Co-simulation related capabilities

• Only fixed size co-simulation steps are supported by Simcenter Amesim. This means that even if the slave FMU allows variable co-simulation step size, the co-simulation masters do not use it.

• Serial explicit co-simulation scheme (also known as staggered sequential scheme) is not implemented.

• When a fixed step size numerical solver is selected in Simcenter Amesim, the co-simulation step size must be a multiple of the numerical step size. The ratio of the two step sizes, known as the multi-rate factor cannot take non-integer values.

• Event iterations are not supported, albeit being mentioned in the FMI 1.0 specification, it does not make sense in the context of co-simulation. This means that the co-simulation step size – provided by user through a dedicated parameter of the import submodel – must be strictly greater than 0, in spite of the fact that the corresponding capability flag can be enabled in the imported FMU interface description.

• Rejection of co-simulation step is not allowed, since the Simcenter Amesim solver does not allow to do rollbacks.

• Partial stepping is not supported. If the slave FMU computes successfully only a sub-interval of the co-simulation step, the co-simulation is aborted and an error message is reported.

• Higher order extrapolation of inputs is not implemented, only zero-order hold is used, in spite of the corresponding capability flag is enabled in the slave FMU. On output, higher order polynomial representation of output variables is neither retrieved nor used by Simcenter Amesim. Note Notice that if the FMU enables it explicitly in its Model Description, it is possible to use several instances of the same FMU for Co-simulation on the same Simcenter Amesim sketch.

Note

For FMI version 3.0, the following additional limitations apply:

• Event mode and intermediate update modes are not supported. Therefore FMUs containing clock variables cannot be simulated

• Early return (or partial stepping) is not supported