Types of Control

Corrective action to restore system to desired condition.

Feedback

On-Off A common example is a home thermostat that turns the heat on when the temperature is too low and turns it off when the desired temperature is reached. Fluctuations above and below the set point are inherent, but performance is acceptable. These fluctuations may be intolerable for many chemical and biological situations.

Proportional The error is measured, and the corrective action is proportional to the error. Gain is the term for the proportionality coefficient. Some error is inherent, but performance tends to be far superior to on-off control.

Integral control Seldom used alone and can greatly improve proportional control because the integral of the error determines the amount of corrective action. As time goes on, the integral increases such that there will be no offset.

Derivative This is not often necessary for bioprocessing where things tend to change slowly but can be essential for some chemical processes. The corrective action depends on the derivative of the error, and some chemical processes are so unstable that errors can get wild. This means that the derivative will be great such that corrective action kicks in quickly.

Feed Forward

Feedback control can be too late to prevent a catastrophe. For example, driving an automobile using feedback control would apply corrective acion after hitting an obstacle. All drivers anticipate and evaluate such that control is applied to prevent errors. The logic for doing so is in the brain of the driver.

An algorithm is needed to select the degree of corrective action. This can be simple rules of thumb or models based on understanding of fundamentals.

Other

These are beyond the scope of this simplified introduction. When there is need for cascades of controllers and networks of controls for integrated processes,bioprocess control can be highly complicated and challenging.
B A C K

Course Schedule