Comparison of Neural Network and PID Control Techniques Based on a Case Study

The performance of a developed Neural Network Controller (NNC) was compared with that of a classical Proportional-Integral-Derivative (PID) algorithm embedded in a Distributed Control System (DCS) using data from a local refinery distillation column. The control techniques were compared through experimental studies. Data collected from a working DCS using PID algorithm was used to develop the Neural Network Controller (NNC) under identical conditions with respect to set-point regulation and load disturbance regulation in real time. It was found that the NNC demonstrated better performance than the classical PID and offered some advantages. It could be inferred from the study that the better performance of the NNC is due to its behavioral characteristics.