Short-term memory-based control of wind energy conversion systems

Variable speed wind turbine control is essential in extracting maximum electric power out of available wind power. The paper presents a memory-based method for variable speed control of wind energy conversion systems. The fundamental idea behind the method is to use certain memorized information (i.e., current rotor speed tracking error, most recent speed tracking error, and previous control experience) to directly modify the control command. The salient feature of the proposed approach lies in its simplicity in design and implementation. Furthermore, the total required memory space does not grow with time and is much smaller than most existing learning control methods. It is shown that this method, when applied to firing angle control of wind turbines, is able to ensure rotor speed tracking in the presence of varying operation conditions, as verified via computer simulation.