PID control
Thread Content
How can PID control ensure that the process variable fluctuates little, while also keeping the fluctuations in the controller’s output at a minimum? It is easy to keep the process variable stable under PID control, but it’s more challenging to ensure that the controller’s output does not fluctuate significantly. I have a case here where temperature and water supply are adjusted using valves (since adding water to the furnace causes it to cool down, this is a reverse effect). The water coming out of the furnace returns to the large tank via an inverter-driven pump; therefore, there are two PID controls in this system. When the first set of PID controllers is used, the process value is well controlled, but the valve opening remains unstable, fluctuating between high and low values, which in turn causes the speed of the inverter-driven pump to also vary. They’re worried about damaging the equipment, so they suggest adjusting it manually: lol
After all, the stability of one side is maintained through the instability of the other side – so why is such a requirement? If you want both to be stable, it’s only possible if the input conditions are already stable. Furthermore, control like yours is actually used to control temperature; since the precision requirements aren’t high, the integration time can be set longer and the dead zone can be made larger.
The stability they refer to is in comparison to manual adjustment: when the valve is set at 70%, the frequency of the variable-frequency pump at 35 Hz is generally sufficient, of course this assumes that operating conditions remain relatively stable; if the conditions change, then manual adjustment will be necessary. My idea is to try to stay as close as possible to their ideas so that the regulators don’t experience large fluctuations. Let’s discuss whether this is possible
This should be considered from the perspective of balanced control.
Neither the controlled variable nor the control variable (manipulated variable) should exhibit excessive oscillation, so uniform control can be considered. The sensitivity of the control system is then reduced, that is, the proportional gain of the controller is decreased (the proportionality setting is increased) until satisfactory results are achieved.
I’ve heard of equilibrium control for the first time; I searched Baidu for information on this topic but found not much. Are there any specific cases that can be used for analysis? After all, resolving problems in production and assisting with production processes is the key to success.
It is uniform control, not balanced control
It seems that uniform control is the only option; increase P and I to make the regulator’s output changes slower. Of course, process value control will also lead to a decline in quality.