4.7 Article

Deep-learning-based reduced-order modeling to optimize recuperative burner operating conditions

相关参考文献

注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article Nanoscience & Nanotechnology

A new fluid flow approximation method using a vision transformer and a U-shaped convolutional neural network

Hyoeun Kang et al.

Summary: Numerical simulation of fluids is crucial in modeling various physical phenomena, and solving the Navier-Stokes equations at a large scale can be computationally expensive. This paper presents CFDformer, a deep learning-based surrogate model that improves fluid flow approximations in computational fluid dynamics. CFDformer performs well under different velocities and pressures, outperforming other baseline models such as U-Net and TransUNet.

AIP ADVANCES (2023)

Article Engineering, Multidisciplinary

A non-linear non-intrusive reduced order model of fluid flow by auto-encoder and self-attention deep learning methods

R. Fu et al.

Summary: This paper presents a new NL-NIROM that outperforms the traditional POD-based ROM by using AE and self-attention based deep learning methods. It introduces a new model reduction neural network architecture for fluid flow problems and a L-NIROM based on POD and self-attention mechanism. In the NL-NIROM, the SAE network compresses high-dimensional physical information into smaller sized representations and uses self-attention based deep learning methods to predict fluid dynamics.

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING (2023)

Article Engineering, Multidisciplinary

POD-DL-ROM: Enhancing deep learning-based reduced order models for nonlinear parametrized PDEs by proper orthogonal decomposition

Stefania Fresca et al.

Summary: DL-ROMs are proposed to overcome limitations of conventional ROMs, but require expensive training. The proposed method combines POD and multi-fidelity pretraining to avoid the costly training stage of DL-ROMs.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2022)

Article Computer Science, Artificial Intelligence

Investigation of NOx emission under different burner structures with the optimized combustion model

Qian Yao et al.

Summary: As restrictions on NOx emission become stricter, the development of control strategies has been a focus. In this study, an efficient optimization procedure using a surrogate model and evolutionary algorithm is proposed to optimize combustion model parameters, leading to consistent temperature and concentration fields with measurements. It is found that the length of pilot wall and jet wall has an impact on NOx emission, with the optimal condition achieved when their lengths are equal and further reduction seen when the pilot wall is 30 mm longer than the jet wall.

NEUROCOMPUTING (2022)

Article Thermodynamics

Application of an automated machine learning-genetic algorithm (AutoML-GA) coupled with computational fluid dynamics simulations for rapid engine design optimization

Opeoluwa Owoyele et al.

Summary: AutoML-GA is an automated active learning approach for surrogate-based optimization of internal combustion engines. It utilizes Bayesian optimization technique and genetic algorithm to find the best machine learning hyperparameters, leading to a better optimum with a lower number of CFD simulations. This framework offers a hands-off approach that can be readily utilized by researchers and engineers without extensive machine learning expertise.

INTERNATIONAL JOURNAL OF ENGINE RESEARCH (2022)

Article Mechanics

A non-intrusive reduced order model with transformer neural network and its application

Pin Wu et al.

Summary: This paper proposes a novel method to construct a non-intrusive reduced order model (ROM) using proper orthogonal decomposition and transformer neural network. The method effectively captures flow dynamics and reduces online calculation time. Experimental results demonstrate that the proposed model outperforms other ROMs in capturing flow details and has higher accuracy.

PHYSICS OF FLUIDS (2022)

Article Energy & Fuels

Development of an Efficient Modelling Approach for Fin-Type Heat-Exchangers in Self-Recuperative Burners

Nicolas Dinsing et al.

Summary: Self-recuperative burners are a common solution for efficient combustion in industrial furnaces, but detailed CFD simulations of the recuperators are computationally expensive. This paper presents an innovative approach to model a fin-type recuperator, reducing geometric complexity and enabling accurate replication of flow and temperature profiles in the recuperator.

ENERGIES (2021)

Article Mathematical & Computational Biology

Binary genetic algorithm for optimal joinpoint detection: Application to cancer trend analysis

Seongyoon Kim et al.

Summary: The genetic algorithm-based joinpoint model demonstrated efficient detection of multiple joinpoints without the need to preassign the number of joinpoints, making it suitable for cancer trend analysis.

STATISTICS IN MEDICINE (2021)

Article Energy & Fuels

A novel machine learning-based optimization algorithm (ActivO) for accelerating simulation-driven engine design

Opeoluwa Owoyele et al.

Summary: A novel design optimization approach (ActivO) utilizing an ensemble of machine learning algorithms was proposed, where a weak learner and a strong learner are used within an active learning loop to identify promising regions and determine the exact location of the optimum. Comparison with other optimizers showed that ActivO reduced the number of function evaluations and time-to-design by 80%. Energy consumption savings of around 1.9% were achieved while maintaining operability and reducing pollutant emissions in internal combustion engine optimization.

APPLIED ENERGY (2021)

Article Mathematics, Applied

Non-intrusive framework of reduced-order modeling based on proper orthogonal decomposition and polynomial chaos expansion

Xiang Sun et al.

Summary: The proposed method utilizes POD and PCE to construct an efficient stochastic representation model through non-intrusive methods, significantly reducing computational costs and storage requirements for high-dimensional physical and random spaces, while demonstrating similar accuracy in predicting statistical quantities as classical sparse PCE.

JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS (2021)

Article Engineering, Mechanical

CFD Optimization of the Pre-Chamber Geometry for a Gasoline Spark Ignition Engine

Haiwen Ge et al.

Summary: In this study, an efficient optimization method was developed based on Bayesian updating strategy for a spark-ignition engine equipped with pre-chamber. Utilizing 3D computational fluid dynamics simulation and machine learning methods, the pre-chamber was optimized to achieve the desired combustion phasing. The study identified the radius of the pre-chamber as the most influential design parameter for engine performance through multiple iterations and combustion analysis.

FRONTIERS IN MECHANICAL ENGINEERING-SWITZERLAND (2021)

Article Green & Sustainable Science & Technology

Advanced design optimization of combustion equipment for biomass combustion

Joseph D. Smith et al.

RENEWABLE ENERGY (2020)

Article Computer Science, Interdisciplinary Applications

Non-intrusive reduced order modeling of unsteady flows using artificial neural networks with application to a combustion problem

Qian Wang et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2019)

Article Engineering, Mechanical

Evaluating Optimization Strategies for Engine Simulations Using Machine Learning Emulators

Daniel M. Probst et al.

JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER-TRANSACTIONS OF THE ASME (2019)

Article Engineering, Mechanical

Uncertainty Quantification of NOx Emission Due to Operating Conditions and Chemical Kinetic Parameters in a Premixed Burner

Sajjad Yousefian et al.

JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER-TRANSACTIONS OF THE ASME (2018)

Article Engineering, Chemical

A Computational Fluid Dynamics Study of Turbulence, Radiation, and Combustion Models for Natural Gas Combustion Burner

Yik Siang Pang et al.

BULLETIN OF CHEMICAL REACTION ENGINEERING AND CATALYSIS (2018)

Article Environmental Sciences

Numerical analysis of an entire ceramic kiln under actual operating conditions for the energy efficiency improvement

Massimo Milani et al.

JOURNAL OF ENVIRONMENTAL MANAGEMENT (2017)

Article Energy & Fuels

Heat recovery potentials and technologies in industrial zones

Feng Huang et al.

JOURNAL OF THE ENERGY INSTITUTE (2017)

Article Environmental Sciences

NO formation analysis of turbulent non-premixed coaxial methane/air diffusion flame

S. Poozesh et al.

INTERNATIONAL JOURNAL OF ENVIRONMENTAL SCIENCE AND TECHNOLOGY (2016)

Article Thermodynamics

Effects of furnace chamber shape on the MILD combustion of natural gas

Yaojie Tu et al.

APPLIED THERMAL ENGINEERING (2015)

Article Thermodynamics

Optimization of a finned concentric pipes heat exchanger for industrial recuperative burners

Marco Cavazzuti et al.

APPLIED THERMAL ENGINEERING (2015)

Article Thermodynamics

Energy saving in ceramic tile kilns: Cooling gas heat recovery

Ana Mezquita et al.

APPLIED THERMAL ENGINEERING (2014)

Article Thermodynamics

Optimization of gas mixing system of premixed burner based on CFD analysis

Tian-Hu Zhang et al.

ENERGY CONVERSION AND MANAGEMENT (2014)

Article Engineering, Mechanical

CFD analysis of a syngas-fired burner for ceramic industrial roller kiln

Marco Cavazzuti et al.

PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART C-JOURNAL OF MECHANICAL ENGINEERING SCIENCE (2013)

Article Biology

Nested Latin hypercube designs

Peter Z. G. Qian

BIOMETRIKA (2009)

Article Computer Science, Interdisciplinary Applications

A new procedure for predicting NOx emissions from furnaces

T Faravelli et al.

COMPUTERS & CHEMICAL ENGINEERING (2001)