Traffic State Index Prediction Model Based on Hybrid Intelligent Methods

Abstract:

Traffic state index (TSI) is a quantitative indicator to evaluate the degree of traffic congestion. Accurate prediction of the TSI can effectively ease the traffic pressure. This paper presents a prediction model based on the hybrid intelligent method. Firstly, use the cross operator and mutation operator to generate the particles and use the simulated annealing algorithm (SA) to prevent the particles falling into local optimum for the basic particle swarm algorithm (PSO). Secondly, use the improved PSO algorithm to optimize the weights and thresholds of the BP neural network (BPNN). Finally, Train the BPNN to obtain the optimal solution. The hybrid intelligent methods prediction model, named IPSOBP, is verified by using the actual data. The results show that the prediction model has higher accuracy to predict TSI compared with BPNN improved by genetic algorithm (GA) and BPNN improved by PSO.

• Periodical: Applied Mechanics and Materials (Volumes 687-691)
• Pages: 1508-1513
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.687-691.1508
• Online since: November 2014
• Authors: Zhou Qi Shi, Yu Fang*, Hong Zhou Chen
• Keywords: BP Neural Network (BPNN), Cross Operator, Mutation Operator, PSO, SA, TSI

References

[1] Zhang Y, Ye Z. Short-term traffic flow forecasting using fuzzy logic system methods [J]. Journal of Intelligent Transportation Systems, 2008, 12(3): 102-112.

DOI: 10.1080/15472450802262281

[2] Xue J, Shi Z. Short-time traffic flow prediction based on chaos time series theory [J]. Journal of Transportation Systems Engineering and Information Technology, 2008, 8(5): 68-72.

DOI: 10.1016/s1570-6672(08)60040-9

[3] Wang Y H. Nonlinear neural network forecasting model for stock index option price: Hybrid GJR–GARCH approach [J]. Expert Systems with Applications, 2009, 36(1): 564-570.

DOI: 10.1016/j.eswa.2007.09.056

[4] Wang J J, Wang J Z, Zhang Z G, et al. Stock index forecasting based on a hybrid model [J]. Omega, 2012, 40(6): 758-766.

DOI: 10.1016/j.omega.2011.07.008

[5] Zhang Y, Wu L. Stock market prediction of S&P 500 via combination of improved BCO approach and BP neural network [J]. Expert systems with applications, 2009, 36(5): 8849-8854.

DOI: 10.1016/j.eswa.2008.11.028

[6] Hu W, Liu Y, Li L, et al. The short-term traffic flow prediction based on neural network [C]. Future Computer and Communication (ICFCC), 2010 2nd International Conference on. IEEE, 2010, 1: V1-293-V1-296.

DOI: 10.1109/icfcc.2010.5497785

[7] Lin-na C W P. The Research of the Application of GABP Neural Network in Traffic Flow Prediction [J]. Microcomputer Information, 2009, 14: 104.

[8] Cui F. Study of Traffic Flow Prediction Based on BP Neural Network [C]. Intelligent Systems and Applications (ISA), 2010 2nd International Workshop on. IEEE, 2010: 1-4.

DOI: 10.1109/iwisa.2010.5473703

[9] Chungui L, Shu'an X, Xin W. Traffic flow forecasting algorithm using simulated annealing genetic BP network [C]. Measuring Technology and Mechatronics Automation (ICMTMA), 2010 International Conference on. IEEE, 2010, 3: 1043-1046.

DOI: 10.1109/icmtma.2010.483

[10] Yan Y E, Zhilin L U. Neural network short-term traffic flow forecasting model based on particle swarm optimization [J]. Computer Engineering and Design, 2009, 30(18): 4296-4298.

[11] Kennedy J, Eberhart R C. Particle swarm optimization [C]. Proceedings of the 4th IEEE International Conference on Neural Networks, Piscataway: IEEE Service Center, 1995, 4: 1942-(1948).

[12] Ling S H, Iu H, Leung F H F, et al. Improved hybrid particle swarm optimized wavelet neural network for modeling the development of fluid dispensing for electronic packaging [J]. IEEE Transactions on Industrial Electronics, 2008, 55(9): 3447-3460.

DOI: 10.1109/tie.2008.922599