Control of Overload Safety in Wind Turbines Through Blade Pitch Control Implementing Artificial Intelligence

With the growth in world energy consumption and the need for more and more green energy production, the harvest of renewable energy is increasingly popular in the modern world. There are many green energy harvesting methodologies out there, and wind energy is becoming the leader in the green energy sector. Due to ever growing issues with the climate change, wind speeds increase and make it more difficult for wind turbines to operate in energy-efficient conversion regions. Because setting up and operating wind turbines is already costly, it is increasingly necessary to optimize the life cycle and operational efficiency to work in more general environmental conditions. This paper would address an overload safety feature that is engaged over the wind. This system is activated when the wind speed increases above the nominal wind speed for the windmill. The pitch angle of the wind turbine blades would be reduced to accommodate excess energy and to keep the wind turbine in the safe operating zone. To do so, we use the Artificial Neural Network (ANN). Many different activation layers, optimizers and loss functions were reviewed to achieve optimal performance and minimal loss. Results were documented for all combinations of ANN configurations.