Computational Intelligence Assisted Engineering Design Optimization (using MATLAB).pdf
The expanding field of complex engineering design optimization has garnered considerable interest in improving efficiency across various industrial applications. This abstract explores the application of metamodel-based design optimization techniques within the context of intricate engineering systems, specifically addressing the challenges posed by uncertainty in optimization models. Metamodels, serving as surrogate models approximating the behavior of complex systems, provide an efficient alternative for optimizing engineering processes under uncertain conditions.
Traditional approaches to optimizing complex engineering systems often grapple with uncertainty, leading to variability in responses. To address this, the study incorporates hybrid metamodel techniques, coupling metamodels with robust design optimization strategies. These strategies aim to handle uncertainty in the model and mitigate the variability of responses, offering a more resilient solution.
The investigation focuses on identifying optimal configurations for enhancing performance in complex engineering systems while considering and mitigating uncertainties. Metamodels, including Response Surface Methodology (RSM), Radial Basis Function (RBF), and Gaussian Process Regression (GPR), are employed to construct efficient approximations of the underlying system models. The integration of robust design optimization techniques further enhances the ability to handle uncertainty and mitigate variability in responses.
The proposed metamodel-based design optimization framework, enriched with robust design techniques, is applied to diverse engineering scenarios. The study considers factors such as design parameters, operational variables, and geometric configurations, all within the context of uncertainty. The results underscore the effectiveness of the hybrid metamodel approach in guiding the exploration of the design space, facilitating the identification of optimal parameters, and providing resilience against uncertainties that contribute to variability in responses.
This research contributes to the growing body of knowledge in complex engineering design optimization, offering a practical and computationally efficient approach for engineers and researchers seeking to improve processes in various engineering applications while explicitly addressing and mitigating the impact of uncertainty. The integration of hybrid metamodel techniques with robust design optimization not only accelerates design exploration but also provides valuable insights into the complex relationships governing engineering system optimization under uncertain conditions.
