Modeling of the Tonal Noise Characteristics in a Foil Flow by using Machine Learning
A machine learning approach for prediction the characteristics of tonal noise formed in a foil flow is tested. Experimental data are used to construct and analyze the mathematical models of pressure amplitude regression and models of classification of regimes of high-level tonal noise coming from the dimensionless parameters of the flow. Different families of algorithms are considered: from linear models to artificial neural networks. It is shown that a gradient boosting model with a determination coefficient 95% is the most accurate for describing and predicting the spectral curves of acoustic pressure on the entire interval of values of amplitudes and characteristic frequencies.
Keywordsmachine learning foil flow tonal noise
Unable to display preview. Download preview PDF.
- 4.T. F. Brooks, D. S. Pope, and M. A. Marcolini, Airfoil Self-Noise and Prediction. NASA Technical Report, NASA-RP-1218 (NASA Reference Publication, 1989).Google Scholar
- 9.S. M. Lundberg and S. I. Lee, “A Unified Approach to Interpreting Model Predictions,” in Proc. of the Conf. on Advances in Neural Information Processing Systems, Long Beach, USA, 4–9 Dec., 2017.Google Scholar
- 10.J. Bergstra, D. Yamins, and D. Cox, “Hyperopt: A Python Library for Optimizing the Hyperparameters of Machine Learning Algorithms,” in Proc. of the 12th Python in Science Conf. (SCIPY 2013), Austin, Texas, 24–29 June, 2013.Google Scholar