Elevator Traffic Pattern Recognition Based on Fuzzy BP Neural Network with SOM Algorithm

Elevator Traffic Pattern Recognition Based on Fuzzy BP Neural Network with SOM Algorithm

Zhenshan Yang Wenjiao Yue

Department of electrical automation, Engineering school, Bohai University, 19, Keji Rd., New Songshan District, Jinzhou City, Liaoning Province, 121013, China

Asia-Pacific Institute of Construction SciTech Information Co., Ltd., No. 36, Suite A, China construction technology group, outer street of Deshengmen,100120, Beijing, China

Corresponding Author Email: 
ydlut@163.com, mailauthor2@domain
Page: 
630-645
|
DOI: 
https://doi.org/10.18280/ama_b.600401
Received: 
1 January 2017
|
Accepted: 
15 July 2017
|
Published: 
31 December 2017
| Citation

OPEN ACCESS

Abstract: 

Elevator traffic pattern recognition (ETPR) is the prerequisite for effectively implementing the strategies of elevator group control system (EGCS). In view of the time-varying, nonlinear and uncertain characteristics of elevator traffic, an ETPR method based on fuzzy BP neural network with self-organizing map (SOM) algorithm is proposed, in which the fuzzy logic (FL) is introduced into BP neural network and, the SOM algorithm is employed to both determine the membership functions and merge the fuzzy rules. Thus as a result, the network structure is optimized, at the same time, the self-learning function of BP neural network enables the weighting coefficients of the FL membership functions to vary with different traffic patterns (TPs) and, the elevator traffic demand information is processed by fuzzy reasoning to realize ETPR and, therefore, to provide effective support to scheduling EGCS. Simulation experiments show the validity of the proposed method.

Keywords: 

Elevator traffic demand, elevator traffic pattern recognition (ETPR), fuzzy neural network, expert experience, self-organizing map (SOM) algorithm

1. Introduction
2. Classification of Elevator TP
3. Construction of the Fuzzy Neural Network
4. Implementation Procedures of ETPR
5. Simulation Experiment
6. Conclusion and Discussion
Appendix: Algorithm Implementation of ETPR
  References

[1] Zhenshan Yang, Cheng Shao, Guizhi Li, Multi-objective optimization for EGCS using improved PSO algorithm, American Control Conference, ACC’ 2007, New York, USA, July, 2007, Proc. pp. 5059-5063.

[2] J. Fernandez, P. Cortes A Survey of elevator group control systems for vertical transportation, a look at recent literature, IEEE Control Systems Magazine, vol.35, no.4, pp.38-55, 2015,

[3] Zhenshan Yang, Yunli Zhang, Dewen Zhu, Research on elevator traffic dispatching using genetic algorithms, AMSE conference, MS’2002, Girona, Spain, Jun 2002, Proc. pp. 627-634.

[4] Pablo Cortés, R.F. Joaquín, José Guadix, et al., Fuzzy logic based controller for peak traffic detection in elevator systems, Journal of Computational and Theoretical Nanoscience, vol. 9, no. 2, pp. 310-318, 2012.

[5] Zhenshan Yang, Cheng Shao, Haifeng Ma, LS-SVM base determination of critical time ranges for elevator TPs, Elevator World, vol. 56, no. 5, pp. 110-115, 2008.

[6] Fujun Ma, Traffic pattern identification of elevator group control based on Bayes decision theory, Mechanical & Electrical Engineering Magazine (in Chinese), vol. 26, no. 9, pp. 104-107, 2009.

[7] Albert T P So, J R Beebe, W L Chan, et a1., An artificial neural network based traffic pattern recognition system, Int. Journal of Elevator Engineering, vol. 1, no. 1, pp. 35-46, 1996.

[8] 岩田稚史, 匹田志朗, 駒谷喜代俊, ニューラルネットワークによるエレベータ交通流の検出, 計測自動制御学会論文集. 1997, vol. 33, no. 3, pp. 210 – 215, 1997.

[9] Zhenshan Yang, Cheng Shao, A new method for multi-objective control of elevator group in high-rise building, AMSE Journals, Series Advances D, vol. 13, Nº 2, pp. 15-27, 2008.

[10] G C Barney, Lutfi Al-Sharif, Elevator Traffic Handbook: Theory and Practice, Taylor & Francis Ltd., 2016.

[11] James C. Bezdek, James Keller, Raghu Krisnapuram, Fuzzy models and algorithms for pattern recognition and image processing, Springer Science+Business Media, Inc., 2005.

[12] Buyun Xu, He Ni, Comparative analysis of SOM and K-means clustering, The Journal of New Industrialization (in Chinese), vol. 4, no. 7, pp. 63-69, 2014.

[13] George R. Strakosch, Robert S. Caporale, The vertical transportation handbook (4th ed.), John Wiley & Sons, Inc., 2010.

[14] CIBSE, Transportation system in buildings, GIBSE Guide D 2010 (4th ed.), Page Bros. (Norwich) Ltd., 2010.

[15] Zhenshan Yang, Yunli Zhang, Traffic demand forecasting for EGCS with grey theory based multi- model method, Int. Journal of Computer Science Issues, vol. 10, no. 1, pp. 61-67, 2013.

[16] Yuanli Gu, Xingchuan Wang, Jianxiang Xu, Traffic data fusion research based on numerical optimization BP neural network, Applied Mechanics and Materials, vol. 513-517, no. 1, pp. 1081-1087.

[17] Alexander I. Galushkin, Neural networks theory, Springer-Verlag Berlin Heidelberg, 2010.

[18] P. Cortés, J. Fernández, J. Guadix et al., Fuzzy logic based controller for peak traffic detection in elevator systems, Journal of Computational and Theoretical Nanoscience, vol. 9, no. 2, pp. 310-318, 2012.

[19] Zhenshan Yang, Wei Ren, Elevator traffic planning under complex building conditions, Journal of Bohai University (in Chinese), vol. 35, no. 1, pp. 17-22, 2014.