Simulink model parameter update and observations at discrete times

I have a Simulink model M which includes a parameter p. Assume that I want to simulate M over the the time period [0, T]. Also assume that I get an output y from the model M.

Now consider discrete time points [0 t_1, t_2, ..., t_n] in the time interval [0, T]. At each t_i I want to change p in the model and want to get the value of y.

How can I do this?

Answer 1

[Paul]

I think you can do exactly what you want using the new (since 2024a) simulation object.

[EA]

Is it possible to give a concrete example?

[Paul]

Upon further reflection, using the simulation object might not be what you want; it depends on how the simulation is configured.

Does the simulation use a fixed step solver (in which case the t_i must be exact multiples of the solver step size)?

If the simulation uses a variable step solver, is the simulation configured such that each t_i is at a major step?

Depending on what p actually is, another option might be to use logic internal to the simulation to compute p as a function of time and then use a Parameter Writer to update p.

[EA]

Simulation uses a variable step solver. t_i's sare time steps determined by the user in the interval [0, T]. t_i's can be assumed to be larger than the the minimum time step required for numerical simulation.

[Paul]

If the t_i values are not currently forced to be at major time steps, the you can change add logic to the model to force that to be the case and adjust p at those times. What happens after that depends on what p actually is and how it's used in the model. Or, if you force t_i to occur at major time steps then you can use the simulation object.

Or, you can use the workflow desribed at Use Model Operating Point for Faster Simulation Workflow to simulate up t_1, adjust p programatically, simulate to t_2, etc.