Adaptive Control System

Adaptive control system adjusts to changes and modifies its parameters to fit the prevailing circumstances. The adaptive control system is based on the use of a microprocessor as the controller. Such a device enables the control mode and the control parameters employed to be adapted to fit the prevailing circumstances, modifying them as the circumstances change.

An adaptive control system can be considered to have three stages of operation:

1. Starts to operate with controller conditions set on the basis of a presumed condition.
2. The desired performance is continuously compared with the actual system performance.
3. The control system mode and parameters are automatically and continuously adjusted in order to minimise the difference between the desired and the actual system performance. For instance, in a control system operating in the proportional mode, the proportional constant K_p may be automatically adjusted to fit the circumstances, changing as they do.

Adaptive control systems can be implemented in a number of forms. Three often used forms are:

• Gain-scheduled control
• Self-tuning control
• Model-reference adaptive system

Gain-Scheduled Control/Pre-programmed Adaptive Control

This form of adaptive control has preset changes in the parameters of the controller made on the basis of some auxiliary measurement of some process variable. This technique is illustrated in the figure below:                      

The shortcoming of this system is that the control parameters have to be determined for many operating conditions so that the controller can select the one to fit the prevailing conditions. Nevertheless, the advantage of this system is that the changes in the parameters can be made quickly when the conditions change.

Also read: Ratio Control

Self-Tuning Control

This is where the system continuously tunes its own parameters based on monitoring the variable that the system is controlling and the output from the controller. This is demonstrated in the figure below:

You will typically find self-tuning/auto tuning in commercial PID controllers. When the operator presses a button, the controller injects a small disturbance into the system and measures the response. This response is compared with the desired response and the control parameters adjusted, by say, a modified Ziegler-Nichols rule, to bring the actual response to the desired response.

Related: How to Tune the PID Controller

Model-Reference Adaptive System

This technique involves an accurate model of the system being developed. The set value is then used as an input to both the actual and the model systems and the difference between the actual output and the output from the model compared. The difference in these signals is then used to adjust the parameters of the controller to minimize the difference.

 The figure below demonstrates this technique:

Also read: Cascade Control