ONLINE TRAINING

modeFRONTIER Metamodeling (RSM – Response Surface Modeling)

Go to Italian version | English version

Course Code: online-mf7-en

Training on demand: please contact us!

Send an information request

 

We use a web platform, which does not require local software installation.
You can participate in the session through: MAC, PC or any mobile device. The user will receive the link and the credentials to participate from the Training Secretariat.

Subscription fee

The subscription fee amounts to:
1800 Euro per person


The course is provided only upon reaching the minimum number of participants. The participant will receive a confirmation email one week before the course starts.

Would you like further information on the on-demand courses, the inhouse options, the English language, special needs training, or do you have any specific requests or require any further information?

Request information

EnginSoft Training Secretariat
Barbara Manzoni
Phone +39 035 368 711
corsi@enginsoft.it

 

Level

Advanced

Duration

15 hours of online training + 6 hours of homework, reviewed together with the tutor.
6 modules in total, divided into 2 modules per day of 2h30’ each, over 3 days.

Language

Italian/English (slide in English)

Tutors

Vito Primavera

Description

The Metamodeling course, divided into 6 modules over three days, is aimed at modeFRONTIER users who wish to acquire in-depth knowledge in the use of Response Surface Modeling (RSM) algorithms.

Unlike the techniques studied in the basic and advanced courses, this course provides an extended and detailed study of the typical algorithms in each family (i.e. Classical, Statistical, Advanced, Machine Learning, Support Vector Machine). In addition to the theoretical explanation of the individual techniques, the user will be guided through a path that, following an approach to "separate variables", will allow him to understand the behavior of the algorithms with respect to a series of benchmark functions of increasing complexity. During this process, the related "best practices" will be discussed, not only for the construction and validation of the metamodels obtained, but also for the qualitative and quantitative analysis of the initial data.

The overall goal is to enable the user to apply RSM techniques in a very deliberate way, enabling a "reasoned" approach that complements the "autonomous" approach available in the platform.

The course, although online, will be made interactive through a series of hands-on exercises, to be done immediately after the theoretical explanation of the topics covered.

Target Audience

The course is aimed at those users (engineers of all specializations, designers, researchers) who need to:

  • have predictive models pertaining to the performance of the application of interest
  • replace laboratory tests with numerical activities using predictive models based on experimental tests
  • accelerate the process of optimization, especially for models that are prohibitively expensive with respect to calculation times
  • implement 1D models or models with focused parameters by using highly accurate "black boxes" to characterize some of the constituent elements, however

Pre-requisites

Basic theoretical proficiency in RSM techniques, basic knowledge of modeFRONTIER and the core techniques of statistical analysis.

Agenda

Module 1 – RSM Methodology (duration 2h 30’) – Day 1/3

  • RSM Methodology & RSM Tool
  • Interpolating algorithms

Module 2 – RSM Methodology (duration 2h 30’) – Day 1/3

  • Interpolating algorithms
  • Application examples with benchmark functions & hands-on exercises

Module 3 – RSM Methodology (duration 2h 30’) – Day 2/3

  • Approximation algorithms

Module 4 – RSM Methodology (duration 2h 30’) – Day 2/3

  • Approximation algorithms
  • Application examples with benchmark functions & hands-on exercises

Module 5 – RSM Methodology (duration 2h 30’) – Day 3/3

  • Machine Learning (H2O)
  • Application examples with benchmark functions & hands-on exercises

Module 6 – RSM Methodology (duration 2h 30’) – Day 3/3

  • Automated RSM creation - RSM Trainer Node
  • Application examples with benchmark functions & hands-on exercises

*Discussion of Best Practices will take place during Hands-on sessions.