Many industrial applications involve checking for flaws in places where worker cannot act due to immense heat, toxic gases & limited space. Hence it is essential to monitor such parameters in high value engineering sectors such as electrical substations and chemical sectors such as oil and gas power plant in order to protect human health and environment, and in certain cases preventing accidents.
Even though tools exists they at bulky, heavy and highly unyielding which consumes a lot of time and energy during an emergency. Also, there is a lack of thermal camera in our country and the existing ones are a bit costly (1000$).
In case, a fire accident occurs it would be very difficult for the fire fighters to locate the injured people in the plant.
EXISTING SYSTEM :
1. DAQRI Smart Helmet
- Augmented Reality based Thermal Vision
- Provides guided work information
- Expensive in comparison to other solutions
- Difficult to reach critical areas
2. Electrical Substation Inspection Robot
- 360° pan-and-tilt positioner
- Autonomous Vehicle Control System
- Detection of gas is not possible ( Requires additional system )
- Large size ( 54''x32''x62'' )
I, Varun, a final year engineering student from Chennai, India. I've always been into Robotics and combining electronics and Product Design.
Inspired after watching the movie ' The 33 ', I realized the hurdles and challenges faced by Industrial Workers in critical areas. Speaking to a Miner personally, I realized that they mostly require an assistive device to help them if the situation goes critical. Hence, I build up an AR based helmet for Miners to record the site Temperature and Gas content. While presenting this idea to Mr. Kshir Sagar, a Senior Manager in Fire & Safety Department at CPCL ( Chennai Petroleum Corporation Limited ), he gave a positive feedback and few ideas for making this better.
After seeing the unbelievable progress in Machine Learning for the past few years, empowering Industrial Workers with Machine Learning for fault detection and testing is my current goal. Along the way I've read statistics on number of fire accidents that occur in chemical power plants and hence I wanted to do my path. I've also been seeking help from industries to make this dream, a reality.
OUR SOLUTION :
We propose a novel solution that could possibly enhance the current industrial safety protocol.
A semi-autonomous miniature multi-spectral diagnostic wireless controlled tank based rover which can monitor thermal characteristics of valves using Thermal Camera and monitor gas leaks in oil and gas power plants.
It is designed to be modular to add custom sensors with it. Currently, it houses a thermal, visual camera along with a gas sensors.
If an abnormal temperature is detected, the IR Image Processing system identifies an inappropriate heat which in turn sends the thermal data to the control centre through VNC, a Wi-Fi server with which alerts or guidance can be in turn provided to the technician, so that necessary steps can be taken.
Block Diagram :
System Design :
Setting up FLIR Lepton with Raspberry Pi
Once the connections are done, move to the software part :
- Install Raspbian OS from NOOBS on your Raspberry Pi
- Enable SPI and I2C interfaces
- Download and Install QT4 Dev Tools
- Go to the Pure Engineering and retrieve the raspberrypi_video directory
- Unzip the Raspberry Pi video directory ( LeptonModule-master ) and raspberrypi_video
- "make" Lepton SDK, so into the “LeptonSDKEmb32PUB” directory and run cd "make”
- Get back out to the “raspberrypi_video” directory and run "qmake && make"
- Type sudo ./raspberrypi_video in your command line
And, Voila :) Thermal images will start pouring out.
Setting up VNC with Raspberry Pi
Now, that's not over. In order to visualize the thermal data remotely, it is necessary to install VNC Server ( Virtual Networking...Read more »