Specify preferences for how Diagnostic Feature Designer performs computations and where the app stores results. Some options are visible only under specific conditions.
By default, the app processes entire signals. You can also divide your signal into uniform segments — or frames — for sequential processing. Frame-based processing allows you to localize abrupt signal changes, and to compute the full set of prognostic ranking features for remaining-useful-life calculations.
Full signal — Process entire signal in one continuous time series. For example, suppose that you have an ensemble of 20 vibration signals and select the feature Time Domain Features > Signal Features > Mean. You create 20 values of the mean, one for each member signal.
Frame-based — Process signal as a set of individual segments defined by frame size and rate.
Frame size (FS) — Specify the time interval in seconds over which data is provided.
Frame rate (FR) — Specify the time interval in seconds between frame start times. This interval is equivalent to the frequency with which new frames begin.
Select a frame policy — Use a previously
specified frame size and frame rate pair. The app stores each frame setting pair
in a frame policy. You can maintain multiple frame
policies, and select from them using this menu. To create a new frame policy
from your current frame size and frame rate entries, select
For example, suppose that you have an ensemble of 20 vibration signals, and
each signal is at least 100 seconds long. If you enter Frame size
50 seconds and Frame rate
40 seconds, each signal contains at
least three frames:
Frame 1: 0–50 seconds
Frame 2: 40–90 seconds
Frame 3: 80–L, where L is the minimum of 130 seconds or the signal end time.
If you compute the mean feature in frame-based data handling mode, each signal contains at least three mean values, with one mean value for each frame.
Whenever you perform a processing operation on your data, you create a new derived variable or feature. By default, the app writes derived variables and features to the same dataset or folder as your original data.
When you import an ensemble datastore object, you have a choice of where to store results. The following options are available only for ensemble datastore objects.
Write results to the same folder as the original data —
Write derived variables and features to the external folders referenced in your
ensemble datastore object. If you are using a
fileEnsembleDatastore object, the object must include a reference to a
write function specific to your data structure. You do not need a
write function for a
Return all results in a local dataset ("InAppData") — Write derived variables and features to the local workspace of the app. Select this option if, for example:
You want to keep your source files pristine at least until you have finalized your processing and feature generation.
You do not have write permission for your source files.
You do not have a
write function, and you are using a
The process of writing the results back to the source files is slow.
To retain your local results at the end of the session, use Save
Session. You can also export your results as a
table to the MATLAB® workspace. From the workspace, you can store the results in the file, or
integrate results selectively using ensemble datastore commands. For more information
on using ensemble datastore objects, see:
A key feature of Diagnostic Feature Designer is that the app performs operations on all ensemble members collectively. For most of these operations, member computations can be performed independently of one another. When you have many members, using parallel processing can improve performance significantly.
If you have the Parallel Computing Toolbox™ installed, you have an option to invoke parallel computing in the app. Select Use parallel computing.