PI Vision Collection Bug

We have a number of wells, and I have set up a template for each in the Asset Framework.  One of the Attributes of each well is whether it is Flowing or not.  The attribute is populated in an expression analysis template, and the Data Type for the Attribute is an enumeration set.  

 

I have set-up a PI Vision display, and I have used a collection to dynamically generate the graphics for all the wells.  I want to filter-out any wells that are not flowing.  I therefore "Edit Collection Criteria", and add a filter on the Asset Attribute for "FLOWING".  The PI Vision display appears to recognize that the various status codes are part of an enum, given me the correction options in the drop-down.  But, the filter doesn't work.

 

FLOWING = TRUE => No Items match the Collection Criteria

FLOWING != FALSE => No Items match the Collection Criteria

FLOWING != TRUE => Returns all assets, flowing or not.

FLOWING = FALSE => Returns all assets, flowing or not.

 

I've tried various solutions - originally the attribute was a String, I've also tried deleting the "Facility" attribute - that makes no difference (except that for the full list, I get a list of all elements from all facilities, so it proves that the Facility filter is working).

 

Can anyone offer any insights?

 

Thanks,

 

Andrew

Parents
  • We've run across this as well. I believe the problem is due to binding an asset attribute that is set to tag create when tied to a template, where the tag is being outputted to by an analysis. For example we have an analysis that outputs to an attribute that determines a boolean attribute. The boolean attribute is bound to a PI tag. When bound to a tag and set to tag create, the attribute will not filter as expected.

     

    When not bound to a tag and is a simple attribute, filtering on the display works fine.

Reply
  • We've run across this as well. I believe the problem is due to binding an asset attribute that is set to tag create when tied to a template, where the tag is being outputted to by an analysis. For example we have an analysis that outputs to an attribute that determines a boolean attribute. The boolean attribute is bound to a PI tag. When bound to a tag and set to tag create, the attribute will not filter as expected.

     

    When not bound to a tag and is a simple attribute, filtering on the display works fine.

Children
No Data