So how accurate is SOLIDWORKS Simulation?
Customers often ask us this question; either on training or when they are interested in the software. The answer we always give is ‘very accurate’, obviously taking into account the correct set up of their problem.There is no standard method to validate any piece of FEA software, but you can test it yourself and get accurate results in comparison to hand calculations. SOLIDWORKS uses 3 different methods to prove that their software gives good results.
The first method used by SOLIDWORKS utilises some verification problems, which are run using the software and the Simulation results are compared to either an analytical solution or a physical test. They supply the file along with the associated setup for the analysis and these can be found in the Simulation help files or in the tutorial examples. To find these, turn on the Simulation Add in and then:
Click Help > SOLIDWORKS Simulation > Validation > Verification Problems
Here you
will find plenty of problems verifying Static, Frequency, Non-Linear, Thermal, Drop
Test and Buckling simulations.
Here’s an example - a Deflection of a Cantilever Beam of Length 10”,
Height 1”, Thickness 0.1” and a load vertically applied at the end of 1lb. This
was done as a static simulation with a shell mesh and the results along with
the percentage error are below.
| Analytical | SOLIDWORKS Simulation | % Difference | |
|---|---|---|---|
| Deflection at free edge (inch) | 0.00133 | 0.001344 | 0.83 |
| Average Shear Stress (psi) | 10.000 | 9.877 | 1.23 |
The second used
are the NAFEMS (National
Agency for Finite Element Methods and Standards) Benchmarks, which is an independent non-profit
organisation whose aim is to promote the effective use of engineering
simulation methods.
NAFEMS Benchmarks
are available in the exact same place as the verification problems and are
available for Linear Static, Non-Linear, Thermal, Frequency, Forced Vibration
and Composites.
An example
from the NAFEMS Benchmarks is a Static Cylindrical Shell under uniform edge
moment of 1.0 KNm /m.
| Analytical | SOLIDWORKS Simulation | % Difference | |
|---|---|---|---|
| Tangential Stress a point E (MPa) | 60.00 | 60.05 | 0.08 |
The final method is using the French standard AFNOR (Association Française de NORmalisation,
French Standardization Association). These benchmarks can be found in theSOLIDWORKS Knowledge Base under article S-069850. The report includes 100
examples of verification, one of which can be seen below - a Fixed Thin Arc
under out of plane bending. The object is fixed at one end and a load is applied
at the other, with the deflection compared at location B and bending moments at
15 degrees on the arc.
| Physical Testing | SOLIDWORKS Simulation | % Difference | ||
|---|---|---|---|---|
| Deflection at B (m) | 0.001333 | 00.001344 | 0.09 | |
| At 15deg | Mt (Nm) | 74.1180 | 74.1830 | 0.09 |
| Mf (Nm) | -96.5925 | -96.4050 | 0.19 | |
As you can see, SOLIDWORKS Simulation is accurate to these 3 benchmarking techniques. Have a look at some of the other examples yourself in your own time. SOLIDWORKS have also produced some design validation white papers for customers here:
http://www.solidworks.com/sw/design-validation-whitepapers.htm.
If you’re interested in Simulation and want to do some hand calculations, physical testing and simulation results comparisons yourself, have a look at our specially designed Simulation Fundamentals training course to get up to speed with the basics of Static Simulation.
https://www.solidsolutions.co.uk/solidworks-Training/Courses/Simulation/Fundamentals.aspx
Look out for the upcoming blogs on the accuracy of Flow Simulation and Plastics Simulation.
Chris Boyles