Seth on Simulink
August 5th, 2008
Advanced Masking Concepts
Masking does more than just put a professional interface on
your algorithm. Simulink blocksets provide elaborate graphical user interfaces
(GUIs) to control the behavior of blocks. The block dialog can dynamically
enable and disable its GUI elements. Blocks can sprout additional ports in
order to accept parameters as input signals. The algorithm can rewire itself
based on the setting in the dialog. Did you ever ask yourself, “how can I do
that?”
This is the first in a series of posts that will introduce these
advanced masking concepts. I have created an example that has:
- A block dialog that enables and disables parameters based on user settings
- The option to add and delete ports to match dialog settings
This example builds on top of the basic concepts introduced
in my
recent post on masking. That post covered how to make a mask icon, and
provide a mask dialog to control the parameters used in an algorithm.
Two saturation blocks
The example I have created combines the dynamic saturation
block from a recent
post on libraries and the fixed saturation block in my
recent post on masking. The algorithms for these two blocks differ
slightly. The dynamic saturation block uses inports to provide the limits, and
the fixed saturation uses constant blocks to provide the same signals.
The Fixed Saturation block has a mask that sets the values
of the uplim and lowlim variables in the mask workspace.
A combined saturation block interface
To combine these blocks, I present the user with a simple
interface that allows them to control the source of their upper and lower
saturation limits. The value can be internal or external.
When internal limits are used, the value comes from the
Upper Limit and Lower Limit fields. When using external limits, those fields
are disabled, and inports provide the limit signals.
How does this combined block work?
Here is a test model that exercises some of the different
modes of this saturation block. The three saturation blocks in this model are
actually three instances of the same library block.
The top block uses the fixed limits. The middle block uses
an external source for the upper limit and a fixed value for the lower limit.
The bottom block uses external sources for both the upper and lower limit.
How does the same block behave in these three different
ways? To see how, you can look under the masks of these blocks and see the
blocks involved in the algorithm.
All three instances of the subsystems get their upper limit
from a block named system/up, and their lower limit from a block named system/lo.
In the case where the limits are fixed, the block is a constant block. In the
case where the limit is dynamic (external source), the block is an inport. To
achieve this, the block runs some M-code that replaces the Constant block with
an Inport when the source is external and replaces the Inport with a Constant
block when the source is internal.
In future posts we will examine:
- Mask dialog callbacks
- Mask initialization and self modifiable library blocks
- Best practices for masking
Now it’s your turn
Have you implemented masks using these concepts? Share your
experience here.
5:15 pm |
Posted in Libraries, Masking |
Permalink |
You can follow any responses to this entry through the RSS 2.0 feed.
You can skip to the end and leave a response. Pinging is currently not allowed.
Leave a Reply
|
I’ve done a few more advanced blocks like this. While a bit more complex than regular blocks, I’ve found them much easier to debug than doing the equivalent in S-Functions.
Although, I tend to hide unused parameters rather than just grey them out.