Wind speed-tip speed ratio controlled wind turbine apparatus

A wind speed-tip speed ratio neural network (NN) trained wind turbine control apparatus, the neural network (NN) trained wind turbine control apparatus comprising:

• a controller module, the controller module comprising:
• a memory and
• a processor, wherein the processor includes instructions to control the wind turbine,
• wherein the wind turbine comprises a drive train comprising a generator, a high-speed shaft, a gearbox, a low-speed shaft, and a rotor and one or more blades connected to the rotor, wherein the gearbox connects the low-speed shaft to the high-speed shaft which is attached to the generator;
• wherein the blades have a length R equivalent to a radius of the wind turbine, wherein each blade of the one or more blades has a blade tip, a blade pitch angle, β, a blade swept area, and a tip speed, and a tip speed ratio (TSR),
• wherein the TSR is a ratio between a tangential speed of the blade tip and a wind velocity,
• wherein the controller module is configured to implement a machine learning neural network method to control a rotor speed of the wind turbine to achieve an optimum rotor speed and a maximum power as determined by the neural network at varying wind velocities, Vw, the method including
• generating a data set including at least 120 samples each of wind velocities and tip speed ratios at wind speeds between 3 to 19.4 m/s;
• determining the maximum power and the optimum rotor speed for every sample of wind velocity in the data set;
• training a neural network (NN) model using the data set in which each sample of wind velocity in the data set is fed as an input and the determined optimum rotor speed and the determined maximum power are output from the NN model,
• testing the NN model with random input wind speed;
• and
• controlling the rotor of the wind turbine based on a current wind velocity and the optimum rotor speed determined by the NN model.