Dither adds noise to your recording. The different algorithms are merely different ways to bias and distribute that noise in such a way as to be less noticeable. Triangular dither isn't "smart", it's just one way to distribute the noise so it's less audible. However, given that dither noise is at such a low level that it's normally inaudible anyway, there's not much point in stressing over which dither algorithm to use.
So yeah, go ahead and pick Triangular and don't lose any sleep over your decision. And you can skip it altogether when exporting to 24 bits.