A Deep Learning Approach to Jet Noise Prediction

Tenney, Andrew; Glauser, Mark; Lewalle, Jacques; Ruscher, Christopher J.

doi:10.2514/6.2018-1736

Cited by 15 publications

(9 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tenney et al adopted MLPs to predict the noise of a rectangular jet. It was clearly shown that the essence of the aerodynamic response predictions is multivariable nonlinear regression [84]. These studies demonstrate that the architecture of MLPs does work, but the results of these experiments show that the accuracy of the predicted data still has much room for improvement.…”

Section: ) Steady Nonlinear Aerodynamic Response Predictions Based Omentioning

confidence: 95%

Neural Networks-Based Aerodynamic Data Modeling: A Comprehensive Review

Zhang

Xiang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

This paper reviews studies on neural networks in aerodynamic data modeling. In this paper, we analyze the shortcomings of computational fluid dynamics (CFD) and traditional reduced-order models (ROMs). Subsequently, the history and fundamental methodologies of neural networks are introduced. Furthermore, we classify the neural networks based studies in aerodynamic data modeling and illustrate comparisons among them. These studies demonstrate that neural networks are effective approaches to aerodynamic data modeling. Finally, we identify three important trends for future studies in aerodynamic data modeling: a) the transformation method and physics informed models will be combined to solve high-dimensional partial differential equations; b) in the research area of steady aerodynamic response predictions, model-oriented studies and data-integration-oriented studies will become the future research directions, while in unsteady aerodynamic response predictions, radial basis function neural networks (RBFNNs) are the best tools for capturing the nonlinear characteristics of flow data, and convolutional neural networks (CNNs) are expected to replace long short-term memories (LSTMs) to capture the temporal characteristics of flow data; and c) in the field of steady or unsteady flow field reconstructions, the CNN-based conditional generative adversarial networks (cGANs) will be the best frameworks in which to discover the spatiotemporal distribution of flow field data. INDEX TERMS Aerodynamics, convolutional neural networks, neural networks, generative adversarial networks, recurrent neural networks.

show abstract

Section: ) Steady Nonlinear Aerodynamic Response Predictions Based Omentioning

confidence: 95%

Neural Networks-Based Aerodynamic Data Modeling: A Comprehensive Review

Zhang

Xiang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…It can be shown that on compact subsets of n-dimensional Euclidean space R n , a perceptron with only a single hidden layer and a finite number of parameters can be used to approximate any continuous function with arbitrary precision. 35 However, single-layer networks quickly become unwieldy for complex problems. Consequently, the addition of hidden layers increases the efficiency of the network compared to that of a single layer, allowing for fewer overall nodes.…”

Section: Direct Dnn-based Prediction Methodsmentioning

confidence: 99%

“…Consequently, the addition of hidden layers increases the efficiency of the network compared to that of a single layer, allowing for fewer overall nodes. 35 In training these systems to predict an output or set of outputs from a set of inputs, the weights associated with the node connections must be adjusted, usually done using a simple update algorithm. For example, a backpropagation algorithm can be used to train the DNN, the algorithm feeding input data into the network and computing the output at each node.…”

Section: Direct Dnn-based Prediction Methodsmentioning

confidence: 99%

“…Gradient descent optimization can then be used to minimize the overall error of the network based on the knowledge of these gradients. 35 First, we present a data-driven DNN-based prediction method (DDNN) for jet noise. The schematic diagram of this DNN-based prediction method is shown in Figure 2.…”

Section: Direct Dnn-based Prediction Methodsmentioning

confidence: 99%

“…More recently, machine learning has made its way into the field of fundamental sciences and shown promising results. For example, Tenney et al 35,36 applied neural networks to jet noise modeling and prediction and Brown et al 37 applied deep neural networks (DNNs) to the prediction of jet surface interaction noise modeling.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A physics merged deep neural network-based prediction method for jet turbulent mixing noise

Bai,

Zhang,

et al. 2024

International Journal of Aeroacoustics

View full text Add to dashboard Cite

Turbulent mixing noise is a vital component of jet noise, and its rapid, accurate prediction has always been persistently pursued. Recent advancement in machine learning has been applied to jet noise prediction. However, these applications are pure curve fitting and lack physical constraints. In this study, a physics-merged deep neural network (PMNN)-based prediction method was developed for turbulent mixing jet noise by merging the physics of the jet noise. This deep neural network (DNN)-based method employed recent advancements in jet turbulent mixing noise containing large- and fine-scale turbulence structures. Two simple rational functions for large- and fine-scale turbulent noise similarity spectra were proposed to replace the original complex similarity spectra functions and incorporated into the DNN-based prediction method. For comparison, we present two data-driven DNN-based prediction methods (DDNN). The first DDNN method used the sound pressure level (SPL) as the output of neural networks, directly establishing the nonlinear relationship between the input features and SPL. In the second DDNN method, the dominant modes of the jet noise spectra extracted using the proper orthogonal decomposition method were merged with DNN. These DNN-based methods were then trained using a set of experimental data over a wide range of jet operating conditions. Their performance was evaluated and compared. It was evident that all these DNN-based methods were capable of predicting turbulent mixing noise reasonably well. In contrast to the DDNN methods, the PMNN method could provide insights into the jet turbulent mixing noise components. It demonstrates that the turbulent mixing jet noise spectra at the mid polar angle is generated by the large-scale noise component at low-frequency range and by the fine-scale noise component at high-frequency range.

show abstract

Review on Metrics and Prediction Methods of Civil Aviation Noise

Feng

Zhou

Zeng

et al. 2023

Int. J. Aeronaut. Space Sci.

View full text Add to dashboard Cite

Civil aviation noise is one of the main factors hindering the growth of the civil aviation industry. With the increase in global air traffic demand, the problem of aviation noise pollution will become more and more serious. It is of great significance to carry out research in aviation noise. First, by summarizing the characteristics of aviation noise metrics, this paper divides them into three categories: single event noise metrics, cumulative exposure metrics, and daily metrics. Representative metrics of each category are selected for explanation and in-depth analysis. Second, according to the principles of aviation noise prediction models, this paper classifies these existing models into three categories: best practice models, scientific models, and machine learning models. Relevant academic research results are summarized. The best practice model regards the aircraft as noise point source, and its specialty is to predict noise under complex air traffic conditions. The scientific model considers the noise from the level of aircraft components and reflects the underlying physical effects. Based on data, the machine learning model uses algorithms to mine the hidden relationship between various factors and noise to achieve the purpose of noise prediction. Then, this paper introduces two kinds of aviation noise simulation software based on the best practice and scientific models, and lists their access addresses. Finally, challenges and prospects are presented from three aspects: metrics, prediction models and simulation software.

show abstract

A Deep Learning Approach to Jet Noise Prediction

Cited by 15 publications

References 7 publications

Neural Networks-Based Aerodynamic Data Modeling: A Comprehensive Review

Neural Networks-Based Aerodynamic Data Modeling: A Comprehensive Review

A physics merged deep neural network-based prediction method for jet turbulent mixing noise

Review on Metrics and Prediction Methods of Civil Aviation Noise

Contact Info

Product

Resources

About