Previous studies have identified the benefits of transferring the well-known aeroacoustic effects of leading edge serrations from single airfoils to rotating machines. However, since rotating systems are much more complex by comparison, the underlying aeroacoustic and aerodynamic mechanisms also operate under more complex boundary conditions. Moreover, concerning the industrial application, a combined assessment of aerodynamic and aeroacoustic performance becomes inevitable. On the other hand, the description and optimization of multiple target values requires a comprehensive knowledge of the underlying mechanisms and emphasizes the need to develop models of high functional complexity. In this paper, a novel modelling strategy is proposed which allows high-accuracy predictions of aerodynamic and aeroacoustic target values in both, the time and the spectral domain. Inspired by machine learning processes, the sampling of the experimental space is done by use of a Latin hypercube design plus a factorial design, providing highly diverse information on the analyzed system. For spectral modelling, this is extended by a filtering approach, which takes into account the physical origin of the acoustic sources by means of discrete and broadband components. Based on this pre-processing of the gathered data, the models are developed by training artificial neural networks for an exemplary low-pressure axial fan, equipped with serrated leading edges. The effects of four influencing parameters are tested, characterizing the inflow conditions as well as the serration geometry. A total of 116 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 1/3rd-octave bands, proving the ability of the model to generalize. The general trends are consistent with the results of previous measurement campaigns and also indicate decisive parameters for the reduction of single signal components with respect to broadband and discrete noise. A meta-heuristic 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. Due to the approximation capability in the spectral range, filtering is also possible and enables, for example, the consideration of A-weighting effects, which leads to broad application possibilities in the industry.


Multi-Objective Modelling of Leading Edge Serrations Applied to Low-Pressure Axial Fans

Division / Sponsor:
International Gas Turbine Institute (IGTI)

Paper Proceedings: https://asmedigitalcollection.asme.org/GT/GT2020/volume/84058

Turbo Expo - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.