A novel sensorless MPPT for wind turbine generators using Very Sparse Matrix Converter based on hybrid intelligent control

The maximum power of Wind Turbine Generators (WTGs) can be obtained by adjusting the rotor speed at the optimum Tip Speed Ratio (TSR), so that the Maximum Power Point Tracking (MPPT) algorithms is needed to obtain the reference speed. The conventional MPPT such as Pertubartion and Observation (PO), a reference speed is determined by a constant step. Choosing appropriate the step size is a problem, because it may cause problems in the response and the oscillations at the maximum point. To solve this problem, the flexible step size is proposed by using Least Squares-Support Vector Machine (LS-SVM). Then, the Adaptive Type-2 Fuzzy Sliding Mode Control (AFSMOC) is applied to control a rotor speed at a reference speed through Very Sparse Matrix Converter (VSMC). To replace a rotor speed sensors, the Model Reference Adaptive System (MRAS) observers is proposed. The simulation results shows that the MPPT based LS-SVM is remarkably faster and more efficient than the PO algorithm. LS-SVM can generate the flexible steps with the accuracy of classification 94.29 %. The AFSMOC can regulating the rotor speed at reference point, so that the maximum power can be achieved at all wind speeds.