Sai Saketha Chandra, Athkuri and V. Eswaran, A New Auxiliary Volume-Based Gradient Algorithm for Tri and Tetrahedral Meshes, Journal of Computational Physics, vol. 422, p. 109 780, Dec. 2020, issn: 00219991. doi: 10.1016/ j.jcp.2020.109780

Sai Saketha Chandra, Athkuri, M. R. Nived, and V. Eswaran, “The mid-point green-gauss gradient method and
its efficient implementation in a 3d unstructured finite volume solver,” International Journal for Numerical Methods in Fluids, vol. 94, no. 5, pp. 395–422, 2022.  doi: https://doi.org/10.1002/fld.5059. eprint: https://onlinelibrary.
wiley.com/doi/pdf/10.1002/fld.5059.

M. Nived, Sai Saketha Chandra, Athkuri, and V. Eswaran, “On the application of higher-order backward difference (bdf) methods for computing turbulent flows,” Computers & Mathematics with Applications, vol. 117, pp. 299–311, 2022,
issn: 0898-1221. doi: https://doi.org/10.1016/j.camwa.2022.05.007.

M. R. Nived, B. Sai Mukesh, Sai Saketha Chandra, Athkuri, and V. Eswaran, “On the performance of rans turbulence models in predicting static stall over airfoils at high Reynolds numbers,” International Journal of Numerical Methods for Heat & Fluid Flow, vol. 32, no. 4, 2022.  doi: https : / / doi . org / 10 . 1108 / HFF - 08 - 2021 - 0519. eprint: https://www.emerald.com/insight/content/doi/10.1108/HFF-08-2021-0519/full/html.

Sai Saketha Chandra Athkuri, M. Nived, R. Aswin, and V. Eswaran, “Computation of drag crisis of a circular cylinder using hybrid rans-les and urans models,” Ocean Engineering, vol. 270, p. 113 645, 2023, issn: 0029-8018. doi: https://doi.org/10.1016/j.oceaneng.2023.113645

Sai Saketha Chandra, Athkuri, A. R, M. R. Nived, and V. Eswaran, “Investigation of lift crisis phenomenon in flow over a rotating cylinder at reynolds numbers 60,000 and 140,000 using a transition model,” in AIAA SCITECH 2022 Forum. doi: 10.2514/6.2022-0736. eprint: https://arc.aiaa.org/doi/pdf/10.2514/6.2022-0736.

Sai Saketha Chandra, Athkuri, M. R. Nived, A. Naidu, and V. Eswaran, “Performance investigation of Physics Informed Neural Networks (PINNs) in comparison to an in-house finite volume Navier-Stokes solver,” in Proceeding of 14th Asian Computational Fluid Dynamics (ACFD), 2023

Sai Saketha Chandra, Athkuri, V. Sharma, A. Assam, and V. Eswaran, “Performance Of Convective Schemes In
Density Based Solver,” in Proceeding of Proceedings of the 24th National and 2nd International ISHMT-ASTFE Heat and Mass Transfer Conference (IHMTC-2017), Connecticut: Begellhouse, Mar. 2018, pp. 1–4.  doi: 10.1615/IHMTC-2017.10

• Delivered lectures on the courses namely, Statistics for Data Science, Python for Data Science, Machine Learning (ML), Artificial Neural Networks (ANN), Deep Learning (DL), and Natural Language Processing (NLP).
• Guided 4 M.Tech students of Narsee Monjee Institute of Management Studies (NMIMS) in their dissertations in the fields of NLP and Computer Vision (CV).
• Mentored 8 trainees in their internship projects which include projects like sentiment analysis, image classification, etc.
• Built/deployed various user-interactive tools for exploring structured data (Exploratory Data Analysis (EDA) tool), building ML models (without the need to code), and real time stock price forecasting, short term and long term portfolio management, hypothesis testing (t, z, χ2 and F ) using the Streamlit interface.
• Participated in more than 20 Kaggle competitions where I have worked with real-time data. Some of these include problems like anomaly detection/ fraud detection (using Auto-encoders), Price prediction, stock price forecasting, and portfolio risk analysis and so on.
• Currently engaged in a challenging SciML project at NIT Calicut where we impose physics on the Neural Networks during training so that they can be used for simulating physics-based flows

As a part of PhD, I have developed three-dimensional general purpose compressible flow hybrid RANS-LES solver at IIT Hyderabad. Some of the publications in this regard can be found in publications section.

Research areas/interests:

1. Hybrid RANS-LES and Transition Models Using In-House Code (Pravaha-IITH, Formerly Pravaha)

Overview: Explore the development and implementation of hybrid Reynolds-Averaged Navier-Stokes (RANS) and Large Eddy Simulation (LES) models using the Pravaha-IITH software. This research focuses on the accurate prediction of transitional and turbulent flows in complex engineering applications.

Key Objectives:

• To enhance the accuracy of flow predictions in transitional regimes using hybrid RANS-LES approaches.
• Integration of advanced turbulence models within the Pravaha-IITH framework to handle a wider range of flow conditions.
• Validation of models with experimental data and high-fidelity simulations.

Methodology:

• Development of transition-sensitive models that can dynamically switch between RANS and LES methodologies based on local flow characteristics.
• Optimization of computational resources to enable efficient simulations of large-scale problems.
• Simulate transonic/supersonic flows using these methods.

Applications:
Aerospace: Simulation of flow over aircraft components for improved design and performance.
Compressible flows: Simulation of supersonic/hypersonic flows, turbulence boundary layer interactions 

2. Development of Overset Mesh Solver for Moving Body Simulations

Overview: This area focuses on developing robust and efficient overset grid techniques to handle simulations involving moving bodies, which are crucial for applications such as vehicle dynamics, biomechanics, and robotics.

Key Objectives:

• To create a versatile solver capable of handling complex geometries and motions.
• Enhance the accuracy and efficiency of simulations involving dynamic interactions between multiple bodies.

Methodology:

• Implementation of advanced interpolation schemes for seamless grid interfacing.
• Automation of grid generation and adaptation processes to reduce setup time and improve simulation fidelity.

Applications:

• Marine: Simulating the dynamic interaction of water with moving ships or underwater vehicles.
• Sports: Analyzing the aerodynamic effects on moving athletes or sports equipment.

3. Fluid Flow Simulation Using Physics Informed Neural Networks (PINNs)

Overview: Investigating the use of PINNs to simulate fluid flow problems, integrating traditional fluid dynamics models with machine learning techniques to improve prediction accuracy and computational efficiency.

Key Objectives:

• To develop neural network architectures that are informed by underlying physical laws.
• Reduce the dependency on extensive simulation data through the integration of physical constraints in training.

Methodology:

• Design and training of PINNs that incorporate conservation laws (e.g., mass, momentum) directly into the learning process.
• Comparison of PINN outputs with traditional CFD methods to evaluate performance enhancements.

Applications:

• Environmental Science: Modeling complex fluid interactions in natural systems, such as rivers or atmospheric layers.
• Chemical Engineering: Optimizing flow processes in reactors and pipelines for enhanced safety and efficiency.

4. AI Applications in Computer Vision and NLP

Overview: Exploring cutting-edge applications of artificial intelligence in the fields of computer vision and natural language processing to develop systems that can interpret visual and textual data with high accuracy.

Key Objectives:

• To leverage AI for advanced image recognition, object detection, and semantic analysis.
• Apply NLP techniques to improve machine understanding and generation of human language.

Methodology:

• Utilization of deep learning models to process and analyze large datasets.
• Integration of multimodal data to enhance the AI's decision-making capabilities.

Applications:

• Healthcare: Automated diagnosis systems using medical imaging and patient data analysis.
• Customer Service: Development of intelligent chatbots and virtual assistants capable of understanding and responding to user inquiries effectively.