A5062068123- Advanced Multi-Objective Optimization Algorithms
- Probabilistic and Robust Engineering Design
- Ship Hydrodynamics and Maneuverability
- Manufacturing Process and Optimization
- BIM and Construction Integration
- Fluid Dynamics Simulations and Interactions
- Metaheuristic Optimization Algorithms Research
- Engineering Applied Research
- Model Reduction and Neural Networks
- Additive Manufacturing and 3D Printing Technologies
- Fluid Dynamics and Vibration Analysis
- Smart Cities and Technologies
- Maritime Transport Emissions and Efficiency
- Gaussian Processes and Bayesian Inference
- Maritime Navigation and Safety
- Acoustic Wave Phenomena Research
- 3D Shape Modeling and Analysis
- Reservoir Engineering and Simulation Methods
- Topology Optimization in Engineering
- Electromagnetic Scattering and Analysis
- IoT and Edge/Fog Computing
- Digital Transformation in Industry
- Turbomachinery Performance and Optimization
- Infrastructure Maintenance and Monitoring
- Wind and Air Flow Studies
National Research Council
2014-2024
Institute of Marine Engineering
2019-2024
Laboratoire de Recherche Hydrodynamique, Energétique et Environnement Atmosphérique
2023
École Centrale de Nantes
2023
Centre National de la Recherche Scientifique
2023
National Research Council
2018-2021
Istituto Nazionale per Studi ed Esperienze di Architettura Navale
2014-2017
Roma Tre University
2014-2017
University of Padua
2014
Ca' Foscari University of Venice
2014
The paper presents a study on four adaptive sampling methods of multi-fidelity (MF) metamodel, based stochastic radial basis functions (RBF), for global design optimisation expensive CFD computer simulations and grid refinement. MF metamodel is built as the sum low-fidelity-trained an error difference between high- low-fidelity simulations. adaptively refined using dynamic criteria, prediction uncertainty in combination with objective optimum computational cost evaluations. are demonstrated...
Abstract A multi-fidelity (MF) active learning method is presented for design optimization problems characterized by noisy evaluations of the performance metrics. Namely, a generalized MF surrogate model used design-space exploration, exploiting an arbitrary number hierarchical fidelity levels, i.e., coming from different models, solvers, or discretizations, accuracy. The intended to accurately predict while reducing computational effort required simulation-driven (SDD) achieve global...
Abstract This paper presents a comparison of two multi-fidelity methods for the forward uncertainty quantification naval engineering problem. Specifically, we consider problem quantifying hydrodynamic resistance roll-on/roll-off passenger ferry advancing in calm water and subject to operational uncertainties (ship speed payload). The first four statistical moments (mean, variance, skewness, kurtosis), probability density function such quantity interest (QoI) are computed with methods, i.e.,...
The paper presents a multi-objective derivative-free and deterministic global/local hybrid algorithm for the efficient effective solution of simulation-based design optimization (SBDO) problems. objective is to show how hybridization two global local algorithms achieves better performance than separate use in solving specific SBDO problems hull-form design. proposed method belongs class memetic algorithms, where exploration capability particle swarm enriched by exploiting search accuracy...
The paper presents a multi-fidelity global metamodel for expensive computer simulations, developed as an essential part of efficient simulation-based design optimization under uncertainty. High- and low-fidelity solvers are managed through adaptive sampling procedure. approximation is built the sum low-fidelity-trained difference (error) between high- simulations. metamodels based on dynamic stochastic radial basis functions, which provide prediction along with associated New training points...
An adaptive N-fidelity approach to metamodeling from noisy data is presented for design-space exploration and design optimization. Computational fluid dynamics (CFD) simulations with different numerical accuracy (spatial discretization) provides metamodel training sets affected by unavoidable noise. The approximation built an additive correction of a low-fidelity metamodels differences (errors) between higher-fidelity levels whose hierarchy needs be provided. encompasses two core techniques,...
This paper presents a comparison of two methods for the forward uncertainty quantification (UQ) complex industrial problems. Specifically, performance Multi-Index Stochastic Collocation (MISC) and adaptive multi-fidelity Radial Basis Functions (SRBF) surrogates is assessed UQ roll-on/roll-off passengers ferry advancing in calm water subject to operational uncertainties, namely ship speed draught. The estimation expected value, standard deviation, probability density function (modelscale)...
The paper presents the use of a supervised active learning approach for solution simulation-driven design optimization (SDDO) problem, pertaining to resistance reduction destroyer-type vessel in calm water. is formulated as single-objective, single-point problem with both geometrical and operational constraints. latter also considers seakeeping performance at multiple conditions. A surrogate model used, based on stochastic radial basis functions lower confidence bounding, approach....
The paper presents a multi-fidelity extension of local line-search-based derivative-free algorithm for nonsmooth constrained optimization (MF-CS-DFN). method is intended use in the simulation-driven design (SDDO) context, where computations are used to evaluate objective function. proposed starts using low-fidelity evaluations and automatically switches higher-fidelity based on line-search step length. driven by suitably defined threshold initialization values length, which associated each...
A multi-fidelity global metamodel is presented for uncertainty quantification of computationally expensive simulations.The approximation built as the sum a low-fidelity-trained and difference (error) between high-and low-fidelity based on dynamic stochastic radial basis functions, which provide prediction along with associated uncertainty.New training points are added where largest, according to an adaptive sampling procedure.The both error considered low-and high-fidelity metamodels,...
Lifting hydrofoils are gaining importance, since they drastically reduce the wetted surface area of a ship hull, thus decreasing resistance.To attain efficient hydrofoils, geometries can be obtained from an automated optimization process, based on simulations.However, hydrofoil high-fidelity simulations computationally demanding, fine meshes needed to accurately capture pressure field and boundary layer hydrofoil.Moreover, immersed depth varies dynamically, which makes simulation...
Despite the increased computational resources, simulation-based design optimization (SBDO) procedure can be very expensive from a viewpoint, especially if high-fidelity solvers are required. Multi-fidelity metamodels have been successfully applied to reduce cost of SBDO process. In this context, paper presents performance assessment an adaptive multi-fidelity metamodel based on Gaussian process regression (MF-GPR) for noisy data. The MF-GPR is developed to: (i) manage arbitrary number...
Uncertainty Quantification (UQ) is vital to safety-critical model-based analyses, but the widespread adoption of sophisticated UQ methods limited by technical complexity. In this paper, we introduce UM-Bridge (the and Modeling Bridge), a high-level abstraction software protocol that facilitates universal interoperability with simulation codes. It breaks down complexity advanced applications enables separation concerns between experts. democratizes allowing effective interdisciplinary...