Sensorless fractional order composite sliding mode control design for wind generation system.

A high-performance control system is essential to transfer maximum power from wind power generation system (WPGS) to the utility grid. In this paper, a fuzzy fractional-order terminal sliding mode control (Fuzzy-FOSMC) is presented based on the boundary layer approach. This boundary layer approach leads to the trade-off between chattering elimination and control performances. Initially a fractional order terminal sliding mode control (FOTSMC) is designed in this paper. Then, the reaching control part of the FOTSMC is replaced by a fuzzy system that eliminates the chattering even in the presence of lumped parametric uncertainties. The fuzzy control part is designed such that:(a) it maintains the stability of the system by introducing a non-linear slope inside the thin boundary layer near the sliding surface, and (b) it eliminates the chattering by acting like a saturation function. A novel wind speed estimation technique is also proposed in this paper based on Gaussian process regression (GPR). The inputs to the GPR framework are selected as the wind turbine power and its rotational speed. The superior performance of the proposed wind speed estimation technique is verified using error comparison with pre-existing techniques. The stability of the proposed GPR-based Fuzzy-FOSMC control paradigm is ensured by using the Lyapunov stability theorem. The proposed paradigm is compared with benchmark sliding mode control (SMC) and FOTSMC strategies. The proposed Fuzzy-FOSMC performance in terms of chattering elimination and stability is validated under normal conditions and lumped parametric uncertainties using extensive simulations in Matlab/SIMULINK and processor in the loop based experimental workbench.