Transfer learning based variable-fidelity surrogate model for shell buckling prediction

2021 ◽  
pp. 114285
Author(s):  
Kuo Tian ◽  
Zengcong Li ◽  
Jiaxin Zhang ◽  
Lei Huang ◽  
Bo Wang
Keyword(s):  
Transfer Learning ◽  
Surrogate Model ◽  
Shell Buckling ◽  
Variable Fidelity

A Rational Design Approach to Gaussian Process Modeling for Variable Fidelity Models

2011 ◽  
Author(s):  
Roxanne A. Moore ◽  
David A. Romero ◽  
Christiaan J. J. Paredis
Keyword(s):  
Gaussian Process ◽  
Surrogate Model ◽  
Rational Design ◽  
Sequential Sampling ◽  
Accurate Determination ◽  
Surrogate Modeling ◽  
Sampling Strategy ◽  
New Approach ◽  
Variable Fidelity ◽  
The Cost

Computer models and simulations are essential system design tools that allow for improved decision making and cost reductions during all phases of the design process. However, the most accurate models tend to be computationally expensive and can therefore only be used sporadically. Consequently, designers are often forced to choose between exploring many design alternatives with less accurate, inexpensive models and evaluating fewer alternatives with the most accurate models. To achieve both broad exploration of the design space and accurate determination of the best alternatives, surrogate modeling and variable accuracy modeling are gaining in popularity. A surrogate model is a mathematically tractable approximation of a more expensive model based on a limited sampling of that model. Variable accuracy modeling involves a collection of different models of the same system with different accuracies and computational costs. We hypothesize that designers can determine the best solutions more efficiently using surrogate and variable accuracy models. This hypothesis is based on the observation that very poor solutions can be eliminated inexpensively by using only less accurate models. The most accurate models are then reserved for discerning the best solution from the set of good solutions. In this paper, a new approach for global optimization is introduced, which uses variable accuracy models in conjuction with a kriging surrogate model and a sequential sampling strategy based on a Value of Information (VOI) metric. There are two main contributions. The first is a novel surrogate modeling method that accommodates data from any number of different models of varying accuracy and cost. The proposed surrogate model is Gaussian process-based, much like classic kriging modeling approaches. However, in this new approach, the error between the model output and the unknown truth (the real world process) is explicitly accounted for. When variable accuracy data is used, the resulting response surface does not interpolate the data points but provides an approximate fit giving the most weight to the most accurate data. The second contribution is a new method for sequential sampling. Information from the current surrogate model is combined with the underlying variable accuracy models’ cost and accuracy to determine where best to sample next using the VOI metric. This metric is used to mathematically determine where next to sample and with which model. In this manner, the cost of further analysis is explicitly taken into account during the optimization process.

Multidisciplinary and Multifidelity Design Optimization of Electric Vehicle Battery Thermal Management System

2018 ◽  
Vol 140 (9) ◽  
Author(s):  
Xiaobang Wang ◽  
Yuanzhi Liu ◽  
Wei Sun ◽  
Xueguan Song ◽  
Jie Zhang
Keyword(s):  
Fluid Dynamics ◽  
Model Selection ◽  
Design Optimization ◽  
Thermal Management ◽  
Surrogate Model ◽  
Management System ◽  
Battery Thermal Management ◽  
Thermal Management System ◽  
Battery Thermal Management System ◽  
Variable Fidelity

Battery thermal management system (BTMS) is a complex and highly integrated system, which is used to control the battery thermal conditions in electric vehicles (EVs). The BTMS consists of many subsystems that belong to different disciplines, which poses challenges to BTMS optimization using conventional methods. This paper develops a general variable fidelity-based multidisciplinary design optimization (MDO) architecture and optimizes the BTMS by considering different systems/disciplines from the systemic perspective. Four subsystems and/or subdisciplines are modeled, including the battery thermodynamics, fluid dynamics, structure, and lifetime model. To perform the variable fidelity-based MDO of the BTMS, two computational fluid dynamics (CFD) models with different levels of fidelity are developed. A low fidelity surrogate model and a tuned low fidelity model are also developed using an automatic surrogate model selection method, the concurrent surrogate model selection (COSMOS). An adaptive model switching (AMS) method is utilized to realize the adaptive switch between variable-fidelity models. The objectives are to maximize the battery lifetime and to minimize the battery volume, the fan's power, and the temperature difference among different cells. The results show that the variable-fidelity MDO can balance the characteristics of the low fidelity mathematical models and the computationally expensive simulations, and find the optimal solutions efficiently and accurately.

scholarly journals Improving surrogate model accuracy for the LCLS-II injector frontend using convolutional neural networks and transfer learning

2021 ◽  
Vol 2 (4) ◽  
pp. 045025 ◽  
Author(s):  
Lipi Gupta ◽  
Auralee Edelen ◽  
Nicole Neveu ◽  
Aashwin Mishra ◽  
Christopher Mayes ◽  
...  
Keyword(s):  
Neural Networks ◽  
Transfer Learning ◽  
Convolutional Neural Networks ◽  
Surrogate Model ◽  
Model Accuracy
Sign in / Sign up

Export Citation Format

Share Document