weisang.com > Documents Articles > Event Isolation Analysis Object *

FlexPro Documentation

Event Isolation Analysis Object *

09.03.2021

With this analysis object, you can search data sets for various events. Two basically different types of events are supported. Either search for all occurrences of the events within a database or determine a sequence of different events starting at a particular position to reach the desired target position.

The appearance of the Events list depends on the selected event (see below). When you search for a sequence of events, this sequence is displayed as a list that is consecutively numbered. Otherwise, you can link any number of events. You can select the following operations by clicking in the Logical operation column:

Logical operation	Description
And	Only the positions where the two events are located are selected.
Or	All positions are adopted where at least one event exists.
Before	Only those positions are adopted where event 1 immediately precedes event 2.
After	Only those positions are adopted where event 1 immediately follows event 2.

Use the logical operators Before and After to evaluate periodic sequences of events. For instance, you can detect all maxima that occur after a slope. You can change the order in which the logical operations are carried out by using parentheses. If you do not use parentheses, the logical AND operation has priority over all other operations.

Double-click on an event to edit it or click on the button Add event.

The Event dialog box FlexPro offers the following events to choose from:

Event

Description

Level crossings

Points at which a specified level is crossed. In the Level field you can specify the level as an absolute value or in a percentage of the maximum value in the data set. The hysteresis specifies the minimum amount by which the signal amplitude has to rise or fall once a level crossing has been recognized in order for the level crossing to be accepted. You can specify this as an absolute value or in a percentage of the maximum value in the data set.

Extrema

Local maximum or minimum values. The Hysteresis is the minimum amount by which the signal has to rise or fall before and after an extreme so that it is accepted. You can specify this as an absolute value or in a percentage of the maximum value in the data set.

Global extrema

Global maximum or minimum.

Slopes

Points at which there are ascents of a specified minimum steepness. For this, the amplitude must ascend or descend within the X difference Delta-X by at least the amount Delta-Y. The Hysteresis is the maximum amount by which the signal amplitude is allowed to swing in the opposite direction within a slope. The values Delta-Y and Hysteresis can be specified as absolute values or in a percentage of the maximum value in the data set.

Slopes at level

Level crossings with specified minimum steepness. The references of the parameters correspond to those of the Level crossings and Slopes events.

Burst beginnings and/or ends

Beginnings or ends of periodic signal sections where a minimum amplitude is exceeded repeatedly within a Delta-X period. In the Level field you can specify this as an absolute value or as a percentage of the maximum value in the data set. The Delta-X field determines the interval [-Delta-X, Delta-X], within which at least one value must surpass the level so that the value to be examined is assigned to the burst. Use the event Values in bursts to extract the bursts in their entirety.

Values in bursts

All values that belong to a burst.

Positive peak beginnings and/or ends
Negative peak beginnings and/or ends

Beginnings or ends of positive or negative peaks. These are the areas where the signal amplitude is above or below a certain level, respectively. Values that are exactly on the level are also found. In the Level field you can specify this as an absolute value or in a percentage of the maximum value in the data set. This event finds the points where the peaks begin or end. Use the event Values above/below level to extract the peaks in their entirety.

Values above level
Values below level

All values above or below a given level.

Values in interval

All values within a given interval. Values that are exactly on the interval limit are also found. You can specify the interval limits as absolute values or as a percentage of the maximum value in the data set.

Void Values

Void floating point values.

Values in Spikes

Spikes are points in the signal where the amplitude increases/decreases by more than the specified minimum amount and decreases/increases again by more than the minimum amount at the latest after the specified number of data points. The minimum height can be specified as an absolute value or as a percentage of the maximum value in the data set. You specify the maximum width as the number of data points.

Time Period

Periodic time segments in a data set with calendar time values. For signals, a search is made in the X component. The data must either be calendar times or a timestamp must be stored in the data object that provides the data.

The following criteria are available:

Criterion	Range of Values
Second in the minute	0 to 59
Minute in the hour	0 to 59
Hour of the day	0 to 23
Day of the month	1 to 31
Month of the year	1 to 12
Day of week	Monday through Sunday
Year	1970 to 3000
Day of year	1 to 366

In the Orientation list box, specify whether you want to search for positive or negative orientation, or events with any orientation. In a level crossing, the positive event is, for instance, an ascending crossing through the level, and, accordingly, the negative event is a descending crossing.

Use the fields Delta-X min and Delta-X max to optionally determine the required minimum duration of an event must be present in order to be accepted. You can use this, for instance, to search peaks with specified minimum and/or maximum widths. When searching within a signal, these values must be specified as an X interval, but if searching within a data series, these vales will be interpreted as numbers of values.

In the Search in field, specify whether the selected events are to be found in the data set specified on the Data tab or in a different data set.

The Remove steady component option subtracts the constant trend from the data before the actual event search. All parameters with absolute amplitude values that you specify for the event then refer to the corrected data. Y values in the result are, however, output without this correction.

The Form complement option reverses the test statement. Here you can search for all values in the data set that do not apply to the selected event.

Select one of the following results for the search:

Result	Description
Extract values	From the data set specified on the Data tab, all values are extracted that satisfy the logical combination of individual events. The result is a data set with the extracted data. This selection is not possible for two-dimensional data, since the result may be different numbers of values for each column. You can display this output as a table, for instance, or use symbols, for instance, to mark events in a diagram.
Set others to void	Keeps all values of the data set specified on the Data tab, but sets the values to void that do not pass the event filter. You can also use this variant on data matrices and signal series, since the number of values is not changed.
Indices of the values	Does not pass the extracted values, but instead passes the indices of all values that fulfill the logical combination of individual events. The result is a data series. This selection is also not permitted for two-dimensional data. This variant is generally used to process the events in formulas. Via their indices you can create, for instance, the signal section between two zero crossings.
Event count	Passes the number of events that satisfy the logical combination of the single events. The result is a scalar value. This selection is also not permitted for two-dimensional data.
Boolean values	Generates a data set with Boolean values, which specify at which points the logical combination of individual events were or were not fulfilled. The data structure corresponds to the data set specified on the Data tab. Two-dimensional data are allowed in this case. This output is ideal, for instance, for visualization as an extra curve in a diagram.
X value of next event	From a specified X starting position a sequence of individual events is processed. The position of the previous event is used as the starting position for the search for the next event. If you do not specify a starting position, the search will start at the beginning of the data set. Two-dimensional data are not permitted here.
X value of previous event	Corresponds to the previous selection, but in this case the search starts from the end of the data set and heads to the left. If you do not specify a starting position, the search will start at the end of the data set.
Interpolate values	Keeps all values of the data set specified on the Data tab and interpolates the values that pass the event filter using linear interpolation. Use this selection to remove outliers, for instance. You can also use this variant on data matrices and signal series, since the number of values is not changed.
Set values to void	Keeps all values of the data set specified on the Data tab, but sets the values that pass the event filter to void. Use this selection to remove outliers, for instance. You can also use this variant on data matrices and signal series, since the number of values is not changed.
Extract segments Extract segments with X correction	Interprets the indices of the events found as a sequence of beginnings and ends of segments extracted from the data set and returned as a list. This also works for events that provide related sequences of indices, such as the event Values above level. Only the indices at which such a sequence begins or ends are then considered here. The variant with X correction corrects the X components of the segments so that they all start at zero.
Split into segments Split into segments with X correction Split into segments without borders Split into segments without borders with X correction	Interprets the indices of the events found as divisions at which the data set is split into segments, and returns the segments as a list. Here again, for events that provide the related sequences of indices, only the indices are considered at which such a sequence begins or ends. The variant with X correction corrects the X components of the segments so that they all start at zero. The variant without borders does not include the border segments at the beginning and at the end of the data set in the result.