On utilising structural information for adaptive control of a PH neutralisation process

Control of the pH level is of vital importance in many industrial applications. However, pH control has only met with limited success. A stumbling block is the severe non-linearity of titration processes. One control strategy that have been employed to overcome the process non-linearity exploits the knowledge that certain pH processes can be casted in the Wiener model structure. In practice, it is very likely that the process characteristics will be affected by the presence of unknown buffers. This paper aims at examining whether there are merits in using structural information to design the architecture of adaptive controllers in situations where the knowledge may be inaccurate due to changing buffering conditions. The main finding of this work is that a “black box” controller is better able to adapt to the changing dynamics caused by the introduction of an unknown buffer.