Hi @supersharp,
with sequence of words do you mean the sequence in the cell or sequence of rows? If the change in the order number always occurs in the same location you might, for string containing seven characters after "# ", remove the first one and then search for duplicates via duplicate row filer.
I read that you sometimes have near identical duplicates, like the example of yours “Invoice# 123456” and “Invoice# 0123456”, which can be filtered out very easily but then I assume also other where you really need some more novel approaches.
Though, breaking down complexity accelerates finding a solution. Would be great if you can provide some example data too making it less abstract.
Cheers
Mike