The Data Fusion in Multi-Sensors Grain Information Monitoring System Based on Improved BP Neural Networks

Abstract:

According to the architecture characteristics of the grain storage system and the main factors influencing the safety of the grain storage, this paper has proposed a two layers data fusion system based on multi-sensors grain storage monitoring technique, which is used to identify the safety of the storage. The first layer including data extraction and encoding, the second layer is mainly built up by the improved BP NN. The results show that the system can fully fuse the data collected by the sensors and realize the real-time safety monitoring of the grain storage.

• Periodical: Applied Mechanics and Materials (Volumes 263-266)
• Pages: 269-276
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.263-266.269
• Online since: December 2012
• Authors: Zhe Mao, Zhuo Ran Zhang, Ya Ling Lu
• Keywords: Grain Depots Monitoring System, Improved BP NN, MATLAB, Multi-Sensors Data Fusion

References

[1] Rumelhart D E，Hinton G E，Williams R J．Learning representations by back-propagation error[J]．Nature，1986，323(9)：533-536

[2] Xinxin Weng，Feng Lu，Chuanxian Wang．Discriminating and Quantifying Potential Adulteration in Virgin Olive Oil by Near Infrared Spectroscopy with BP-ANN and PLS[J]．Spectroscopy and Spectral Analysis，2009，29(12)：3283-3287

[3] Guanghui Yan，Li Yang，Chunli Fan．Inverse Heat Conduction Problem Identification of Fault Components in the Electrical Cabinet Based on Infrared Temperature Measurement[J]．Laser and Infrared，2012，42(2)：151-156

[4] Yubo Tian．Hybrid Neural Networks Techniques[M]．Beijing：Science Press，2009：22-30．

[5] Yuwen Wu， Chenz Chen，Wenha Kang．Face Orientation Recognition Based on BP Neural Network [J]．Image Technology，2012，24(1)：29-64