VibScope

Vibrational Analyser

DEVELOPMENT OF IoT BASED VIBRATION MONITORING AND SPECTRUM ANALYSIS SYSTEMS FOR TECHNICAL OBJECTS

//Done as a course project//

Table of contents

• Objective
• Screenshots
• Technologies
• Bill of Materials
• Getting Started
• Setup
• Features
• Status
• Gallery
• Collabrations
• Contact

Objective

• To design a low cost alternative to vibration monitor and spectrum analyzer using the Raspberry Pi microcomputer and 3-axis digital MEMS MPU6050 accelerometer.
• To study if inexpensive vibration monitoring systems could be suitable for condition monitoring
• To highlight the impact of different components of the signal chain to the measured vibration signal itself and familiarize the reader with the signal chain found in vibration monitoring.
• To analyze vibration parameters in order to predict and prevent possible accidents, thus reducing the costs associated with the failure of the cutting tools, expensive parts and assemblies of the CNC machine.

Screenshots

LabVIEW UI

LabVIEW UI for FFT based Vibrational analysis

Technologies

• Python & Matlab
• IoT integration

Bill of materials

Electronic components

Component	Description	Source	Price (Rs.)
Raspberry Pi 3 Model B	Version 1.2	Amazon	2750
MicroSD Card	8GB	Amazon	230
SD Card Reader	SD card adapter for writing	Amazon	40
Jumper wires (F-F)	Connects the RPi to Sensor	Local shop	30 /10 wires
MPU6050	Accelerometer and Gyroscope	Sparkfun	190

Getting started

Let's get started! First thing first, solder the MPU6050 with breakout board pins and connect to the Raspberry Pi for the best results. Other means of connection, such as jumper wires or connectors are discouraged as they might disconnect during usage.

Now, we are also going to need a few tools, so downloading them now is a good idea.

Tools

• Etcher - SD Card flasher
• Advanced IP Scanner - IP Scanner
• Putty - SSH client
• VNC Viewer - RDP client

Raspbian Jessie Lite

Download and flash Raspbian Jessie Lite on a micro SD card (preferably a 8GB one) with Etcher, or an alternative flasher. We will be using the lite version of Raspbian since we will not need a video output or many of the software packages that come with the Desktop version.

After flashing the OS on the SD card, we need to enable the SSH server and connect it to a Wireless Access Point in order to communicate with it. Open the micro SD card directory from your File Explorer and create an empty file called ssh. Create another file called wpa_supplicant.conf with the following text:

ctrl_interface=DIR=/var/run/wpa_supplicant GROUP=netdev
update_config=1
network={
    ssid="YOUR_SSID"
    psk="YOUR_PASSWORD"
    key_mgmt=WPA-PSK
}

Where YOUR_SSID and YOUR_PASSWORD are the SSID and Password of your WiFi router.

This is what your /boot directory should look like now

The Raspberry Pi Zero W on boot will read these two files and automatically enable the SSH server and connect it to your WiFi router.

Now eject the micro SD card, put it in the Raspberry and power it on with a micro USB cable.

Now your Pi should have connected to your router, and we need to find out its IP address. An easy way to do this is to use Advanced IP Scanner.

Now that we have found out the IP address, let's SSH into it. We'll use Putty for that.

You'll be asked username and password.

The default user name is:

pi

And the default password is:

raspberry

Now you should be logged in. It is advised to change these credentials for safety reasons.

We're all set! Let's move forwards by installing some essentials modules.

Note that all the following commands will be executed on the Raspberry Pi Zero W through the Putty SSH session.

---

Python

Python will be used to read from the sensors and transmit the data to Node-RED via mqtt

sudo apt-get install python-2.7 python-pip

Now let's install the Scipy, Numpy, Matplotlib and pyplot library for Python

sudo apt-get install build-essential gfortran libatlas-base-dev python-pip python-dev
sudo pip install --upgrade pip
sudo apt install python-numpy python-scipy python-matplotlib

In order to read from the I2C from Python, we need to install the smbus module

sudo apt-get install python-smbus

This library is needed so that Python can read from the MPU6050 sensor

sudo apt-get install build-essential python-pip python-dev python-smbus git
sudo pip install mpu6050-raspberrypi

---

Setup

Clone this repository in your /home/pi directory of your Raspberry Pi.

To clone and run this application, you'll need Git installed on your computer. From your command line:

# Clone this repository
$ git clone https://github.com/CoDeRgAnEsh/viBscope

To run the program for callibrating MPU6050 accerleometer, open the Src folder

cd /home/pi/vibscope/Src

And run the following program

python /home/pi/vibscope/Src/g.py

To run the program for realtime MPU6050 accerleometer readings and graph, open the Src/live folder

python /home/pi/vibscope/Src/live/record.py

To run the program for realtime logging MPU6050 accerleometer readings and graph, open the Src/log folder

python /home/pi/vibscope/Src/log/Start.py

To run the program for Vibrational analysis MPU6050 accerleometer readings and graph, open the Src/Vibrational Analysis folder

python /home/pi/vibscope/Src/Vibrational analysis/MPU.py

Note use your own Pi IP address that you found with Angry IP Scanner

Code Examples

Python implementation of RMS time plot

#Compute RMS and Plot
tic = time.clock()
w = np.int(np.floor(Fs)); #width of the window for computing RMS
steps = np.int_(np.floor(N/w)); #Number of steps for RMS
t_RMS = np.zeros((steps,1)); #Create array for RMS time values
x_RMS = np.zeros((steps,1)); #Create array for RMS values
for i in range (0, steps):
	t_RMS[i] = np.mean(t[(i*w):((i+1)*w)]);
	x_RMS[i] = np.sqrt(np.mean(x[(i*w):((i+1)*w)]**2));  
plt.figure(2)  
plt.plot(t_RMS, x_RMS)
plt.xlabel('Time (seconds)')
plt.ylabel('RMS Accel (g)')
plt.title('RMS - ' + file_path)
plt.grid()
toc = time.clock()
print("RMS Time:",toc-tic)

Matlab FFT & Power Spectral Density plot

    %Compute FFT & PSD
    Fs = fActual;
    x = datalist(:,2);     
    N = length(x);
    freq = 0:Fs/length(x):Fs/2;
    xdft = fft(x);
    xdft = xdft(1:floor(N/2)+1);
    psdx = (1/(Fs*N)) * abs(xdft).^2;
    psdx(2:end-1) = 2*psdx(2:end-1);
    psdx = psdx';
    xdft = 1/length(x).*xdft;
    xdft(2:end-1) = 2*xdft(2:end-1);
    xdft = xdft';
    phase = unwrap(angle(xdft));
    xdft = abs(xdft);

Features

List of features ready and TODOs for future development

• FFT plot GUI for input signals
• Accerleometer data logging
• Matlab implementation of Analysis

To-do list:

• Instead of Signal simulation, use data from accerlometer.
• Hardware and Software bindings
• IoT implementation.

Status

Project is: in progress !

Gallery

Check out the Snaps from here

Collabrations

 Ganesh Kumar T K (MSM17B034)
 Dhilipan S (MSM17B002)
 Chandralekha R (MSM17B027)
 Ext. Avital Bhaptakri (3rd Yr, Mech, NIT Raipur)

Contact

Created by @coderganesh - feel free to contact me!