Optimization

JMAG offers a number of functions to provide strong support for design optimization. JMAG optimization calculation functions are characterized by the integration of both analysis and optimization. This means that factor analysis and optimization can be carried out with ease, allowing for the effective evaluation of a large number of design proposals. Result analysis and sensitivity analysis functions suggest points of improvement for designs. Embedded third-party optimization engines such as HEEDS, Optimus, modeFRONTIER and optiSLang, as well as user-defined optimization engines, are additionally available.

|  Parameter Optimization  |  Topology Optimization  |

Parameter Optimization

  • Realize optimization calculations with JMAG-Designer without any reductions in usability
    • Optimization calculations can be performed simply by broadening settings in parametric analyses
    • Configuring settings is made easy by selecting just design variables and objective functions, constraint conditions, and the optimization engine
    • Setup to results verification in just three simple steps
    • The cases and values of optimum solutions can be confirmed immediately from solution distribution
    • Users can then move straight onto analysis cases with ease
    • It is also possible to directly confirm response surfaces, Pareto curves, and the correlation graphs of objective functions for design variables

Displaying results with response surfaces

  • Optimization algorithms can be selected to match the features and environment of the design optimization
    • The quadratic response surface method, genetic algorithms, multi-objective genetic algorithms, and MATLAB global optimization can be selected
    • Users can choose algorithms appropriate to optimizations that match the single-objective or multi-objective characteristics of the design space
    • The MATLAB Global Optimization Toolbox can be used as an optimization engine

Motor torque optimization with genetic algorithms

Multi-objective optimization
Performs searches for design proposals that minimize torque ripple and maximize average torque

Correlation graphs
Confirm the correlation between design variables and objective functions

Range Finder in Optimization Calculation

  • Search for the range in which the geometry is feasible and acquire the feasible geometry at high speed
  • Acquire feasible geometries at high speed
  • Pareto front for parameter optimization advanced
  • Use the Geometry Evaluation Tool, Judge whether a Range Finder is necessary or not

JMAG Function Videos

Run Optimization Calculation
Optimization Calculation with Multi-Objective Genetic Algorithm
Creating a Pareto curve
Filtering a Response Graph

Use of Range Finder for shape collapse problemUse of Range Finder for shape collapse problem

Use of Range Finder for shape collapse problemEfficient evaluation method for multi-case calculation results

Topology Optimization

  • Design optimum magnetic circuits with unrestricted ideas free from existing designs
    • Support for magnetic field analysis and structural analysis
    • Analysis methods include the evolutionary on / off method using NGnet, and the density method

Example 1: The optimization of an IPM motor with the evolutionary on / off method

IPM motor optimization

Example 2: The optimization of a magnetic shield with the density method

Magnetic shield optimization