modeFRONTIER PDF

 modeFRONTIER

 

modeFRONTIER is a multi-objective optimization and design environment, written to allow easy coupling to almost any computer aided engineering (CAE) tool, whether commercial or in-house. As the name suggests, modeFRONTIER provides an environment which allows product engineers and designers to integrate their various CAE tools, such as CAD, Finite Element Structural Analysis and Computational Fluid Dynamics (CFD) software.

 

Using a variety of state-of-the-art optimization techniques, ranging from gradient-based methods to genetic algorithms, the process or design of interest can be optimized by specifying objectives and defining variables which affect factors such as geometric shape and operating conditions. modeFRONTIER in effect becomes a wrapper around the CAE tool, performing the optimization by modifying the value assigned to the input variables, and monitoring the outputs.

 

modeFRONTIER can be used for products or process design giving to the design team a tool to rationalise the design process linking specialised simulation software available in the company or at sub-contractor location using intranet/internet network. It is able to exploit any parallel computing platform from PC networks to large mainframes.


System Features :
  • addresses multiple objective optimisations with both continuous and discrete variables
  • uses off the shelf design evaluation tools on whatever platforms are suitable
  • independent of any particular CAD system or other application package
  • global distributed architecture
  • design evaluation process specification capability
  • general approach to parameterisation
  • range of optimisers to provide robustness and efficiency
  • response surfaces for data modelling
  • decision support and visualisation
Optimization Technology :
  • Genetic Multi Objective Algorithms :
    • Options for initial selection of candidate designs:
      • User-defined set
      • Restart from a previous set
      • Random choice
      • Sobol algorithm
    • Options for selection operators:
      • Local Geographic Pareto Tournament
      • Local Geographic Directional Tournament
      • Global Pareto Tournament
      • Global Directional Tournament
    • Options for crossover:
      • Directional crossover
      • Classical two-point
    • Options for mutation:
      • Constant mutation rate
      • Variable mutation rate
  • Hill climbers
    • Sequential Quadratic Programming
    • Conjugate Gradient
    • Nelder&Mead Simplex
  • Response surface definition
    • Polynomial option
    • K-nearest option
    • Gaussian process option
    • Neural networks