Design Explorer in Inspire

The Design Explorer (DE) workflow, previously available only in HyperWorks, is now available in Inspire as well, exposing exploration to an entirely new set of users, even earlier in the design cycle where more advantage may be realized by the exploration results and insights. 

If you’re reading this blog, and if you’re a reader of or subscriber to this forum, you likely have some ideas about, understanding of, or even some level of expertise in data science.  But how can engineers and product designers make use of, and benefit from, this technology?  At Altair, we’re working to add machine learning-based tools in many places in our software and tools.  That said, the design explorer is the ideal environment for these tools because of its intuitive easy-to-use workflow, which is seamlessly integrated into users’ modeling and visualization environment.  Let’s take a quick look at DE, within Altair Inspire.

Design Explorer

The Design Explorer in a beginning-to-end workflow for creating, managing, evaluating, and interpreting explorations (DOEs and optimizations).  Because of its seamless integration into Inspire and graphics-driven features, it can greatly flatten the learning curve associated with typical exploration tools.

Inspire Advantages

One of the main tasks in setting up explorations is the definition of design variables, or aspects of the model that can change – within pre-defined tolerances – in a design.  Inspire allows users to parameterize their CAD, and this parameterization can be directly used as design variables in explorations.  There’s literally nothing else for a user to do, besides checking the box next to the desired variables. 

Responses and goals are similarly simple to create, by selecting directly on the model graphics.

By leveraging the speed of SimSolid (OptiStruct is also supported), users can run even large explorations very efficiently.  DE provides a whole host of post-processing and interpretive tools to help users gain insight from their explorations.

DE already includes some AI and analytics tools, like linear effects plots, which indicate which variables most influence responses, and trade-offs, where users can see in real-time how changes to design variables are predicted to change response values.  But where are we headed, and what additional and more advanced AI tools are possible?

The Convergence of AI and Simulation

Hopefully you already agree that the benefits of exploration are substantial.  And hopefully you’ve now seen that using DE can bring those benefits to bear, especially in an environment like Inspire.  But this is only the beginning.  The EDS team continues to develop additional tools, and the plan is to continue to expose those tools within the DE workflow.  DE in HyperWorks already has field prediction and clustering.  For a quick look at how these tools are already being used in HW DE, if you haven’t already, check out this recent blog post by Jonathan Ollar: How you can get started with machine learning for CAE in 1 minute.  These same tools, along with others, are planned to be brought to DE in Inspire soon.  So, stay tuned!