Databases tend to support (limited) regex these days. So, perl is better, but databases are not out of the picture.

Anyway, the OP says:

• the data comes from power plants
• the identifiers are not the same
• she loaded the data into a spreadsheet

All this points to many tables (one per plant) with fields (spreadsheet), at the very least, id in one and the rest in another. Matching those ids to each other can be done either through a manual list, or other provided information such as by matching addresses. This can be very easy for a database to do. He should use the best tool for the situation. Perl is the best tool when it comes to text manipulation, any in many, many other cases. Though, a database might be better in this particular case.