It’s often tempting with systems modelling to want to reach a solution as quickly as possible, effectively “throwing” all the components together, parameterising and simulating the scenario of interest and getting the results we need.
However, doing this is generally more time consuming than taking a steadier approach. The steadier approach involves:
- The parameterisation and validation of each newly developed and/or newly parameterised component model on virtual test beds with known boundary conditions.
- Once each newly parameterised component has been validated in terms of results and simulation performance, it can be integrated with other components in lager subsystem models and these in turn get validated and tested virtually.
- The validated subsystems can now be integrated into a larger system model.
- The larger system model can now be simulated with minimal debugging necessary, if any, as this will have mostly been done in stages 1 and 2.
A few considerations reinforcing the importance of the points above: If you consider a component might be reused several times in a large system model, for example a valve or a pipe, when this component is not defined and functioning in an optimal way, these inefficiencies will be replicated throughout the system model and scaled up depending on the number of instances of the component there are. Often the inefficiencies have an exponential compounded effect on model robustness and simulation performance rather than a linear effect.
If you look at any of our Modelica libraries, their structure will reflect the stage by stage model development and testing philosophy that I described above.
Figure: Snapshot of a section of the VeSyMA – Powertrain library highlighting some of the component experiment packages.
Sounds like a well known old story but in summary, take it steady and you’ll get there quicker.
Written by: Alessandro Picarelli – Engineering Director
Please get in touch if you have any questions or have got a topic in mind that you would like us to write about. You can submit your questions / topics via: Tech Blog Questions / Topic Suggestion