Robustness analysis of Airborne Wind Energy Systems Using Neurofuzzy Control Approach

Renewable energy has shown great potential over the last few decades as an excellent alternative to fossil fuels. Renewable energy technology, if properly implemented will reduce global carbon emissions and cost of energy production without a physical limit in supply as opposed to fossil fuels. However, certain aspects of this technology continue to pose challenges such as the enormous quantities of materials required to set it up, large amount of capital, human resource requirements, little or no structure of automation, and most importantly, the negative environmental impact such a system may constitute which are in contrast to the recently revised Sustainable Development Goals (SDGs). Airborne wind energy (AWE) is an emerging wind energy technology which requires minimal material when compared to the existing wind energy technologies especially wind turbines. This new approach has shown great potential in terms of cost reduction, negligible environmental impact and theoretically higher power generation when compared to the wind turbine. In this work, an intelligent controller is designed for AWES to alleviate some of the aforementioned problems associated with wind energy technologies. The intelligent controller is designed using neurofuzzy techniques which would reduce the hazards of implementation of this system as the system will automatically vary its altitude to harvest optimal power and also ground the system to a halt when the wind conditions render the system to become inoperable. The response of this controller when compared to an optimally tuned Proportional Integral Derivative (PID)controller showed that the performance of the genetic-Adaptive Neurofuzzy Inference System (ANFIS) is better in terms of the response when subjected to disturbances and it is more adaptable and stable. The neurofuzzy controller showed 71% improvement in rise time, 65% improvement in settling time and 41% improvement in overshoot over the PID controller.