Topology Optimization of Complex Structures and Architected Metamaterials (Ongoing, Role: PI, Partner PI: Jarkko Niiranen, Aalto University, Finland)
The architected cellular metamaterials that exhibit low density as well as high stiffness and strength are inspired from nature. The primary objective of the study is to develop computational methods for the design optimization of materials/structures with complex cellular architecture for various applications of different scales. Applications of such metamaterials can be found in civil, mechanical, aerospace, healthcare – for example, microlattices with graded porosity, sandwiched beams and plates, pyramidal truss structures, light trapping solar cells, artificial cartilages and bones, among others.
Flood Risk and Resilience of Infrastructure Systems (Ongoing)
This project involves quantification of risk and resilience of infrastructure systems in a community. A localized data-driven approach is adopted based on damage/recovery of infrastructure against flood hazard to get a more realistic estimates for decision making. Leveraging on probabilistic graphical models and GIS technologies, a spatial and dynamic risk and resilience index is quantified.
Model Reduction in Large Scale Structural Topology Optimization (Ongoing, Role: Collaborator, Co-collaborators: Prof. Piotr Breitkopf, UTC, France; Prof. Balaji Raghavan, INSA Rennes, France; Prof. Manyu Xiao, NPU, China)
Big data problems in engineering design, such as topology optimization of large-scale systems, suffers from huge computation demand of the numerical solver. The main objective of this research work is to project the solutions of original space to a reduced dimensional space so that heavy computation can be performed in that space. The reduced model is obtained from selected original snapshots based on the orthonormality of basis solution vector, sensitivity and error criteria.
Big-Data Analytics for Prognostics and Health Management of Infrastructure Systems (Ongoing)
It is an interdisciplinary project, in which a modeling framework is proposed and validated for diagnosis and prognosis of infrastructure systems based on real-time data. The major contributions of the work lies in the development of a data-driven physics informed equation free prognosis method. Such robust predictions of future health state for the existing infrastructure will be beneficial to stakeholders for a sustainable development.
Dimensionality Reduction and Surrogate modelling for Uncertainty Quantification Problems (Ongoing, Role: Collaborator, Co-Collaborator: Ziqi Wang, Guangzhou University, China)
The uncertainty quantification and optimization for large-scale problems under uncertainty involving millions of degrees of freedom is typically computation-intensive in nature. In this project, a modeling framework is proposed using a sequential/coupled dimensionality reduction of original model and constructing a surrogate on the reduced basis. Developed algorithms are tested on applications like, optimization of large scale structures under uncertainty, random vibration problems etc.
Form-finding and load Analysis of Tensile Membrane Structures (Ongoing, Role: Collaborator; PI: Prof. Siddhartha Ghosh, IIT Bombay)
Form-finding and analysis of tensile membrane structures is complex due to the form and force interaction. In particular, existing numerical form-finding algorithms are not robust and highly parametric, and high computational demand for analyzing large-scale TMS. In this work, stochastic optimization-based methodologies are proposed for form-finding and load analysis adhering to the laws of equilibrium and structural stability.
Form-finding and Reliability-based Design Optimization of Tensile Membrane Structures (Completed)
This work has focused on design optimisation of flexible tensile membrane structures under uncertainties using stochastic optimisation techniques, advanced probabilistic simulation methods and uncertainty quantification with stochastic metamodels.