Aarooj Yashin Email & Phone Number

Edmonton, Ab, Ca

Edmonton, Alberta, Canada

Edmonton, Alberta, Canada

• Started with creating voxels of the simulation box of reaction system in biomass processing to capture the spatial and temporal distribution of the system.• Generated occupancy information of atoms of interest in the voxels and made a binary classifier to predict the solvent reorganization effects with time• Implemented an LSTM Autoencoder to get the latent space, and use this latent space to build the dense neural classifier• Currently have switched to transformer -based model for time series forecasting of co-ordinates of the atoms. • At this moment the prediction horizon (H) of 1 and 100 are giving acceptable results.• Target is to improve the H for the system with hyperparameter tuning.

Montreal, Quebec, Canada

Project: Improvement of atom-scale models of clay minerals using machine learning and molecular simulations.- Carried out DFT calculations in Quantum Espresso (QE) for aluminum (a part of bentonite).- Ran the self-consistent field calculations to prepare the input files for further steps- Using the results to find Bulk Modulus, Elastic Constants, and other properties to verify with existing literature results.- Creating a dataset for MLIP (Machine Learning Inter-atomic Potential) batch learning using QE and implementing the MTP (Moment Tensor Potential) algorithm to create an MLIP.

Jamnagar, Gujarat, India

Project: Stabilisation of LCO Hydro-cracker while processing Dryer Overhead Gasoline from DHDS UnitsThis project aims to solve the following problem:There are two diesel dryers producing gasoline, and the properties of this gasoline is unstable due to inefficiency in packing and maldistribution of reflux liquid over packing. Also the dryers often underperform because of high capacity. This high capacity arises due to high content of naphtha in DHDS feed. The gasoline streams enters the LCO Hydro-cracker Fractionator at the Heavy Naphtha Section, where it is an impurity as it has high FBP. This stream has a very high end point ( FBP) than expected due to inefficient separation in the packing of the dryers, so we have to compensate with optimum re-boiling and pump around duty in the LCO Hydro-cracker section. Here the problem arises as the re-boiler and pump around duty are limited (re-boiler is process to process heat exchanger and its surface area is fixed) so the heavy naphtha stream remains contaminated with impurities (gasoline) and the end point is higher than what it should have been.

Master'S Degree (Msc Thesis Based), Chemical Engineering

Bachelor'S Degree, Chemical Engineering, 8.74

Class 12, Pcm, 91.2%

Class 10, Cgpa: 10.0/10.0