[TRNSYS-users] Enquiry for the Type 683 Problem

[Diego A. Arias]

Dear Ling,

During each time step, TRNSYS iterates over all of the components and 
only until all of the inputs and outputs have converged it advances to 
the next time step.  You can take advantage of this feature by creating 
loops of information.  By using a loop you can calculate an input by 
controlling an output.

I will assume that the input that you want to calculate is Input_1_683, 
by controlling Output_1_683 (I am not using the actual numbers assigned 
to Type683 for clarity reasons).

Lets define a new type, TypeAAA where you can specify as inputs and outputs:
Input_1_aaa = User defined stream inlet temperature
Input_2_aaa = Output_1_683
Output1_aaa = updated process air outlet humidity ratio

The equations inside TypeAAA can be of this kind:
if (Input_2_aaa - Input_1_aaa) > DELTA
  Output_1_aaa = Output_1_aaa + DELTA
endif
Although you will have to work on the details, the idea is to update the 
process air outlet humidity ratio in the direction in which the output 
from Type683 has changed.

Finally, you connect Output_1_aaa to the Input_1_683.

To make this strategy work, you need to give good initial values to Type683.
I hope this explanation was not too confusing.  Please feel free to 
email me directly if you want to implement this strategy or if you 
encounter any problems while implementing it.
Best regards,

Diego


HuiLing.Tay at csiro.au wrote:

> Dear Trnsys Users.
>  
> Hi! I am trying to model a rotary desiccant wheel dehumidifier, hence 
> Type 683 was chosen. In this model, the required inputs are the 
> process air and regeneration air stream mass flow rates, the process 
> stream inlet conditions (dry bulb temperature and absolute humidity 
> ratio), the desired process stream outlet humidity ratio, and the 
> regeneration stream inlet absolute humidity ratio.
>  
> However, I would like to have the regeneration stream 
> inlet temperature to be specified by the user and the process air 
> outlet humidity ratio to be calculated by the model (i.e. not a set 
> point). This could not be done in this model. Have anyone come across 
> this problem before and developed a new model to suit this requirement?
>  
> Thank you in advance for any help.
>  
>  
> Best regards,
> Ling
>
>------------------------------------------------------------------------
>
>_______________________________________________
>TRNSYS-users mailing list
>TRNSYS-users at engr.wisc.edu
>https://www.cae.wisc.edu/mailman/listinfo/trnsys-users
>  
>


-- 

Diego A. Arias
TRNSYS Coordinator
Solar Energy Laboratory
University of Wisconsin - Madison
1500 Engineering Drive
Madison, WI 53706