The Indian Institute of Technology, Delhi (IIT Delhi) has added another achievement to its list of researches and innovative projects. The researchers of the institute have developed a first of its kind Machine Learning (ML) software that can predict and optimise glass compositions. With this innovation, addressing the shattered windscreens and cracked phone screens might become easier soon.
Desiring window panes, glass utensils, and phone screens that can resist damage will now be an easier thing to achieve. IIT Delhi researchers have come up with a solution that provides a mechanism that can predict glass compositions and can be used for developing products with tailored properties
The unique Machine Learning software is named Python for Glass Genomics (PyGGi) and it has been made for enabling companies and researchers to easily predict glasses that have superior properties like crack resistance and scratch resistance.
Also Read: IIT Delhi Collaborates with 5 CSIR Laboratories for Promoting Cooperative Research
One of the project investigators, Professor N. M. Anoop Krishnan who belongs to the Department of Civil Engineering, spoke about this new research project and said that the key of the entire project is understanding and predicting the relationship of composition-structure-property. He said that through this software, it will be easy to determine whether the glass is scratch-free or breakable and then it can be used for several things such as car screens, phones, window panes, etc. Krishnan said that this software is the first of its kind in the world and it will also help in designing bulletproof glass or unbreakable glass.
Also Read: IIT Kanpur Collaborates with Ericsson for Combating Pollution
Another Project Investigator (PI) from the Department of Mechanical Engineering, Professor Hariprasad Kodamana, said that PyGGi will be updated constantly in order to meet all the challenges that the field has to offer. He also said that the team of researchers is open to creating raw modules based on the requirements of the user. These modules can be exclusively given to those users who support the research in PyGGi.