Local Latin hypercube refinement for multi-objective design uncertainty optimization
نویسندگان
چکیده
Optimizing the reliability and robustness of a design is important but often unaffordable due to high sample requirements. Surrogate models based on statistical machine learning methods are used increase efficiency. However, for higher dimensional or multi-modal systems, surrogate may also require large amount samples achieve good results. We propose sequential sampling strategy solution multi-objective robust optimization problems. Proposed local Latin hypercube refinement (LoLHR) model-agnostic can be combined with any model because there no free lunch possibly budget one. The proposed method compared stationary as well other strategies from literature. Gaussian process support vector regression both models. Empirical evidence presented, showing that LoLHR achieves average better results tested examples.
منابع مشابه
Uncertainty Quantification Based Multi- Objective Optimization for Crashworthiness Design
This paper presents a methodology for uncertainty quantification based multi-objective optimization of automotive body components under impact scenario. Conflicting design requirements arise as one tries, for example, to minimize structural mass while maximizing energy absorption of an automotive rail section under structural and occupant safety related performance measure constraints. Uncertai...
متن کاملMulti-objective Optimization Approach for Robust Design under Uncertainty
This paper presents a new strategy for synthesizing an appropriate process with multi-objective under uncertainty. The uncertainty is classified depending on its sources and mathematical model structure as deterministic or stochastic. The proposed methodology is a two-layer algorithm. In the outer layer, the synthesis problem is represented by a multi-objective optimization problem considering ...
متن کاملMulti Objective Optimization under Uncertainty for Catamaran Preliminary Design
The decision tools for high speed craft HSC design have improved due to the increasing computer ability to treat more complex problems. Genetic algorithms and multi objective problems where the mathematical models used are complex and take computer time to evaluate and select the feasible designs are very common nowadays. Uncertainty also is very frequent in high speed vessel design variables a...
متن کاملMulti-Objective Optimization for Multi-Product Multi-Period Four Echelon Supply Chain Problems Under Uncertainty
The multi-objective optimization for a multi-product multi-period four-echelon supply chain network consisting of manufacturing plants, distribution centers (DCs) and retailers each with uncertain services and uncertain customer nodes are aimed in this paper. The two objectives are minimization of the total supply chain cost and maximization of the average number of products dispatched to custo...
متن کاملHierarchical Refinement of Latin Hypercube Samples
In this article, a novel method for the extension of sample size in Latin Hypercube Sampling (LHS) is suggested. The method can be applied when an initial LH design is employed for the analysis of functions g of a random vector. The article explains how the statistical, sensitivity and reliability analyses of g can be divided into a hierarchical sequence of simulations with subsets of samples o...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Applied Soft Computing
سال: 2021
ISSN: ['1568-4946', '1872-9681']
DOI: https://doi.org/10.1016/j.asoc.2021.107807