Comparison between backstepping and input–output linearization techniques for pH process control

In this work performances of adaptive backstepping controller (BSC) and globally linearizing controller (GLC) are compared for pH control. First, based on the system full order model a GLC has been designed and it has been shown that this controller is identical to BSC proposed in the literature. Next in order to avoid state estimator design, BSC and GLC are designed based on pH reduced order model and their identities have been established. Through computer simulations, it has been shown that the performance of non-adaptive GLC designed based on reduced order model is better than that of adaptive BSC designed based on pH full order model which requires state measurement for implementation. Finally, the effectiveness of GLC designed based on the reduced order model in load rejection and set-point tracking has been demonstrated through simulation and experimental studies.

► Backstepping (BS) technique used for pH process control is identical to global linearizing controller (GLC).
► Non-adaptive GLC based on reduced order model is better than adaptive BS controller designed based on full order model.
► The proposed controller does not need observer and works well experimentally.