11.8 Summary

A number of enhancements to standard PID controllers have been presented in this chapter. All commercial controllers implement ARW methods of one form or another. It is particularly important for cascade controllers to use external reset windup protection.

A number of controllers also have autotune features, but some of these are proprietary. We have presented the approach that is most widely presented in the literature. Using a relay function in the closed loop provides a closed-loop oscillation similar to the Ziegler-Nichols closed-loop oscillation method. There are several of advantages to the relay-based approach: the magnitude of the oscillations is bounded by proper selection of the manipulated input constraints, and the time required to develop several periods of the oscillation is much shorter than the time it takes to laboriously change a proportional gain to create a closed-loop oscillation.

One of the primary ways of creating a nonlinear PID controller is to make the controller gain a function of the absolute value of the error. This approach works well for processes where small deviations from setpoint are acceptable, such as surge vessel level control. Gain scheduling is a approach that can be used when the model and controller gain can be characterized as a function of the operating condition.

The proper selection of measurements and actuators can help linearize the input-output relationship. Equal percentage valves often make flow systems appear linear over a wider range. Similarly, using the logarithm of distillation product compositions can make the input-output behavior more linear.