Accelerated Multi-Fidelity Emulator Modeling for Probabilistic Rotor Response Study

2019 ◽  
Author(s):  
Harok Bae ◽  
Ian M. Boyd ◽  
Emily Carper ◽  
Jeff Brown
Keyword(s):  
Prediction Accuracy ◽  
Point Cloud ◽  
Functional Relationship ◽  
Matrix Inversion ◽  
Numerical Examples ◽  
Computational Costs ◽  
Inversion Methods ◽  
Training Samples ◽  
Probabilistic Study ◽  
Modal Solution

Abstract This paper presents an efficient methodology to build a modal solution emulator for the probabilistic study of geometrically mistuned bladed rotors by using the newly developed localized-Galerkin multi-fidelity modeling and eigensolution reanalysis with the symmetric successive matrix inversion methods. The key idea of the mistuned blade emulator is to establish a reduced functional relationship between the stochastic geometric variations and the disturbed modal responses. The prediction accuracy of an emulator depends on how many training samples of modal solutions are available and how well the potential modal switching due to stochastic mistuning is captured. To reduce the computational costs of generating training samples without sacrificing accuracy, this paper introduces the collaborative framework of the new approaches of multi-fidelity modeling and eigensolution reanalysis. The proposed framework is demonstrated for its computational benefits with several numerical examples including the point-cloud scanned mistuned blade problem.

Accelerated Multifidelity Emulator Modeling for Probabilistic Rotor Response Study

2019 ◽  
Vol 141 (12) ◽  
Author(s):  
Harok Bae ◽  
Ian M. Boyd ◽  
Emily B. Carper ◽  
Jeff Brown
Keyword(s):  
Prediction Accuracy ◽  
Point Cloud ◽  
Functional Relationship ◽  
Matrix Inversion ◽  
Numerical Examples ◽  
Computational Costs ◽  
New Approaches ◽  
Training Samples ◽  
Probabilistic Study ◽  
Modal Solution

Abstract This paper presents an efficient methodology to build a modal solution emulator for the probabilistic study of geometrically mistuned bladed rotors by using the newly developed localized-Galerkin multifidelity (LGMF) modeling and eigensolution reanalysis (ER) with the symmetric successive matrix inversion (SSMI) methods. The key idea of the mistuned blade emulator is to establish a reduced functional relationship between the stochastic geometric variations and the disturbed modal responses. The prediction accuracy of an emulator generally depends on how many training samples of modal solutions are available and how well the potential modal switching due to stochastic mistuning is captured. To reduce the computational costs of generating training samples without sacrificing accuracy, this paper introduces the collaborative framework of the new approaches of multifidelity (MF) modeling and ER. The proposed framework is demonstrated for its computational benefits with several numerical examples including the point-cloud scanned mistuned blade problem.

scholarly journals Weight-Selected Attribute Bagging for Credit Scoring

2013 ◽  
Vol 2013 ◽  
pp. 1-13 ◽  
Author(s):  
Jianwu Li ◽  
Haizhou Wei ◽  
Wangli Hao
Keyword(s):  
Credit Risk ◽  
Prediction Accuracy ◽  
Financial Risk ◽  
Evaluation Method ◽  
Credit Scoring ◽  
Principal Component ◽  
Support Vector ◽  
Training Samples ◽  
Attribute Evaluation ◽  
Bagging Method

Assessment of credit risk is of great importance in financial risk management. In this paper, we propose an improved attribute bagging method, weight-selected attribute bagging (WSAB), to evaluate credit risk. Weights of attributes are first computed using attribute evaluation method such as linear support vector machine (LSVM) and principal component analysis (PCA). Subsets of attributes are then constructed according to weights of attributes. For each of attribute subsets, the larger the weights of the attributes the larger the probabilities by which they are selected into the attribute subset. Next, training samples and test samples are projected onto each attribute subset, respectively. A scoring model is then constructed based on each set of newly produced training samples. Finally, all scoring models are used to vote for test instances. An individual model that only uses selected attributes will be more accurate because of elimination of some of redundant and uninformative attributes. Besides, the way of selecting attributes by probability can also guarantee the diversity of scoring models. Experimental results based on two credit benchmark databases show that the proposed method, WSAB, is outstanding in both prediction accuracy and stability, as compared to analogous methods.

scholarly journals TAILORED FEATURES FOR SEMANTIC SEGMENTATION WITH A DGCNN USING FREE TRAINING SAMPLES OF A COLORED AIRBORNE POINT CLOUD

2020 ◽  
Vol XLIII-B2-2020 ◽  
pp. 339-346
Author(s):  
E. Widyaningrum ◽  
M. K. Fajari ◽  
R. C. Lindenbergh ◽  
M. Hahn
Keyword(s):  
Deep Learning ◽  
Point Cloud ◽  
Semantic Segmentation ◽  
Blue Color ◽  
Point Cloud Data ◽  
Dynamic Graph ◽  
Cloud Data ◽  
Training Samples ◽  
Map Generation ◽  
Fast Development

Abstract. Automation of 3D LiDAR point cloud processing is expected to increase the production rate of many applications including automatic map generation. Fast development on high-end hardware has boosted the expansion of deep learning research for 3D classification and segmentation. However, deep learning requires large amount of high quality training samples. The generation of training samples for accurate classification results, especially for airborne point cloud data, is still problematic. Moreover, which customized features should be used best for segmenting airborne point cloud data is still unclear. This paper proposes semi-automatic point cloud labelling and examines the potential of combining different tailor-made features for pointwise semantic segmentation of an airborne point cloud. We implement a Dynamic Graph CNN (DGCNN) approach to classify airborne point cloud data into four land cover classes: bare-land, trees, buildings and roads. The DGCNN architecture is chosen as this network relates two approaches, PointNet and graph CNNs, to exploit the geometric relationships between points. For experiments, we train an airborne point cloud and co-aligned orthophoto of the Surabaya city area of Indonesia to DGCNN using three different tailor-made feature combinations: points with RGB (Red, Green, Blue) color, points with original LiDAR features (Intensity, Return number, Number of returns) so-called IRN, and points with two spectral colors and Intensity (Red, Green, Intensity) so-called RGI. The overall accuracy of the testing area indicates that using RGB information gives the best segmentation results of 81.05% while IRN and RGI gives accuracy values of 76.13%, and 79.81%, respectively.

The Effect of Solidification Models on the Prediction Results of the Temperature Change of the Aluminum Cylinder Head Estimated by FDM Solidification Analysis

2007 ◽  
Vol 561-565 ◽  
pp. 1967-1970 ◽  
Author(s):  
Hiroshi Onda ◽  
Kazunari Sakurai ◽  
Tatsuya Masuta ◽  
Katsunari Oikawa ◽  
Koichi Anzai ◽  
...  
Keyword(s):  
Aluminum Alloy ◽  
Temperature Change ◽  
Cylinder Head ◽  
Prediction Accuracy ◽  
Functional Relationship ◽  
Solidification Process ◽  
Solidification Model ◽  
Fraction Solid ◽  
Aluminum Cylinder ◽  
Selection Of

This paper presents the prediction results of the temperature change during the solidification process of the cylinder head made of the AC2A aluminum alloy. Prediction results have been obtained by using the FDM solidification analysis based on two different solidification models were investigated. Here, the solidification model means functional relationship between the Temperature and the Fraction Solid. The first model is a simple Linear function and the second model is estimated from DSC measurement. The comparison between the simulated and measured temperatures of the aluminum cylinder head revealed that the selection of solidification models significantly reflects the prediction results. The DSC model gives higher prediction accuracy of the temperature change than the Linear model. The solidification models estimated by using Thermo- Calc and UMSA [3] were also investigated.

scholarly journals A New High-Order Stable Numerical Method for Matrix Inversion

2014 ◽  
Vol 2014 ◽  
pp. 1-10 ◽  
Author(s):  
F. Khaksar Haghani ◽  
F. Soleymani
Keyword(s):  
Iterative Method ◽  
Numerical Method ◽  
Matrix Inversion ◽  
High Order ◽  
New Method ◽  
Order Convergence ◽  
Initial Value ◽  
Numerical Examples ◽  
Stable Numerical Method

A stable numerical method is proposed for matrix inversion. The new method is accompanied by theoretical proof to illustrate twelfth-order convergence. A discussion of how to achieve the convergence using an appropriate initial value is presented. The application of the new scheme for finding Moore-Penrose inverse will also be pointed out analytically. The efficiency of the contributed iterative method is clarified on solving some numerical examples.

scholarly journals Adaptive Variable Weight Accumulation AVWA-DGM(1,1) Model Based on Particle Swarm Optimization

2019 ◽  
pp. 1-17
Author(s):  
Lang Yu
Keyword(s):  
Higher Education ◽  
Particle Swarm Optimization ◽  
Prediction Accuracy ◽  
Particle Swarm ◽  
Data Series ◽  
Particle Swarm Algorithm ◽  
Swarm Optimization ◽  
Raw Data ◽  
Numerical Examples ◽  
Variable Weight

The development of higher education is an extremely important issue. It is the source of the country's technological innovation and the realization of innovation and development, especially in China, where higher education is still at an exploratory stage. Aiming at the shortcoming that the classical DGM (1,1) model accumulates the raw data series with the weight of constant "1", this paper proposes an adaptive variable weight accumulation optimization DGM (1,1) model, abbreviated as AVWA-DGM (1,1) model. Taking the enrollment numbers of postgraduate, master degree, undergraduate and junior college student and undergraduates students in China as numerical examples, the DGM (1,1) model and AVWA-DGM (1,1) model are established to simulate and predict respectively, and the weighted coefficients of AVWA-DGM (1,1) model are optimized and solved by particle swarm algorithm. The results show that the AVWA-DGM(1,1) model has higher simulation and prediction accuracy than the classical DGM(1,1) model in the four numerical examples provided in this paper. It can be seen that the adaptive accumulation of the raw data sequence by the particle swarm optimization algorithm can make the first order accumulation sequence more in line with the requirements of the DGM (1,1) model on the data features, thereby improving the simulation and prediction accuracy.

Error Measures for Functional Product Testing

1999 ◽  
Author(s):  
Uichung Cho ◽  
Kristin L. Wood ◽  
Richard H. Crawford
Keyword(s):  
Error Estimation ◽  
Prediction Accuracy ◽  
Estimation Procedure ◽  
Functional Testing ◽  
Partial Knowledge ◽  
Product Testing ◽  
Numerical Examples ◽  
Error Measures ◽  
Functional Product ◽  
Similarity Method

Abstract During product development, testing of models and prototypes offers significant advantages over direct product testing, including easier, cheaper, and faster fabrication. However, two issues prevent effective functional testing with prototypes: prediction accuracy and confidence in scale testing results. The traditional similarity method, which is based on dimensional analysis, is commonly applied to perform scale testing. However, the method may not provide accurate scale testing results, especially when available model materials are different from the final product materials. The authors have developed a new empirical similarity method, wherein specimen pairs and partial knowledge of systems are systematically utilized, to improve the prediction accuracy. In this paper we describe the construction of error measures to utilize scale testing results with confidence. In practice, scale testing results are validated based on experiences with previous testing results. This approach to predicting accuracy is difficult to formalize. We develop and simulate a systematic two-level error estimation procedure. Realistic numerical examples demonstrate the feasibility of the approach.

Sign in / Sign up

Export Citation Format

Share Document