As accuracy of optimization can not be guaranteed without high-quality samples, the distribution of a finite number of evaluation points where experiments should be conducted in design space is an important issue, particularly when the experiment to obtain sample is expensive. To utilize limited number of evaluation points to represent the design space, optimal latin-hypercube design (OLHD), with considerable space-filling quality, is widely used as a methodology in design of experiments (DOE). However, OLHD generation requires significant time. This study focuses on further improvement of efficiency in generation of OLHD in terms of both time and latin-hypercube design (LHD) optimization. Two modified algorithms, namely the modified enhanced stochastic evolutionary (MESE) and translational propagation modified enhanced stochastic evolutionary (TPMESE) algorithms, based on existing algorithms, are proposed. The MESE algorithm is modified from the enhanced stochastic evolutionary (ESE) algorithm by using a new update method for “temperature,” while the TPMESE algorithm optimizes the LHD via translational propagation (TPLHD) instead of optimizing a random LHD like the MESE algorithm does. Their performance is evaluated by comparison with several famous heuristic algorithms and each original algorithm using optimization tests of LHDs with various sizes. For all cases, proposed algorithms show better performance of convergence than other heuristic algorithms participated in our comparison. For large and medium LHDs, the MESE algorithm faster converges to a solution with the same level as original algorithm (ESE). For large LHDs, the TPMESE algorithm is the most time efficient algorithm in obtaining near-optimal or sufficient near-optimal designs.

It is common that bluff bodies form complex wake structures and act on themselves with significant influence on aerodynamic forces. A square cylinder was studied as a typical model that makes complex flowfields. A magnetic suspension and balance system (MSBS) was used to support the model with magnetic forces and eliminate the influence of mechanical supports on the flowfield of wind-tunnel testing. The aerodynamic forces and base pressure of the square cylinder models with fineness ratios (L/D) of 1.0, 1.2, 1.68, and 2.0 were measured using the MSBS. The employed Reynolds numbers were from [ ] to [ ]. The drag coefficients were independent of the Reynolds number Re and could be divided into two regions between [ ] and 1.68. The base pressure coefficient decreased monotonically from 1.0 to 1.68. The drag did not significantly change for [ ] and 1.2 with angles of attack from -2 to 2 deg. In contrast, it quadratically increased from [ ] and 2.0. The lift–curve slopes were negative from [ ] to 1.2 and positive from [ ] to 2.0, confirming the reversal. The moment–curve slopes were negative for all fineness ratios. Notably, these changed linearly from [ ] to 1.2 and nonlinearly from [ ] to 2.0. The visualization test using the surface-oil-flow method revealed that the square cylinders had a reattachment position near [ ] for [ ] and [ ] for [ ]. This suggests that the reversal of the lift–curve slope was caused by reattachment.

The correction of the blockage effect is crucial in wind tunnel experiments, particularly for the drag forces on bluff bodies with a large flow separation at the leading edge. Maskell’s wake blockage correction is a widely used method for bluff bodies. However, the correction method is limited to the separated flow that does not reattach. In this study, the blockage effect and correction methods were investigated due to the presence or absence of the leading-edge-separated shear layer reattachment in a flow field with non-support interference produced using a magnetic suspension and balance system (MSBS). The drag coefficient without a wall constraint ($C_{\mathrm{DL}}$) was evaluated by varying the fineness ratio in the models with $L/D=1.0$ and $2.0$. For $L/D=1.0$, where the leading-edge-separated shear layer does not reattach, the solid blockage correction following the wake blockage correction reduced the difference to within $\pm 0.01$. However, as $S/C$ changed, the corrected drag coefficient was not constant. Further investigation is needed to evaluate the possibility of overcorrection. For $L/D=2.0$, the reattachment occurs; the wake blockage correction alone is insufficient; therefore, a solid blockage correction must apply along with the wake blockage correction. It leads the difference to become less than $\pm 0.02$ compared to $C_{\mathrm{DL}}$. The results of this study suggest that both wake and solid blockage corrections are effective regardless of the presence or absence of the reattachment of the leading-edge-separated shear layer on the cylinder.