Opposition-based discrete action reinforcement learning automata algorithm case study: optimal design of a PID controller

In this paper, the discrete action reinforcement learning automata (DARLA) method is expressed. The performance of the reinforcement learning algorithm is improved using the opposite concepts. This is an automatic method that can find the global optima without any knowledge about the parameters of the research space. To find the global optimal point, the interval that contains the optima is determined by DARLA as the cost function is minimized. In the opposition-based DARLA method, learning is performed based on opposition. The main idea in the opposition is to consider the search direction and its opposite at the same time to reach the candidate solution. This concept has increased the convergence rate and accuracy, and this algorithm can be used for many real-time applications. To prove this, the opposition-based DARLA is proposed to design a proportional-integral-derivative (PID) controller for the automatic voltage regulator system. The experimental results for the optimizing PID controller problem demonstrate the superior performance of the proposed approach.

Opposition-based discrete action reinforcement learning automata algorithm case study: optimal design of a PID controller

In this paper, the discrete action reinforcement learning automata (DARLA) method is expressed. The performance of the reinforcement learning algorithm is improved using the opposite concepts. This is an automatic method that can find the global optima without any knowledge about the parameters of the research space. To find the global optimal point, the interval that contains the optima is determined by DARLA as the cost function is minimized. In the opposition-based DARLA method, learning is performed based on opposition. The main idea in the opposition is to consider the search direction and its opposite at the same time to reach the candidate solution. This concept has increased the convergence rate and accuracy, and this algorithm can be used for many real-time applications. To prove this, the opposition-based DARLA is proposed to design a proportional-integral-derivative (PID) controller for the automatic voltage regulator system. The experimental results for the optimizing PID controller problem demonstrate the superior performance of the proposed approach.

___

  • F. Mohseni Pour, A.A. Gharaveisi, A. Afroomand, S.M.A. Mohammadi “Optimizing a fuzzy logic controller for a photovoltaic grid independent system”, 1st Annual Clean Energy Conference on International Center for Science, High Technology & Environmental Sciences pp. 237–241 2010.
  • M. Kashki, A. Gharaveisi, F. Kharaman, “Application of CDCARLA technique in designing Takagi-Sugeno fuzzy logic power system stabilizer (PSS)”, IEEE International Conference on Power and Energy, pp. 280–285, 2006 G. Heydari, A.A. Gharaveisi, M. Rashidinejad, “Optimized PI controller design in motor speed control by composition reinforcement learning automata approach” 1st Annual Clean Energy Conference on International Center for Science, High Technology & Environmental Sciences, pp. 69–76, 2010.
  • G. Heydari, A.A. Gharaveisi, S.M.R Rafie, “Optimized PID controller design in voltage control of boost regulator by composition reinforcement learning automata approach”, 1st Annual Clean Energy Conference on International Center for Science, High Technology & Environmental Sciences, pp. 204–211, 2010.
  • T. Fukuda, Y. Hasegawa, K. Shimojima, F. Saito, “Reinforcement learning method for generating fuzzy controller”, IEEE International Conference on Evolutionary Computation, Vol. 1, pp. 273–278, 1995.
  • HR. Tizhoosh, “Opposition-based learning: a new scheme for machine intelligence”, Proceedings of the International Conference on Computational Intelligence for Modeling, Control and Automation, Vol. 1, pp 695–701, 2005.
  • HR. Tizhoosh, “Opposition-based reinforcement learning”, Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol. 10 pp. 579–586 2006.
  • HR. Tizhoosh, “Reinforcement learning based on actions and opposite actions”, International Artificial Intelligence and Machine Learning Conference pp. 94–98 2005
  • H.R. Tizhoosh, M. Ventresca, Oppositional Concepts in Computational Intelligence, New York, Springer, 2008. X Yao, Y Liu, G. Lin, “Evolutionary programming made faster”, IEEE Transactions on Evolutionary Computation, Vol. 3, pp. 82–102, 1999.
  • Z.L. Gaing, “A particle swarm optimization approach for optimum design of PID controller in AVR system,” IEEE Transactions on Energy Conversion, Vol. 19, pp. 384–394 ,2004.
  • H. Yoshida, K. Kawata, Y. Fukuyama, Y. Nakanishi, “A particle swarm optimization for reactive power and voltage control considering voltage security assessment,” IEEE Transactions on Power Systems, Vol. 15, pp. 1232–1239, 2000.
  • F. Naderi, A.A. Gharaveisi, M. Rasidinejad, “Optimal design of type 1 TSK fuzzy controller using GRLA for AVR system”, Large Engineering Systems Conference on Power Engineering, pp. 106–111, 2007.
  • K.H. Ang, G. Chong, Y. Li, “PID control system analysis, design, and technology”, IEEE Transactions on Control System Technology, Vol. 13, pp. 559–576, 2005.
Turkish Journal of Electrical Engineering and Computer Science-Cover
  • ISSN: 1300-0632
  • Yayın Aralığı: Yılda 6 Sayı
  • Yayıncı: TÜBİTAK