Travaux collaboratifs sur la mise au point de surfaces de réponses pour l’étude de la dispersion des polluants en milieu urbain
Publications scientifiques au M2P2
V. Mons, Luca Margheri, J.-C. Chassaing, Pierre Sagaut. Data assimilation-based reconstruction of urban pollutant release characteristics. Journal of Wind Engineering and Industrial Aerodynamics, Elsevier, 2017, 169, pp.232 - 250. ⟨10.1016/j.jweia.2017.07.007⟩. ⟨hal-01631036⟩ Plus de détails...
Xun Wang, Shahram Khazaie, Luca Margheri, Pierre Sagaut. Shallow water sound source localization using the iterative beamforming method in an image framework. Journal of Sound and Vibration, Elsevier, 2017, 395, pp.354 - 370. ⟨10.1016/j.jsv.2017.02.032⟩. ⟨hal-01527615⟩ Plus de détails...
Shallow water is a complicated sound propagation medium due to multiple reflections by water surface and bottom, imprecisely measured sound speed, noisy environment, etc. Therefore, in order to localize a shallow water sound source, classical signal processing techniques must be improved by taking these complexities into account. In this work, the multiple reflections and uncertain reflectivity of water bottom are explicitly modeled. In the proposed model, a measured signal is a mixture of the direct propagation from the source and the multiple reflections. Instead of solving the Helmholtz equation with boundary conditions of reflections, each signal is interpreted as a superposition of signals emitting from the physical source and its image sources in a free space, which results in a fast computation of sound propagation. Then, the source location, along with its amplitude, reflection paths and power loss of bottom reflection, is estimated via the iterative beamforming (IB) method, which alternatively estimates the source contributions and performs beamforming on these estimates until convergence. This approach does not need to compute the sound propagation for all the possible source locations in a large space, which thus leads to a low computational cost. Finally, numerical simulations are introduced to illustrate the advantage of the proposed model and the source estimation method. The sensitivity of the proposed method with respect to model parameter uncertainties is also investigated via a full uncertainty quantification analysis. The localization error of IB is proved to be acceptable in the given error range of sound speed and water depth. Besides, the IB source estimate is more sensitive to the sound speed while the matched-field processing methods have a stronger sensitivity to the water depth: this result can guide the choice of source localization method in different cases of model parameter uncertainties.
Xun Wang, Shahram Khazaie, Luca Margheri, Pierre Sagaut. Shallow water sound source localization using the iterative beamforming method in an image framework. Journal of Sound and Vibration, Elsevier, 2017, 395, pp.354 - 370. ⟨10.1016/j.jsv.2017.02.032⟩. ⟨hal-01527615⟩
Luca Margheri, Pierre Sagaut. A hybrid anchored-ANOVA - POD/Kriging method for uncertainty quantification in unsteady high-fidelity CFD simulations. Journal of Computational Physics, Elsevier, 2016, 324, pp.137-173. ⟨10.1016/j.jcp.2016.07.036⟩. ⟨hal-01461789⟩ Plus de détails...
To significantly increase the contribution of numerical computational fluid dynamics (CFD) simulation for risk assessment and decision making, it is important to quantitatively measure the impact of uncertainties to assess the reliability and robustness of the results. As unsteady high-fidelity CFD simulations are becoming the standard for industrial applications, reducing the number of required samples to perform sensitivity (SA) and uncertainty quantification (UQ) analysis is an actual engineering challenge. The novel approach presented in this paper is based on an efficient hybridization between the anchored-ANOVA and the POD/Kriging methods, which have already been used in CFD-UQ realistic applications, and the definition of best practices to achieve global accuracy. The anchored-ANOVA method is used to efficiently reduce the UQ dimension space, while the POD/Kriging is used to smooth and interpolate each anchored-ANOVA term. The main advantages of the proposed method are illustrated through four applications with increasing complexity, most of them based on Large-Eddy Simulation as a high-fidelity CFD tool: the turbulent channel flow, the flow around an isolated bluff-body, a pedestrian wind comfort study in a full scale urban area and an application to toxic gas dispersion in a full scale city area. The proposed c-APK method (anchored-ANOVA-POD/Kriging) inherits the advantages of each key element: interpolation through POD/Kriging precludes the use of quadrature schemes therefore allowing for a more flexible sampling strategy while the ANOVA decomposition allows for a better domain exploration. A comparison of the three methods is given for each application. In addition, the importance of adding flexibility to the control parameters and the choice of the quantity of interest (QoI) are discussed. As a result, global accuracy can be achieved with a reasonable number of samples allowing computationally expensive CFD-UQ analysis. (C) 2016 Elsevier Inc. All rights reserved.
Luca Margheri, Pierre Sagaut. A hybrid anchored-ANOVA - POD/Kriging method for uncertainty quantification in unsteady high-fidelity CFD simulations. Journal of Computational Physics, Elsevier, 2016, 324, pp.137-173. ⟨10.1016/j.jcp.2016.07.036⟩. ⟨hal-01461789⟩