How to understand cascade control?

The cascade control system is named due to its structural characteristics. It operates by two controllers, a main one and a secondary one, working in series. The output of the main controller serves as the setpoint for the secondary controller, whose output is used to operate the control valve in order to achieve constant-value control of the variable.

The principles for selecting the primary variable are the same as those for selecting the controlled variable in a simple control system. The selection principle for secondary variables is: (1) There should be a certain inherent relationship between primary and secondary variables, and changes in the secondary variables should be able to significantly affect changes in the primary variable ； (2) By selecting the auxiliary variables, the resulting auxiliary loop can incorporate the main disturbances of the system ； (3) Where possible, the secondary loop should include as many major disturbances as possible, but the secondary variables should not be too close to the primary variables ； (4) The selection of secondary variables should take into account the matching of the time constants of the primary and secondary objects to prevent resonance

The purpose of a cascade control system is to better stabilize the primary variable and keep it equal to a set value; the primary variable is the output of the main loop, which is why the main loop constitutes a constant-value control system. The output of the secondary loop is the secondary variable, while the setpoint for the secondary loop is the output of the main controller. Therefore, in a cascade control system, the secondary variable is not required to remain constant; rather, it is supposed to change in accordance with the output of the main controller. Hence, it constitutes a follow-up control system.

The purpose of a cascade control system is to stabilize the primary variable with high precision. High requirements are placed on this primary variable, and any residual error is generally not allowed; therefore, a proportional-integral control law is usually chosen for the main controller. When there is significant lag in the system, an appropriate differential action can also be introduced. In a cascade control system, the requirements for the secondary variable are not strict. During the control process, the secondary variable continuously changes in response to the output of the main controller; therefore, a proportional control law is generally sufficient for the secondary controller, with appropriate integral action being introduced if necessary, while differential action is usually not required.

Our facility has a cascade control loop that consists of a bottom-of-tower level control system and a flow regulation circuit; the level control serves as the main loop, while the flow regulation circuit functions as the secondary loop. Due to the small size of the liquid accumulation tank at the bottom of the tower, the liquid level fluctuates significantly, resulting in alerts for either high or low levels from time to time. Is it possible to add differential control to the main loop, while avoiding the use of integral control in the secondary loop? I also have a minor question: would using a level controller and a control valve together to form a circuit yield better control results? Is such a design feasible?

Everyone has spoken very professionally; I will explain what cascade control is from a different perspective. Please correct me if I’m wrong: The essence of cascade control is to address each key issue separately. For example, in a level control system, I can adjust the liquid level by controlling the inlet flow rate. However, if large fluctuations occur, it turns out that there is an interfering factor that has a significant impact on the liquid level – temperature. If we use only simple feedback control, then we proceed as follows to identify and eliminate the effect of temperature on the liquid level: temperature change --> flow rate change --> liquid level change --> comparison with the set value --> detection of deviation and correction. This raises a problem: we control the liquid level by means of changes in flow rate caused by temperature changes, which is like trying to scratch an itch through thick boots. This type of control lag can lead to oscillations in severe cases. Well, let’s analyze each issue specifically now. Temperature has a significant impact; fine, we will address temperature control separately. We will compare the output of the temperature control loop (the secondary loop) with the output of the main loop, so as to respond more quickly to changes in temperature.

The purpose of a cascade control system is to stabilize the primary variable with high precision; high requirements are placed on this primary variable, and any residual error is generally not allowed. Therefore, a proportional-integral control law is usually chosen for the main controller. When there is significant lag in the system, an appropriate differential element can also be introduced

I feel like what was explained on the 7th floor was good, but I still don’t quite understand it. Is it because my foundation is too weak? It seems I need to work harder

Simply put, it uses two inputs to adjust a value

In boiler control, water volume, steam production rate, and liquid level are the three impulses for control.