Log In
Home
02/12/2016
Offline
Hello,
Considering Count aggregate operating on discrete samples, I would like to know how to compute the status code (quality of result) for an aggregate interval.
I have configured TreatUncertainAsBad="False" PercentDataGood=XX and PercentDataBad=YY
I have the following information for a given interval
Count of all raw data samples
Count of good samples
Count of bad samples
Count of Uncertain samples
Using this information how to compute the result status code for final result?
Don't we need to consider count of uncertain samples? If so how?
Appreciate your response. Thank you.
Regards
Samba
04/01/2014
Offline
To determine the status code follow these two steps for non-time based aggregates.
The first step is to determine which of all available raw data points should be included in the calculation. Count uses all good values so anything uncertain or bad is excluded. If there is no raw data the aggregate specifically says the value of the result will be 0 (zero). You will not use uncertain samples in the count because the aggregate states that only good samples will be used.
The second step is to determine the status code by using the PercentValues calculation. Clause 5.4.3.2.1 and clause 4.2.1.2 describe what happens when calculating an aggregate status code. The Count aggregate also states that the calculated bit will always we set and the Partial bit may be set but only on an interval that is not complete.
Rod Stein Manager of Technology Matrikon OPC http://www.matrikonopc.com
02/12/2016
Offline
Hello Rod Stein,
Thank you for response.
Lets consider Count aggregate only which processes samples with good quality.
The PercentValues criteria operates based on the number of good samples, number of bad samples and total number of samples.
What about number of uncertain samples? (resulting from TreatUncertainAsBad=True)
Don't we have to consider uncertain data point count for the status code calculation?
or
For example : An interval having only good and uncertain data points with TreatUncertainAsBad=True? How to compute status code for this case?
Thanks
Samba
04/01/2014
Offline
The calculation of the status code as listed in the Count aggregate depends on the PercentGood and PercentBad settings as well as the TreadUncertainAsBad setting. For Historian1 the values are both %100. This means that in the individual interval if ALL entries are bad then the status is Bad. If ALL entries are good then the status is Good. If they are not ALL good or ALL bad then the status is Uncertain. When looking to see if all entries are either good or bad look at the TreadUncertainAsBad setting and modify the Uncertain entries as needed.
In your example the interval would not have ALL entries as Good and therefore would have an Uncertain status.
Regards,
Rod
Rod Stein Manager of Technology Matrikon OPC http://www.matrikonopc.com
1 Guest(s)
Usage Policy