3 adjacent nodes outside the feeder.

The classification of two faults at buses 7 and 64 using the Gaussian kernel as an example is shown below:

SVM Illustration 1

where the hyperplane in red distinctly classifies these two faults using the slope rates of V-I trajectories. For the SVM under multiple HIFs, we can project the input to a two-dimensional space by keeping only two slope rates (s1 and s2). The performance of SVM with a linear kernel on three different HIFs is demonstrated below:

SVM Illustration 2

These faults are simulated at buses 7, 64, and 82, respectively. Because they are located in different areas of the system, even with partial inputs, the linear SVM can efficiently identify these HIFs.

Owner

  • Name: Open Energy Data Initiative
  • Login: openEDI
  • Kind: organization

The OEDI Data Lake is a centralized repository of high-value energy research datasets aggregated from the U.S. Department of Energy’s Programs and Offices.

Citation (CITATION.cff)

# This CITATION.cff file was generated with cffinit.
# Visit https://bit.ly/cffinit to generate yours today!

cff-version: 1.2.0
title: >-
  High Impedance Fault Localization Algorithm Using Support
  Vector Machine
message: >-
  If you use this software, please cite it using the
  metadata from this file.
type: software
authors:
  - given-names: Yuqi
    family-names: Zhou
    affiliation: NREL
  - given-names: Yuqing
    family-names: Dong
    affiliation: NREL
  - given-names: Rui
    family-names: Yang
    affiliation: NREL
identifiers:
  - type: doi
    value: 10.48550/arXiv.2311.15168
  - type: other
    value: SWR-24-13
    description: NREL Software Record
repository-code: 'https://github.com/openEDI/oedisi-hif-identification'
abstract: >-
  Accurate and quick identification of high-impedance faults
  is critical for the reliable operation of distribution
  systems. Unlike other faults in power grids, HIFs are very
  difficult to detect by conventional overcurrent relays due
  to the low fault current. Although HIFs can be affected by
  various factors, the voltage current characteristics can
  substantially imply how the system responds to the
  disturbance and thus provides opportunities to effectively
  localize HIFs. In this work, we propose a data-driven
  approach for the identification of HIF events. To tackle
  the nonlinearity of the voltage current trajectory, first,
  we formulate optimization problems to approximate the
  trajectory with piecewise functions. Then we collect the
  function features of all segments as inputs and use the
  support vector machine approach to efficiently identify
  HIFs at different locations. Numerical studies on the IEEE
  123-node test feeder demonstrate the validity and accuracy
  of the proposed approach for real-time HIF
  identification. 
keywords:
  - Systems and Control
  - Machine Learning
  - Distribution System Algorithms
  - OEDISI
license: BSD-3-Clause

GitHub Events

Total
  • Fork event: 1
Last Year
  • Fork event: 1

Committers

Last synced: 7 months ago

All Time
  • Total Commits: 14
  • Total Committers: 3
  • Avg Commits per committer: 4.667
  • Development Distribution Score (DDS): 0.214
Past Year
  • Commits: 1
  • Committers: 1
  • Avg Commits per committer: 1.0
  • Development Distribution Score (DDS): 0.0
Top Committers
Name Email Commits
zhouyuqisconvex 1****x 11
Joseph McKinsey 3****y 2
Joseph McKinsey j****y@n****v 1
Committer Domains (Top 20 + Academic)

Issues and Pull Requests

Last synced: 7 months ago


Dependencies

Dockerfile docker
  • python 3.12.1-slim-bullseye build
requirements.txt pypi
  • matplotlib *
  • numpy *
  • pandas *
  • scikit-learn *