The last thing you've done  

It doesn't have to be a closed triangle, does it?
Couldn't you just bend the aluminum into a triangular shape and leave the vertex at the top open? It looks to me as though the brace will be loaded in tension across its bottom edge, not in compression from above.

Well . . .
assuming that some other part performs the function of keeping the two halves of the thing together.
Yes, the vital part would be the base of the triangle. But a single bent beam (or rather a pair of them) crossing the middle and doing everything, would seem the simplest correction to what Heath has sketched ? I assume the leg panels would be let into those beams via dados -- and if the dados were wedge-shaped to match the panels, then the whole thing would be locked together when the beams were connected to the seat panels.
But maybe I'd better leave something for Heath to do . . . !

I really appreciate the...
I really appreciate the thoughts and consideration, thankyou. SDR I had thought of that boomerang idea with the dados too and will try it out tommorow and the other idea, I tried something similar with some sprung steel as a beam with screws coming down from the underside of the seat but it wasn't a success, the difficult art of the simple huh?
Have walked away from it for today, had enough.