Signals_Classification_Model

Overview

In this example, a convolutional neural network (CNN) model will be trained to identify 16 types of electrical signals common in the field of electrical power distribution by analyzing the signal that contains noise and identifying the pattern of the main signal (Accuracy + 95% ).

This model can be used to classify any similar data given that some minor modifications must be made in the code.

The model is built and Trained using popular open source Machine Learning libraries such as Tensorflow and Keras on Jupyter-Notebook Platform.

The Signals Dataset Generated Using This MATLAB Dataset Generator project : https://www.sciencedirect.com/science/article/abs/pii/S0378779621001334?dgcid=author

The 16 Type of Signals :

1. Flicker
2. Flicker+Harmonics
3. Flicker+Sag
4. Flicker+Swell
5. Harmonics
6. Impulsive Transient
7. Interruption
8. Interruption+Harmonics
9. Normal
10. Notch
11. Oscillatory transient
12. Sag
13. Sag+Harmonics
14. Spike
15. Swell
16. Swell+Harmonics

Please Check the Attached PDF file (Model_Report.pdf) for the results and Code preview and other details.