Discrete flexibility! Its a wonderful thing. It may souns a bit lewd, but its not what you think. Imagine, for instance, an inflatable bouncy castle with a bunch of kids jumping up and down on it. Then you know why its called a bouncy casle. The thing bounces and flexes all over the place! Now imagine you had to calculate the flexibility of that castle. Mesh it using very small elements and then have a supercomputer process the whole model, solving the millions of equations. It'll take a long time. trust me, I know!Now why is calculating that bouncy castle so damn complex? Perhaps we are over-calculating? imagine you are a bouncy castle designer, do you really want to know at exactly what angle and amplitude every wrinkle and crevice occurs at which forces? No! You want to know wether one of the towers can collapse, and hit a kid in the baby-curled head!
Same with inflatable wings, I want to know how far it flexes. I dont need to know that at x = 0.65434 the third wrinkle is 0.263cm high. And of course, then only to find out I made a small error in one of the boundary conditions which renders the entire solution useless. All these numbers behind the decimal point give a fake sense of accuracy. When measuring the distance from Amsterdam to Paris, you don't use a 20cm ruler!
In comes discrete flexibility. Divide the beam up into a number of stiff discrete elements and join them together with springs. Write smart equations for the behavior of the springs, making them behave as inflatable beams. Et Voila! A fast running simulation! It may not be as exact as a FEM analysis can be, but its a hell of a lot quicker, and thus, far more intuitive.
Yes... I love discrete flexibility! I have some programming to do. Catch you later :)
Jeroen Breukels
