M-Target for Simulink®

Both process (controlled system) as well as controller and control programs are modeled in Simulink^®. Using application specific toolboxes reduce the time spent developing the system. Through the numerous domain-specific extensions for physical modeling, such as Simscape Electrical^TM, Simscape Fluids^TM, Simscape Multibody^TM etc., this can also be done directly in a description environment familiar to the respective process expert. For mathematical modeling, MATLAB^® and Simulink^® also offer a range of ready-to-use algorithms. Established service companies offer additional expertise in all domains.

Already at this phase, interfaces to IEC 61131-3 programs, C/C++ modules, and other software components can be taken into account and modeled. This creates a complete digital representation of the later application program, including its networking.

The resulting complete model is now simulated offline on the computer. Comprehensive test series of possible operating states and error situations can already be run through at this phase. Iterative model adjustments and new simulations follow seamlessly.

The powerful methods/solvers of MATLAB^® and Simulink^® enable the numerical computation of complex dynamic systems. Extensive visualization functions such as curve plots, surface plots, and animated 3D-CAD models support the analysis, validation, and optimization of simulation models.

In this step, the process model is first separated from the actual control part. Then a mouse click triggers automatic code generation and preparation of the application program for the real-time system.

The finished application program is loaded directly from the Simulink^® user interface onto the controller. In this process, in addition to the actual application, the required communication and program interfaces are also taken into account automatically.

Alternatively, libraries for IEC 61131-3 programs or for C/C++ can be generated. In contrast to directly executable software modules, these are used in the respective development environments for further application creation.

The generated real-time program now runs in the controller. Via the communication interfaces integrated during code generation, it exchanges data directly with Simulink^® on the PC. In what is known as "External Mode", the actual process values (variables, channel values) are available online in Simulink^®. At the same time, variable values and internal parameters of the Simulink^®-blocks can be changed in the real-time program.

In this mode, the Simulink^® model is only the graphical front end for visualization and parameter specification, while the algorithms run on the control system. In the real execution environment, the solution found can thus be verified and further optimized.

A suitable test option is available for each development phase: In addition to software-in-the-loop tests (SIL), hardware-in-the-loop tests (HIL) as well as co-simulations can be set up. This allows functions and system behavior to be safeguarded at an early stage and risks during commissioning to be significantly reduced. For iterative improvement, it is possible to switch to a previous process step and make refinements there.

As soon as a satisfactory result is achieved, the project engineering PC can be separated from the controller. This continues to run autonomously in real time. All interfaces to parallel-running application programs on the same (or other) controller(s) remain in place. The published process variables can be depicted through the engineering tool SolutionCenter or in visualizations.

The application program generated from Simulink^® can be operated permanently in parallel with existing IEC 61131-3 and C/C++ applications. Updates or the exchange of application programs can be performed without restarting the controller. This increases plant availability and minimizes downtime during maintenance and further development.

Comprehensive online diagnostics are also available directly from Simulink^® during operation. This allows process data to be analyzed, parameters to be adjusted, and optimizations to be made without leaving the modelbased development workflow.