Yikun Wei

Two representative axial wake distributions are considered in this study. They are derived from the DARPA (Defense Advanced Research Projects Agency) Suboff submarine model and the Korea Research Institute of Ships and Ocean Engineering Container Ship (KCS), respectively. Both distributions are imposed as nonuniform inflow conditions on an identical propeller to investigate how hull-wake morphology is converted into downstream propeller-wake dynamics. The DARPA Suboff wake represents a relatively balanced submarine-type inflow, whereas the KCS wake exhibits a pronounced velocity deficit localized within a limited circumferential sector. Four advance coefficients are considered for the evaluation of the integral hydrodynamic response, while J = 0.591 and J = 0.984 are further examined through velocity and vorticity fields, cross-plane wake evolution, Q-criterion structures, the root mean square (RMS) velocity fluctuations, kinetic-energy spectra, and tip-vortex-core vorticity. The KCS wake yields lower thrust and torque coefficients than the Suboff wake, with the reduction increasing with advance coefficient and reaching approximately 30.2% in thrust coefficient (KT) and 28.7% in torque coefficient (KQ) at J = 0.984. In contrast, the amplification of unsteady wake dynamics is more pronounced at J = 0.591. The localized KCS velocity deficit produces azimuthally varying blade loading, leading to directional wake deformation, enhanced RMS fluctuations, stronger low-frequency and noninteger spectral components, and a nonmonotonic evolution of vortex-core vorticity in the affected sector. At J = 0.984, these unsteady responses are substantially weakened despite the larger difference in integral loading between the two inflows. The integral hydrodynamic response becomes more dependent on inflow morphology, whereas the amplification of inflow nonuniformity into unsteady vortex dynamics becomes weaker.

Existing research has established that high-altitude, low-Reynolds-number (low-Re) conditions cause deterioration in centrifugal compressor efficiency, pressure ratio, and operational stability; however, the aerodynamic matching between sub-components and the energy transfer and loss mechanisms within a centrifugal compressor stage under varying altitude conditions strongly depend on the stage configuration and operating condition. In this study, numerical simulations of flow in the High Efficiency Centrifugal Compressor (HECC) stage were conducted at altitudes ranging from 0 km to 10 km. The influence of altitude on stage and sub-component aerodynamic performance, energy transport characteristics, and loss mechanisms was examined. Results indicated that as altitude increased from 0 km to 10 km, the stage near-peak adiabatic efficiency decreased from 81.98% to 77.89%, while the stage total pressure ratio at peak efficiency increased from 4.60 to 6.16 due to the rise in corrected rotational speed. The Reynolds number decreased to approximately 42% of its sea-level value, enhancing viscous effects that thickened boundary layers and expanded the wake region at the impeller outlet. The total energy flux magnitude within the impeller decreased by more than half owing to reduced air density, while the local dissipation coefficient increased monotonically with altitude, with turbulent viscous dissipation identified as the dominant loss mechanism. The diffuser total pressure loss coefficient increased systematically with altitude, driven by the expansion of high-loss regions on the shroud side originating from intensified tip leakage flow interaction and endwall boundary layer thickening. The diffuser peak static pressure recovery coefficient remained approximately 0.66 across all altitudes, but the high-efficiency operating range narrowed considerably at high altitudes, indicating increased sensitivity to off-design conditions.

The effects of winglet angle on the flow characteristics in the tip clearance region of the AXF950 axial fan are studied in depth by numerical and experimental methods at different winglet angles (0°, 15°, 30°, 45°, 60°, and 90°). A comparison between the results of numerical simulations and those of experiments confirms the reliability of the simulations. The turbulent energy and tip leakage vortex area of the axial fan are effectively reduced by the winglet at the rated condition, while the uniformity of the blade surface pressure distribution is effectively improved and sharp pressure peaks are avoided at 15° and 30° winglet angles. The flow field is also influenced by the winglets at high flow conditions, which leads to lower and more uniform pressures at the leading edge. Airflow separation is mitigated and the aerodynamic performance characteristics of the axial fan are significantly improved. The application of winglets effectively suppresses the secondary flow at the blade midstream and trailing edges, whereas the area of the secondary flow region at the leading edge expands under low-flow-rate conditions. Our comprehensive analysis reveals that the aerodynamic performance of the axial fan is notably improved at a 15° winglet angle and that the application of winglets leads to a substantial enhancement of the efficiency of the fan. The effects of the winglet angle on the internal flow loss and aerodynamic performance of axial fans are elucidated, which has significant implications for the design and optimization of these fans.

Oblique inflow modifies propeller wake transition by breaking the symmetry of the helical tip-vortex system and altering vortex interaction. A high-fidelity numerical investigation of the wake dynamics of a propeller under oblique inflow is conducted using the lattice Boltzmann method–large eddy simulation. Simulations are performed at four representative advance coefficients (J = 0.394, 0.591, 0.787, and 0.984) and four inflow angles, including an axial inflow case (β = 0°) and three oblique inflow conditions (β = 15°, 20°, and 25°), to examine the coupled effects of loading condition and inflow angle on wake evolution. The results show that oblique inflow breaks the symmetry of the wake and induces lateral deflection, leading to spatially asymmetric evolution of tip vortices. As the inflow angle increases, vortex trajectories are distorted and brought into closer proximity, which enhances vortex interaction and promotes earlier onset of pairing, merging, and breakdown. Consequently, the unstable wake region shifts upstream and expands in spatial extent. Quantitative analyses of tip-vortex characteristics show that the advance coefficient primarily controls the strength of vortex interaction, whereas the inflow angle governs the asymmetry and instability development. Increasing J weakens vortex interaction and delays vortex breakdown, while increasing β intensifies asymmetric vortex evolution and accelerates the transition toward instability. The unsteady flow characteristics are further quantified in the near-, intermediate-, and far-field wake regions. Velocity fluctuation intensity and power spectral density analyses show that strong vortex interaction leads to amplified fluctuations and rapid spectral redistribution, whereas weaker interaction results in a more gradual transition toward broadband unsteadiness. The results establish that the transition from coherent tip-vortex structures to disordered wake flow is governed by vortex interaction strength controlled by loading, while the inflow angle determines the asymmetric evolution pathway.

Semi-open centrifugal impellers are crucial parts of industrial compressors, as their aerodynamic performance directly influences the energy efficiency and operational stability of the system. The current optimization research faces major challenges: substantial computational costs for high-dimensional multi-objective optimization, susceptibility to local optima, and insufficient elucidation of the physical mechanisms connecting blade angle modifications to flow loss reduction, all of which severely hinder enhancements in compressor performance. The main contribution of this work is the comprehensive revealing of the physical mechanism by which blade profile modification suppresses tip clearance-induced flow losses, as well as the efficient 9-dimensional blade angle parameterization system. The geometric parameters of the centrifugal impeller are sampled using Latin hypercube sampling (LHS), and computational fluid dynamics (CFD) is employed to execute high-precision full-flow passage numerical simulations for each sample. To significantly reduce the computational costs and overcome the local optimum defect of traditional optimization algorithms, the design space is approximated via artificial neural networks (ANNs) and optimized using the non-dominated sorting genetic algorithm-III (NSGA-III). The optimization results reveal that the polytropic efficiency and pressure ratio of the optimized impeller are increased by 0.33% and 1.56% at design conditions, respectively. Furthermore, the aerodynamic performance of the compressor has been enhanced across the whole range of operations. At the peak efficiency point (1.65 kg/s), the polytropic efficiency and pressure ratio have been enhanced by 1.68% and 3.08%, respectively. Finally, this study highlights that semi-open centrifugal impeller blades can be angled appropriately in the reverse direction of rotation to enhance flow field uniformity and mitigate flow losses, indicating a comprehensive physical understanding of flow characteristics and the aerodynamic performance of the centrifugal compressor.

A study on the aerodynamic multi-objective optimization and internal flow analysis of the AXF950 axial flow fan with blade deflection is presented in this paper. A refined numerical model was systematically constructed to investigate the influence of blade deflection on the internal flow characteristics of the fan. The effects of blade geometric variations on flow structures, pressure distribution, and tip leakage vortex intensity were comprehensively examined. For the optimization process, a multi-objective genetic algorithm, combined with an artificial neural network, was effectively employed. Two optimized models, designated as axial flow fan I (efficiency-dominant) and axial flow fan III (pressure-dominant), were subsequently developed. Under rated operating conditions, improvements in static pressure efficiency of 2.3% and 1.84%, respectively, were observed, along with corresponding increases in design point static pressure of 5.9% and 8%. Furthermore, the impact of blade deflection geometrical parameters on fan performance was elucidated across cross-sections S1–S5, revealing how specific geometric adjustments alter the radial pressure gradient and secondary flow development. The optimized designs were demonstrated to reduce flow separation and leakage vortex intensity significantly, while blade tip loading and static pressure were concurrently enhanced through improved aerodynamic matching. This research provides a theoretical foundation and technical support for the design of high-performance axial flow fans by clarifying the underlying mechanisms through which blade deflection modifies internal flow features and overall fan performance.

The double-suction centrifugal fans are designed under a standard atmospheric condition but are usually operated under off-design conditions where the ambient temperature varies, in which the density and viscosity of the air are different from those of the standard condition, thereby affecting the fan’s performance. This work performed a surrogate-based multi-objective reliability-based design optimization (RBDO) on a double-suction centrifugal fan considering the ambient temperature ( T ) uncertainty. The RBDO intends to improve the static pressure rise ( P _s ) and static pressure efficiency ( η _s ) of the fan by optimizing the design variables of outlet installation angle, camberline radius and number of the blades, and the volute pitch. The global sensitivity analysis using Sobol’ and Borgonovo indices reveal that the volute pitch dominates η _s , while the blade camberline radius and volute pitch dominate P _s . The Borgonovo index highlights the non-negligible effect of blade outlet installation angle on the probability distribution of P _s and η _s . The effectiveness of RBDO was validated through the Reynolds-Averaged Navier-Stokes (RANS) simulations. The RBDO improves P _s and η _s by 10.44% and 1.87%, respectively, at T = 20°C where the fan is designed, and by 8.62% and 2.00% at T = 0°C, and 9.39% and 2.06% at T = 40°C, revealing a robust effectiveness of RBDO. The working capability of the optimized fan within a range of ambient temperature is quantified to evaluate the trade-off between performance improvement and feasible reliability. The design feasibility domain reduces significantly due to the temperature sensitivity which determines the safety margin as the target static pressure rise exceeds 84 Pa. An analysis on flow physics demonstrates that the optimized fans effectively suppress the separation and recirculating vortices, weaken the entropy generation, and generate a better uniform distribution of flow angle and velocity in multiple blade passages of the impeller, which contributes to the improved performance of the fan.