This paper presents an effective scheme for tracking the optimal power extracting from the natural wind of a Doubly-Fed-Induction-Generator based on Wind Turbine (DFIG-WT). The identification and the operation of the model components on dynamic behaviour, and vector control concerning the DFIG, in a synchronously rotating d-q framework are introduced. Thus, the desired control signals of the electromagnetic torque and the reactive power are regulated by controlling the machine side converter, combined with the implementation of the flexible scheme of the optimal torque algorithm to follow continually the trajectory of the reference electromagnetic torque during rapid fluctuation of the wind speed, and under different supplying scenarios of reactive power generation. The maximum power point is achieved directly by the value of wind speed, estimated by the artificial neural network through real measured data of two input samples as the mechanical power of a wind turbine and its mechanical rotor speed. Particularly, this scheme leads to better dynamic efficiency in wind generation, which does not need an anemometer, which is available, or a wind sensor. In essence, the analysis and the comparison between the proposed scheme and the conventional algorithm at the similar conditions are provided via demonstrative simulation studies. Obviously, the effectiveness and the feasibility of a smart-sensorless controller in terms of good performance and fast convergence are confirmed.