HyperStudy

HyperStudy is a design exploration tool for engineers and designers. It automatically creates intelligent design variants, manages runs, and collects data. Users are then guided to understand data trends, perform trade-off studies and optimize design performance and reliability.

“HyperStudy allows flexible, multi-platform optimisation management and is generic and applicable to any weld distortion project without any adaptation.”
–Adrian Chapple, Analysis Supervisor
Gestamp Tallent Ltd.

HyperStudy enables users to explore, understand and improve their designs using methods such as design-of-experiments, response surface modelling and optimization. Results from these studies can be easily analyzed and interpreted using HyperStudy’s advanced post-processing and data mining capabilities. HyperStudy’s intuitive user interface combined with its seamless integration to HyperWorks for direct model parameterization and CAE result readers simplifies the study setup.

 

Benefits

Improve Design Performance and Quality

HyperStudy includes state-of-the-art, innovative optimization, design of experiments and stochastic methods for rapid assessment and improvement of design performance and quality.

Image

26% warpage reduction for a molded part

 

Perform Trade-off Studies

HyperStudy’s fit capability allows users to create response surface models. These efficient surrogates can then be used to perform trade-off studies. They can also be exported as spreadsheets for field engineers’ use.

 

Reduce Development Time and Costs

HyperStudy helps engineers reduce trial-and-error iterations and hence helps to reduce both the design development and testing time.

Image

60% Development time reduction for advanced oil well liners (Courtesy of Baker Hughes)

 

Higher Productivity through Easy-to-use Environment

HyperStudy’s step-by-step process guides the user in setting up and carrying out design studies. Its open architecture allows easy integration with 3rd party solvers.

 

Powerful Dataset Analyses

Comprehensive set of post processing and data mining methods simplify and aid an engineer’s job of analyzing and understanding large simulation datasets.

 

Improve Simulation Correlation

HyperStudy's optimization capabilities can be applied to improve correlation of analysis models with test results or with other models.

 

Image

80% reduction in material calibration time

Mars lander design for reliability

7% seat mass reduction with HyperStudy

Bottle optimization: Increased performance and reduced mass

DeWalt Cooling Fan Optimization

Optimizing for increased durability using CAEfatigue and HyperStudy

8% weight reduction and 30% performance improvement in Renault powertrain

Toshiba reducing development time by 70%

Turbine blades optimization using HyperStudy with AcuSolve

Capabilities

Design of Experiments

Design of Experiments (DOE) methods in HyperStudy include:

  • Full factorial
  • Plackett-Burman
  • Central composite design
  • Hammersley
  • Fractional factorial
  • Box-Behnken
  • Latin hypercube
  • User defined and direct input of external run-matrix.

The study matrix can consist of continuous or discrete variables that can be either controlled or uncontrolled. DOE studies can be performed using exact simulation or the fit model.

 

Response Surface Method (Fit)

Available response surface methods are:

  • Least squares regression
  • HyperKriging
  • Moving least squares
  • Radial basis functions.

Response surfaces can be used for performing trade-off, DOE, optimization and stochastic studies.

 

Optimization

HyperStudy’s comprehensive optimization methods solve different types of design problems including multi-objective and reliability/robustness based design optimization. These methods are:

  • Adaptive response surface method (ARSM)
  • Sequential quadratic programming
  • Genetic algorithm
  • Sequential optimization and reliability analyses (SORA)
  • Single loop approach
  • Global response surface method (GRSM)
  • Multi-objective genetic algorithm
  • ARSM based SORA
  • User-defined optimizer.

Optimization studies can be performed using either exact simulation or fit model. In addition, HyperStudy provides an API to incorporate external optimization algorithms.

 

Stochastic

The stochastic approach in HyperStudy allows engineers to assess reliability and robustness of designs and provide qualitative guidance to improve and optimize based on these assessments. HyperStudy sampling methods are:

  • Simple random
  • Hammersley
  • Latin hypercube

Stochastic studies can be performed using either exact simulation or the fit model.

 

Post-Processing and Data Mining

HyperStudy helps engineers to gain a deeper understanding of a design through extensive post-processing and data-mining capabilities. This significantly simplifies the task of studying, sorting and analyzing results. Study results can be post-processed as statistical data, correlation matrices, scatter plots, box plot, interaction effect plots, histograms, and parallel coordinates among others. Furthermore, HyperStudy guides the user in the selection of post processing methods to use based on the design objectives.