Sharlinda Salim

Hi Nurhaz,

In Version 63 onwards, you can use Periodic Condition >> Custom as in this link https://support.seeq.com/kb/R63/cloud/what-s-new#id-(R63)What'sNew?-CustomfrequencyperiodicconditionswithoutFormula

If you are in older version, at the Formula tool you can write as below:

periods(5d)

Hi Dayyan,

For the second question:-

1. Create a 6 month interval condition using formula

periods(6mo)

2. Split the 6month when there's replacement condition using the Composite Condition > A minus B as below screenshot.

3. Use the count function for individual manual draining

4. Use runningCount() for the tracking purpose. At the below screenshot, after the replacement condition, it will automatically reset to zero. Please let me know if there's any criteria here did not meet your requirement.

Hi Dayyan,

On your first question: How can I effectively create a capsule or representation for this seal replacement?

Seems you have the derivative signal, you can :-

1. Use value search to create the condition when the derivative < 0,
2. Use value search to create the condition when the derivative > 0
3. Use Composite Condition > Join to join (1) and (2)

On your second question, are you currently using runningCount() to track the occurrences of manual draining? 

Hi David,

Can you clarify which part you are experiencing slowness, is it when pushing the script or when you are doing the analysis in Workbench? If its the analysis, can you please elaborate on the analysis you are doing. You can also book a slot in our Office Hour, https://info.seeq.com/office-hours to discuss further on this issue.

Hi Olav,

There is a workaround in XY Plot but it may not able to plot exactly the same as your Power BI screenshot.

Step 1: Create the binning conditions using value search. As an example I created 4 conditions with bin size of 10.

Step 2: Create the condition for June and July

join(
months(Month.June, "Asia/Kuala_Lumpur"),
months(Month.July, "Asia/Kuala_Lumpur"),90d)

Step 3: Calculate the average value during each bin in June&July and combine them together. The move is to avoid overlap at the same startkey.

combinewith(
$t.within($b010).aggregate(average(),$june_july,startKey(),0s).move(1h),
$t.within($b102).aggregate(average(),$june_july,startKey(),0s).move(2h),
$t.within($b203).aggregate(average(),$june_july,startKey(),0s).move(3h),
$t.within($b304).aggregate(average(),$june_july,startKey(),0s).move(4h))

Step 4: Used the setproperty() to create the corresponding X value for each of the average value.

combinewith(
$t.within($b010).aggregate(average(),$step2,startKey(),0s).move(1h).toCapsules().setProperty('Bin',10),
$t.within($b102).aggregate(average(),$step2,startKey(),0s).move(2h).toCapsules().setProperty('Bin',20),
$t.within($b203).aggregate(average(),$step2,startKey(),0s).move(3h).toCapsules().setProperty('Bin',30),
$t.within($b304).aggregate(average(),$step2,startKey(),0s).move(4h).toCapsules().setProperty('Bin',40))
.toSignal('Bin',startKey()).toDiscrete()

Step 5: Switch to the XY Plot and expand the date range. Create the individual conditions for each year and you can then color based on these conditions.

Let me know first is this workaround suits you.

Hi Ivan,

At the moment this is still the best way to calculate median with multiple signals. We did logged a feature request to simplify this method. If you are interested to receive an email when there's an update on this, please do submit a seeq ticket at https://seeq.atlassian.net/servicedesk/customer/portal/3.

Thank you.

Hi mKaiser,

The main constraint here is calculation based on display date range is possible via simple metric only.

I have one suggestion and please let me know your thought and correct me too if I'm not capturing correctly your criteria.

Step 1 & 2 similar as yours.

Step 3 : Set the date range of interest using the manual condition.

Step 4 : Calculate the average and percentile using Signal from condition and bounded it to Step 3 condition.

Step 5: This is another tricky step as you need to define the date range in identifying the final condition. Example in the last 6month.

condition(6mo,capsule(now()-6mo,now()))

Step 6 : Use value search and composite condition to define the final output. I have these functions in one formula. I am not 100% sure whether I understand this final criteria but let me know if you need further help.
 

//isbetween two percentile signals or certain range above average

   ($step2 <= $percentile_90.aggregate(startValue(true), $step5, durationKey())
and $step2 >= $percentile_80.aggregate(startValue(true), $step5, durationKey()))
or
   ($step2 > $average.aggregate(startValue(true), $step5, durationKey()))

So each time you change the date range at the manual condition, the whole calculation will automatically recalculate limited to step5 definition.

Hi Taylor,

Before we discuss the Table part, I would just like to point out that displaying similar result in a trend view is easier by utilizing our 'Capsule' view. KB article

In order to visualize in Table, there are more steps depending on how many samples you have in a day. 

Step 1: Create the daily condition.

Step 2: Create capsules for each sampling point

$a.toDiscrete().toCapsules()

Step 3: From Step (2), extract the first to the nth capsule. Notice that I had also embedded all the samples value into the daily condition. Hopefully you are dealing with lab data or similar where you are not expecting too many sampling in a day.

//find each sampling point
$capsule1 = $daily.transform($w ->$all_sampling.toGroup($w).first())
$capsule2 = $daily.transform($w ->$all_sampling.toGroup($w).pick(2))
$capsule3 = $daily.transform($w ->$all_sampling.toGroup($w).pick(3))

//Embedded the value into daily condition and label Sample 1,2,3 accordingly.
$daily
.setproperty('Sample 1', $signal.aggregate(startValue(true), $capsule1, startKey(), 0s), startValue())
.setproperty('Sample 2', $signal.aggregate(startValue(true), $capsule2, startKey(), 0s), startValue())
.setproperty('Sample 3', $signal.aggregate(startValue(true), $capsule3, startKey(), 0s), startValue())

Step 4: Switch to the Table view and set accordingly as per below screenshot.

In case you did not able to view all the features I had used here, please check out the lower left corner of your Seeq browser window to see which version you have.

Hi Koki,

Please do book a slot in our Open Office hour, https://info.seeq.com/office-hours, so that you can share your Seeq screen and we can brainstorm from there.

Thank you.

Regards,

Sharlinda

Is there any commonality in the equipment's tag name or description? 

Hi Kate,
Unfortunately not at the moment but we do have this feature request logged. If you submit a ticket to our support portal, we can add your name into the ticket to ensure you will get an automatic update on this request. Thank you.
 

Hi Nurhaz,

You need to look for the particular folder in the "Shared" and ensure you have the necessary access.

Regards,

Sharlinda

Hi Nurhaz,

This is one possible method:-

Step 1 : Create a new formula where we convert the month format based on your preference

$month_numeric = 

$f
.toSignal('End')
.toString()
.replace('/(?<year>....)-(?<month>..)-(?<day>..)T(?<hour>..):(?<minute>..):(?<sec>..)(?<dec>.*)Z/' , 
          '${month}')

//changing the name format

$month_alphabet =

experimental_lookupTable(
"[
 ['01', 'Jan'],
 ['02', 'Feb'],
 ['03', 'March'],
 ['04', 'April'],
 ['05', 'May'],
 ['06', 'June'],
 ['07', 'July'],
 ['08', 'August'],
 ['09', 'September'],
 ['10', 'October'],
 ['11', 'November'],
 ['12', 'December']  
 ]",
InterpolationMethod.Discrete)
$month_alphabet.experimental_lookup($month_numeric, 'B')

Step 2 : Configure the sequence/format based on your preference

$extract = $f.toSignal('End').toString()

$day = $extract
       .replace('/(?<year>....)-(?<month>..)-(?<day>..)T(?<hour>..):(?<minute>..):(?<sec>..)(?<dec>.*)Z/' , 
                '${day}')



$YYYYhhmm = $extract
            .replace('/(?<year>....)-(?<month>..)-(?<day>..)T(?<hour>..):(?<minute>..):(?<sec>..)(?<dec>.*)Z/' , 
                    '${year} ${hour}:${minute}')



$final = $day + ' ' +$month_alphabet+ ' ' +$YYYYhhmm
return $final

Hope this helps.

Yes, understood but it will not be straightforward though thus I recommended formatting the header. Can you please let me know the background of your use case? It it still in the context you would like to display the time of the last value? If yes, may I know why you do not prefer to configure the header.

Hi Nurhaz,

The easiest will be to configure at the header. In this example, I used Simple Table >> Last Value, followed by customizing the header as showed in the second screenshot. If you click at the '?' beside the Date Format it will provide you more formatting options. Finally, you can add another column and name it 'Last Sample Time".

This method is applicable still if you prefer to use Scorecard Metric.

The difference between this method and the one discussed above is this method will generate the last value too.

To Create a rolling linear forecast on data, we can use the formula tool:

$signal.forecastLinear(7d,3d)

As of R63 you can also use the ForecastArima function in Seeq Formula as well to create a rolling forecast. Please see the following knowledgebase article for more information on how to use ForecastArima: https://support.seeq.com/kb/latest/cloud/best-practices-for-using-forecastarima

Please do have a look at few other discussions as below links that applied the forecastlinear() function.

The cumulative of time as shown in Step 3 can be simplified to timesince($condition,1min). 

Content Verified DEC2023

Hi Nurhaz,

You may also book a slot in Seeq Office Hours to discuss further in a call. (https://info.seeq.com/office-hours)

Firstly, I observe your condition for the $run can be optimize as the start of the capsule seems slightly off than the peak of your signal.

If you could not join the office hour can you share how you create the $run condition.

Secondly, did you use exact formula as below including the 10d setting? Your $run seems longer than 10d so you might want to expand that, probably 3mo.

timeSince($runs.growEnd(10d), 1h)

Interesting question from a Seeqer who would like to create a BatchID for each batch with the format YYYYMMDDS# where YYYY is the year, MM is month, DD is day, S is the shift and # is the order of the batch in that shift. 
 

This is one of the possible method.

STEP 1: Create the signal for YYYYMMDD

//STEP1: Trend the "Start" date of each capsule

$StartDate =  $batch
              .move(7h) //based on time zone use in workbench.
              .toSignal('Start')
              .toString()
              .move(-7h) //move back the signal to the original position
              
//STEP2: Replace the format to YYYYMMDD

$YYYYMMDD = $StartDate
             .replace('/^(..........).*/' , '$1') //example of the output is 2022-09-02
             .replace('-','') //20220902
             .validValues()
             .toStep(1d)
$YYYYMMDD

STEP 2: Create the signal for S which represent the shift number

combinewith(
shifts (6, 12, "Asia/Kuala_Lumpur").setProperty('Shift','1'),
shifts(18, 12, "Asia/Kuala_Lumpur").setProperty('Shift','2'))

Followed by another formula,

$s.toSignal('Shift')

STEP 3 : Create the # which represent batch sequence in the respective shift
 

$count = $batch.aggregate(count(),$batch, startkey(),0s)

$count
.runningCount($shift)
.toStep(1d)
.toString()

STEP 4 : Create the BatchID and set as property
 

$BatchID = $YYYYMMDD + $S + $x

$batch.setProperty('BatchID', $BatchID, StartValue())

Hi Sivaji,

Step 1: Calculate aggregated value of every n-samples. Example here I am calculating the average for every 5 samples.

Use aggregateByCount which is introduced in version R54.

$signal.aggregateByCount(average(), 5, 10d)

Step 2: The difference between the sample and aggregated value is greater than certain threshold

These capsules will be  0s duration as the aggregateByCount generated a discrete signal.

Step 3: Replace the samples with the aggregated value 

Hope this help.

Hi jkasworm,

You need to create access key (link) in order to obtain the username and password.

Hi VpJ,

You mentioned gap and missing data, does it looks similar like below screenshot?
In this case, instead of creating manual condition you can use the function .isNotValid() in the formula tool. With this, you do not need to worry when scaling it across assets. 

Hi KYM,

Please have a look at this article - https://support.seeq.com/space/KB/2168684706/How-To%3A+Export+Signals+and+Conditions+to+OSIsoft+PI

Thank you.

Hi Brian,

Please check out the derivative() function in the Formula. If the current sampling interval in your continuous signal is not 1min, you can incorporate resample() function.

Example:

$signal.resample(1min).derivative()

Hi JWu,

I had seen cases of "Page Unresponsive" due to few issues but not particularly for a high number of worksheets. Roughly how many worksheets do you have?

Best to contact our Support Portal with a downloaded logs when you observed the error so that we can look further into the issue.

Thank you.

Hi Niranjan,

Can you please provide a bit more background on your use case and calculations for us to have a bit more insight.

Thank you.

Regards,

Sharlinda