Open Access

A novel modeling strategy is proposed which allows high-accuracy predictions of aerodynamic and aeroacoustic target values for a low-pressure axial fan, equipped with serrated leading edges. Inspired by machine learning processes, the sampling of the experimental space is realized by use of a Latin hypercube design plus a factorial design, providing highly diverse information on the analyzed system. The effects of four influencing parameters (IP) are tested, characterizing the inflow conditions as well as the serration geometry. A total of 65 target values in the time and frequency domains are defined and can be approximated with high accuracy by individual artificial neural networks. Furthermore, the validation of the model against fully independent test points within the experimental space yields a remarkable fit, even for the spectral distribution in 1/3-octave bands, proving the ability of the model to generalize. A metaheuristic multi-objective optimization approach provides two-dimensional Pareto optimal solutions for selected pairs of target values. This is particularly important for reconciling opposing trends, such as the noise reduction capability and aerodynamic performance. The chosen optimization strategy also allows for a customized design of serrated leading edges, tailored to the specific operating conditions of the axial fan.

Passive air-jet blowing is an effective yet simple technique to control flow-induced noise due to vortex shedding of bluff bodies. The current study investigates the effect of the specific slot angles of passive jets in connection with the suppression capabilities of vortex shedding and the byproduct of vortex-induced noise in the wake region of a circular cylinder. Aeroacoustic tests for a baseline case and 10 different slotted cases with slot angles of 80°≤𝜃≤125° are performed for Reynolds numbers 6.6×103≤R≤3.3×104. This is supplemented by numerical computational fluid dynamics (CFD) analyses to identify the underlying aerodynamic mechanisms. The results obtained reveal that using the current passive control method results in a significant reduction of the vortex shedding tonal noise for slot angles of 90°≤𝜃≤125° and high Reynolds numbers. The numerical results showed good agreement with a remarkably reduced kinetic energy for slot-end angles of 115°≤𝜃≤125°. At low Reynolds numbers, however, the identified aeroacoustic benefits tend to cease.

Leading edge serrations are well known for their ability to reduce turbulence-induced noise of single aerofoils while also providing aerodynamic advantages under certain operating conditions. Continuatively, applying leading edge serrations to rotating machinery such as axial fans proved the validity to generally transfer the obtained aeroacoustic benefits of single aerofoils. However, for the rotating applications the noise reduction potential highly depends on the point of operation. The current work aims at assessing the aeroacoustic effects of serrated leading edges under the increased geometrical complexity of the fan blades through blade skew. Therefore, the question is whether combining two potentially effective noise-reducing treatments through blade skew and leading edge serrations results in leveraging or obstructing effects. By varying the skew angle from 0 deg to 38 deg, four different prototypes of the fan impeller are tested experimentally in a test rig according to ISO 5136 and ISO 5801. All configurations are tested with original blades of straight leading edges plus five sets of serrations each, parameterised by the serration amplitude and the serrations wavelength. The intensity of the incoming turbulence ranges from 2.6% to 12.1%. The results obtained show the skewed blades to improve both the aerodynamic performance and the noise radiation after exceeding an initial skew angle, complemented by a significant onset of stall. Moreover, no contraindication between blade skew and serrated leading edges is encountered, showing the potential to further extend the noise reduction capabilities by combining effects of blade skew and leading edge treatment.

Heavy-duty centrifugal fans require high reliability and first-class performance. Besides, extreme conditions and harsh environments are often encountered, such as in the papermaking process, in steel or cement plants or the chemical and petrochemical industry. Therefore, the design of high-performance heavy-duty industrial fans requires robust yet efficient solutions. The previous work indicates a high aerodynamic and aeroacoustic sensitivity concerning the specific position of the volute cutoff (tongue). This effect will be further investigated, not by directly changing the orientation of the cutoff, but by varying the position of the impeller relative to a fixed volute casing. The initial evaluation is done through a numerical study of three influencing parameters, which allow the aerodynamic dependencies to be modeled using low-layer artificial networks. Subsequently, extensive experimental studies were carried out to validate the aerodynamic dependencies and also to incorporate information on the aeroacoustic performance. The obtained results show that the operating point represents the key factor in determining the optimal positioning, with qualitatively comparable dependencies found for both tested fans. From an aeroacoustic point of view, the determined optimal configuration does not necessarily coincide with the observed aerodynamic desires, so careful analysis and a reasonable compromise are required, motivating for a multi-objective optimization process.

Recent studies have shown that leading edge serrations can effectively reduce broadband noise in axial fans and expand their operational range. However, additional aerodynamic losses have to be considered alongside the benefits of improved acoustic performance. The majority of studies focused on applying leading edge serrations along the entire span of the rotor blades regardless of local differences in radial inflow angles, deflection effects, and blade loading. In this regard, local and spanwise varying applications of leading edge serrations are considered crucial to reconciling maximum aeroacoustic and aerodynamic performance. In making these applications, a fan blade span is divided into five sub-segments. Each segment can then be equipped with either an acoustic or aerodynamic optimum as well as a baseline straight leading edge. Through testing at three different inflow turbulence levels along the full characteristic curve, an extensive statistical experimental design is executed. Ultimately, a simplified statistical model is derived. The model provides information on the optimum spanwise shape of a serration design. The results indicate significant effects of the local spanwise selection of varying serration designs. Discrepancies between aerodynamic and aeroacoustic optimum solutions can be satisfactorily resolved where, primarily, a significant extension of the working area is observed. Moreover, an extended potential in the delay of stall entry and the associated improvement of total pressure as well as a significant reduction of aeroacoustic signature is found. The underlying mechanisms are attributed to the compartmentalization effects of the serrations which restrict separation phenomena to local cells of finite spanwise extension. Furthermore, the spanwise geometrical parametrization of the serration segments correlates well with radial blade loading as well as respective flow conditions. By taking the radial distribution of aerodynamic quantities into account, an individual design of locally applied leading edge serrations can be integrated into the fan design process, enabling improved balancing between aeroacoustic benefits while maximizing aerodynamic performance.