prof. dr inż. Sławomir Kozieł
- Gmach Elektroniki Telekomunikacji i Informatyki pokój 304
- (58) 347 20 19
A Multi-Fidelity Surrogate-Model-Assisted Evolutionary Algorithm for Computationally Expensive Optimization Problems
Integrating data-driven surrogate models and simulation models of different accuracies (or fideli-ties) in a single algorithm to address computationally expensive global optimization problems has recently attracted considerable attention. However, handling discrepancies between simulation models with multiple fidelities in global optimization is a major challenge. To address it, the two major contributions of this paper include:...
Rapid design of miniaturised branch-line couplers through concurrent cell optimisation and surrogate-assisted fine-tuning
In this study, the authors introduce a methodology for low-cost simulation-driven design optimisation of highly miniaturised branch-line couplers (BLCs). The first stage of their design approach exploits fast concurrent optimisation of geometrically dependent, but electromagnetically isolated cells that constitute a BLC. The cross-coupling effects between the cells are taken into account in the second stage, where a surrogate-assisted...
The objective of this paper is to present a robust optimization algorithm for computationally efficient airfoil design under mixed (inherent and epistemic) uncertainty using a multi-fidelity approach. This algorithm exploits stochastic expansions derived from the Non-Intrusive Polynomial Chaos (NIPC) technique to create surrogate models utilized in the optimization process. A combined NIPC expansion approach is used, where both...
seen 485 times