Nonlinear process control using multiple models: Relay feedback approach

In this work the relay feedback autotuning is extended to handle process nonlinearity using multiple local models. Local models from relay feedback tests are scheduled using the Takagi-Sugeno fuzzy model, and local controllers are designed accordingly. This results in a gradual model switching between different operating conditions. The characteristics of the fuzzy implications are explored, and analytical expressions for the fuzzy model are derived. The importance of the selection of the scheduled parameters is emphasized, and the necessity of model scheduling for different loops is also explored. One transfer function example and two recycle plant examples are used to illustrate the advantage of the simple model scheduling method. Performance is evaluated according to the regions of robust performance and/or simulations. Results show that the proposed approach provides a simple and workable scheme for model scheduling large-scale systems.