• Official comment
Jason

The initial value of an Integrator, Reservoir, and Pool element must be static. This means that the value cannot change during the simulation and it must not be a function of a dynamic property of GoldSim. This is typically not a problem but there are times when you must refer to data using an expression that is dynamic. For example, what if you wanted to look up a value in a time series by referring to a specific time? This type of function is dynamic and trying to use this in the Store's initial value would cause an error. To get around this, use the Replace feature of a discrete change and link that to the Store element. This model uses this approach for Store type elements, which can be copied into your model.

• Diahann Butler

This solution appears to address the current challenge with my simulation but I am not understanding how to utilize it properly. In my current model, I have defined a time series element with times and values to represent the discrete items (voters) for a given time period. I am using this time series to drive the count of a Timed Event. For my timed event element, I am using the occurrence type “defined cumulative event count”. Ideally, I would like the number of discrete items for the start of the next time period to be set to my defined amount in my time series. For example, at t=1/ items = 100; at t=2/ items =350, and so forth.  It appears this model would help me achieve that desire (by changing my simulation for run from a pool or reservoir) but I am not grasping how to use your solution. I was anticipating the sum of the volume in your Reservoir_Observed to equate the total volume on the Reservoir1 (if there is no withdrawal) and at each time event the new initial value is set based on the values listed in Reservoir_Observed data. Is that how it should work and if so what else do I need to define in order to utilize your Dynamic Initial Quantity model?

• Jason

Diahann,

The solution mentioned above will allow you to take the final value calculated at the end of a Monte Carlo simulation and then apply it to the starting value of a subsequent simulation for the starting point. This does not have anything to do with propagating values through subsequent time steps. It sounds, based on your description, that you are trying to carry values forward from the previous time step to the next in time. This is a very different situation but I still don't fully understand what you need. Do you mind starting a new Forum post here. With a full description of what you need to accomplish in your own words? We will look at your description and try to provide some advice into how this might be done in GoldSim. Thanks for your patience.

-Jason

• Diahann Butler

Hello Jason,

I started a new forum post entitled "creating discrete events based on a defined distribution". I attempted to explain my simulation in case my question is not understood. Thank you for your time and patience.

-Diahann

• Jason

Diahann,

Thank you for posting more detail in the forum.

-Jason