Parth Harbola

At CEA's Laboratory of Dynamic Studies, I worked on a project aimed at building a high-fidelity material testing database for ductile fracture modeling in nuclear-grade steels. My work combined experiments, simulations, and synthetic data generation into one unified framework. I used DIC to analyze full-field strain and displacement in 316L stainless steel tensile tests and validated the results against FEM simulations in Cast3M.

To extend this experimentally validated framework, I developed a Python-based virtual lab to generate synthetic speckle images using isoparametric mapping, subpixel rendering, and Newton–Raphson solvers. These simulations were run on high-performance computing clusters, reducing computation time by 80% and generating 200+ virtual tests and 500,000 data points.

Ultimately, I developed a comprehensive uncertainty quantification pipeline to statistically evaluate DIC accuracy, verify Gaussian error behavior, and facilitate data-driven calibration of fracture models for nuclear safety applications.

At Wrocław University of Science and Technology, I designed a next-generation modular trailer architecture focused on structural integrity, aesthetics, and rapid assembly. The project applied inventive engineering principles, including TRIZ, Synectics, and the Kano Model, to identify and optimize the connection mechanism as the most critical component influencing load transfer and modularity.

Through comparative analysis of multiple joining strategies, including pins, brackets, interlocks, and flanges, I selected and refined a riveted bracket-based connection in SOLIDWORKS, achieving superior strength, manufacturability, and visual integration. TRIZ principles like segmentation, local quality, and universality guided design improvements that reduced stress concentrations and enhanced fatigue resistance. The final concept delivered a seamless internal bracket system with optimized load paths, minimal external hardware, and strong potential for scalable, customizable trailer configurations.

At Wrocław University of Science and Technology, I performed a detailed numerical study of crack propagation in 3D steel specimens using the extended finite element method (XFEM) in Abaqus. The goal was to capture complex fracture paths without the need for remeshing, enabling realistic simulation of crack initiation and growth under tensile loading.

I modeled nonlinear fracture behavior using principal stress failure criteria and dynamic enrichment functions, ensuring accurate representation of discontinuities. Structured hexahedral meshes (C3D8R) were generated with adaptive stabilization techniques to maintain numerical convergence at high strain levels. Displacement-controlled boundary conditions were applied to replicate tensile tests, allowing the extraction of stress intensity factors and the observation of crack-tip evolution.

Postprocessing through CAE visualization tools enabled monitoring of damage fields and full crack propagation using STATUSXFEM results. The complete workflow established a reliable framework for fracture analysis, supporting the design and validation of safety-critical steel components.

At the Indian Institute of Technology, Kanpur, I designed and simulated advanced microphone systems optimized for compact consumer devices such as smartphones and wearables. The project focused on vibro-acoustic coupling between the diaphragm, enclosure, and surrounding air to enhance sound fidelity and noise isolation.

I modeled the complete microphone assembly in Siemens NX, followed by detailed meshing and structural setup in Abaqus for coupled acoustic-structural simulations. Using Actran, I performed time-domain frequency analyses to study wave propagation, pressure fields, and resonance behavior within the cavity. The design was optimized for acoustic sensitivity, damping, and cavity geometry to improve the signal-to-noise ratio and reduce distortion.

Simulation results were validated against experimental acoustic data, refining boundary conditions and material parameters to achieve accurate predictive models. The final design demonstrated strong potential for integration into early-stage prototyping workflows, supporting the development of miniaturized, high-performance microphone arrays for modern electronic devices.

At IIT Kanpur, I optimized the vibrational and acoustic performance of automotive components using advanced FEM and coupled simulations. I developed detailed Abaqus models with damping, material anisotropy, and complex load paths to capture dynamic behavior under realistic conditions. I also performed frequency response function (FRF) analyses to identify natural frequencies, mode shapes, and harmonic responses, and conducted vibro-acoustic simulations in Actran to study structural–fluid interactions and noise propagation.

Additionally, I achieved a 99% reduction in computation time through workflow optimization and reduced-order modeling. I also used MATLAB to generate 3D acoustic domains and boundary conditions aligned with experimental data, and automated the analysis pipeline using Python scripting for batch runs and post-processing.

Finally, I validated results against experimental vibration and acoustic measurements, refining mesh and damping models to improve predictive accuracy and support automotive NVH design optimization.

At the University of Lille, I developed a finite element simulation framework to optimize the structural and thermal performance of solar power cell assemblies. The objective was to enhance energy efficiency, reliability, and sustainability through advanced numerical modeling. I defined geometric parameters of the support frame and discretized it using 2D planar elements to capture realistic boundary conditions for rooftop and freestanding setups. Uniform gravitational and wind loads were applied to assess deformation, strain, and Von Mises stress. Using a Python-based FEM implementation, I assembled stiffness matrices, solved for nodal displacements, and verified numerical stability through convergence analysis.

The results identified high-stress and deflection-sensitive regions, guiding material-efficient structural refinements. The simulation framework was validated against analytical solutions, ensuring accuracy and physical relevance.

This work contributed to designing lighter, more reliable solar panel structures aligned with green energy and sustainability goals.

At the University of Lille, I designed and analyzed optimized ribbed tire tread patterns for European long-haul commercial vehicles to balance traction, stiffness, rolling resistance, and wear resistance. Three tread configurations were modeled using AutoCAD for 2D parametric design and Siemens NX for 3D geometry generation based on pitch segmentation, followed by export to Abaqus for simulation. Finite element analyses were conducted under realistic radial, lateral, and tangential loads using accurate rubber material models and encastre boundary conditions. Both hex and tetrahedral meshes were employed to ensure convergence and efficiency. Surface loads replicated tire–road interactions during straight-line driving, turning, and traction.

Simulation results revealed how groove geometry influenced stress distribution, deformation, and contact behavior. Among the three designs, the final configuration with lateral grooves and solid shoulders provided the optimal balance between grip, stiffness, and deformation control, demonstrating superior performance for long-haul vehicle applications.

At the University of Lille, I led an experimental mechanics project to design anatomically accurate and mechanically realistic hospital mannequins for clinical training. The work combined numerical modeling, material testing, and data analysis to replicate human tissue response under clinical loading. I performed a literature review on pelvic biomechanics and soft tissue constitutive modeling, then designed simplified CAD geometries of target body regions. Using Abaqus, I simulated large deformations with hyperelastic material laws such as Yeoh and Mooney–Rivlin models to study nonlinear stress–strain behavior. Silicone specimens were fabricated through 3D-printed molds and tested under controlled tensile loading to observe strain-rate sensitivity and Mullins-type softening. Experimental data were processed in Python for curve fitting and model validation.

Results showed that silicone aligned closely with Yeoh model predictions, while human tissue data correlated better with Mooney–Rivlin behavior. The study proposed PVA and composite materials as next-step candidates better to capture anisotropic soft tissue behavior for improved mannequin realism.

At the National Technical University of Athens, I conducted a comparative study on topology optimization methods for modern building structures using BESO and SIMP algorithms. The objective was to minimize compliance and optimize material layout under realistic load conditions. I defined 2D and 3D design domains with boundary constraints, applied live, dead, and wind loads following building design codes. I performed finite element analysis at each optimization step to evaluate displacement, stiffness, and stress distribution. Sensitivity analysis guided material redistribution, while adaptive mesh refinement improved accuracy in stress concentration regions. Iterative updates were applied based on volume fraction and penalization parameters to achieve convergence toward optimal designs. The final optimized topologies were visualized through stress and compliance plots, highlighting that BESO offered more discrete material boundaries. At the same time, SIMP provided smoother and more continuous layouts, enabling structural efficiency comparisons for practical architectural applications.