LS-OPT® is a simulation-based optimization tool which enables the solution of complex, multi-stage design processes or regression/classification tasks. LS-OPT interfaces with LS-DYNA® (e.g. result extraction) and also supports popular pre- and post-processors, e.g. for shape optimization. For visualization of results, graphical pre- and post-processing tools are included in the package.

Tasks:

▀ Multidisciplinary and Multi-Objective Optimization (MDO/MOO)

▀ Discrete and Mixed Optimization

▀ Global Optimization

▀ Robust and/or Reliability-based Optimization

▀ LS-DYNA statistics, including outlier analysis and LS-PrePost® support

▀ Parameter Identification with matching of noisy, steep and hysteretic curves

▀ Full-field calibration using Digital Image Correlation

▀ Uncertainty Quantification

▀ Sensitivity Analysis

LS-OPT界面参数定义流程图

Solvers and Methods:

▀ Sequential Response Surface Method

▀ Genetic Algorithm and Efficient Global Optimization (EGO)

▀ NSGA-II algorithm for MOO

▀ Monte Carlo methods (direct and metamodel-based)

▀ Outlier Analysis

▀ Support Vector Machines (SVMs) for Statistical Classification

▀ Taguchi Method

▀ Curve similarity measures: Dynamic Time Warping (DTW), Partial Curve Mapping and Discrete Fréchet

▀ Experimental Design: Space-filling, Full or Fractional Factorial, Latin Hypercube

▀ Metamodels: Neural networks, Polynomials, Kriging and Support Vector Regression

▀ Network-based job scheduling

LS-OPT is capable of performing optimization with multiple objectives, disciplines and load cases. It can account for uncertainties in the design or loading (stochastic analysis and optimization), and can also be used to optimize tolerances with a multi-level setup.

Graphical Post-Processing:

Result plots (Correlation Matrix, Scatter plots, Parallel Coordinate, Self-Organizing Maps, Time-history, Statistical)

▀ Metamodel plots (Surface, 2D cross-sections, Accuracy, Global sensitivities, History sensitivities)

 Classification boundary plot

▀ Pareto plots (Scatter plots, Parallel Coordinate, Self-Organizing Maps)

▀ Stochastic Analysis (Statistical tools, Correlation, Stochastic Contribution)

▀ Effect plot (Taguchi)

▀ Optimization History

▀ Tables with interactive features