Balance-based adaptive control of a neutralisation process

This paper deals with the application of the general Balance-Based Adaptive Control ( B-BAC) methodology to the control of the nonlinear neutralization process. Since usually in the practice only the flow rates and the pH values are measurable on-line, it is very difficult to derive any model-based control strategy on the basis of the complete nonlinear model, partially because there is always a large uncertainty both on the form and on the values of the parameters of this model and partially because even if this model was known, the resulting control law would have very complex form. Thus, in the paper, the quasi-phenomenological and simplified model of the process is suggested. This input-output model describes the relationship between the measurable parameters, such as flow rates and inlet and outlet pH values, and all the nonlinearities of the process are represented by the only one time-varying parameter. Its value can be easily estimated on-line by the recursive least-squares procedure, which ensures the adaptability of the control law. On the basis of this model, the final explicit form of theB-BAController is derived and it is evaluated by simulation and compared with the conventional PI controller in the application to the example complete nonlinear model of the neutralization process. The simulation results demonstrate very good control performance of the suggested control law in the presence of the disturbances changes. Moreover, the robustness to the large uncertainty on the measurement data of the disturbing parameters included in the control law as the feedforward action is also proved.