Edge Impulse Inc. / Sensorless Drive Diagnosis Feature Classifier Public

Edge Impulse Inc. / Sensorless Drive Diagnosis Feature Classifier

This is your Edge Impulse project. From here you acquire new training data, design impulses and train models.

About this project

This project uses pre-extracted features from the Dataset for Sensorless Drive Diagnosis

To evaluate the ability of a NN classifier to detect different types of AC motor faults based on 48 statistical features. Information on the feature extraction may be found in the citation below.

Note that the pre-computation of these features may be done using a custom DSP block within edge impulse. For more information and an example see: https://github.com/edgeimpulse/edge-impulse-emd-feature-dsp-block

[1] Bator, Martyna & Dicks, Alexander & Mönks, Uwe & Lohweg, Volker. (2012). Feature Extraction and Reduction Applied to Sensorless Drive Diagnosis. 10.13140/2.1.2421.5689.

[2] F. Paschke, C. Bayer, M. Bator, U. Mönks, A. Dicks, O. Enge-Rosenblatt, and V. Lohweg, “Sensorlose Zustandsüberwachung an Synchronmotoren,” in Proceedings 23. Workshop Computational Intelligence, Karlsruhe: KIT Scientific Publishing, 2013, pp. 211–225. [2] C. Bayer, M. Bator, U. Mönks, A. Dicks, O. Enge-Rosenblatt, and V. Lohweg, “Sensorless Drive Diagnosis Using Automated Feature Extraction, Significance Ranking and Reduction,” in 18th IEEE Int. Conf. on Emerging Technologies and Factory Automation (ETFA 2013): IEEE, 2013, pp. 1–4.

Creating your first impulse (100% complete)

Acquire data

Every Machine Learning project starts with data. You can capture data from a development board or your phone, or import data you already collected.

Design an impulse

Teach the model to interpret previously unseen data, based on historical data. Use this to categorize new data, or to find anomalies in sensor readings.

Deploy

Package the complete impulse up, from signal processing code to trained model, and deploy it on your device. This ensures that the impulse runs with low latency and without requiring a network connection.

Download block output

Title Type Size
Raw data training data NPY file 46478 windows
Raw data training labels NPY file 46478 windows
Raw data testing data NPY file 12006 windows
Raw data testing labels NPY file 12006 windows
NN Classifier model (version #1) TensorFlow Lite (float32) 46 KB
NN Classifier model (version #1) TensorFlow Lite (int8 quantized) 15 KB
NN Classifier model (version #1) TensorFlow Lite (int8 quantized with float32 input and output) 16 KB
NN Classifier model (version #1) TensorFlow SavedModel 59 KB
NN Classifier model (version #2) TensorFlow Lite (float32) 3 KB
NN Classifier model (version #2) TensorFlow Lite (int8 quantized) 2 KB
NN Classifier model (version #2) TensorFlow SavedModel 10 KB
NN Classifier model (version #3) TensorFlow Lite (float32) 111 KB
NN Classifier model (version #3) TensorFlow Lite (int8 quantized) 32 KB
NN Classifier model (version #3) TensorFlow SavedModel 328 KB

Clone project

You are viewing a public Edge Impulse project. Clone this project to add data or make changes.

Summary

Data collected
46m 47s

Project info

Project ID 79283
Project version 5