About Me
I am a Ph.D. candidate at Purdue University specializing in Machine Learning, Battery Informatics, and Time Series Analysis.
My research focuses on degradation predictive modeling, synthetic data generation, and advanced analytical methods.
I am passionate about applying my expertise to solve cutting-edge problems in battery and ML technologies.
Download My Resume
Education
- Purdue University - Ph.D. Candidate in Aeronautics and Astronautics (Expected Aug 2025)
- Purdue University - Masters in Aeronautics and Astronautics (2019 - 2021)
- RV College of Engineering - Bachelors in Aerospace Engineering (2015 - 2019)
Experience
- Graduate Research Assistant, Purdue University - Developed ML models for battery degradation prediction.
- Assistant Member of Technical Staff, And Battery Aero (ABA) - Optimized battery materials for electric aircraft.
- Summer Research Intern, Indian Institute of Science - Contributed to research on helicopter inflow equations.
Technical Skills
- Programming: Proficient in Python (Pandas, PyTorch, Flask, Scikit-learn, Matplotlib, TensorBoard, PyBamm), MATLAB, Fortran, Mathematica
- Software \& Tools: Battery Testers (Neware, Arbin), LiB Disassembly, EIS Spectroscopy, other engineering tools such as Abaqus, Ansys, Tecplot, and SolidWorks
- Machine Learning: Neural Networks, Time Series Analysis
Conference Presentations
- Meghana Sudarshan, Jaya Vikeswara Rao Vajja, Mahavir Singh, Low Data Machine Learning Framework for Accelerated Lithium-Ion Battery Degradation Prediction, 247th ECS Meeting, Montréal, Canada, from May 18 – May 22, 2025 (Submitted)
- Jaya Vikeswara Rao Vajja, Meghana Sudarshan, Brian Chuanyu Chang, Vikas Tomar, "Predicting Rapid Degradation Onset in Lithium-Ion Batteries during Real-Time Operation Using Machine Learning", NASA Aerospace Battery Workshop, Huntsville, AL, November 19-21, 2024
- Marco Herbsommer, Sushrut Karmarkar, Meghana Sudarshan, Mahavir Singh, Vikas Tomar, Deep neural network driven combined THz-Raman spectroscopy technique for physical spectroscopic identification station, CBM Fall 2024 IAB Meeting South Bend, IN, Notre Dame, December 3rd, 2024
- Meghana Sudarshan, Vikas Tomar, "Enhancing Temperature-Dependent Li-ion Battery Behavior Predictions with Transfer Learning", 244th ECS meeting Sweden, 2023
- Casey Jones, Meghana Sudarshan, Vikas Tomar, "Data-Driven Prediction of Long- and Short-Term Li-ion Battery Degradation Using Public Datasets and Nail Puncture Testing", NASA battery workshop, 2022
- Meghana Sudarshan, Casey Jones, Vikas Tomar, "Prevention of thermal runaway in Li-ion batteries using machine learning model prediction", TMS 150th Annual meeting, 2022
- Meghana Sudarshan, Alexey Serov, Casey Jones, Atul Prakash, Vikas Tomar, "Data-driven model based comparison of public datasets for online state of charge estimation in Lithium-ion batteries", TMS 150th Annual meeting, 2022
- Meghana Sudarshan, Ayotomi Olokun, Abhijeet Dhiman, Vikas Tomar, "Comparison of deflagration modes in a granular energetic material due to spherical and planar impact", TMS 150th Annual meeting, 2022
- Meghana Sudarshan, Bing Li, Casey Jones, Vikas Tomar, "Use of internal temperature sensors for early detection of thermal runaway in large capacity Lithium-ion pouch cells", NATAS 2021
- Benjamin Rohit, Punya D Gowda, Nandini B Nagaraju, Meghana Sudarshan, "Life and Failure of Aircraft Wheels - A Review", IOP Conference Series: Material Science and Engineering 520 (1), 012002, 2019
- Meghana Sudarshan, Punya D Gowda, Benjamin Rohit, Nandini B Nagraju, "Effect of Temperature on Local Stresses in Magnetoelectric Composites", IEEE 10th International Conference on Mechanical and Aerospace Engineering (ICMAE), 550-554, 2019
Journal Publications
- Meghana Sudarshan, Jaya Vikeswara Rao Vajja, Mahavir Singh, Vikas Tomar, "DegradAI: Long-Term Capacity Degradation Prediction for Lithium-Ion Batteries Using Limited Cycling Data", Journal of Energy Storage (Submitted)
- Jaya Vikeswara Rao Vajja, Alexey Serov, Meghana Sudarshan, Mahavir Singh, Vikas Tomar, In Operando Health Monitoring for Lithium-Ion Batteries in Electric Propulsion Using Deep Learning, July 2024, Batteries 2024, 10(10), 355; doi: 10.3390/batteries10100355
- Meghana Sudarshan, Ritesh Gautam, Mahavir Singh, R. Edwin García, Vikas Tomar, "A comparative analysis of the influence of data-processing on battery health prediction by two machine learning algorithms", Journal of Energy Storage, 2024, vol. 10(10), 355, doi: 10.3390/batteries10100355
- Meghana Sudarshan, Alexey Serov, Surya Mitra Ayalasomayajula, Casey Jones, R. Edwin García, Vikas Tomar, "Data-Driven Autoencoder Neural Network for Onboard BMS Lithium-Ion Battery Degradation Prediction", Journal of Energy Storage, Volume 82, 110575, 2024
- Jaya Vikeswara Rao Vajja, Alexey Serov, Meghana Sudarshan, Mahavir Singh, Vikas Tomar, "In Operando Health Monitoring for Lithium-Ion Batteries in Electric Propulsion Using Deep Learning, July 2024, Batteries 2024, 10(10), 355; doi: 10.3390/batteries10100355
- Casey Jones, Meghana Sudarshan, Vikas Tomar, "Analysis of Capacity Decay, Impedence, and Heat Generation of Lithium-ion Batteries experiencing multiple simultaneous abuse conditions", Frontiers of Energy (Submitted)
- Casey Jones, Meghana Sudarshan, Vikas Tomar, "Predicting the discharge capacity of a Lithium-ion battery after nail puncture using a Gaussian process regression with incremental capacity analysis", Energy, Volume 285, 129364, ISSN 0360-5442, 2023
- Casey Jones, Meghana Sudarshan, R. Edwin García, Vikas Tomar, "Direct measurement of internal temperatures of commercially available 18650 lithium-ion batteries", Scientific Reports 13, 14421, 2023
- Casey Jones, Meghana Sudarshan, Alex Serov, Vikas Tomar, "Investigation of physical effects on prismatic lithium-ion cell electrodes after partial nail puncture using Raman spectroscopy and incremental capacity analysis", eTransportation 12, 100174
- Abhijeet Dhiman, Nolan Simon Lewis, Tyler Dillard, Meghana Sudarshan, Vikas Tomar, "Advancements in mechanical raman spectroscopy for applications in energetic materials", Energetic Materials Frontiers 2 (3), 193-200, 2021
Selected Projects
- Image Transformer with DeiT: Fine-tuned DeiT for classification tasks on identifying species of butterflies and animals outperforming other methods in comparison to other CNN-based image classification methods.
- Neural Network for Lithium-ion Battery Performance: Worked on developing an on-device shallow neural networks for efficient computational performance on gradient descent methods using relaxation convex optimization techniques in comparison to non-convex loss functions. Networks like LSTM and RNN layers with convex optimization outperform the standard layers.
- Regression ML Model for Finite Element Simulations: Achieved less than 4% error on stress test data.
- Multi-objective Optimization of propellant rank for Sounding Rocket: Genetic elistist algorithm was found to be preferable over goal attainment approach by comparing pareto fronts to determine structural properties of sounding rocket on applied loads.
- System modeling of global food security through food loss prevention: Guidelines for decision-making bodies using system dynamics model and ROPE scope showing the importance of having procurement centers instead of middlemen in a food-chain based simulation with supply-chain data from India.
