Research Profile
My research focuses on advancing computational methods for structural engineering through adaptive finite element analysis, high-performance computing, and machine learning integration.
I develop scalable numerical algorithms and open-source software for complex structural problems, with emphasis on GPU acceleration, error-driven mesh refinement, and multi-agent reinforcement learning for optimization tasks.
I am seeking PhD opportunities in computational mechanics, structural optimization, seismic resilience, or AI-assisted engineering simulation with supervisors conducting research at the intersection of numerical methods and structural systems.
Research Interests
Research Direction
Computational Framework
My doctoral research will advance adaptive finite element methods through GPU-accelerated algorithms and machine learning integration. Current work demonstrates 30× performance improvements via parallel computing and establishes reproducible computational frameworks for structural mechanics problems.
Research Alignment
I seek supervision in laboratories focused on computational mechanics, structural optimization, seismic performance simulation, or AI-enhanced engineering analysis. My published work in evolutionary algorithms and adaptive meshing provides foundation for advanced doctoral research in these domains.
Education
Sep 2020 - Jan 2023
Khajeh Nasir Toosi University of Technology
Tehran, Iran · GPA: 14.27/20
Thesis: Two-Dimensional Adaptive Finite Element Analysis Using GPGPU Computing
Relevant coursework: Finite Element Method, Numerical Methods in Structural Engineering, Structural Dynamics, Nonlinear Analysis, Stability of Structures, Elasticity Theory.
2015 - 2020
Islamic Azad University, Tehran South Branch
Tehran, Iran · GPA: 15.30/20
Publications & Research
Published Papers
2026
CMA-MAPPO: Integrating Covariance Matrix Adaptation Evolution Strategy with Multi-Agent Proximal Policy Optimization for Enhanced Exploration in Sparse-Reward Environments
Swarm and Evolutionary Computation (Elsevier)
Authors: Amir Hossein Khatami
DOI: 10.1016/j.swevo.2026.102330 | Code Repository
Hybrid CMA-ES and MAPPO framework for sparse-reward multi-agent reinforcement learning, demonstrating exploration and convergence improvements against baseline algorithms.
2025
PyAdMesh: A Novel High-Performance Software for Adaptive Finite Element Analysis
Simulation Modelling Practice and Theory (Elsevier)
Authors: S. Asil Gharebaghi, Amir Hossein Khatami
DOI: 10.1016/j.simpat.2025.103074 | Code Repository
Open-source h-adaptive finite element software for reducing discretization error, with CPU/GPU parallel workflows using scientific Python tools.
2025
Two-Dimensional Adaptive Finite Element Using GPGPU
Civil Engineering Journal, Amirkabir University of Technology
Authors: Amir Hossein Khatami, S. Asil Gharebaghi
DOI: 10.22060/ceej.2025.23113.8111
Adaptive finite element analysis accelerated with GPGPU-oriented computation for two-dimensional structural mechanics problems.
Under Review
Under Review
Seismic Performance of RC Split Core Walls with Distributed HDR Viscoelastic Links: Nonlinear Dynamic and Collapse Assessment Across Building Heights
Status: Under review
Comprehensive nonlinear dynamic analysis and collapse assessment of reinforced concrete split core walls equipped with high-damping rubber viscoelastic links across various building heights.
Under Review
A Hierarchical Distributed Active Damping Framework with Local Rapid Response and Global Modal Coordination for Structural Vibration Control
Status: Under review
Novel hierarchical control architecture combining local rapid-response damping with global modal coordination for enhanced structural vibration mitigation.
Under Review
A Hierarchical Distributed Active Damping System with Local-Modal Control for Structural Vibration Mitigation: Comparison with LQR, LQG, MPC, and H∞ Controllers
Status: Under review
Comparative evaluation of hierarchical distributed active damping against classical control strategies including Linear Quadratic Regulator, Linear Quadratic Gaussian, Model Predictive Control, and H-infinity controllers.
Research Software
PyAdMesh
High-performance adaptive finite element software focused on inter-mesh data transfer, refinement workflows, and CPU/GPU acceleration.
CMA-MAPPO
Evolutionary multi-agent reinforcement learning framework integrating CMA-ES with MAPPO for sparse-reward exploration.
Project Manager Dashboard
Web-based project management system using Django and MySQL with role-based access control, multi-level approval workflows, and financial tracking for distributed operations. Deployed using Gunicorn and production-oriented static-file serving.
Technical Expertise
Scientific Computing
Python, C, NumPy, SciPy, CuPy, Numba, PyTorch, CUDA/OpenCL-oriented workflows, Linux, Git, Docker.
Computational Mechanics
Finite element analysis, adaptive meshing, numerical methods, structural dynamics, nonlinear analysis, OpenSeesPy, Abaqus.
AI and Optimization
Reinforcement learning, multi-agent RL, CMA-ES, metaheuristic optimization, scientific machine learning, benchmark design.
Engineering Software
SAP2000, ETABS, SAFE, AutoCAD, Revit, BIM workflows, and planned Python automation for ETABS, RFEM 6, and Tekla workflows.
Professional Experience
Nov 2024 - Present
- Review structural design documents and engineering calculations for branch facilities across multiple cities.
- Developed an internal digital dashboard for project monitoring, documentation, and administrative workflow management.
- Coordinate engineering review processes with attention to code compliance, documentation quality, and implementation requirements.
Mar 2021 - Sep 2021
- Supported structural design and analysis of steel and reinforced concrete systems using SAP2000, ETABS, and SAFE.
- Prepared technical drawings, calculations, reports, and project documentation for civil and structural projects.
- Assisted with site monitoring, quality control, material assessment, and structural safety checks.