Design of a Nonlinear Model Predictive Controller for a Hydro-turbine Governing System

• Published in: 发电技术 Vol. 41; no. 4; pp. 407 - 414
• Main Author: Wenjun ZHOU
• Format: Journal Article
• Language: Chinese; English
• Published: Editorial Department of Power Generation Technology 30-08-2020
• Subjects: hydro-turbine governing system; model predictive control; quadratic programming; system constraint; takagi-sugeno fuzzy control
• ISSN: 2096-4528
• DOI: 10.12096/j.2096-4528.pgt.19097

The nonlinear terms of the hydro-turbine governing system (HTGS) were approximated based on Maclaurin series. Takagi-Sugeno (TS) fuzzy control was introduced to handle the nonlinear terms of the Maclaurin series, and TS models of the HTGS were developed resultantly. Based on the TS fuzzy models and the defined cost function, a nonlinear model predictive control (NMPC) strategy was designed for the HTGS. Considering the constraints of the HTGS, quadratic programming (QP) was utilized to solve the optimal controller output under the system constraints. Simulations were conducted to assess the performance of the designed QP-based NMPC (QPNMPC) controller. Simulation results reveal that the proposed QPNMPC controller can effectively regulate the output power of the HTGS and that compared with the conventional PID controller the advantages of the proposed QPNMPC controller are demonstrated.