Title | ||
---|---|---|
Self-tuned Model-Based Predictive Control Using Evolving Fuzzy Model of a Non-linear Dynamic Process |
Abstract | ||
---|---|---|
Increasing demand for data stream processing algorithms and evolving identification methods warrants research focused into new (fuzzy) evolving identification algorithms. The resulting models can increasingly be used in control system scenarios, resulting in an effectivly self-tuned closed loop system. These trends have motivated us to combine a recently developed evolving, fuzzy, FRLS-based identification technique with the model-based predictive control algorithm Predictive Functional Control in state space. The included nonlinear identification algorithm and the resulting model structure are investigated and shown to be suitable for direct use in model based control scenarios. The entire approach has a low number of tuning parameters. Since the model is evolved and then directly usable in a control scenario, the approach results in a self-tuned closed loop system. The approach is validated on a Wienner-Hammerstein dynamic process - a non-linear second order dynamic process. The evaluation is performed in an assortment of scenarios including reference tracking, closed loop dynamics analysis and both input- and output-disturbance rejection. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1007/978-3-030-82099-2_37 | EXPLAINABLE AI AND OTHER APPLICATIONS OF FUZZY TECHNIQUES, NAFIPS 2021 |
DocType | Volume | ISSN |
Conference | 258 | 2367-3370 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ziga Strzinar | 1 | 0 | 0.34 |
Igor Skrjanc | 2 | 354 | 52.47 |