ModeFRONTIER

From Wikipedia, the free encyclopedia
Jump to: navigation, search
modeFRONTIER
Developer(s) ESTECO s.p.a
Stable release 4.5 / September, 2013
Development status Active
Operating system Cross-platform
Type Technical computing
License Proprietary
Website [1]

modeFRONTIER is a multi-objective optimization and design environment, written to couple CAD/computer aided engineering (CAE) tools, finite element structural analysis and computational fluid dynamics (CFD) software. It is developed by 'ESTECO SpA [2]' and provides an environment for product engineers and designers. modeFRONTIER is a GUI driven software written in Java that wraps around the CAE tool, performing the optimization by modifying the value assigned to the input variables, and analyzing the outputs as they can be defined as objectives and/or constraints of the design problem.

History[edit]

ESTECO was created in 1999 to transfer the knowledge acquired by its founders while working on a European Union sponsored project on Design Optimization (FRONTIER, started in 1996) into a commercial product, called modeFRONTIER. In 2001, modeFRONTIER version 2.4 become a global player among the MDO/PIDO tools, being one of the first to enable true multi-objective optimization through Pareto dominance criteria.

Process integration[edit]

The logic of the optimization loop can be set up in a graphical way, building up a "workflow" structure by means of interconnected nodes. Serial and parallel connections and conditional switches are available. modeFRONTIER builds automatic chains and steers many different external application programs using scripting (DOS script, UNIX shell, Python programming language, Visual Basic, JavaScript,etc...) and direct integration nodes (with many CAE/CAD and other application programs).

Design optimization[edit]

modeFRONTIER includes design of experiments (DOE), optimization algorithms and robust design tools, that can be combined and blended to build up the most efficient strategy to solve complex multi-disciplinary problems.

Design of experiments[edit]

Different strategies are available, including random generator sequences, Factorial DOEs, Orthogonal and Iterative Techniques, as like as D-Optimal or Cross Validation. Monte Carlo and Latin hypercube are available for robustness analysis.

Multi objective algorithms[edit]

Among the others, different implementations of heuristic optimization methods such as the Genetic Algorithm, Game Theory, Simulated Annealing, Evolution strategies are able to manage continuous, discrete and mixed variable problems.[1] More classical mono-objective algorithms are as well available, as like as Gradient-based methods or Simplex algorithm. Three multi-strategy methods were also added: the FAST algorithm (using the Response surface methodology to accelerate the iterations to reach the optimum),[2] the Hybrid method[3] (combining global search and local refinement) and the SAnGeA [4](adding an automatic screening phase to manage high-dimension and unconstrained problems).[5]

Response surfaces[edit]

Different Response surface methodology techniques are available to generate reliable meta-models able to approximate the multivariate input/output behavior of complex systems; this is particularly useful when the optimization applies to problems where every fitness function evaluation is time-expensive.[6]Single Value Decomposition and Polynomial Responses are implemented, as well as the more sophisticated Kriging, Neural Network and Gaussian process ones.

Data processing and multiple criteria decision making (MCDM)[edit]

This set of tools enables the user to explore, filter and rank the set of optimal solutions of a multi-objective problem (the so-called Pareto frontier), to perform sensitivity analyses, robustness verifications and also to produce standard and customizable reports of the optimization project (RTF, PDF, HTML formats). The MCDM algorithms enable what-if analysis by providing runtime a ranking of design also when importance of attributes is not made explicit or the probability associated with alternatives cannot be calculated by the decision maker.[7]

Robust design optimization[edit]

This is the latest step of MDO towards 6 Sigma: optimizing a design taking into account uncertainties and tolerances. This consists of investigating the noise factors in the neighborhood of a sample design with a given probability distribution through a multi-objective optimization algorithm aimed at optimizing mean values while minimizing their variations.[8] RSM techniques can be used to overcome the increase in time-expense due to this extensive statistic approach.

External links[edit]

References[edit]

  1. ^ "Optimization Algorithms in modeFRONTIER". 
  2. ^ Rezou, M.; Meouche, R.E., Hamzaoui,R., Feng, Z.-Q. (June 2013). "Using the Fast Multi-Objective Genetic Algorithm to Improve the Urban Flood Modeling". IACSIT International Journal of Engineering and Technology, Vol. 5, No. 3. 
  3. ^ "Using mode FRONTIER to integrate GT-Valve train and GT-Power models for valve event optimization". 
  4. ^ "Morphological Shape Optimization of a Three Dimensional Piston Bowl using modeFRONTIER and Sculptor". 
  5. ^ Radhi, H.E.; Barrans, S.M. "Comparison between Multiobjective Population-Based Algorithms in Mechanical Problem". Applied Mechanics and Materials (Volumes 110 - 116). 
  6. ^ "Response Surface Methodology (RSM) in modeFRONTIER". 
  7. ^ "Decision making in modeFRONTIER". 
  8. ^ "Multi-objective Robust Design Optimization with modeFRONTIER".