5 Replies Latest reply on Nov 19, 2018 3:20 AM by Eugene Lee

How to utlize PI Performance Equation for a 2-dimension interpolation

Dr.Zhou Nov 16, 2018 10:36 AM

Dear All,

I'm trying to employ PI Perfromance Equation to create a 2-dimension interpolation, as showed below. Say, to get to know the exact "condenser pressure" when "thermal load of condenser" were set to be 40%, as where the red circle shows.

As for a 1-dimension interpolation I may utilize the "curve" function, where the arguments should be "numerical constants" referring to the manual. After getting 5 curves by the "curve" PE functions, how could I proceed to the 2-dimension interpolation?

Thanks!

ZYY

2018.11.16

Re: How to utlize PI Performance Equation for a 2-dimension interpolation

Eugene Lee Nov 16, 2018 11:56 AM (in response to Dr.Zhou)

You can try out bilinear interpolation using the AF Table Lookup Data Reference. You will need to create an AF Table with 3 columns as your reference.

Thermal Load, Efficiency (or Performance?),Condenser Pressure

Then you can feed in the Thermal Load and Efficiency in the Table Lookup DR to get your Pressure.

Please see this link for an example
PI Server
1 of 1 people found this helpful
Like Show 0 Likes(0)

Actions
- Re: How to utlize PI Performance Equation for a 2-dimension interpolation
  
  Dr.Zhou Nov 16, 2018 12:14 PM (in response to Eugene Lee)
  
  Hi Eugene,
  Thanks for the answer. AF must stand for the "PI Analysis Framework", which I'm not familiar with. I was wondering whether I could utilize Performance Equation only to make the bilinear interpolation. Regards, ZYY
  
  Like Show 0 Likes(0)
  
  Actions
  - Re: How to utlize PI Performance Equation for a 2-dimension interpolation
    
    Eugene Lee Nov 16, 2018 12:19 PM (in response to Dr.Zhou)
    
    Actually it stands for "PI Asset Framework"
    PI AF - Overview
    
    I would really encourage you to start adopting AF to do things the "right way" and not accumulate technical debt.
    Many of our newer products are all based on AF. AF provides context to data which is quite important in my opinion. As you know, PE scheduler is deprecated too.
    PI Server 2018 is released
    For a new installation of the PI Data Archive, the default installation does not include PI AF Link Subsystem, PI Alarm Subsystem, PI Performance Equation Scheduler, PI Recalculation Subsystem and PI Batch Subsystem.
    
    So i wouldn't encourage to continue using it.
    2 of 2 people found this helpful
    
    Like Show 1 Likes(1)
    
    Actions
    - Re: How to utlize PI Performance Equation for a 2-dimension interpolation
      
      Dr.Zhou Nov 16, 2018 1:37 PM (in response to Eugene Lee)
      
      Dear Eugene,
      I read through the AF Overview, wow...., it is really an excellent new technology.
      I get back to my "PI System Management Tools" - "PI Version" with the finding that all the PI version is 3.4.380.70, while all my menual refers AF to "Analysis Framework" after looking up all the Help documents I have.
      Is the "Asset Framework" involved in my PI installation?
      
      Best Regards
      ZYY
      
      Like Show 0 Likes(0)
      
      Actions
      - Re: How to utlize PI Performance Equation for a 2-dimension interpolation
        
        Eugene Lee Nov 19, 2018 3:20 AM (in response to Dr.Zhou)
        
        It is a separate install kit during the old versions (not the integrated install kit like now). You can also install the newest version which is AF 2018 (to have the bilinear interpolation Table Lookup Data Reference) if you have access to it. The release notes for the newest version AF 2018 states that
        
        The PI AF Client supports access to PI Data Archive version 3.4.380 or later.
        
        If possible, I will also advise you to upgrade your PI Data Archive as 3.4.380 is quite old that I can't find it in the roadmap. This is to minimize compatibility issues with the AF Server.
        PI Data Archive - Roadmap Details
        
        Like Show 0 Likes(0)
        
        Actions

Go to original post