Mach number prediction models based on Ensemble Neural Networks for wind tunnel testing

Jin, Zhiwei; Zhao, Li; Rao, Zhengzhou

doi:10.1109/ccdc.2014.6852430

Cited by 8 publications

(6 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the purpose of this paper is not to develop Mach number predictive model, the effectiveness of ANFIS has still been validated. In [5], an ensemble neural network (ENN) predictive model is proposed to forecast Mach number in wind tunnel. Projections of the original training data on several subspace subsets are firstly used to learn base neural network models, which are combined then by a fusion rule.…”

Section: Related Workmentioning

confidence: 99%

“…Because the bootstrap sampling technique used by Bagging sample data points with replacement, the probability that a subset contains outliers is much lower than normal points. We express this statement in Formula (5).…”

Section: A Robustness Enhancementmentioning

confidence: 99%

“…(( , ) ∈ ∩ { }) ≪ (( , ) ∈ ∩ { }) (5) where ( , ) represents any data point in . This equation indicates that for most training subsets, the fraction of outliers is smaller than that in original training set.…”

Section: A Robustness Enhancementmentioning

confidence: 99%

“…Accordingly, a parametric predictive model seems appropriate for nonlinear plants. In literature, several machine learning algorithms have been used for forecasting Mach number in wind tunnel, such as neural network (NN) [5], least squares support vector machine (LSSVM) [6], adaptive network-based fuzzy inference system (ANFIS) [7], and Gaussian process regression (GPR) [8]. Note that all these predictive models share the same objective, i.e.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Enhancing the Prediction of Mach Number in Wind Tunnel With a Regression-Based Outlier Detection Framework

et al. 2021

View full text Add to dashboard Cite

Model predictive control (MPC) has been successfully applied in practical wind tunnel systems. As a critical component of MPC, the accurate prediction of Mach number becomes hence very significant. Real measurement data records often contain outliers or noisy data because of the harsh working conditions of practical systems. Such datasets are detrimental to most data-driven Mach number predictors. This paper focuses on improving Mach number prediction via a regression-based outlier detection framework. The development of a regression-based outlier detector for enhancing Mach number prediction is natural since only regression analysis needs to be implemented for both prediction and outlier detection. In contrast to existing regression-based detectors, we have taken into account three problems that may greatly impact the detecting performance. Specifically, we use Bagging technique to reduce the negative influence of unknown outliers during the training phase. Then, the notion of extreme value analysis is used to determine an appropriate threshold for the calculated outlier scores. A robust scaling method based on Hampel identifier is also used to alleviate the influence of outliers on data scaling. Several datasets stemming from a real-world wind tunnel are used in experiments to verify the effectiveness of our detector. The experimental results show that regression-based outlier detectors often perform better if the prior knowledge about the dependent variable is used. In addition, the proposed detector has outperformed other regression-based detectors.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: A Robustness Enhancementmentioning

confidence: 99%

Section: A Robustness Enhancementmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Enhancing the Prediction of Mach Number in Wind Tunnel With a Regression-Based Outlier Detection Framework

et al. 2021

View full text Add to dashboard Cite

show abstract

“…On the other hand, some researchers have proposed data-driven methods to model Information Technology and Control 2017/3/46 404 wind tunnel systems. For instance, Jin et al [7] developed a feature subsets based ensemble neural networks (ENN) nonlinear model and Rui et al [22] presented a BP neural network based NARMAX model. The above methods both suffer heavy computational complexity problems [10], which may narrow their applicability for real-time control tasks.…”

Section: Introductionmentioning

confidence: 99%

A Nonlinear Block-oriented Model for Wind tunnel System

Zhang

Mao²,

Yuan³

2017

ITC

View full text Add to dashboard Cite

This paper develops a novel nonlinear block-oriented model for the wind tunnel system. Based on the available signals, the wind tunnel system can be divided into three parts, namely, the exhaust valve loop, the choke finger loop and the flow field. Then the considered plant is described as a nonlinear block-oriented model. The exhaust valve subsystem and the flow field subsystem are both expressed by linear dynamic models, whereas the choke finger subsystem exhibits a nonlinear characteristics and is approximated by a pseudo-Hammerstein model. Based on the above parameterization model, the recursive identification algorithms are presented for three subsystems. Interestingly, the adaptive weighted recursive least squares algorithm is applied to the pseudo-Hammerstein model, and the hierarchical recursive least squares algorithm is used to reduce the computational complexities. Both simulations and experiments are carried out to verify the effectiveness of the proposed method.

show abstract

Integrating Mach Number Prediction with Outlier Detection for Wind Tunnel Systems

Wang

Mao

2019

J. Aerosp. Eng.

View full text Add to dashboard Cite

Mach number prediction models based on Ensemble Neural Networks for wind tunnel testing

Cited by 8 publications

References 5 publications

Enhancing the Prediction of Mach Number in Wind Tunnel With a Regression-Based Outlier Detection Framework

Enhancing the Prediction of Mach Number in Wind Tunnel With a Regression-Based Outlier Detection Framework

A Nonlinear Block-oriented Model for Wind tunnel System

Integrating Mach Number Prediction with Outlier Detection for Wind Tunnel Systems

Contact Info

Product

Resources

About