:::

News & Events

:::

Speaker:Ph.D. Ching-Chi Yang (Pennsylvania State University, USA)

Seminar
Poster:Webmasters事件日期:2019-02-22
 
Topic:Dimensional Analysis for Response Surface Methodology

Speaker:Ph.D. Ching-Chi Yang (Pennsylvania State University, USA)

Date Time:FRI. FEB 22, 2019, 10:00 AM - 10:50 AM 
 
Place: 4F-427, Assembly Building I

Abstract

Response surface methodology (RSM) is widely used in chemistry and industrial processes. RSM mainly consists of experimental design, model building, and optimization. The objective is to obtain an experiment, in a small number of experiments, such that the response is optimal. Dimensional analysis (DA) is a well-developed methodology in physical sciences and engineering studies. DA can extract the essential information (such as geometric similarity and the scale factors) from the original variables. The extraction facilities experimental design as well as model fitting because the number of the essential variables is less than the number of original variables.
If a system/model is dimensionally homogeneous, the system/model can be re-expressed by its simplest form via dimensional analysis. The concept can also be applied in RSM. Our proposed procedure of employing DA for RSM is with theoretical supports. The proposed procedure extracts information from the original variables before conducting experiments. The design over the region in the DA space can obtain the essential information to approximate the real‐world model. Besides, the design in the DA space requires a smaller number of experiments than the design in the original space do. Moreover, the surrogate model follows universal laws, maintains the physical interpretability, and is dimensionally homogeneous. The proposed procedure, thus, is favorable to be used in obtaining the optimal response.
 
Last modification time:2019-02-22 AM 10:22

  • recruiting animation-EN
cron web_use_log