If all you have is the file, without knowing where you got it, then you'd need more than a download manager to resume it. You'd need software that heuristically analyzed the contents of the partial file you have, searched for the file (which would probably be very difficult), downloaded pieces of possible candidates (it wouldn't even start out knowing the proper length for the file, after all, so only by guessing and checking could it figure out what file to download), and eventually download the rest of your file most of the time.
No such utility exists. And no such utility is likely to be developed, because even if you've forgotten a lot, as the person who originally decided to download the file, you fundamentally know more about where to look for it than anyone else, including any artificial intelligence.
So, the real-world solution in your situation is for you to search for the file, applying whatever knowledge you have to find what you believe to be the download link for it (or some other download link, if this is a file that is hosted in multiple places).