We are about to rename our tags (for various reasons including standardisation across our buisiness, correcting errors and more) and I was wondering if anyone else has come across any pitfalls that we can avoid.

Changes being proposed.

Our Tag name is made up as follows:
Region~Substation~Circuit~Nomenclature~Units~Type
Region~Voltage~Substation~Circuit~Nomenclature~EngUnits~Type

e.g. "W~132~Flint~Prestatyn/Colwyn Bay~MW~TAI"

The 'Region' and 'Type' is being changed in all tags. 'Voltage' added to all tags. Minor edits to 'Substation' & 'Circuit' in many tags. Changes to 'Units' (and the EngUnits attribute) in around 2000 cases.

Things we have identified so far:

(a) ProcessBook is quite happy with tag renaming (up to a point!) and will pick up the
(b) Tags in PB Datasets are not automatically changed and will need to be done by us.
(c) Tags in Excel (Datalink) are not automatically changed and will need to be done by us.
(d) MDB. Not investigated yet but suspect it may be OK although our aliases will need edits too.

My approach:
============
(1) Import tags into Excel. Approximately 13,000
(2) Excel Macro makes ALL changes to Tag Name (& EngUnits in some cases) in the sheet.
(3) Export tags NewName to PI Server.
(4) Distribute macro to Datalink users for them to make the exact same changes to their sheets.

The above is now in place and (almost) ready to roll.

The ProcessBook Support Tool can carry out Find/Replace in Datasets but our changes are not that simple and would involve thousands of R/R actions! So:

(5) Create ProcessBook VBA to identify tags in Datasets.
(6) For each tag find it's replacement in (from the sheet produced in 2 above).
(7) Replace tag.
(8) Repeat for each tag in the Datasets.
(9) Repeat for each Datasets.

All the above seems very simplistic and I would very much appreciate everyones experiences or advice on this matter.

OSIsoft vCampus is a subscription-based, online offering that consists of providing everything people need to develop applications on the PI System.
We invite you to take a "tour" of the OSIsoft Virtual Campus - also feel free to consult the FAQ  or contact OSIsoft vCampus for more details.

squatty,

using the PointID; marvelous idea and one I have just passed on to the other users. Thanks. Saw that formula but never understood the need for it. Got me thinking that there are other features there that I must make time to explore.

I certainly agree that calulated tags would better in the long run than datasets. However many datasets refer to data already in the historian and not just to data from the data of creation onwards. Calculated tags would be of little help there. I believe!

(5). I agree that the code must reside in a display in a processbook but it need not be in every one. I already have code based on Rhys' Framework http://www.rjksolutionsl...-for-PIProcessBook.aspx that uses code in one processbook display(PDI) to modify trends on Dislpays in another Book. My opo in Scotland has also produced a bit of code that does almost what you suggest. I need to modify it to lookup the found tag and replace with the new one.

It's good to talk (as The British Telecom ad used to say) - thanks.