A novel control of a small wind turbine driven generator based on neural networks

This paper presents a novel control strategy of turbine directly driven permanent magnet synchronous generator (PMSG) for a small wind generation system. Compared to the traditional techniques, this new method has following advantages: 1) the proposed neural networks provides a fast and accurate estimation of actual wind velocity without anemometer; 2) the maximum mechanical power of small wind turbine can be well tracked at both dynamic and steady states. The neural networks are trained offline by Matlab software and the control algorithms then will be implemented by DSP controller. Therefore, the new small wind generation system is capable of delivering maximum electric power to customer applications with low cost, high efficiency and high reliability