Material Synthesis using ML
Synthesis of materials of high end applications with zero compromise on quality is a tedious process generally and involves large amounts of trial and errors which lead to higher production costs. So in this project we aim to optimize these complex synthesis problems using various Machine learning algorithms which lead to better quality of products at lower costs. For our analysis we have used an aluminium bronze casting, that is coated with PVD (Physical vapour deposition). We want to optimise the SN ratio of the casting by altering the different coating parameters of PVD.
We built a regression model that predicts the SN ratio of the material for different conditions. Using SN ratio we can predict the quality of our material. The hardness, surface roughness, material cost etc. can be predicted.
Python, Jupyter notebook
We have used the data set from the given research paper on effects of Different Parameters of
PVD Coated Turning on Aluminium Bronze Casting. Link to paper
The methodology used in predicting the SN ratio of the material is shown here
We found out that by using a polynomial regression model we were able to accurately predict the
outcomes quite accurately.
● Data cleaning
● Machine learning algorithms like regression.
● Understanding of machining, metal forming and how different parameters affect it.
Using Machine learning we can optimise various other processes and can apply in:
-Physical Vapour Deposition
-Thin films synthesis
-Coatings in aerospace materials
-Powder metallurgy(to get the powder of desired quality)
-Process Kinetics of various metallurgical reactions.
-Production of polymers
The inspiration behind this idea was a research paper
● Priyank Agarwal (8971114411) - email@example.com
● Karan Jain (978495951) - firstname.lastname@example.org
● Ansar Nadaf (6363074088) - email@example.com
● Ken kaushal (9535977944) - firstname.lastname@example.org
● Rachana (9538817799) - email@example.com