Dynamic Modeling Laboratory

Chapter Three

3. Implementation of model

Although the improvement of complex design system involves the development of representations for design tools and mappings, design tool environment models, and project flow definitions, these significant improvements in systems cannot be technologically advanced in isolation. After outlining the various frameworks, the developer(s) of any given model must be able to examine what is under creation to prove that the prototype meets its required purpose standards. For results to be achieved, the proposal must commence with test data from the core device that it will be used in. Mappings between different plans must be tried to guarantee that the outcome of plotting from the prototype is comparable to the original. Design device environment prototypes must be tested to guarantee that the design implement begins and end as anticipated and that they experience no problem with the figures obtainable from the end product.

Get Help With Your Essay

If you need assistance with writing your essay, our professional essay writing service is here to help!
Find out more about our Essay Writing Service

To make the modelling possible it involves the presence of testing environments that permit prototypes to be speedily instantiated and confirmed to warrant that interactions and reliance are as anticipated. To test plotting requires a system, which mimics the functionality of a complete plotting system, joined with prototype visualizers to enable the accuracy of the plotting to be determined. Alike testing is needed for scheme prototypes and design tool environment prototypes.

To test prototypes in each of these disparities would entail very dissimilar testing environments. However, if a particular model conceptualization tool were generated for testing all joined design systems during development; the final product could be itemised in any prototype with the inventor being sure of the accuracy of the requirement. Having a single set of testing devices also permits faster integration between the testing devices, with several benefits from the different phases of testing.

3.1 Setting and Principle of model

For a simulation to take place, there be existence of an environment. This environment is usually achieved through configuration that take place during the simulation process. An example of the configuration will be explained in the FMU export from Simulink that clearly explain the environment under which the FMU export is facilitated.

3.2 Implementation in Dymola

Introduction

Dymola with refers to the Dynamic Modeling Laboratory is a device used for modeling and simulation of incorporated and complex systems used in industries such as aerospace, automotive , robotics and other applications. With its state of the art engineering, Dymola’s abilities display novel and innovative answers for prototyping and simulation, as it is probable to simulate the vibrant conduct and intricate relations amongst structures of various production fields, such as mechanical, electrical and other control systems. This implies that users of Dymola can construct prototypes that are more joined and have simulations results that depict reality. Other highlights that can be realized of Dymola are those of Handling of large and complex multi-engineering models. Faster modelling by graphical model composition and faster simulation through symbolic pre-processing can also be achieved for purposes of increased productivity. Other befits of Dymola are it support for Open user defined prototype modules, Open interface to other applications, 3D Simulation and Real-time simulation, but just to mention a few of its benefit.

3.2.1 Environment

The Dynamic Modeling Laboratory (Dymola) setting practices the open Modelica demonstrating semantic, which implies it is open to its users. Dymola users are, therefore, free to develop their own model libraries or modify the ready-made model libraries as desired to satisfy their individual user’s unique modeling and simulation needs. With Dymola being flexible, it makes more of an adaptable device. Flexibility, therefore, brands Dymola seamless for prototyping and simulation of novel substitute strategies and skills now and in the future.

4. FMU export from Dymola

The objective of this sub-topic is to illustrate the steps that one would take when he/she is intending to export prototypes from physical simulation settings as FMUs. To be able to perform an export from a Dymola an individual would need to perform two very crucial steps. One is that of adjusting the simulation model interface and secondly perform an export the simulation model as an FMU.

To achieve the export from Dymola, proceed as explained in the following:

1. First, adjust the interface (ports) of your simulation model in existence from a physical modeling tool. It is important to note that this process of adjusting the interface must be performed in a signal-based way for purposes of properly exporting a model/ models as an FMU. The interface of the desired simulation model will be defined by input and output signals. For purposes of reliability, efficiency and better results, an individual install sensors. The installed sensors are also used to amount certain prototype conditions and actuators in order to put on physical aspects to the prototype.
2. The second step is to Export the simulationn model as an FMU by using the FMI export functionality of your physical simulation tool. For example, the exporting functionalities of Dymola (the options for the FMI export) can be found in the Simulation Setup GUI. In export process, there are usually three settings that need to be performed. The first setting is that of sectioning a Type. The second setting is that of choosing an FMI Version and finally choosing further Options.

For Type, one can set either the environment for Model exchange (FMI-ME) or

Co -simulation (FMI-CS) as Model exchange exports. This is because export can either be performed using model exchange or co- simulation. In the model-exchange setting, the FMU comprises only the prototype and no slave. Therefore, the slave of the introducing simulator is used. In the co-simulation setting, the FMU comprises both the prototype and a slave. Here the importing simulator performs as the main of the co-simulation. The prototype without slaves and Co-simulation exports a summarized prototypee and slave.

For Version, selecting 1.0 will ensure compatibility with V1.7 of the Modelon FMI Toolbox. In the case of Choices, it is not necessary to include the basis cypher or mass outcome in mat file.

3.3 Implementation in Simulink

As in the event of Dymola, Simulink can also be used to implement different prototypes in different environs. For example in the application of control procedure, the control procedures is established in a simulation setting (MATLAB/Simulink) and verified on simulation prototypes. After that, MATLAB/Simulink can be coupled with a PLC, and the procedure is verified on a physical prototype. This linking offers real-time communication amongst MATLAB/Simulink and the PLC (B&R 2005).

Control procedures have to be written in a worldwide programmable language supported by both MATLAB and PLC, because of its broadcast into the PLC.

The Control procedures established in the simulations can be used in a different area such as in the control of heating devices at home for purposes of temperature regulation. the control algorithms can also be used in industries among other place. The presence of Simulink has been a major boost in innovations.

3.3.1 Environment

Many advantages can be associated with Simulink. The advantages experienced by Simulink users are in its ability to provide the right set of tools for fast, accurate modeling and simulation. Simulink is designed to facilitate extensive features of block library for developing complex models. It is also designed to be convenient tools for monitoring simulation results, and tight integration. This is facilitated by the presence of MATLAB, which aids in accessing the most comprehensive collection of design and analysis tools.

3.3.3 FMI-Toolbox

The FMI Toolbox for MATLAB fits in Modelica-based physical prototyping into the MATLAB/Simulink surroundings.

FMI Toolbox offers the following core features, FMI toolbox permits the Simulation of assembled vibrant prototypes, FMUs, in Simulink. An FMI-compliant device such as OPTIMICA Studio by Modelon, SimulationX or Dymola, may generate fMUs. The Simulink FMU block offers realization of limits and input values as well as block results. FMI toolbox also enables Export of Simulink prototypes to FMUs. FMUs may also be simulated in FMI accommodating simulation device such as SimulationX or Dymola.

FMI toolbox may also be used for the Simulation of assembled vibrant prototypes, FMUs; using MATLAB’s built-in integrators, for example, ode45 and ode15s. This piece makes FMI Toolbox beneficial for operators without them having to contact to Simulink.

The other advantage of an FMI toolbox is that it facilitates the Static and dynamic analysis of FMUs through design-of- experiments (DoE) functions for optimization, calibration, control design, and robustness analysis. The dynamic analysis features require the MATLAB Control System Toolbox. The FMI Toolbox supports FMI import for Model Exchange and FMI for Co-Simulation. FMI Toolbox also supports FMI export and a DoE analysis for Model Exchange 1.0. In an FMI Toolbox, Simulink models can be exported as Model Exchange.

•FMI Toolbox also supports improved handling of FMU blocks that are supported by Simulink Coder/Real-Time Workshop usually stored in a Simulink library.

3.3.4 FMU export from Simulink

A Simulink prototype can be pass on as an FMU and introduced in an FMI-compliant device such as OPTIMICA Studio by SimulationX, Dymola or Modelon. This section describes how a Simulink model can be exported as an FMU.

Code from a Simulink model is generated by Simulink Coder/Real-Time Workshop and is then wrapped in an FMU for Model Exchange 1.0 or Co-Simulation 1.0.

There are various steps required to export an FMU for Model Exchange from Simulink. The first step is usually to select the build target. This is usually done by opening the Configuration Parameters dialog. Then go to the Real-Time Workshop/Code Generation tab depending on the MATLAB version an individual is using at that particular time. From Browse button, select the System target file. The final step that takes place before exportation can take place is that of selecting fmu_me1.tlc from the browser dialog for exporting the FMU as Model Exchange or fmu_cs1.tlc for Co-Simulation, when either of the two is selected, click OK to export.

However, FMU export limitations such as The FMU target uses the code format S-function and target type non real time. This means in general that the same limitations of Simulink Coder´s native S-function target, rtwsfcn is applied to the FMU target.

Complex input and output ports are not supported. There is no corresponding data type in the FMI standard.Another limitation of FMU is that Enumeration data types are not supported for example the Enumerated Constant block is not supported. Discrete variables (variability attribute set to discrete) may change the value at instants other than during initialization or at event instants. Support for precompiled S-functions is only supported for export of Model Exchange FMUs and not co- simulation.

3. Co-simulation

The main aim of co-simulation is to come up with a user-friendly interface type for connecting/joining simulation tools in its environs. The data exchanged in this models subsystems is limited to distinct communication targets. The communication interval among two sub-systems is controlled autonomously by respective subsystem solvers. The master algorithm is usually responsible for controlling the exchange of data among subsystems and the harmonization of complete simulation slaves (solvers). In this case, basic master algorithms and complex ones are supported. It is imperative nonetheless to note that the master algorithm is not a part of the FMI standard.

Dymola 2013 and later supports export of prototypes (slaves) with built-in numerical slaves according to the FMI for Co-simulation specification. The SUNDIALS suite of numerical slaves (version 2.4.0) is used in the co-simulation FMUs. In Dymola 2013 and later, the translateModelFMU command will produce an FMU that supports both the FMI for Model Exchange requisite and the FMI for Co-Simulation slaves interface where by all responsibilities will be present in the DLL.

Model simulation is also supported in Simulink. It is, however, important to note that when Simulink FMU block co-simulation FMUs with modelDescription attribute canRunAsynchronuously is set to true, they are usually not supported.

References

[Jak2003] Johan Åkesson. Operator Interaction and Optimization in Control Systems. ISRN LUTFD2/TFRT–3234–SE. Lund University. Sweden. 2003.