Gaurav Kumar

Postgraduate(M.Tech, Computational Mechanics) [email protected]| Ó+91-9557230496

Department of Aerospace Engineering [email protected]
Indian Institute Of Technology,Kanpur ‡ Gaurav-Kumar | ¯ Gaurav-kumar

ACADEMIC QUALIFICATIONS

Year Degree/Certificate Institute CPI/%

2022 - Present Master Of Technology Indian Institute of Technology, Kanpur 7.57/10
2019 Bachelor of Technology National Institute Of Technology, Jamshedpur 6.77/10
2015 Higher Secondary(XII) Benhur Public School, Pilibhit 90.6%
2013 Secondary(X) Krishna Public School, Bareilly 8.8/10

INDUSTRIAL TRAINING
• Pantech.ai (17 July 2022-17 August 2022 )
– Worked on attendance system using face recognition.
• TATA MOTORS, Jamshedpur (June 2018-July 2018 )
– Manpower mapping of contractual labor in frame line and assembly line.
– Learned about Industrial safety and security.
ACADEMIC PROJECTS
• Master’s Thesis: Data-driven machine learning approach in elastoplastic FEM model and in crystal plasticity problems.
• Mentor: Dr. Pritam Chakarborty, Department of Aerospace Engineering, IIT Kanpur. (Ongoing)
To establish and explore the use of machine learning in solid mechanics problems:-
– Elastoplastic constitutive equations were formulated using MATLAB, generating the foundational dataset for our subsequent
machine-learning model.
– ANN model was trained from the above data and deployed in FEM code which replaces the integration of coupled non-linear
constitutive equations.
• B.Tech Project: CFD analysis of flat plate solar water heater and comparison to experimental results.
• Mentor: Dr. MK Paswan, Department of Mechanical Engineering, NIT Jamshedpur. (Jan 19 to May 19 )
The project aims to analyze the system’s performance and its efficiency. The conclusions are as follows:-
– Conducted experiment for single tube flat plate solar water heater.
– Performed CFD analysis of single tube flat plate solar water heater under similar conditions using ANSYS.
– Heat transfer coefficient for experimental setup was found to be 11% less than the results obtained from CFD analysis.
– In conclusion gray scale formation inside the pipe leads to a decrease in the heat transfer coefficient in the experimental
setup which leads to a decrease in the efficiency of the setup.
• B.Tech Project: Design, Analysis, and optimization of SI Engine air intake manifold for FSAE.
• Mentor: Dr. Sanjay, Department of Mechanical Engineering, NIT Jamshedpur. (Jan 18 to May 18 )
The project aims to design a new air intake manifold according to the SAE rule book. The conclusions are as follows:-
– New Air intake manifold was required of 20mm diameter according to SAE rules in place of the original 33mm diameter.
– CFD analysis was conducted on different geometries of the new air intake manifold.
– The spherical plenum configuration yielded the most optimal parameters for enhancing the performance of our KTM 390
engine.
– Proposed air intake manifold was prototyped.
• Term Projects at IIT Kanpur :
1. Rolling force prediction
− A MATLAB code was developed to generate rolling force data using mathematical relationships and was stored in an Excel
file. This data was subsequently employed for training the predictive model.
− Three machine learning algorithms Linear Regression, Artificial Neural Network and SVM were employed and their
R2 score was compared.
− SVM gave the best results with R2 score of 0.905. (AI/ML in Materials Engineering MSE643A)

2. Developed MATLAB code to solve a 2D partial differential equation using the Galerkin finite element method.
− Defined a triangular mesh over the domain and implemented Gaussian quadrature for accurate numerical integration within
each element.
− Local stiffness matrices and force vectors were computed for each element using linear shape functions. These local contri-
butions were then assembled to form the global stiffness matrix and force vector.
− Solved the linear system to obtain a nodal solution vector and visualized the solution by plotting results on the triangular
mesh. (FEM AE675A)
3. Micro-mechanical analysis of Composite material
− The MATLAB code was developed focusing on characterizing a composite material made up of fiber and matrix components.
Various methods for determining effective mechanical properties are explored including Voigt, Reuss, Strength of Materials
(SOM), Concentric Cylinder Approach (CCA), Halpin-Tsai, Hashin-Strikman, Mori-Tanaka, Self-consistent approach, and
Hill Concentration Factor Approach.
− The effective properties of the composite material, such as Young’s modulus (E1, E2), Poisson’s ratios (nu12, nu23), and
shear modulus (G12, G23) were evaluated for the volume fraction of fiber ranging from 0 to 1.
− The different methods offer a spectrum of effective properties, highlighting the importance of selecting an appropriate method
based on specific application requirements. (Composite Materials (AE681A))
4. Comparative Study of Numerical Methods for Solving Partial Differential Equations: FDM, FVM, and FEM
− Implemented FDM, FVM, and FEM algorithms for the given PDE problem.
− Both FEM and FVM solutions exhibit remarkable proximity to the analytical solution. (CM (AE703A))

SELF PROJECTS
• House price predictor
– Designed a model based on four algorithms namely Linear Regression, Random Forest, Decision Tree, and XGBoost
Regression and compared the accuracy of algorithms using root mean square value.
– Excel was used for data preprocessing and cleaning.
– Explored data statistics and calculated correlation matrix to understand relationships between variables.
– Among the various algorithms assessed, XGBoost Regressor demonstrated superior performance, yielding the lowest RMSE
of 0.0138 during training and 3.2326 during testing. This underscores its robustness in capturing underlying patterns and
its ability to generalize effectively to unseen data.
• Movie recommender system
– Designed a movie recommender system using Python, Json, Pandas, and Scikit-learn.
– Processed and merged movie data, including genres, keywords, cast, and crew, to generate informative tags for each movie
after this applied text preprocessing and stemming techniques to clean and format movie tags. Calculated cosine similarity
between movie tags to determine movie similarities, facilitating accurate recommendations.
– Developed a user-friendly web application using Streamlit, allowing users to select a movie and receive personalized movie
recommendations based on their preferences.
• Spam Email classifier
– Developed an email spam classifier using Python and various libraries like pandas, numpy, nltk, and sklearn.
– The dataset was cleaned by removing duplicates and unnecessary columns. Text data underwent tokenization, stemming,
and filtering out stopwords and non-alphanumeric characters.
– Three Naive Bayes classifiers were trained and evaluated on preprocessed text data after feature extraction using TfidfVec-
torizer. The Multinomial naive Bayes model was selected and saved. A Streamlit web application was developed to showcase
the model’s functionality.
• Laptop price predictor regression project
– Analysed different supervised learning algorithms on a laptop price data set and evaluated their performance parameters.
TECHNICAL SKILLS
• Data analytics and Machine learning.
• Data Structures and Algorithms
• Programming Languages: Python, MATLAB, HTML*, JavaScript*.
• Libraries and frameworks Numpy, Matplotlib, Pandas, sklearn, Tensorflow.
• Database management system MySQL, Microsoft excel.
• Other Software tools: SolidWorks, ANSYS, CATIA, LATEX. (* elementary proficiency)
RELEVANT COURSEWORK
• Artificial intelligence and machine learning in Materials engineering (MSE643A), Instructor: Dr. Krishanu Biswas.
• Computational Methods (AE703A), Instructor: Dr. Ashoke De.
• Introduction to Finite Element Method (AE675A), Instructor: Dr. CS Upadhyay.
• Composite Materials (AE681A), Instructor: Dr. PM Mohite.
• Continuum Mechanics (AE702A), Instructor: Dr. Pritam Chakarborty.
• Mathematics for Aerospace engineers (AE602A), Instructor: Dr. AC Mandal.
• Deformation and Fracture (AE704A), Instructor: Dr. Rajesh Kitey.
POSITIONS OF RESPONSIBILITY
• Teaching Assistant: Assisted course ESO202A (Strength of materials) in odd semester, 2023-2024.
• Event head in OJASS’19.
• Event head in Cognitio’19.
• Design head in formula student racing team from July-2017 to July-2018.
EXTRA-CURRICULAR ACTIVITIES
• Secured silver medal in volleyball for mechanical engineering department during Annual athletics and sports meet Urja’19.
• Third position in the Auto quiz competition Organized by OJASS’18.
• First position in PANORAMA-2013 a quiz competition Organized by Future group of Institution.
HOBBIES & INTERESTS
• Hobbies : Playing volleyball and cricket.
• Area of Interest : Swimming.