Study on the choices of design parameters for inverse design of metasurface using Deep leargning

2020 ◽  
Author(s):  
Junjie Hou ◽  
Hai Lin ◽  
Lijie Chen ◽  
Feng Deng ◽  
Zhonghua Fang
Keyword(s):  
Inverse Design ◽  
Design Parameters

scholarly journals Artificial neural networks for inverse design of resonant nanophotonic components with oscillatory loss landscapes

Nanophotonics ◽  
2020 ◽  
Vol 10 (1) ◽  
pp. 385-392
Author(s):  
Joeri Lenaerts ◽  
Hannah Pinson ◽  
Vincent Ginis
Keyword(s):  
Machine Learning ◽  
Optimization Procedure ◽  
Linear Mapping ◽  
Inverse Design ◽  
Design Parameters ◽  
Second Step ◽  
Input Parameters ◽  
Novel Variant ◽  
Resonant Systems ◽  
The Fourier Transform

AbstractMachine learning offers the potential to revolutionize the inverse design of complex nanophotonic components. Here, we propose a novel variant of this formalism specifically suited for the design of resonant nanophotonic components. Typically, the first step of an inverse design process based on machine learning is training a neural network to approximate the non-linear mapping from a set of input parameters to a given optical system’s features. The second step starts from the desired features, e.g. a transmission spectrum, and propagates back through the trained network to find the optimal input parameters. For resonant systems, this second step corresponds to a gradient descent in a highly oscillatory loss landscape. As a result, the algorithm often converges into a local minimum. We significantly improve this method’s efficiency by adding the Fourier transform of the desired spectrum to the optimization procedure. We demonstrate our method by retrieving the optimal design parameters for desired transmission and reflection spectra of Fabry–Pérot resonators and Bragg reflectors, two canonical optical components whose functionality is based on wave interference. Our results can be extended to the optimization of more complex nanophotonic components interacting with structured incident fields.

Parametric Design of a Waterjet Pump by Means of Inverse Design, CFD Calculations and Experimental Analyses

2010 ◽  
Vol 132 (3) ◽  
Author(s):  
Duccio Bonaiuti ◽  
Mehrdad Zangeneh ◽  
Reima Aartojarvi ◽  
Jonas Eriksson
Keyword(s):  
Flow Field ◽  
Parametric Study ◽  
Parametric Design ◽  
Inverse Design ◽  
Design Parameters ◽  
Pump Design ◽  
Hydrodynamic Efficiency ◽  
Numerical Predictions ◽  
High Level ◽  
Suction Performance

The present paper describes the parametric design of a mixed-flow water-jet pump. The pump impeller and diffuser geometries were parameterized by means of an inverse design method, while CFD analyses were performed to assess the hydrodynamic and suction performance of the different design configurations that were investigated. An initial pump design was first generated and used as baseline for the parametric study. The effect of several design parameters was then analyzed in order to determine their effect on the pump performance. The use of a blade parameterization, based on inverse design, led to a major advantage in this study, because the three-dimensional blade shape is described by means of hydrodynamic parameters, such as blade loading, which has a direct impact on the hydrodynamic flow field. On the basis of this study, an optimal configuration was designed with the aim of maximizing the pump suction performance, while at the same time, guaranteeing a high level of hydrodynamic efficiency, together with the required mechanical and vibrational constraints. The final design was experimentally tested, and the good agreement between numerical predictions and experimental results validated the design process. This paper highlights the contrasting requirements in the pump design in order to achieve high hydrodynamic efficiency or good cavitation performance. The parametric study allowed us to determine design guidelines in order to find the optimal compromise in the pump design, in cases where both a high level of efficiency and suction performance must simultaneously be achieved. The design know-how developed in this study is based on flow field analyses and on hydrodynamic design parameters. It has therefore a general validity and can be used for similar design applications.

An Inverse Design Based Methodology for Rapid 3D Multi-Objective/Multidisciplinary Optimization of Axial Turbines

2011 ◽  
Author(s):  
Pietro Boselli ◽  
Mehrdad Zangeneh
Keyword(s):  
Design Method ◽  
Turbine Blades ◽  
Multidisciplinary Optimization ◽  
Inverse Design ◽  
Design Parameters ◽  
Multi Objective ◽  
Work Distribution ◽  
Blade Shape ◽  
Inverse Design Method ◽  
Axial Turbines

Design of axial turbines, especially LP turbines, poses difficult tradeoffs between requirements of aerodynamic design and structural limitations. In this paper, a methodology is proposed for 3D multi-objective design of axial turbine blades in which a 3D inverse design method is coupled with a multi-objective genetic algorithm. By parameterizing the blade using blade loading parameters, spanwise work distribution and maximum thickness, a large part of the design space can be explored with very few design parameters. Furthermore, the inverse method not only computes the blade shape but also provides accurate 3D inviscid flow information. In the simple multi-disciplinary approach proposed here the different losses in axial turbines such as endwall losses, tip leakage losses and an indication of flow separation are related through well known correlations to the blade surface velocities predicted by the inverse design method. In addition, geometrical features such as throat area, lean angles and airfoil cross sectional area are computed from the blade shape employed during the optimization. Also, centrifugal stresses and bending stresses are related to the blade geometry. The methodology is then applied to the redesign of an LP turbine rotor with the aim of reducing the maximum stresses while maintaining the performance of the rotor. The results are confirmed by using the commercial CFX CFD (Computational Fluid Dynamics) code and Ansys FEA (Finite Element Analysis) codes.

Multiobjective Optimization Design of a Pump–Turbine Impeller Based on an Inverse Design Using a Combination Optimization Strategy

2013 ◽  
Vol 136 (1) ◽  
Author(s):  
Wei Yang ◽  
Ruofu Xiao
Keyword(s):  
Response Surface ◽  
Optimization Design ◽  
Design Method ◽  
Inverse Design ◽  
Design Parameters ◽  
Surface Model ◽  
Optimization Strategy ◽  
Pump Turbine ◽  
Combination Optimization ◽  
Final Design

This paper presents an automatic multiobjective hydrodynamic optimization strategy for pump–turbine impellers. In the strategy, the blade shape is parameterized based on the blade loading distribution using an inverse design method. An efficient response surface model relating the design parameters and the objective functions is obtained. Then, a multiobjective evolutionary algorithm is applied to the response surface functions to find a Pareto front for the final trade-off selection. The optimization strategy was used to redesign a scaled pump–turbine. Model tests were conducted to validate the final design and confirm the validity of the design strategy.

On the Coupling of Inverse Design and Optimization Techniques for Turbomachinery Blade Design

2006 ◽  
Author(s):  
Duccio Bonaiuti ◽  
Mehrdad Zangeneh
Keyword(s):  
Mechanical Design ◽  
Design Method ◽  
Three Dimensional ◽  
Axial Compressor ◽  
Optimization Techniques ◽  
Operating Conditions ◽  
Inverse Design ◽  
Design Parameters ◽  
Optimization Strategy ◽  
Design And Optimization

Optimization strategies have been used in recent years for the aerodynamic and mechanical design of turbomachine components. One crucial aspect in the use of such methodologies is the choice of the geometrical parameterization, which determines the complexity of the objective function to be optimized. In the present paper, an optimization strategy for the aerodynamic design of turbomachines is presented, where the blade parameterization is based on the use of a three-dimensional inverse design method. The blade geometry is described by means of aerodynamic parameters, like the blade loading, which are closely related to the aerodynamic performance to be optimized, thus leading to a simple shape of the optimization function. On the basis of this consideration, it is possible to use simple approximation functions for describing the correlations between the input design parameters and the performance ones. The Response Surface Methodology coupled with the Design of Experiments (DOE) technique was used for this purpose. CFD analyses were run to evaluate the configurations required by the DOE to generate the database. Optimization algorithms were then applied to the approximated functions in order to determine the optimal configuration or the set of optimal ones (Pareto front). The method was applied for the aerodynamic redesign of two different turbomachine components: a centrifugal compressor stage and a single-stage axial compressor. In both cases, both design and off-design operating conditions were analyzed and optimized.

A Geometric Design Model for the Circolimaçon Positive Displacement Machine

2008 ◽  
Vol 130 (6) ◽  
Author(s):  
Ibrahim A. Sultan
Keyword(s):  
Case Studies ◽  
Design Process ◽  
Geometric Parameters ◽  
Design Model ◽  
Inverse Design ◽  
Design Parameters ◽  
Design Models ◽  
Positive Displacement ◽  
Circular Arcs ◽  
Design Requirements

A circolimaçon positive displacement machine is driven by a limaçon mechanism, but the profiles of its rotor and housing are circular arcs. As such, its design models are different from those of the limaçon-to-limaçon machines, whose profiles are cut to the limaçon equations. For the benefit of the reader, the paper starts with a brief background on the general geometric aspects of the limaçon fluid processing technology. However, the focus is then turned to the circolimaçon machine, where its design parameters are introduced and geometric models are proposed to assist with the design process. Also, a computational inverse design model has been employed to work out a set of congruent geometric parameters to meet certain design requirements. Case studies are presented at the end of the paper to give the reader a numerical perspective on the design process of this class of positive displacement machines.

scholarly journals Three-Dimensional Inverse Design of Low Specific Speed Turbine for Energy Recovery in Cooling Tower System

Energies ◽  
2018 ◽  
Vol 11 (12) ◽  
pp. 3348
Author(s):  
Wei Yang ◽  
Xiaoyu Lei ◽  
Benqing Liu
Keyword(s):  
Response Surface ◽  
Three Dimensional ◽  
Inverse Design ◽  
Design Parameters ◽  
Design Criteria ◽  
Blade Loading ◽  
Turbine Runner ◽  
Lean Angle ◽  
Low Specific Speed ◽  
Specific Speed

A three-dimensional inverse design of a low specific speed turbine is studied, and a set of design criteria for low specific speed turbine runner is proposed, including blade loading distributions and blade lean angles. The characteristics of the loading parameters for low specific speed turbine runner are summarized by analyzing the suction performance of different loading positions, loading slopes and blade lean angles based on the orthogonal experiment design and range analysis. It is found that the blade loading distribution at the band plays a more important role than it does at the crown and it should be fore loaded for both band and crown. The blade lean angle at the blade leading edge should be negative. Then, the blade is optimized through the inverse method by fixing blade lean angle, based on the response surface method. After seeking the optimal value of the response surface function, the optimal result of the design parameters is obtained, which is in conformity with the design criteria and verifies the rationality of the established design criteria for low specific speed turbine.

A Fast 3D Inverse Design Based Multi-Objective Optimization Strategy for Design of Pumps

2009 ◽  
Author(s):  
M. Zangeneh ◽  
K. Daneshkhah
Keyword(s):  
Pareto Front ◽  
Design Method ◽  
Leading Edge ◽  
Inverse Design ◽  
Design Parameters ◽  
Optimization Strategy ◽  
Multi Objective ◽  
Multi Objective Genetic Algorithm ◽  
Pump Stage ◽  
Meridional Shape

A methodology for designing pumps to meet multi-objective design criteria is presented. The method combines a 3D inviscid inverse design method with a multi-objective genetic algorithm to design pumps which meet various aerodynamic and geometrical requirements. The parameterization of the blade shape through the blade loading enables 3D optimization with very few design parameters. A generic pump stage is used to demonstrate the proposed methodology. The main design objectives are improving cavitation performance and reducing leading edge sweep. The optimization is performed subject to certain constraints on Euler head, throat area, thickness and meridional shape so that the resulting pump can meet both design and off-design conditions. A Pareto Front is generated for the two objective functions and 3 different configurations on the Pareto front are selected for detailed study by 3D RANS code. The CFD results confirm the main outcomes of the optimization process.

Redesign of a Transonic Compressor Rotor by Means of a Three-Dimensional Inverse Design Method: A Parametric Study

2007 ◽  
Author(s):  
Duccio Bonaiuti ◽  
Abeetha Pitigala ◽  
Mehrdad Zangeneh ◽  
Yansheng Li
Keyword(s):  
Design Method ◽  
Thickness Distribution ◽  
Three Dimensional ◽  
Tip Clearance ◽  
Inverse Design ◽  
Design Parameters ◽  
Geometrical Parameters ◽  
Blade Loading ◽  
Inverse Design Method ◽  
The Impact

In the present paper, the redesign of a transonic rotor was performed by means of a three-dimensional viscous inverse design method. The inverse approach used in this work is one where the pressure loading, blade thickness distribution and stacking axis are specified and the camber surface is calculated accordingly. The design of transonic and supersonic axial compressors strongly relies on the ability to control the shock strength, location and structure. The use of an inverse design method allows one to act directly on aerodynamic parameters, like the blade loading, and provides an efficient tool to control the shock wave and its interaction with the boundary and secondary flows and with the tip clearance vortex. In the present study, the parametric investigation of the blade loading distribution was carried out. Few design parameters, with immediate physical meaning, were required to control the three-dimensional blade loading, and their impact on the design and off-design performance of the rotor was assessed by means of CFD calculations. Further investigations were then performed in order to study the impact on the rotor performance of the geometrical parameters (meridional channel and thickness distribution), which must be imposed in the design with the inverse method. As a result, it was possible to develop guidelines for the aerodynamic design of transonic rotors that can be exploited for similar design applications.

3D Multi-Disciplinary Inverse Design Based Optimization of a Centrifugal Compressor Impeller

2014 ◽  
Author(s):  
Mehrdad Zangeneh ◽  
Fred Mendonça ◽  
Youngwon Hahn ◽  
Jack Cofer
Keyword(s):  
Response Surface ◽  
Centrifugal Compressor ◽  
Inverse Design ◽  
Design Parameters ◽  
Operating Range ◽  
Multi Objective ◽  
Development Times ◽  
Stable Operating ◽  
Compressor Impeller ◽  
3D Fea

Design of centrifugal compressors in different applications from industrial to turbochargers to aeroengine is subject to difficult multi-disciplinary ( aerodynamics and mechanical) and multipoint/multi-objective requirements. These multi-disciplinary and multi-point requirements have to be met by iterations between aerodynamics and mechanical design, leading to long development times and bottlenecks in the design process. In this paper, for the first time, a commercially available solution, compatible with industrial development times, is presented for 3D multi-disciplinary and multi-point design optimisation of turbomachinery blades. The methodology combines 3D inverse design method, automatic optimizers, 3D CFD and 3D FEA codes. The key aspect of the approach is to parameterise the 3D geometry through the blade loading distribution used in 3D inverse design code TURBOdesign1, which results in ability to access large part of design space with very few design parameters. The Design of Experiments method is used to generate a number of geometries which are then analysed by 3D CFD code STAR-CCM+ and 3D FEA code Abaqus. Different performance parameters related to aerodynamics (efficiency, stable operating range etc) and structural integrity (maximum principal stress, etc) are then evaluated. The data is then used to create a response surface. The validity and accuracy of the response surface is evaluated by CFD and FEA and then once confirmed a Multi-objective Genetic Algorithm is run on the response surface to explore the trade-offs between different design parameters, such as peak efficiency, stable operating range and mechanical stress. In this paper the methodology is applied to the redesign of the well-known Eckardt centrifugal compressor impeller.

Sign in / Sign up

Export Citation Format

Share Document