Research On Gain Scheduling Control Of Wind Turbine Generation System Based On T-S Fuzzy Linearization

With the futher generalization of wind energy in scale utilization, modern Wind Turbine Generation System (WTGS) is anticipated to be grid-friendly. It means that the enhencement of load management ability of WTGS both from mechanical side and eletrical side becomes very necessary. Meanwhile, different from the original phase expanded in quantity, the wind power market has stepped into a developing phase ascended in quality. More and more attention has been paid on the cost per kilowatt hour of the electricity generated by WTGS. It leads that the optimization and amelioration of WTGS operation level is popularly focused on. In order to improve the process control performance of WTGS, the research works are carried out from the aspects of control task, operation strategy and control method as follows:With the control task transition of modern WTGS and based on the traditional operation strategy, a kind of widely-ranged power-adjustable operation strategy compatible with the tradtional one is presented via regulating the power coefficient of aerodynamic system. Especially, a decoupling coordinated method is proposed to deal with the coupled problem between variable-speed control loop and variable-pitch control loop with two degrees of freedom below the rated wind speed. As a result, the deep load management under full range of wind speed is preliminarily realized. The simulations and analysis under different operation regions are executed which validate that, via the above operation strategy, the scheduling values from the power gird can be well tracked which fulfills the expected response characteristics.Considering the direct data measuring points of wind power generation process, the derivation of mechanism model and the data-driven algorithm are combined in the paper and a solution scheme of estimation of effective wind speed is proposed. Utilizing the pratical measurement data for training and testing, it is validated by simulations that the nonlinear inverse mapping model with high accuracy between the aerodynamic torque and the wind speed can be obtained through this approach. Then, the wind speed can be accurately estimated with required estimation error.Compared with the traditional gain scheduling technique, in the paper, the Takagi-Sugeno (T-S) fuzzy gain scheduling algorithm is brought in and the T-S fuzzy Proportion Integration Differentiation (PID) controller is adopted to realize the control design around a single operation point. Then, the gain scheduling technique with two-level T-S fuzzy structure is proposed to complete the process control design of WTGS. The corresponding stability condition and the parameter tuning method of controller are given. Compared with the tradtional PID gain scheudling technique about the process control of WTGS via simulation, the gain scheduling technique with two-level T-S fuzzy structure has better process control performance.Based on a kind of Affine Nonlinear Parameter Varying (ANPV) model, the T-S fuzzy linearization approach is adopted in the paper and the Linear Parameter Varying (LPV) T-S fuzzy model is obtained. Based on this model, the piecewise quadratic Lyapunov stability is studied and the H∞ control problem using state-feedback controller or output-feedback controller are explored, respectively. The sufficient conditions in Linear Matrix Inequality (LMI) form are deduced. Subsequently, utilizing the LPV T-S fuzzy modeling and control approaches, the H∞ control problems abstracted from the wind power generation process are solved and the particular control design procedures are derived. Compared with the LPV gain scheduling control, the LPV T-S fuzzy control is more conveniently to be solved and has better process control performance.