perhaps you could treat the exponents as a seperate number, so rather (example only) having " 1.9816e-36" as number, you could have "1.9816" and "-36" as numbers. When the first number gets too large or too small, you then change the second number by a preferred magnitude like 9. Then you could assign the second number different names for various small and large units.
But I can't code at all so I don't know how complicated it would be to implement something like that.