A Geometry-Based Deep Learning Feature Extraction Scheme for Airfoils

The perception of geometry-features of airfoils is the basis in aerodynamic area for performance prediction, parameterization, aircraft inverse design, etc. There are three approaches to percept the geometric shape of an airfoil, namely manual design of airfoil geometry parameter, polynomial definition and deep learning. The first two methods can directly extract geometry-features of airfoils or polynomial equations of airfoil curves, but the number of features extracted is limited. While deep learning algorithms can extract a large number of potential features (called latent features), however, the features extracted by deep learning are lacking of explicit geometrical meaning. Motivated by the advantages of polynomial definition and deep learning, we propose a geometry-based deep learning feature extraction scheme (named Bézier-based feature extraction, BFE) for airfoils, which consists of two parts: manifold metric feature extraction and geometry-feature fusion encoder (GF encoder). Manifold metric feature extraction, with the help of the Bézier curve, captures features from tangent space of airfoil curves, and GF encoder combines airfoil coordinate data and manifold metrics together to form a novel feature representation. A public UIUC airfoil dataset is used to verify the proposed BFE. Compared with classic Auto-Encoder, the mean square error (MSE) of BFE is reduced by 17.97% ~29.14%.

Global Sensitivity Analysis of Chosen Harmonic Drive Parameters Affecting Its Lost Motion

The aim of the presented study was to perform a global sensitivity analysis of various design parameters affecting the lost motion of the harmonic drive. A detailed virtual model of a harmonic drive was developed, including the wave generator, the flexible ball bearing, the flexible spline and the circular spline. Finite element analyses were performed to observe which parameter from the harmonic drive geometry parameter group affects the lost motion value most. The analyses were carried out using 4% of the rated harmonic drive output torque by the locked wave generator and fixed circular spline according the requirements for the high accuracy harmonic drive units. The described approach was applied to two harmonic drive units with the same ratio, but various dimensions and rated power were used to generalize and interpret the global sensitivity analysis results properly. The most important variable was for both harmonic drives the offset from the nominal tooth shape.

GEP- and MLR-based equations for stable channel analysis

For decades, research on stable channel hydraulic geometry was based on the following parameters: river discharge, dimensionless discharge, the median size of bed material and the slope. Although significant research has been conducted in this area, including applied machine learning to increase the geometry model prediction accuracy, there has been no remarkable improvement as the variables used to describe the geometry relationship remain the same. The novelty of this study is demonstrated by the parameters used in the stable channel geometry equations that outperform the existing equation's accuracy. In this research, sediment transport parameters are introduced and analysed by applying the multiple linear regression (MLR) and gene expression programming (GEP) methods. The new equation of the width, depth and bed slope can give much-improved results in efficiency and lower errors. Furthermore, a new parameter B/y is introduced in this study to solve the restriction issue, either in width or depth prediction. The results from MLR and GEP show that in addition to the existing hydraulic geometry parameter, the B/y parameter is also able to give high accuracy results for width and depth predictions. Both calibration and validation for the B/y parameter yield high R2 and NSE values with low mean squared errors and mean absolute errors.

Normalized Partial Scattering Cross Section for Performance Evaluation of Low-Observability Scattering Structures

The development of diffusion metasurfaces created new opportunities to elevate the stealthiness of combat aircraft. Despite the potential significance of metasurfaces, their rigorous design methodologies are still lacking, especially in the context of meticulous control over the scattering of electromagnetic (EM) waves through geometry parameter tuning. Another practical issue is insufficiency of the existing performance metrics, specifically, monostatic and bistatic evaluation of the reflectivity, especially at the design stage of metasurfaces. Both provide limited insight into the RCS reduction properties, with the latter being dependent on the selection of the planes over which the evaluation takes place. This paper introduces a novel performance metric for evaluating scattering characteristics of a metasurface, referred to as Normalized Partial Scattering Cross Section (NPSCS). The metric involves integration of the scattered energy over a specific solid angle, which allows for a comprehensive assessment of the structure performance in a format largely independent of the particular arrangement of the scattering lobes. We demonstrate the utility of the introduced metric using two specific metasurface architectures. In particular, we show that the integral-based metric can be used to discriminate between the various surface configurations (e.g., checkerboard versus random), which cannot be conclusively compared using traditional methods. Consequently, the proposed approach can be a useful tool in benchmarking radar cross section reduction performance of metamaterial-based, and other types of scattering structures.

Metamodel-driven data mining model to support three-dimensional design of centrifugal compressor stage

The design of a centrifugal compressor with high efficiency and a wide operating range is a challenging task. A great effort has been undertaken to solve the three-dimensional design problem with the assistance of a metamodel. However, the published works lack any study that systematically performs the data mining between the performance and three-dimensional geometry variation due to two unsolved issues, i.e., lack of reliable systematic data mining model and unresolved high-dimensional data problem in the centrifugal compressor community. To tackle these issues, a systematic metamodel-driven data mining model including six general modules has been proposed and implemented to the well-known Radiver stage. In this data mining task, four specific techniques were used to go through the general to specific data mining. The results showed the performance improvement probabilities, the trade-off relationships between performance parameters, the characteristic variation of the performance, and the correlations between performance and the most sensitive two geometry parameter variation. The appropriate variation ranges for wide operating range design of the two sensitive geometry parameters were recommended and the flow mechanism behind them was clarified. The statistical results showed that over 90% of the design stages in the recommended variation ranges had a wide operating range. A design case was chosen randomly in the recommended range to verify the performance improvement via CFD simulations. The outcomes of this work are particularly relevant for the advanced design of compressors with high efficiency and a wide operating range for flexibility.

Generating Component Designs for an Improved NVH Performance by Using an Artificial Neural Network as an Optimization Metamodel

In modern vehicle development, suspension components have to meet many boundary conditions. In noise, vibration, and harshness (NVH) development these are for example eigenfrequencies and frequency response function (FRF) amplitudes. Component geometry parameters, for example kinematic hard points, often affect multiple of these targets in a non intuitive way. In this article, we present a practical approach to find optimized parameters for a component design, which fulfills an FRF target curve. By morphing an initial component finite element model we create training data for an artificial neural network (ANN) which predicts FRFs from geometry parameter input. Then the ANN serves as a metamodel for an evolutionary algorithm optimizer which identifies fitting geometry parameter sets, meeting an FRF target curve. The methodology enables a component design which considers an FRF as a component target. In multiple simulation examples we demonstrate the capability of identifying component designs modifying specific eigenfrequency or amplitude features of the FRFs.

Approximate Bayesian computation for railway track geometry parameter estimation

The quality of track geometry is directly linked to vehicle safety, reliability and ride quality. The performance of track is therefore considerably hindered when track geometry indicators deviate from the specified and approved limits due to loads and continuous usage. Information obtained from the analysis of track geometry data can inform the prompt application of preventive and corrective maintenance measures like tamping, to increase the lifespan of the track and provide higher train speeds, optimizing track performance. Recently, there has been the application of Bayesian statistical methods in track degradation models. However, most models rely heavily on likelihood functions which are intractable. The aim of this paper is to apply Approximate Bayesian Computation (ABC), also known as the likelihood-free method, in estimating Track Quality Indices (TQIs) which are essential for track degradation modeling. ABC is applied using methods like the rejection algorithm and Markov Chain Monte Carlo (MCMC). In ABC, it is essential that summary statistics are computed from the observed data followed by the simulation of summary statistics for different parameter values. Two ABC-MCMC algorithms were used for parameter estimation in this paper. Models were developed using 200 ft. TQIs and 150 ft. TQIs with different priors and model choice was performed with the aid of Bayes Factors.

Top Quark Pair-Production in Noncommutative Standard Model

The differential cross-section of the top quark pair production via the quark-antiquark annihilation subprocess in hadron collision is calculated within the noncommutative standard model. A pure NC analytical expression for the forward-backward asymmetry at the tree level is obtained. Moreover, using recent Tevatron results from the full RUN2 data, a new lower bound on the noncommutative geometry parameter is deduced.