, Duration: 17:14*Enlace a Spanish version
This video is a continuation of [
Here, apart from multicriteria-optimization issues discussed in the cited video, we now consider which process variables should be chosen as “controlled variables”, i.e., variables for which a constant setpoint is set so closed-loop control tries to keep such a setpoint despite external disturbances trying to move away such variables.
The basic idea is that a given process variable is a good candidate to be a “controlled” variable if its optimal value (according to given design criteria and tradeoffs) remains constant even if exogenous factors (disturbances) do change.
Three “thought experiments” are presented, considering possible changes of outside temperature and different “importance” of energy-saving issues in the comfort vs. power cost trade-off.
In a first case, temperature will be a good option as a controlled variable; in a second case, operation at constant power would be the recommended option and, in a third case, neither constant temperature nor constant power are optimal choices, so a “setpoint optimization” block would be recommended to generate temperature and power operating points as a function of outside temperature (assumed measurable, i.e., this optimizer would amount to something very similar to a feedforward term, generalising the concept and the realm of application of basic feedforward ideas to complex systems).
NOTE: no one will do this in his living room (at least no one will be actually aware he is doing this); this video is a very complex way of telling that ”prioritising comfort amounts to controlling temperature”, ”prioritising energy saving amounts to setting a given cooling power when comfort is plainly unacceptable”.
*Link to my [ whole collection] of videos in English. Link to larger [ Colección completa] in Spanish.