This modeFRONTIER course provides a general introduction to modeFRONTIER, the multi-objective optimisation and design environment, produced by ESTECO Srl and distributed in Europe by EnginSoft SpA .
The modeFRONTIER course starts with a brief introduction on multi-variable and conflicting objective challenge solving in multi-disciplinary scenarios
After that, four main training modules are available:
Module 1 - Process integration and multi-objective design optimization
Duration: 3h
Objectives: learn how to integrate software models in an optimization loop using the available dedicated interfaces. Set up a multi-objective problem, and solve it going through the selection of the best strategy and result visualization.
Tutorial: optimize a Milk Box for reduced heat exchange and transportation cost: explore the “Process Integration” capabilities, integration through Dedicated Nodes, optimization campaign setup using Strategy Wizard, detect and plot the optima.
MODULE 2 - Advanced workflows, multi-disciplinary optimization, Universal Integration
Duration: 3h
Objectives: learn how to integrate third-party software through Universal Connection. Set up multi-disciplinary complex workflows, handling also vector variables, and solve it.
Tutorial: optimize a structure for weight and dynamic behaviour under maximum stress constraint, connecting the corresponding FEM model.
Practicing modeFRONTIER (guided exercise): Verify Process Integration on a multi-disciplinary optimization real case, both Direct and Universal Integration. Running optimization campaigns using Strategy Wizard. Detecting and plotting the optima.
MODULE 3 - Response Surface Modelling basics
Duration: 4h
Objectives: learn to apply Response Surface Modelling (RSM) to fit, visualize and interpolate data; integrate them in a workflow to speed up optimization on long runtime models.
Tutorial: Milk Box example, showing how to explore the modeFRONTIER RSM theory, the quality judgement techniques, RSM visualization and integration with optimization.
Practicing modeFRONTIER (guided exercise): Verify your ability of running a “virtual optimization” based on a dataset, and propose a “next design to be tested” accordingly to new specifics.
Module 4: Design improvement strategies
Duration: 3h
Objectives: Present a selection of the available optimization algorithms classified by problems where they represent the most efficient approach. Algorithms presented include: NLPQLP – MIPSQP; Simplex – MOGT; MOGA-NSGA; Evolution Strategy; Robust Design; Multi-Variate-Analysis.
Tutorial: Design improvement: strategy exercise. Overview on optimization strategies by means of one example, including virtual optimization.
Practicing modeFRONTIER (guided exercise): Optimization strategy exercise (incl. virtual optimization).
This course is coming soon.
