Benchmarks for mechatronic models / Gemma Ferdinand Kaunang
VerfasserKaunang, Gemma Ferdinand
Begutachter / BegutachterinBreitenecker, Felix
UmfangXIII, 169 Bl. : Ill., graph. Darst.
HochschulschriftWien, Techn. Univ., Mag.-Arb., 2008
Schlagwörter (DE)Informatik/ Mechatronik / Matlab / Dymola / Modellica
Schlagwörter (EN)Informatic / Mechatronic / Matlab / Dymola / Modellica
URNurn:nbn:at:at-ubtuw:1-23121 Persistent Identifier (URN)
 Das Werk ist frei verfügbar
Benchmarks for mechatronic models [3.12 mb]
Zusammenfassung (Deutsch)

Before a new product is launched to the market, a company have to test the product and make sure that the product is ready and presentable, especially if the product is a problem solving system, such as control system, automated system or self-learning system. Testing is a highly cost consuming yet unavoidable activity. Therefore to reduce cost of production, company will use methods like simulation to test their product.

There are so many simulation softwares in the market which offers different abilities and advantages. The various choices has makes it even more difficult for end-users (company) to choose which one is more suitable and useful for the company. On this thesis Three "comparison problem" based on electrotechnic will be compared each other by using four simulations software (Matlab/Simulink, Dymola, Mosilab and SimulationX), with different approaches to model of the system.

The method used for this research is a literature study to have a deeper understanding about the behaviour and algorithm of the code from 4 different simulation softwares, the design model of three comparison problems and simulates these models to find the most suitable plot result.

After a thorough research of these three comparison problems, conclusion can be made as follow: - Matlab is the only simulation software which able to simulate all calculation of matrix.

- Stateflow, stategraph and statechart which can only model the equation with switching state, the harder the equation is, the longer time required to simulate the equation.

- Dymola has the most variation type of modelling that needed in this thesis, the fastest time simulation is by dymola electrical model to simulate task d in comparison 3 =0,015s, - SimulationX took the longest time in simulating task diode C in comparison 20 = 1307,6718s, Type of designer block in simulationX is very useful feature for expert user in defining their code in new element type - Below are the simulation timing ranking from fastest to slowest type of modelling:

a. Textual mode b. Electrical model c. Hybrid model d. Stateflow/stategraph/statechart model