Co-simulation For Heterogeneous Simulation System And Application For Aerospace

With the increasing extension of application in the industry of computer technology, simulationhas become the main technical method in the design of industrial products. With the advent oftechnology and industry, modern industrial products have become more complex systems, whichconsist of control system, hydraulic system, electronical system,mechanical system and so on. Thus,simulation plays a more important role than the past. In this paper, the main focus is multi-disciplineco-simulation. The key technology of co-simulation is design of interface algorithm and workflow ofsimulation. The main contribution of the paper includes the following.First, the concept of co-simulation is described which includes coupled-simulation, co-simulation,coupled-solves and distributor-simulation. A new definition which is called ‘multi-solves-simulation’is provided. The process of multi-solves-simulation is assumed, and then the workflow and calculationflow of multi-solves-simulation is described based on the process assumption. At last, the workflow ofserver and client during multi-solves-simulation will be shown when the multi-solves-simulation isrunning.Second, the form of mathematical description of subsystem models of co-simulation is shown inthis paper. The zero-stability is used to define the stability of mathematical description ofco-simulation. The coupled integration of subsystems is zero-stability when the zero-stability methodis used. The form of mathematical description of coupled simulators can prove that the coupledsimulator system is zero-stability, too.Third, interface algorithm and implement method are provided. Interface algorithm consists ofstrong coupling and loose coupling. The strong coupling requires an iteration procedure to ascertainuser-defined convergence criteria. Multi-rate co-simulation is used to solve the problem of coupledrigid systems. Then two examples are used to show the implement method of multi-solve-simulationinterface. The first one is the example of co-simulation between Matlab/Simulink and Flowmaster. Thesecond one is the calling method of co-simulation of MWorks.Finally, two cases are used to prove multi-solver-simulation is available. The first example is thesimulation of attitude of four-rotor aircraft which is modeled in LMS Motion, AMESim andMatlab/Simulink. The second one is the simulation of fuel control system, hydraulic system, motorand electronic system of aircraft based on Saber, Simplorer, Flowmaster, AMESim andMatlab/Simulink.