Search

Solving a multiple-criteria optimization problem with severe constraints remains a significant issue in multi-objective evolutionary algorithms. The problem primarily stems from the need for a suitable constraint handling technique. One potential approach is balancing the search in feasible and infeasible regions to find the Pareto front efficiently. The justification for such a strategy is that the infeasible region also provides valuable information, especially in problems with a small percentage of feasibility areas. To that end, this paper investigates the potential of the infeasibility-driven principle based on multiple constraint ranking-based techniques to solve a multi-objective problem with a small feasibility ratio. By analyzing the results from intensive experiments on a set of test problems, including the realistic multi-objective car structure design and actuator design problem, it is shown that there is a significant improvement gained in terms of convergence by utilizing the generalized version of the multiple constraint ranking techniques. [ABSTRACT FROM AUTHOR]

Dynamic flows past three tandem elliptic cylinders of equal spacing in parallel arrangement with a moving wall are numerically conducted by the lattice Boltzmann method. The gap ratio (G/D, where G and D are the gap between the wall surface and cylinder center and major axis, respectively) from 0.6 to 2.5 and spacing ratio (L/D, where L is the distance of two adjacent cylinder centers) from 1.5 to 10 are examined at Reynolds number of Re = 150 (based on D). The desired analysis correlates variations of hydrodynamic coefficients, wake structures, and spectral analysis in wide space of G/D and L/D with underlying fluid mechanics. The flow is highly adjustable in G / D − L / D space, dividing into six distinct regimes: overshoot, continuous reattachment, alternative reattachment, wavy, meandering, and coshedding, which are spatially classified into four modes because of flow interference, namely, shear layer, primary, two-layered, and secondary vortex shedding modes. The transition between adjacent modes defines three boundaries. The first boundary always occurs behind the middle or downstream cylinder, whereas the second and third boundaries occur between the upstream and middle cylinders and the downstream cylinder wake. Due to the near-wall flow interference, the hydrodynamic coefficients are highly changeable in L / D − G / D space, signifying the crucially lower drag coefficient of the middle and downstream cylinders than that of a single cylinder. A small G/D determines the increase in lift coefficient of the upstream cylinder as a result of the ground effect, while the shadowing effect of the upstream cylinder induces identical Strouhal numbers of the middle and downstream cylinders. [ABSTRACT FROM AUTHOR]

Indonesia is located in a high-seismic-risk region with a significant number of non-engineered houses, which typically have a higher risk during earthquakes. Due to the wide variety of differences even among parameters within one building typology, it is difficult to capture the total risk of the population, as the typical structural engineering approach to understanding fragility involves tedious numerical modeling of individual buildings—which is computationally costly for a large population of buildings. This study uses a statistical learning technique based on Gaussian Process Regression (GPR) to build the family of fragility curves. The current research takes the column height and side length as the input variables, in which a linear analysis is used to calculate the failure probability. The GPR is then utilized to predict the fragility curve and the probability of collapse, given the data evaluated at the finite set of experimental design. The result shows that GPR can predict the fragility curve and the probability of collapse well, efficiently allowing rapid estimation of the population fragility curve and an individual prediction for a single building configuration. Most importantly, GPR also provides the uncertainty band associated with the prediction of the fragility curve, which is crucial information for real-world analysis. [ABSTRACT FROM AUTHOR]

The composite kernel learning (CKL) method is introduced to efficiently construct composite kernels for Gaussian process (GP) surrogate models with applications in engineering design. The mixture of kernel functions is cast as a weighted-sum model in which the weights are treated as extra hyperparameters to yield a higher optimum likelihood. The CKL framework aims to improve the accuracy of the GP and relieves the difficulty of kernel selection. In this paper, the combination of five kernel functions are studied, namely, Gaussian, Matérn-3/2, Matérn-5/2, exponential, and cubic, with each kernel sharing the same length scale. Numerical studies were performed on a set of engineering problems to assess the approximation capability of GP-CKL. The results show that, in general, GP-CKL yields a lower approximation error and higher robustness as compared to GP models with single kernel and model selection methods. Numerical experiments show that it is worth using the GP-CKL method for achieving higher accuracy with extra burdens on the number of hyperparameters and a longer training time. Besides, experiments were also performed with variable length scales for each kernel and each variable. However, this variant of CKL is not suggested because it is prone to overfitting and is significantly more expensive than the other GP variants. [ABSTRACT FROM AUTHOR]