Weibull Dağılımının Parametrelerini Tahmin Etmek için Kaotik Parçacık Sürüsü Optimizasyon Algoritmasının Uygulanması

Yıl 2025, ERKEN GÖRÜNÜM, 1 - 1

Emre Koçak , Büşra Aksel , Hacı Hasan Örkcü

https://doi.org/10.2339/politeknik.1733077

Öz

Güvenilirlik ve yaşam süresi çalışmaları gibi çeşitli disiplinlerde en iyi bilinen ve yaygın olarak kullanılan dağılımlardan biri olduğu için Weibull dağılımının parametrelerinin istatistiksel çıkarımları üzerine çok sayıda çalışma yapılmıştır. Maksimum olabilirlik, bilinmeyen parametrelerin tahmin sürecinde yaygın olarak kullanılan bir yöntem olmasına rağmen, üç parametreli Weibull dağılımı gibi bazı dağılımlar için, olabilirlik fonksiyonunu maksimize ederek parametreleri tahmin etmek oldukça zordur. Bu sorunu ele almak ve daha iyi sonuçlar elde etmek amacıyla Parçacık Sürü Optimizasyonu (PSO) algoritması incelenmektedir. Ancak, iyi sonuçlar elde etmek ve PSO’nun performansını artırmak için algoritma için farklı parametre değerlerinin ayarlanması gerekmektedir. Bu bağlamda, arama sürecini önemli ölçüde etkileyen atalet ağırlığının belirlenmesi oldukça önemlidir. Bu makalede bir yenilik olarak, PSO’nun yakınsamasını etkileyen faktör olan eylemsizlik ağırlığı için kaotik haritalarele alınmış ve üç parametreli Weibull dağılımının farklı parametre değerlerinin tahmini için ayrıntılı olarak incelenmiştir. Önerilen yöntemin etkinliği kapsamlı bir Monte-Carlo simülasyon analizi ile araştırılmıştır. Simülasyon bulguları, önerilen kaotik harita yaklaşımının klasik doğrusal azalan eylemsizlik ağırlıklarından daha iyi performansa sahip olduğunu göstermektedir.

Anahtar Kelimeler

Kaotik harita , parçacık sürüsü optimizasyonu , weibull dağılımı , maksimum olabilirlik

Implementation of a Chaotic Particle Swarm Optimization Algorithm to Estimate the Parameters of Weibull Distribution

Öz

Numerous studies on the statistical inferences of the Weibull distribution’s parameters have been performed because it is among the most well-known and widely applied distributions in several fields, including lifetime studies and reliability. Although maximum likelihood is a widely used method in the estimation process of unknown parameters, estimating the parameters by maximizing the likelihood function is very challenging for some distributions, like the three-parameter Weibull distribution. The Particle Swarm Optimization (PSO) algorithm is examined in order to address this issue and achieve improved outcomes. However, different parameter values for the algorithm need to be adjusted to achieve good results and increase the performance of PSO. In this context, it is very important to determine the inertia weight, which significantly affects the search process. As a novelty in this paper, chaotic maps for the inertia weight, which is the factor affecting the convergence of the PSO, are examined in detail for the estimation of different parameter values of the three-parameter Weibull distribution. The effectiveness of the suggested method is investigated by a thorough Monte-Carlo simulation analysis. The simulation findings demonstrate that the proposed chaotic map approach outperforms the classic linear decreasing inertia weights.

Anahtar Kelimeler

Chaotic map , particle swarm optimization , weibull distribution , maximum likelihood

Etik Beyan

The author(s) of this manuscript declare that the materials and methods used in their studies do not require ethics committee approval and/or legal-specific permission.

Teşekkür

The authors would like to thank Gazi University Academic Writing Application and Research Center for proofreading the article.

Kaynakça

• [1] Nosal M., Legge A. H. and Krupa S. V., “Application of a stochastic, Weibull probability generator for replacing missing data on ambient concentrations of gaseous pollutants”, Environmental Pollution, (2000).
• [2] Rousu D. N., “Weibull skewness and kurtosis as a function of the shape parameter”, Technometrics, 15(4), (1973).
• [3] Cousineau D., “Fitting the three-parameter weibull distribution: Review and evaluation of existing and new methods”, IEEE Transactions on Dielectrics and Electrical Insulation, 16(1): 281-288, (2009).
• [4] Örkcü H. H., Özsoy V. S., Aksoy E. and Dogan M. I., “Estimating the parameters of 3-p Weibull distribution using particle swarm optimization: A comprehensive experimental comparison”, Appl Math Comput, 268, (2015).
• [5] Acitas S., Aladag C. H. and Senoglu B., “A new approach for estimating the parameters of Weibull distribution via particle swarm optimization: An application to the strengths of glass fibre data”, Reliab Eng Syst Saf, 183, (2019).
• [6] Luus R. and Jammer M., “Estimation of parameters in 3-parameter Weibull probability distribution functions”, Hungarian journal of industrial chemistry, 33, (2005).
• [7] Abbasi B., Eshragh Jahromi A. H., Arkat J. and Hosseinkouchack M., “Estimating the parameters of Weibull distribution using simulated annealing algorithm”, Appl Math Comput, 183(1), (2006).
• [8] Örkcü H. H., Aksoy E. and Dogan M. I., “Estimating the parameters of 3-p Weibull distribution through differential evolution”, Appl Math Comput, 251, (2015).
• [9] Kantar Y. M. and Şenoǧlu B., “A comparative study for the location and scale parameters of the Weibull distribution with given shape parameter”, Comput Geosci, 34(12), (2008).
• [10] Akram M. and Hayat A., “Comparison of Estimators of the Weibull Distribution”, J Stat Theory Pract, 8(2): 238-259, (2014).
• [11] Teimouri M., Hoseini S. M. and Nadarajah S., “Comparison of estimation methods for the Weibull distribution”, Statistics (Ber), 47(1): 93-109, (2013).
• [12] Cohen A. C., Whitten B. J. and Ding Y., “Modified moment estimation for the three-parameter Weibull distribution”, Journal of Quality Technology, 16(3), (1984).
• [13] Chen D. and Chen Z., “Statistical inference about the location parameter of the three-parameter Weibull distribution”, J Stat Comput Simul, 79(3), (2009).
• [14] Garrido A., Caro-Carretero R., Jimenez-Octavio J. R., Carnicero A. and Such M., “A new approach to fitting the three-parameter Weibull distribution: An application to glass ceramics”, Commun Stat Theory Methods, 50(14), (2021).
• [15] Yang X., Xie L., Song J., Zhao B. and Li Y., “On interval estimation methods for the location parameter of the Weibull distribution: An application to alloy material fatigue failure data”, Commun Stat Theory Methods, 53(17), (2024).
• [16] Örgen F. K. D., Altıntaş A., Yaşar S., Öztürk M., Çiftçi E. and Tuncer A. D., “Software-Based Wind Energy Potential Assessment: A Case Study from Western Turkey”, Politeknik Dergisi, 26(1), (2023).
• [17] Nassir N., Acar B. and Fahed A. K. A., “Vertical Axis Wind Turbine Installation Based on Wind Data Collected in Gharyan City, Libya”, Politeknik Dergisi, 25(3), (2022).
• [18] Özsoy V. S., Örkcü H. H. and Bal H., “Particle Swarm Optimization Applied to Parameter Estimation of the Four-Parameter Burr III Distribution”, Iran J Sci Technol Trans A Sci, 42(2), (2018).
• [19] Bard Y., “Comparison of Gradient Methods for the Solution of Nonlinear Parameter Estimation Problems”, SIAM J Numer Anal, 7(1), (1970).
• [20] Eberhart R. C. and Shi Y., “Particle swarm optimization: Developments, applications and resources”, Proceedings of the 2001 Congress on Evolutionary Computation, Seoul, Korea (South), 81-8, (2001).
• [21] Wen X., Yu-xia X., Gui-xiang S. and Ying-zhi Z., “Parameter estimation of three-parameter weibull distribution via particle swarm optimization algorithm”, Proceedings 2011 International Conference on Transportation, Mechanical, and Electrical Engineering (TMEE), Changchun, China, 336–338, (2011).
• [22] Kennedy J. and Eberhart R., “Particle swarm optimization”, Proceedings of ICNN'95 - IEEE International Conference on Neural Networks, Perth, WA, Australia, 1942-1948 (1995).
• [23] Fukuyama Y., “Fundamentals of Particle Swarm Optimization Techniques”, in Modern Heuristic Optimization Techniques: Theory and Applications to Power Systems, IEEE, 71-87, (2008).
• [24] Kennedy J., “Particle Swarm Optimization”, in Encyclopedia of Machine Learning, Boston, MA: Springer US, (2010).
• [25] Shi Y. and Eberhart R., “A modified particle swarm optimizer”, 1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence, Anchorage, AK, USA, 69-73, (1998).
• [26] Tharwat A. and Hassanien A. E., “Chaotic antlion algorithm for parameter optimization of support vector machine”, Applied Intelligence, 48(3), (2018).
• [27] Alatas B., Akin E. and Ozer A. B., “Chaos embedded particle swarm optimization algorithms”, Chaos Solitons Fractals, 40(4), (2009).
• [28] Ren B. and Zhong W., “Multi-objective optimization using chaos based PSO”, Information Technology Journal, 10(10), (2011).
• [29] Wang G. G., Deb S., Gandomi A. H., Zhang Z. and Alavi A. H., “Chaotic cuckoo search”, Soft comput, 20(9), (2016).
• [30] Arora S. and Anand P., “Chaotic grasshopper optimization algorithm for global optimization”, Neural Comput Appl, 31(8), (2019).
• [31] Sayed G. I., Tharwat A. and Hassanien A. E., “Chaotic dragonfly algorithm: an improved metaheuristic algorithm for feature selection”, Applied Intelligence, 49(1), (2019).
• [32] Sayed G. I., Khoriba G. and Haggag M. H., “A novel chaotic salp swarm algorithm for global optimization and feature selection”, Applied Intelligence, 48(10), (2018).
• [33] Kohli M. and Arora S., “Chaotic grey wolf optimization algorithm for constrained optimization problems”, J Comput Des Eng, 5(4), (2018).
• [34] Saremi S., Mirjalili S. and Lewis A., “Biogeography-based optimisation with chaos”, Neural Comput Appl, 25(5), (2014).
• [35] Tiku M. L. and Akkaya A. D., “Robust estimation and hypothesis testing”, New Delhi: New Age International, (2004).
• [36] Yang X. S., “Engineering optimization: an introduction with metaheuristic applications”, John Wiley and Sons, (2010).
• [37] Shi Y. and Eberhart R. C., “Parameter selection in particle swarm optimization”, Evolutionary Programming VII. Lecture Notes in Computer Science, Berlin, 591-600, (1998).