Re:PID Without a PhD (2016) [pdf]
by TeMPOraL on Thursday April 02, 2020 @03:28AM
(#22756943)
Not a process engineer, but I'm guessing that if you're looking at correlation between temperature data vs. AC power, and AC is reacting to changes in temperature (i.e. derivative, D in PID), you'll see AC ramping up power before the temperature meaningfully increases. A naive analysis could then conclude that it's AC that causes high temperature.
[ Parent
| on HN ]
Or more simply than that, if you only ever look at "cold" and "hot" rooms (e.g. analysis of cancerous and non-cancerous patients) you're never going to see the dynamics at all.
But more fundamentally, I believe there is a theorem which asserts that it is in general impossible to extract the open loop behavior of a system (i.e. a model of how the system would respond without a controller) solely from closed loop data (i.e. data collected from the system when the controller is on.)