IRIS: Machine Learning Driven Grease Diagnostics

IRIS uses basic ML and FTIR spectroscopy to detect grease degradation early, learning from every sample to predict machine health before failures occur.

Comment

Inspiration

I wanted to write programs and software that allow people with monotonous or tedious jobs to be more efficient and finish their tasks faster

What it does

This program will make the analysis of grease much easier, as a basic Machine Learning System takes over the monotoues job of reading graphs, interpreting data, or logging customer information.

How we built it

We built this using Python, Principal Component Analysis, and Isolation Forest, all in VSC

Challenges we ran into

Lots of issues with the machine learning part, like understanding how to train it on the data and what that could look like. What did we want it to try and understand about the future data we would give it, and what parameters it would have.

Accomplishments that we're proud of

I am proud that there is a pretty basic Machine Learning system inside the program. While we don't have enough to train it fully, it is awesome that we have the start of something right there.

What we learned

I learned a lot about Python setup and environments, as well as GUIs with Python, alongside machine learning techniques and protocols

What's next for IRIS: Machine Learning Driven Grease Diagnostics

Further development of the machine learning algorithm and absorbing more data, so it is even stronger.

Built With

Share this project:

Updates