Research of Multi-Objective Optimization in Controlled Atmosphere Storage System

Abstract:

According to multi-objective optimization problem in controlled atmosphere storage system, a modified multi-objective particle swarm optimization algorithm based on maximin fitness function was proposed. This algorithm introduced normalization algorithm and the conception of ε-dominance to the computation of maximin fitness function value, which solved the skewed popularity of the particles in the flight process. A modified computation method and an alterable ε-dominance strategy were put forward, which effectively improved the convergence speed of the algorithm and the diversity of the particles. The modified multi-objective particle swarm optimization algorithm was applied to optimize parameters like fruits and vegetables type, gas content in the controlled atmosphere storage system. The experimental results showed that the convergent speed of the algorithm was fast and the algorithm was valid.

• Periodical: Applied Mechanics and Materials (Volume 248)
• Pages: 578-583
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.248.578
• Online since: December 2012
• Authors: Yue Jin Huang, Ming Xu, Xi Shen, Jiang Ping Gu, Hua Qiang Jin
• Keywords: Controlled Atmosphere Storage, Maximin Fitness Function, Multi-Objective Optimization (MOP), Particle Swarm Optimization (PSO), ε-Dominance

References

[1] Y. Liu, Z.M. Wu, Y.F. Li and R.Z. Wang. Food and Fermentation Industries. 4, 13(2006).

[2] P. Zhang and S.H. Chen. Refrigeration and air-conditioning. 1, 21(2008).

[3] D.Q. Yan, H.T. Hua, Z.X. Zhang and Y.M. Liu. Modern agricultural science and technology. 9, 17(2009).

[4] C.A. Coello Coello, D.A. Van Veldhuizen, and G. B Lamont. Kluwer Academic Publishers, New York(2002).

[5] G. Toscano Pulido and C.A. Coello Coello. Using clustering techniques to improve the performance of a particle swarm optimizer. Proceeding of the 2004 Genetic and Evolutionary Computation Conference. (2004) 225-237; Washington, USA.

DOI: 10.1007/978-3-540-24854-5_20

[6] S. Mostaghim and J. Teich. The role of ε-dominance in multi-objective particle swarm optimization method. Proceeding of the 2003 Congress on Evolutionary Computation, (2003), Vol 3, 1764-1771; Canberra Australia.

DOI: 10.1109/cec.2003.1299886

[7] R. Balling. The maximin fitness function; Multi-objective city and regional planning. Proceeding of EMO 2003, (2003), 1-15; Berlin, Heidelberg.

DOI: 10.1007/3-540-36970-8_1

[8] X. Li. Better spread and convergence: particle swarm Multi-objective optimization using the maximin fitness function. Proceeding of GECCO 2004, (2004), 117-128; Washington USA.

DOI: 10.1007/978-3-540-24854-5_11

[9] J. Kennedy and R. Eberheat. A new optimizer using particle swarm theory. Proceedings of the Sixth International Symposium on Micro Machine and Human Science, (1995), 39-43; Portugal.

DOI: 10.1109/mhs.1995.494215

[10] J. Kennedy, and R. Eberheat. Particle swarm optimization. IEEE International Conference on Neural Networks, (1995), 1942-1948; Piscataway USA.

[11] Y. del Valle, G.K. Venayagamoorthy, S. Mohagheghi, J.C. Hernandez and R.G. Harley, IEEE Transactions on Evolutionary Computation , 2, 171 (2008).

DOI: 10.1109/tevc.2007.896686

[12] R.C. Eberhart and Y. Shi. IEEE Transactions on Evolutionary Computation , 3, 201 (2004).

[13] E. Zitzler, K. Deb and L. Thiele. IEEE Transactions on Evolutionary Computation, 2, 173(2000).

[14] L.Q. Shen, J.F. Huang and G.R. Huang. Bulletin Science and Technology, 6, 878(2006). procedures related to the electrophysiology of the heart. U.S. Patent 5, 529, 067, Jun 25 (1995).