抄録
This paper considers the direct adaptive PI control of nonlinear chemical processes. A simple yet effective PI controller parameter tuning algorithm, which has different structures separated by a defined zero-level set of process output error, is derived. This algorithm enables the PI controller to control chemical processes in an interactive and autonomous way by simply observing the process output errors. To guarantee the stability of the resultant PI control system, a rigorous analysis using Lyapunov approach is provided. The effectiveness and applicability of the proposed direct adaptive PI control schemes are demonstrated through the control of a nonlinear CSTR in the presence of plant/model mismatch, unmeasured disturbance and parameter variations. Comparisons with several conventional PI controllers and an indirect adaptive PID controller are included as a basis for performance evaluation. Simulation results reveal that the proposed PI control schemes with the derived parameter tuning algorithm are promising, which provides flexibility and excellent control capability for the direct adaptive control of nonlinear chemical processes.
