Tuesday, April 30, 2013
Impact finite elements for seismic analysis
I'm quite happy, that I got Impact-fem working again! I had a lot of trouble before, and the modelling gave me a headache. After a couple of years, the project appears very active, and it is much improved.
This is just a cantilever springboard and the colours show the strain. You can let it go and watch it go whappy-whappy. This is much like a building on the side on a shake table, and it is showing the fundamental mode of vibration. I encourage everybody to start playing with this. You will soon realize that the existing conventions of seismic engineering are crap, once you add more reality. For example, I will put a stiff building on a stiff half-space (the ground), and propagate a realistic seismic wave. There is no major amplification in the building because seismic energy radiates back out into the ground. It is not a shake table.
Now, why would one care? This old conventions are one reason why the effect of soft ground is grossly underestimated by a factor of 10. They are also responsible for using PGA (peak ground acceleration) for scaling, which is hogwash. Better to use PGV.
Finally, they are rebuilding Darlington nuclear plant from the ground up, because they operated it so badly. This is going to cost billions of dollars. At least one billion will be wasted if they use the 40-year-old seismic conventions to seismically qualify new equipment. Although this will be covered up, it is still a shame. Comon' people, have some fun! Do this. I'll be doing some more modelling in the future.