Giter VIP home page Giter VIP logo

1dcnn's Introduction

1DCNN

Summary

CNN, Convolutional Neural Network, is famous for image recognition, but could be a good modeling framework for time series data with multiple variables. Convolutional operation applied to 1d data sets and graphical interpretation of the logic will be explained. Classification of different types of bearing faults using raw accelerometer data is performed and compared with different approaches.

Datasets and Inputs

The data is available from CaseWestern Reserve University Bearing Data Center. The data is sampled from 3 accelerometers (Drive End, Fan End and Base) on an industrial equipment at 12k hz. However Normal data is missing on Base, so only Drive End and Fan End will be used. For training of auto-encoder, the Normal data will be used with some shifts to create large number of training data. To evaluate the fault classification capability, 30 data files from 10 types of faults and 3 types of Motor Load(HP) will be used as described in paper [1]. The 10 types of faults are Normal, Ball(0.007, 0.014, 0.021-inch faults), Inner Race (same 3 types as above), and Outer Race (same 3 types as above). One sample for the model is the form of 2,048x 2 (Drive End and Fan End) array. There are 19,800 test samples and 750 test samples. The 10 types of faults are completely balanced across training and test dataset. Test samples does not have any shift (overlap).

Prerequisites

  • Create ./data and ./data/download before running functions in data_prep.ipynb

Reference

[1] Wei Zhang,Gaoliang Peng, Chuanhao Li, Yuanhang Chen and Zhujin Zhang, โ€œA New Deep Learning Model forFault Diagnosis with Good Anti-Noise and Domain Adaption Ability on RawVibration Signalsโ€, MDPI Sensors, 2017.

1dcnn's People

Contributors

meitetsu3 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.