SEED_PQD_v1 (SEED - Power Quality Disturbance Dataset ver1)

We are pleased to share the dataset SEED-PQD-v1 (SEED Power Quality Distrubance Dataset v1) used in our study titled  "XPQRS: Expert power quality recognition system for sensitive load applications," published in Elsevier Journal Measurement. This dataset is invaluable for researchers and practitioners in the field of power quality analysis, especially those focusing on sensitive load applications. This dataset can be used in Python as well as in MATLAB. 

Access the published paper:

https://www.sciencedirect.com/science/article/abs/pii/S0263224123004530

Dataset Details:

Fundamental Frequency: 50 Hz

Sampling Rate: 5 kHz

Number of Classes: 17

Signals per Class: 1000

Length of Each Signal (samples): 100

Length of Each Signal (time): 20 ms

Amplitude of Each Signal: Scaled between -1 to 1

Data Format:

The dataset is available in two formats: MATLAB and CSV.

MATLAB File:

Filename: 5Kfs_1Cycle_50f_1000Sam_1A.mat

Structure: A matrix of dimensions (1000 x 100 x 17), where:

1000 = Signals per class

100 = Samples per signal

17 = Number of classes

Class Order:

Pure_Sinusoidal

Sag

Swell

Interruption

Transient

Oscillatory_Transient

Harmonics

Harmonics_with_Sag

Harmonics_with_Swell

Flicker

Flicker_with_Sag

Flicker_with_Swell

Sag_with_Oscillatory_Transient

Swell_with_Oscillatory_Transient

Sag_with_Harmonics

Swell_with_Harmonics

Notch

CSV Files:

Files: 17 CSV files, one for each class.

Structure: Each CSV file has dimensions (1000 x 100), where:

1000 = Signals per class

100 = Samples per signal

Usage:

This dataset is designed to support the development and testing of power quality recognition systems. The 17 classes cover a broad range of power quality disturbances, providing a comprehensive resource for training machine learning models and validating their performance in recognizing various types of power quality issues.

Acknowledgements:

All users of the dataset are advised to cite the following article:

Citation: Muhammad Umar Khan, Sumair Aziz, Adil Usman, XPQRS: Expert power quality recognition system for sensitive load applications, Measurement, Volume 216, 2023, 112889, ISSN 0263-2241, https://doi.org/10.1016/j.measurement.2023.112889. Link to the article

Thank you for your interest in our work. We hope this dataset facilitates further advancements in power quality analysis and related fields.keywords: Power Quality Recognition,  Power Quality Classification, Electrical Signal Analysis, Power System Disturbances, Signal Processing, Power Quality Monitoring