In this paper, we propose a fully sensorless control strategy that guarantees robustness and stability properties. The aim is to maximize the extracted power and to ensure an optimum control without mechanical sensors to reduce the cost. Firstly, a fuzzy logic technique has been used to estimate the wind speed information. Then an improved model reference based on adaptive sliding mode controller has been developed to estimate the permanent-magnet synchronous generator speed. The proposed control strategy includes two control loops, an outer control loop based on fuzzy super twisting algorithm to regulate the generator speed to its optimal value and to provide the optimum DC side current, and an internal control loop based on sliding mode controller to regulate the DC side current to its optimal value and provides the appropriate signal for the DC-DC boost converter. The considered system in this work is supposed supplying a water pumping system for a use in isolated areas. The proposed control strategy has been compared to the conventional MPPT control method based on the concept of perturbation and observation. The obtained results show, a good estimation performance for the wind speed estimator and the improved MRAS speed observer, a good tracking performance and a good stability around the maximum power point. Comparing to the classical MPPT strategy, the current technique can greatly reduce the ripples that can generate vibrations and significant noise on the generator and the motor-pump group. Copyright (C) 2020 The Authors.