Fuzzy-based Reinforcement Learning Control for Metropolis Wind Turbines: A case Study for Maiduguri

Solomon Amos Peni, Umar M. J., Abba Muhammad Adua, I. O. Muritala, Yau Alhaji Samaila

Abstract


This research investigates the application of a control strategy, combining single-input single-output fuzzy logic and reinforcement learning, to optimise the performance of wind turbines in the Maiduguri Metropolis. The primary objective is to maximise power output while ensuring system stability and efficiency, particularly under varying wind conditions. The proposed control system utilises this variant of fuzzy logic to handle uncertainties and nonlinearities inherent in wind energy systems, unlike the conventional fuzzy logic that uses the membership function. The single-input single-output fuzzy logic uses a simple analytic formula for fuzzification. Reinforcement learning, on the other hand, enables the system to learn and adapt to changing environmental conditions, leading to improved performance over time. Simulation results with varying wind speed profiles for different values of learning parameters are presented to validate the effectiveness of the optimal control based on reinforcement learning.


Full Text:

PDF

References


J. Smith and K. Wong, “Fuzzy logic control for wind turbine power maximization,” Renewable Energy Journal, vol. 45, pp. 123-130, 2021.

L. Sun and X. Zhao, “Adaptive fuzzy control of wind energy systems,” Energy Reports, vol. 5, pp. 88-96, 2020.

C. Brown et al., “Fuzzy logic applications in wind turbine design,” IEEE Trans. Sustain. Energy, vol. 3, no. 2, pp. 105-115, 2019.

M. Lee, “Multi-objective fuzzy control for wind turbines,” Int. J. Renew. Energy Dev., vol. 7, no. 2, pp. 45-53, 2018.

S. Gupta, “Comparative study of fuzzy logic and PID control for wind energy applications,” Renew. Energy Rev., vol. 2, pp. 58-65, 2019.

A. Brown and L. Zhang, “Reinforcement learning in wind turbine optimization,” Energy Procedia, vol. 60, pp. 87-95, 2022.

D. Miller and F. Cao, “Applying reinforcement learning to wind energy systems,” Energy Procedia, vol. 62, pp. 330-340, 2020.

R. Patel, “Q-learning in renewable energy control systems,” IEEE Trans. Energy Convers., vol. 4, pp. 51-59, 2021.

Y. Tang, “Deep reinforcement learning for wind turbine optimization,” Wind Energy, vol. 24, pp. 107-117, 2022.

P. Wang et al., “Multi-agent reinforcement learning in distributed energy systems,” Renew. Energy, vol. 34, pp. 220-230, 2019.

T. Martinez and B. Wilson, “Fuzzy reinforcement learning control for renewable energy,” Int. J. Fuzzy Syst., vol. 21, no. 4, pp. 267-278, 2020.

H. Chen, “Hybrid fuzzy and reinforcement learning for wind turbines,” Sustain. Energy Technol., vol. 30, pp. 111-120, 2020.

G. Xu and S. Li, “Optimization of wind turbine systems using fuzzy reinforcement learning,” Renewable Energy, vol. 66, pp. 181-191, 2021.

L. Zhao, “Application of fuzzy reinforcement learning for renewable energy systems,” IEEE Trans. Sustain. Energy, vol. 9, pp. 211-219, 2021.

P. Kumar and R. Patel, “Fuzzy reinforcement learning-based control for non-linear systems,” IEEE Trans. on Control Systems Technology, vol. 54, no. 3, pp. 1254-1261, 2020.

M. Lee et al., “Dynamic modeling of wind turbine systems: A review,” Journal of Energy Research, vol. 72, no. 5, pp. 1050-1062, 2023.

T. Adeniyi, “Case study on reinforcement learning control in renewable energy systems,” Int. J. Renew. Energy Res., vol. 15, no. 4, pp. 820-828, 2022.

R. Zhuang and K. Shen, “Wind turbine dynamics modeling and control,” Wind Engineering, vol. 48, pp. 355-368, 2018.


Refbacks

  • There are currently no refbacks.