Funded projects

Hybrid reduced- and full-space large-scale design optimization

National Science Foundation

Full title: Hybrid reduced-space and full-space architectures for large-scale design optimization

Relevant papers

  1. Anugrah Jo Joshy, John T. Hwang. modOpt: A modular development environment and library for optimization algorithms. Advances in Engineering Software, 2026.
    DOIPDF
  2. Ryan C. Dunn, Anugrah Jo Joshy, Jui-Te Lin, Cédric Girerd, Tania K. Morimoto, John T. Hwang. Scalable enforcement of geometric non-interference constraints for gradient-based optimization. Optimization and Engineering, 2024.
    DOIPDF
  3. Anugrah Jo Joshy, Ryan Dunn, Mark Sperry, Victor E. Gandarillas, John T. Hwang. An SQP algorithm based on a hybrid architecture for accelerating optimization of large-scale systems. AIAA AVIATION 2023 Forum, 2023.
    DOIPDF
  4. Anugrah Jo Joshy, Jiayao Yan, John T. Hwang. A hybrid architecture for large-scale system design optimization of PDE-based models. AIAA SCITECH 2022 Forum, 2022.
    DOIPDF
  5. Bingran Wang, Anugrah Jo Joshy, John T. Hwang. Equality-Constrained Engineering Design Optimization Using a Novel Inexact Quasi-Newton Method. AIAA Journal, 2022.
    DOIPDF
  6. Anugrah Jo Joshy, John T. Hwang. Unifying Monolithic Architectures for Large-Scale System Design Optimization. AIAA Journal, 2021.
    DOIPDF
  7. Bingran Wang, Anugrah Jo Joshy, John T. Hwang. An Adaptive, Inexact Gradient-based Algorithm for Multidisciplinary Design Optimization. AIAA AVIATION 2021 FORUM, 2021.
    DOIPDF
  8. Anugrah Jo Joshy, John T. Hwang. A new architecture for large-scale system design optimization. AIAA AVIATION 2020 FORUM, 2020.
    DOIPDF

This research examines optimization architectures and numerical algorithms for coupled engineering systems. The work studies how analysis and optimization can be organized to reduce computational cost.

Paired chart comparing optimality convergence over time and final turbine locations for reduced-space optimization and SURF.
On a smaller nine-turbine wind-farm benchmark, SURF reaches the same local optimum about 25% faster than reduced-space optimization; the matching layouts verify the comparison. [1]

Research connections

Related to Hybrid reduced- and full-space large-scale design optimization