Model-Based Design for Automation

Project-oriented
system development is currently the most common approach among OEMs. As a
result, existing automation structures are rarely integrated in the development
cycle as a normal part of the design process. Nevertheless, model-based design methods are finding
their way into industrial automation.

Despite
the initial costs involved in system analysis and model design, the investment
quickly pays off for the OEM. For example, once simulation models have been
created, they can be updated for developing future machine generations at a
relatively low cost. An added benefit is that expert knowledge spanning
multiple departments can be represented with a simulation model, providing the
company documentation of its accumulated know-how. Furthermore, simulation
helps detect conceptual errors early in the development process without having
to construct expensive prototypes.

Development
Tools Partnership

Collaboration
between B&R and The MathWorks on tools for model-based design,
simulation and automatic code generation is one example of the technology
partnerships being created to provide OEMs with the tools needed for more
efficient product development.

Model-Based Design for Automation

Click here for a larger version

B&Ra euro ~s open architecture project development tool -
Automation Studio - is connected to The MathWorks Simulink, allowing the
transfer of automatically generated source codes to machine control systems. To
bridge the worlds of simulation and project development on the industrial
controller, B&R developed Automation Studio Target for Simulink. This tool
provides product and systems designers with a uniform interface between
MathWorks software such as MATLAB, Simulink and Stateflow, and B&R's
Automation Studio.

Consistent
Workflow

Connecting
automation systems design tools and model-based design software allows for the
creation of interface blocks to handle communication between the automatically
generated program unit and the rest of the automation system components used in
the project. Standard process variables are used to exchange data with other
program units, visualization elements, hardware I/Os and drives.

A
common challenge for automation systems designers is selecting the correct
settings for the code generator - a process which take hours and interrupt the
designer's workflow. Configuration blocks in the automation systems design
software handle all the necessary settings once it has been pulled into the
simulation model.

The
automatic code generation itself only takes a few moments, with the code
commonly being completed automatically as the system is designed. As a result,
the generated PLC source code is integrated into the project without requiring
the user to copy the generated files or create process variables. Even for
complex Simulink models in the range of several thousand individual blocks, the
process only takes a few minutes. To implement this manually in PLC code would
require months of work.

Using
such combined design tools, it is recommended that any changes or additions to
the project be made in the simulation model. Doing so keeps the entire project
up-to-date and documented via the graphic structure of the simulation software.
Changes can then be transferred to the PLC using the automation systems design
environment.

The
biggest advantage, however, is not just in the speed with which these models
can be created and deployed into a systems design project, but in the

Comments (0)

Please log in or register to post comments.
By submitting this form, you accept the Mollom privacy policy.
  • Oldest First
  • Newest First
Loading Comments...