In PID control, what is a self-balancing object?
Thread Content
In PID control, when setting the PID parameters, for systems that lack self-balancing capability, the integral term is not included; instead, proportional control or proportional-differential control is used. I’ve seen this statement in many sources, but in actual DCS setups, most systems use PI control, with a smaller number using PID control; pure proportional control or PD control is very rare. I would appreciate it if experts in automatic control could explain this, thank you.
This post was last edited by ylb913 on 2015-12-14 06:56. 1. When setting PID parameters, for objects that lack \"self-balancing capability\". ——What this sentence means is that there is no pattern whatsoever in the changes in parameters; therefore, it makes no sense to use integration to eliminate cumulative errors. Proportional control is the most fundamental form of control, and adding derivative control helps to regulate the speed of these changes, preventing them from happening too rapidly, as rapid changes can lead to problems. 2. In actual DCS settings, most use PI control, a small number use PID control, and there are very few cases of pure proportional control or PD control. ——P is of course the most important factor. For I, a value is usually assigned; in most cases, no one really cares whether it has an effect or not. The higher the value of I, the weaker its effect. If you don’t want the impact of these points to be too strong, it’s better to set a high value for I. I’ve never seen anyone set I to 0. Generally, D is needed for temperature control, as there is a significant lag in temperature control. But for other parameters such as pressure and liquid level control, when a stable flow rate is more important than stable pressure or liquid level, I often use differentiation.
A process without self-balancing capability refers to a situation in which, after the operating parameter remains constant and the measured value stabilizes, gradually changing this operating parameter fails to restore equilibrium, causing the measured value to continue changing. Liquid level is a representative example; some pressure control processes are also processes without self-balancing capability.
As a simple example, if a water tank is equipped with a control valve at its outlet and a process valve at its inlet, then the liquid level control circuit for that tank is generally self-balancing. If a pump is added in front of the control valve, then it becomes non-self-balancing at this point.