Prediction of Salinity in Qiantang Estuary Based on Wavelet Neural Network Optimized by Particle Swarm Optimization

Abstract:

Saltwater intrusion is a serious problem for most of macro-tidal estuaries, especially when it happened at the downstream region of drinking water sources. Therefore, accurate prediction of salinity variation becomes absolutely necessary. A three-layer salinity prediction model is established in this paper, based on wavelet neural network (WNN) optimized by particle swarm optimization (PSO). The optimized model overcomes the local minimum problem and accelerates the convergence speed of WNN method. Hourly observed salinity data at Ganpu station in Qiantang Estuary was used for case study. Additionally, we compare this model with another two models, WNN and BP, by predicting the salinity at Ganpu station with the same sample. The results show that PSO-WNN model has higher convergence rate and prediction accuracy.

• Periodical: Applied Mechanics and Materials (Volumes 353-356)
• Pages: 2683-2687
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.353-356.2683
• Online since: August 2013
• Authors: Guang Hui Li, Zhi Lin Sun, Shi Xiang Hu
• Keywords: Particle Swarm Optimization (PSO), Qiantang Estuary, Saltwater Intrusion, Wavelet Neural Network (WNN)

References

[1] Rajkumar T., Johnson M.,. Prediction of salinity in San Francisco bay delta using neural network, Proceedings of IEEE SMC., Arizona, (2001).

[2] Huang, W. and S. Foo, Neural network modeling of salinity variation in Apalachicola River. Water Research, 2002. 36(1): pp.356-362.

DOI: 10.1016/s0043-1354(01)00195-6

[3] Dong, X., et al. Prediction of salt water intrusion using BP-RAGA coupled neural network model[C]. 2010 3rd International congress on Image and Signal Processing - CISP , 2010: 4244-4247.

DOI: 10.1109/cisp.2010.5646891

[4] Dan Xu, Zhilin Sun, Delu Pan. Neural network modeling of salinity in Qiantang Estuary. Journal of Zhejiang University(Sciences Edition)[J]. 2011(2): 234-238.

[5] Kenney J, Eberhart R C. Particle Swarm Optimization[C]. Proceedings IEEE international conference on Neural Networks. Perth, USA, 1995: 1942-(1948).

[6] Eberhart R C, Kennedy J . A new optimizer using particle swarm theory[C]. Proc. the Sixth Int. Symposium on Micro Machine and Human Science. Nagoya, Japan, 1995, 39-43.

DOI: 10.1109/mhs.1995.494215

[7] Eberhart R C, Shi Y. Particle swarm optimization: developments, applications and resources. Proc 2001 Congress on Evolutionary Computation. Seoul, South Korea, 2001, 81- 86.

DOI: 10.1109/cec.2001.934374

[8] Abdul Talib Bon, Jean-Marc Ogier, Ahmad Mahir Razali, Ihsan M. Yassin: Feature Selection in Extrusion Beltline Moulding Process Using Particle Swarm Optimization. Computer and Information Science 1(2): 20-26 (2008).

DOI: 10.5539/cis.v1n2p20

[9] Wei Sun, Ying Zou. Short term load forecasting based on BP neural network trained by PSO[C]. Proceedings of the Sixth International Conference on Machine Learning and Cybernetics, Hong Kong, 19-22 August 2007: 2863-2866.

DOI: 10.1109/icmlc.2007.4370636

[10] Du, L., Z. Jia and L. Xue. Human face recognition based on principal component analysis and particle swarm optimization-bp neural network[C]. Third International Conference on Natural Computation, Hainan, China, 24-27 August 2007: 287-291.

DOI: 10.1109/icnc.2007.418

[11] Wang, P., et al., Mechanical property prediction of strip model based on PSO-BP neural network[J]. Journal of iron and steel research, international, 2008. 15(3): pp.87-91.

DOI: 10.1016/s1006-706x(08)60132-6

[12] Zhao, H., et al. Analysis of energy consumption prediction model based on genetic algorithm and wavelet neural network[C]. 2011 3rd International Workshop on Intelligent Systems and Applications (ISA), Wuhan, China, 28-29 May 2011: 1-4.

DOI: 10.1109/isa.2011.5873468

[13] Zheng, Y., et al. Prediction of gathering pipeline internal corrosion rate based onimproved wavelet neural network model[C]. ICPTT 2011: Sustainable Solutions For Water, Sewer, Gas, and Oil Pipelines, Beijing, China, 26-29 October 2011: 90-99.

DOI: 10.1061/41202(423)11