Redirigiendo al acceso original de articulo en 20 segundos...
Inicio  /  Algorithms  /  Vol: 13 Par: 5 (2020)  /  Artículo
ARTÍCULO
TITULO

Uncertainty Quantification Approach on Numerical Simulation for Supersonic Jets Performance

Carlo Cravero    
Davide De Domenico and Andrea Ottonello    

Resumen

One of the main issues addressed in any engineering design problem is to predict the performance of the component or system as accurately and realistically as possible, taking into account the variability of operating conditions or the uncertainty on input data (boundary conditions or geometry tolerance). In this paper, the propagation of uncertainty on boundary conditions through a numerical model of supersonic nozzle is investigated. The evaluation of the statistics of the problem response functions is performed following ?Surrogate-Based Uncertainty Quantification?. The approach involves: (a) the generation of a response surface starting from a DoE in order to approximate the convergent?divergent ?physical? model (expensive to simulate), (b) the application of the UQ technique based on the LHS to the meta-model. Probability Density Functions are introduced for the inlet boundary conditions in order to quantify their effects on the output nozzle performance. The physical problem considered is very relevant for the experimental tests on the UQ approach because of its high non-linearity. A small perturbation to the input data can drive the solution to a completely different output condition. The CFD simulations and the Uncertainty Quantification were performed by coupling the open source Dakota platform with the ANSYS Fluent® CFD commercial software: the process is automated through scripting. The procedure adopted in this work demonstrate the applicability of advanced simulation techniques (such as UQ analysis) to industrial technical problems. Moreover, the analysis highlights the practical use of the uncertainty quantification techniques in predicting the performance of a nozzle design affected by off-design conditions with fluid-dynamic complexity due to strong nonlinearity.

 Artículos similares

       
 
Xuan Di, Rongye Shi, Zhaobin Mo and Yongjie Fu    
For its robust predictive power (compared to pure physics-based models) and sample-efficient training (compared to pure deep learning models), physics-informed deep learning (PIDL), a paradigm hybridizing physics-based models and deep neural networks (DN... ver más
Revista: Algorithms

 
Mingzhi Li, Xianjun Yu, Dejun Meng, Guangfeng An and Baojie Liu    
Studies on the geometry variation-related compressor uncertainty quantification (UQ) have often used dimension reduction methods, such as the principal component analysis (PCA), for the modeling of deviations. However, in the PCA method, the main eigenmo... ver más
Revista: Aerospace

 
Shenren Xu, Qian Zhang, Dingxi Wang and Xiuquan Huang    
Precise and inexpensive uncertainty quantification (UQ) is crucial for robust optimization of compressor blades and to control manufacturing tolerances. This study looks into the suitability of MC-adj-nonlinear, a nonlinear adjoint-based approach, to pre... ver más
Revista: Aerospace

 
Andreas Nugaard Holm, Dustin Wright and Isabelle Augenstein    
Uncertainty approximation in text classification is an important area with applications in domain adaptation and interpretability. One of the most widely used uncertainty approximation methods is Monte Carlo (MC) dropout, which is computationally expensi... ver más
Revista: Information

 
Wenhao Fu, Zeshuai Chen and Jiaqi Luo    
Stochastic variations of the operation conditions and the resultant variations of the aerodynamic performance in Low-Pressure Turbine (LPT) can often be found. This paper studies the aerodynamic performance impact of the uncertain variations of flow para... ver más
Revista: Aerospace