Continuous Systems Validation- Implications for Software Solutions
July 4th 2013
Continuous systems validation brings a whole new level of complication to simulation. In my first post in this series, we looked at how designs progress in various engineering disciplines. In the second one, we looked at how cross-disciplinary simulations are actually assembled out of these different designs. But furthermore, as the design progresses at the system level, that configuration will change. Today, in this post, we'll look at some of the most critical capabilities that a software solution needs to address this reality. Let's dive in.
Flexibly Connecting Design Representations
An obvious first step is being able to run these simulations. Obvious, right? But this capability isn't as widely available as you might think. A lot of progress has been made insofar as Hardware in the Loop (HIL) for controllers. Today, with some simulation solutions, you can plug physical mechanical hardware and physical boards and processors into controller software running on a desktop (as opposed to controller hardware). You can test out the configuration before you ever receive your controller hardware.
But the opposite isn't quite as available. The idea of connecting a mechanical simulation to a board simulation and a physical controller isn't very accessible. Every possible permutation of digital concept, detailed design and physical item for each engineering discipline needs to be covered. This is mainly due to the frequent reuse of items in the design cycle. Complete system clean sheet design is rare.
This marks the first area that needs improvement.
Comprehensive Management of Discipline Specific Simulations
I've written before about the complexities of simulation data management and how connected they need to be to CAD data management over at ENGINEERING.com. Some of the simulation data management (SDM) solutions available today provide such capabilities for mechanical design. However, I know of no SDM solution that offers such capabilities across mechanical, board, processor and controller design disciplines.
You really do need to know which version of the design model was used to create the abstracted and simplified model and, of course, the simulation model. The same applies across all engineering disciplines.
This is the second area that needs improvement.
System Simulation Configuration Management
If you've read this blog series, then you can probably guess the last area for improvement that I think needs improvement. It's tracking the configuration of the system that is simulated and tested. And that not only covers the digital side of things, but also physical possibilities as well. Remember, today's system simulations are complex combinations of Hardware in the Loop and Software in the Loop.
This is the third area that needs improvement.
Summary and Questions
Let's recap. Here's the list of three areas that need to see improvement in terms of software solutions that enable system simulations.
- First, such software systems need to enable the flexible connection of conceptual, detailed or physical representations of different designs from any engineering discipline.
- Next, such software systems also need to be able to manage the progression of how a discipline specific simulation was created, starting with the version of the design model, the abstracted and simplified model and finally the simulation model.
- Finally, these software systems need to track and manage the configuration of digital and physical representations used to create a system simulation across all engineering disciplines. This tells us what configuration was used to gain a specific simulation and test result.
Well, those are my thoughts folks. I'm interested to hear from you on this. How are you tracking this today (or are you)? What other capabilities are needed? Sound off and let us know what you think.
Take care. Talk soon. And thanks for reading.