By default, Proficy Portal samples the data every 10 seconds for
each historical tag you select from a data source. You can control
the sample interval by configuring a time property called the data
point interval. This property controls the time range to use when
calculating data points in the component (e.g., chart or grid). For
example, if the data point interval is one minute, values are
interpolated in one-minute intervals. If you set the data point
interval to 30 minutes, data points represent interpolated values
within a 30-minute interval.
Data values are interpolated because the Historian collectors
typically compress data and store only significant data changes.
The data's deadband determines what constitutes a significant
change. For example, if the current value is 50 and the deadband is
10, the data value must exceed 55 or drop below 45 in order for the
Historian server to store the value change.
When Historian data is compressed, every value that the
Historian collector received is not stored in the Historian
archive. Consequently, when the data is retrieved, Historian
interpolates the values it sends to the Proficy Portal server. When
Proficy Portal receives the data, it plots the values in one or
more charts and connects (interpolates) each point plotted.
Interpolation:
-
Smooths out the gaps that result from using
compressed data.
-
Allows Proficy Portal to display a value
regardless of where you place the time cursor in a chart.
-
Is calculated on actual data collected and
stored in Historian archives.
When Proficy Portal interpolates data, it assumes that the line
connecting two values is a straight line. All points along that
line are estimated except the starting point and the ending
point.
NOTE: When setting a
historical data source for a chart, if the data retrieval mode is
set to Std Dev or Raw Std Dev, two values are required for the time
calculation. If the collection interval and data point interval are
both set to 1 second, no data will display on the chart. You can
set the collection interval to a value faster than 1 second and the
data point interval to 1, or set the collection interval to 1
second and change the data point interval to a time period greater
than 1 second.
The following Historian retrieval modes can be selected:
Historian Retrieval Modes
|
Retrieval Mode
|
Description
|
|
Current Value
|
Retrieves the raw sample value with the
newest timestamp, of any quality, received by the archiver. The
timestamp is not necessarily the current time – rather it will be
the timestamp on the sample sent to the archiver. If you have
collector compression and have not exceeded if for some time, the
timestamp may be much earlier than the current time.
|
|
Interpolated
|
Displays samples spaced on intervals other
than the collection rate. Interpolation is used to produce an
evenly space list of the most likely real-world values. Typically,
you would use interpolated queries when data is not collected on a
set time schedule, or if you want to see the results returned in an
interval that is less than the collection rate. The following
examples describe instances where you can use interpolated mode to
make evenly spaced out values:
-
A tag is collected as "unsolicited". In
this case, you do not know what the time interval is between
collected values.
-
The deadband and/or archive compression for
a tag results in non-evenly spaced collection intervals.
-
A tag is collected once per 8-hour shift,
but you want to see it displayed in 1-hour intervals with a linear
slope between points.
NOTE: For numeric data
in charts, the start time point is calculated.
|
|
Trend
|
Displays the raw minimum and raw maximum
value received during the data point interval. Available for
Historian 2.0 and greater.
|
|
Lab
|
Displays actual collected raw values. The
data value is never interpolated. Interval timestamps are the same
as Interpolated sampling – it is the values that can be different.
Since lab sampling will return real collected values, it is more
accurate when a sufficient number of raw samples are stored. Use
Interpolated sampling for highly compressed data.
|
|
Average
|
Displays the average of the raw samples, but
there is special logic for time weighting and for computing the
value at the start of the interval. This is useful for computing an
average on compressed data.
|
|
Count
|
Displays the number of raw samples in the
interval. This only addresses the count and not values or
quantities of samples. Count is useful for analyzing the
distribution of raw data samples.
|
|
Total
|
Retrieves the time-weighted total for each
calculation interval. The collected value must be a rate. This
calculation mode determines a count from the collected rate.
|
|
Minimum
|
Displays the minimum value in the interval.
It may be a raw or an interpolated value. The minimum must be a
good data quality sample.
|
|
Maximum
|
Displays the maximum value in the interval.
It may be a raw or an interpolated value. The maximum must be a
good data quality sample.
|
|
Std Dev
|
Retrieves the time-weighted standard
deviation of each calculation interval. This is useful for
computing a standard deviation on compressed data.
|
|
Raw By Time
|
Returns all raw samples of all qualities
with a timestamp greater than or equal to the start time and less
than or equal to the end time.
For charts, if no data is received for the
requested time range, the first value previous to the start time,
and the next value after the end time are retrieved, allowing a pen
to be drawn from the two values.
NOTE: For numeric data
in charts, if there is no point with a timestamp that matches the
start or end time, start and end time points are calculated. If
there is an actual start or end time, they are provided as is.
|
|
Raw By Number
|
Displays all raw samples of all qualities
with a timestamp greater than or equal to the start time. The end
time is ignored. Unlike the RawByTime, this will return a sample if
it has the same timestamp as the start time.
|
|
Raw Average
|
Displays the arithmetic average of the good
data quality raw points in the interval. This is useful only when a
sufficient number of raw samples are collected.
|
|
Raw Std Dev
|
Displays the arithmetic standard deviation
of raw values for each calculation interval.
|
|
Time Good
|
Displays the amount of time (in
milliseconds) during the interval when the data is of good quality
and matches filter conditions if the filter tag is used. Available
for Historian 2.0 and greater.
|
NOTE: The Maximum Points
field has been removed from the iHistorian Edit Source dialog box
(this field existed in Proficy Portal (infoAgent) 1.1). To limit
the amount of data that is passed from the Historian archive, use
the Retrieve Mode (General tab) and time range parameters (Time
tab). Additionally, you can limit the data by setting the Max query
tag in the config.xml file associated with the data source. The
config.xml file can be found in the datasources directory
(InstallDir\GE Fanuc\Proficy
Portal\webapps\infoAgentSrv\config\datasources\iHistorian). Locate
the particular Historian data source config.xml file. Open the file
in a text editor such as Notepad, and change the value of the
MAXQUERY tag. The default value is 100000. If the data request
exceeds this amount, you will receive an error message.
See Also
