To allow consistent implementation of control applications, fuzzy logic control may be implemented as a function block. The fuzzy rules and membership functions can be predefined except for the scaling of the membership functions for error, change in error, and change in output. As addressed in Chapter 6 of Advanced Control Foundation – Tools, Techniques, and Applications, when fuzzy logic control is structured in the way it becomes possible to provide a tuning capability to automatically set the fuzzy logic control block’s scaling. The tuner interface can be designed to look and behave similar to that provided for PID control, except that the tuning parameter is for the membership functions’ scaling. An example of such an interface as implemented in the DeltaV control system is shown below.
Similar to auto-tuning provided for PID, the process dynamics are automatically determined when the Test button is selected. The process is under two-state control during testing. The manipulated parameter is changed from its initial value by the specified step size. Based on the process dynamics, the recommended membership functions’ scaling is calculated automatically. The recommended tuning is transferred to the fuzzy logic block by selecting Update. Thus, it is possible to make the steps involving in commissioning fuzzy logic control as easy as those followed to commission PID.