Seeq R21 is a combination of ease of use features, improved analytical tools, and user requested features.
The big changes in R21 include an updated user interface (UI), frequency analysis with Fast Fourier Transform (FFT) with the new Frequency Analysis tool, new formulas, and new functions.
Ease of Use
Expanded Color Palette
Quickly identify Workbench Analysis (green) and Organizer Topic (blue) with new color scheme.
Quickly focus on specific data
Enable Show only selected items from the Dimming dropdown on the toolbar of any Workbench Analysis to trend only items selected in the Details Pane. When this is not selected, previous signal dimming remains.
Multicolumn Scorecard Metrics
Create multicolumn tables with Scorecard Metric with each column based on a capsule or particular time range within the display range. Thresholds can be signals or strings, and will display boundaries along with the signal of interest when the scorecard is viewed in the trend viewer. Read more on configuring the flexibility this tool provides here.
Change Workbook Ownership
Users can now assign a workbook to another person when they shift roles.
Quickly note what folder work is stored in, or jump to a parent folder with links at the top of the header section of the screen. Right click on these to open the parent folder in a new tab!
Investigate range update
The capsule preview in the investigate range now defaults to OFF. You can turn it back on when you want to have that functionality with the button to the right of the investigation range. Leaving this off can reduce calculation time to load the worksheet.
Access full list of Seeq Functions
A full list of Seeq functions is available to review, in addition to previous search functionality. The function search has been updated to improve search result priority.
Improved Analytical Tools
Frequency Analysis Tool
Users can now perform frequency analysis on signals with Fast Fourier Transform (FFT) with the new Frequency Analysis tool. Read more on the tool here.
New formula function correlationOffset() improves regression analyses with time-delay impacts. Using this formula first will identify time delay between two signals and shift one so that the correlation is maximized in subsequent regression analysis.
As an example, Signals 1 and 2 (shown at right) are highly correlated but Signal 1 leads Signal 2. As a result, the strong correlation is not evident in a scatter plot. Signal 1 can be aligned with Signal 2 using the correlationOffset() function, which identifies the positive or negative offset (time shift) needed to maximize correlation. The offset value is then used by the delay() function to create a new aligned signal. The Formula for this example is shown below:
Note that the offset time (a negative value when shifting the signal to the left on the time axis) can be returned from the formula and used in other calculations.
This new functionality can be useful in many ways, including alignment of product quality data and analyzing cause/effect relationships between signals in different processing sections.
Signal 1, Signal 2 trends and scatter plot before applying correlationOffset()
Signal 1 Aligned, Signal 2 trends and scatter plot after applying correlationOffset()
Histograms for Conditions
Conditions can now be used as inputs into histograms, enabling counting of batches and summing of durations during other other conditions.
Other new functions
|timeSince()||Users can use this function to assist with modeling fouling or activity decay with respect to time. A long formula block could achieve this in R20 and prior, but the new timeSince() function makes it so much easier|
|replace()||Replace string values in a signal with another string. Maybe whenever your compressor is in "Stage 1" or "Stage 2", you just want the signal to show "ON". This function supports regular expression searches.|
|isValid() & isNotValid()||Streamline the discovery of periods of invalidity using ValueSearch() in formula, along with the new isValid() and isNotValid() functions.|
|sin(), cos(), tan(), asin(), acos(), atan()||New support for trigonometry functions|
|agileFilter()||Smooth a signal using the Loess method, also known as Lowess, local polynomial regression, or moving regression.|
|sgFilter()||Smooth a signal using the Savitzky-Golay method, also known as least-squares or DISPO (Digital Smoothing Polynomial).|
|runningAggregate()||Visualize stats and how they accumulate per capsule in a condition. See how average, stdDev, min/max change over time.|
|combineWith()||Combine samples from 2 or more signals into 1 without any interpolation effects.|
|aggregate()||The parameters to aggregate() are significantly reworked to make it easier for advanced users to write expressive aggregations of signals and conditions|