Model Validation Methods for Multiple Correlated Responses via Covariance-Overlap Based Distance

2019 ◽  
Vol 142 (4) ◽  
Author(s):  
Jiexiang Hu ◽  
Ping Jiang ◽  
Qi Zhou ◽  
Austin McKeand ◽  
Seung-Kyum Choi
Keyword(s):  
Engineering Design ◽  
Model Validation ◽  
Computational Models ◽  
Engineering Problem ◽  
Cumulative Probability ◽  
Global Evaluation ◽  
Correlated Responses ◽  
Validation Methods ◽  
Engineering Design Problems ◽  
Practical Engineering

Abstract Model validation methods have been widely used in engineering design to provide a quantified assessment of the agreement between simulation predictions and experimental observations. For the validation of simulation models with multiple correlated outputs, not only the uncertainty of the responses but also the correlation between them needs to be considered. Most of the existing validation methods for multiple correlated responses focus on the area metric, which only compares the overall area difference between the two cumulative probability distribution curves. The differences in the distributions of the data sets are not fully utilized. In this paper, two covariance-overlap based model validation (COMV) methods are proposed for the validation of multiple correlated responses. The COMV method is used for a single validation site, while the covariance-overlap pooling based model validation (COPMV) method can pool the evidence from different validation sites into a scalar measure to give a global evaluation about the candidate model. The effectiveness and merits of the proposed methods are demonstrated by comparing with three different existing validation methods on three numerical examples and a practical engineering problem of a turbine blade validation example. The influence of sample size and the number of partitions in the proposed methods are also discussed. Results show that the proposed method shows better performance on the uncertainty estimation of different computational models, which is useful for practical engineering design problems with multiple correlated responses.

Model Validation via Uncertainty Propagation Using Response Surface Models

2002 ◽  
Author(s):  
Lusine Baghdasaryan ◽  
Wei Chen ◽  
Thaweepat Buranathiti ◽  
Jian Cao
Keyword(s):  
Mathematical Models ◽  
Engineering Design ◽  
Response Surface ◽  
Model Validation ◽  
Computational Models ◽  
Uncertainty Propagation ◽  
Generic Model ◽  
Analysis Model ◽  
Element Analysis ◽  
Engineering Design Problems

Model validation has become a primary means to evaluate accuracy and reliability of computational simulations in engineering design. Mathematical models enable engineers to establish what the most likely response of a system is. However, despite the enormous power of computational models, uncertainty is inevitable in all model-based engineering design problems, due to the variation in the physical system itself, or lack of knowledge, and the use of assumptions by model builders. Therefore, realistic mathematical models should contemplate uncertainties. Due to the uncertainties, the assessment of the validity of a modeling approach must be conducted based on stochastic measurements to provide designers with the confidence of using a model. In this paper, a generic model validation methodology via uncertainty propagation is presented. The approach reduces the number of physical testing at each design setting to one by shifting the evaluation effort to uncertainty propagation of the computational model. Response surface methodology is used to create metamodels as less costly approximations of simulation models for uncertainty propagation. The methodology is illustrated with the examination of the validity of a finite-element analysis model for predicting springback angles in a sample flanging process.

scholarly journals Multipopulation Ensemble Particle Swarm Optimizer for Engineering Design Problems

2020 ◽  
Vol 2020 ◽  
pp. 1-30
Author(s):  
Ziang Liu ◽  
Tatsushi Nishi
Keyword(s):  
Engineering Design ◽  
Real World ◽  
Particle Swarm ◽  
Engineering Problem ◽  
Particle Swarm Optimizer ◽  
Design Problems ◽  
Source Codes ◽  
Searching Strategy ◽  
Engineering Design Problems ◽  
Improved Function

Particle swarm optimization (PSO) is an efficient optimization algorithm and has been applied to solve various real-world problems. However, the performance of PSO on a specific problem highly depends on the velocity updating strategy. For a real-world engineering problem, the function landscapes are usually very complex and problem-specific knowledge is sometimes unavailable. To respond to this challenge, we propose a multipopulation ensemble particle swarm optimizer (MPEPSO). The proposed algorithm consists of three existing efficient and simple PSO searching strategies. The particles are divided into four subpopulations including three indicator subpopulations and one reward subpopulation. Particles in the three indicator subpopulations update their velocities by different strategies. During every learning period, the improved function values of the three strategies are recorded. At the end of a learning period, the reward subpopulation is allocated to the best-performed strategy. Therefore, the appropriate PSO searching strategy can have more computational expense. The performance of MPEPSO is evaluated by the CEC 2014 test suite and compared with six other efficient PSO variants. These results suggest that MPEPSO ranks the first among these algorithms. Moreover, MPEPSO is applied to solve four engineering design problems. The results show the advantages of MPEPSO. The MATLAB source codes of MPEPSO are available at https://github.com/zi-ang-liu/MPEPSO.

Wisdom of Microcrowds in Evaluating Solutions to Esoteric Engineering Problems

2019 ◽  
Vol 141 (8) ◽  
Author(s):  
Nurcan Gecer Ulu ◽  
Michael Messersmith ◽  
Kosa Goucher-Lambert ◽  
Jonathan Cagan ◽  
Levent Burak Kara
Keyword(s):  
Social Sciences ◽  
Computational Models ◽  
Engineering Problem ◽  
Wisdom Of Crowds ◽  
Specialized Knowledge ◽  
Main Hypothesis ◽  
Design Problems ◽  
Domain Experts ◽  
Engineering Design Problems ◽  
Engineering Problems

A multitude of studies in economics, psychology, political and social sciences have demonstrated the wisdom of crowds (WoC) phenomenon, where the collective estimate of a group can be more accurate than estimates of individuals. While WoC is observable in such domains where the participating individuals have experience or familiarity with the question at hand, it remains unclear how effective WoC is for domains that traditionally require deep expertise or sophisticated computational models to estimate objective answers. This work explores how effective WoC is for engineering design problems that are esoteric in nature, that is, problems (1) whose solutions traditionally require expertise and specialized knowledge, (2) where access to experts can be costly or infeasible, and (3) in which previous WoC studies with the general population have been shown to be highly ineffective. The main hypothesis in this work is that in the absence of experts, WoC can be observed in groups that consist of practitioners who are defined to have a base familiarity with the problems in question but not necessarily domain experts. As a way to emulate commonly encountered engineering problem-solving scenarios, this work studies WoC with practitioners that form microcrowds consisting of 5–15 individuals, thereby giving rise to the term the wisdom of microcrowds (WoMC). Our studies on design evaluations show that WoMC produces results whose mean is in the 80th percentile or better across varying crowd sizes, even for problems that are highly nonintuitive in nature.

A unified first-year engineering design-based learning course

2016 ◽  
Vol 45 (1) ◽  
pp. 47-58 ◽  
Author(s):  
Saad Odeh ◽  
Shauna McKenna ◽  
Hosni Abu-Mulaweh
Keyword(s):  
Engineering Design ◽  
Real Life ◽  
Engineering Problem ◽  
First Year ◽  
Design Problems ◽  
First Semester ◽  
Group Assignment ◽  
Engineering Design Problems ◽  
Workforce Skills ◽  
Apply Physical

This paper describes an innovative engineering design of a first-year engineering course. The course is offered in the second semester of the academic year to students of different engineering disciplines such as mechanical, mechatronic, electrical, electronics, civil, environmental and manufacturing. The course incorporates a mix of techniques to help students better engage with the subject matter and with one another. A major part of the new course is the practical assessment component requiring students to apply physical, mathematical, mechanical and electrical concepts to real life engineering design problems. Three different engineering design modules were developed. Each module consists of an authentic engineering design problem which has been specially constructed in order to provide students with the opportunity to apply the basic engineering, maths and physics concepts they acquired during the first semester. Depending on the students intended engineering major, they choose one of the three engineering design modules. In order to best prepare students for the design project, they firstly do two small group assignment tasks on a particular engineering problem. This serves as the preparatory work for the engineering design module. The assignments are done in class time so as to promote full collaboration between students and instructors and to encourage the exchange of knowledge and ideas. The course aims to better equip students with workforce skills in problem solving and effective oral and written communication.

A study in applying case-based reasoning to engineering design: Mechanical bearing design

2003 ◽  
Vol 17 (3) ◽  
pp. 235-252 ◽  
Author(s):  
XIAOLI QIN ◽  
WILLIAM C. REGLI
Keyword(s):  
Engineering Design ◽  
Constraint Satisfaction ◽  
Engineering Problem ◽  
Problem Definition ◽  
Case Based Reasoning ◽  
Design Problems ◽  
Engineering Design Problems ◽  
Previous Design ◽  
Bearing Design ◽  
Case Based

Case-based reasoning (CBR) is a promising methodology for solving many complex engineering design problems. CBR employs past problem-solving experiences when solving new problems. This paper presents a case study of how to apply CBR to a specific engineering problem: mechanical bearing design. A system is developed that retrieves previous design cases from a case repository and uses adaptation techniques to modify them to satisfy the current problem requirements. The approach combines both parametric and constraint satisfaction adaptations. Parametric adaptation considers not only parameter substitution but also the interrelationships between the problem definition and its solution. Constraint satisfaction provides a method to globally check the design requirements to assess case adaptability. Currently, our system has been implemented and tested in the domain of rolling bearings. This work serves as a template for application of CBR techniques to realistic engineering problems.

Standards, Codes of Practice and Expert Systems

1988 ◽  
Vol 21 (1) ◽  
pp. 5-9 ◽  
Author(s):  
E G McCluskey ◽  
S Thompson ◽  
D M G McSherry
Keyword(s):  
Expert System ◽  
Expert Systems ◽  
Engineering Design ◽  
Performance Criteria ◽  
Design Problems ◽  
Codes Of Practice ◽  
Engineering Design Problems ◽  
And Performance

Many engineering design problems require reference to standards or codes of practice to ensure that acceptable safety and performance criteria are met. Extracting relevant data from such documents can, however, be a problem for the unfamiliar user. The use of expert systems to guide the retrieval of information from standards and codes of practice is proposed as a means of alleviating this problem. Following a brief introduction to expert system techniques, a tool developed by the authors for building expert system guides to standards and codes of practice is described. The steps involved in encoding the knowledge contained in an arbitrarily chosen standard are illustrated. Finally, a typical consultation illustrates the use of the expert system guide to the standard.

A Robust Error-Pursuing Sequential Sampling Approach for Global Metamodeling Based on Voronoi Diagram and Cross Validation

2014 ◽  
Vol 136 (7) ◽  
Author(s):  
Shengli Xu ◽  
Haitao Liu ◽  
Xiaofang Wang ◽  
Xiaomo Jiang
Keyword(s):  
Engineering Design ◽  
Voronoi Diagram ◽  
Cross Validation ◽  
Sequential Sampling ◽  
Voronoi Cell ◽  
Sensitive Region ◽  
Engineering Design Problems ◽  
Simulation Based ◽  
Global Metamodeling ◽  
Sampling Approach

Surrogate models are widely used in simulation-based engineering design and optimization to save the computing cost. The choice of sampling approach has a great impact on the metamodel accuracy. This article presents a robust error-pursuing sequential sampling approach called cross-validation (CV)-Voronoi for global metamodeling. During the sampling process, CV-Voronoi uses Voronoi diagram to partition the design space into a set of Voronoi cells according to existing points. The error behavior of each cell is estimated by leave-one-out (LOO) cross-validation approach. Large prediction error indicates that the constructed metamodel in this Voronoi cell has not been fitted well and, thus, new points should be sampled in this cell. In order to rapidly improve the metamodel accuracy, the proposed approach samples a Voronoi cell with the largest error value, which is marked as a sensitive region. The sampling approach exploits locally by the identification of sensitive region and explores globally with the shift of sensitive region. Comparative results with several sequential sampling approaches have demonstrated that the proposed approach is simple, robust, and achieves the desired metamodel accuracy with fewer samples, that is needed in simulation-based engineering design problems.

Sign in / Sign up

Export Citation Format

Share Document