Research on Two-Dimensional Landslide Model Control System Based on Support Vector Machine Modeling

Abstract:

Two-dimensional landslide model system is an important experimental platform for studying geological disaster, landslide behavior in different conditions are achieved by controlling the hydraulic pump drive model platform uplifting cabinet inclination. During the process of automation transformation of the two-dimensional landslide model system, control system based on computer is used to achieve a smooth lifting and precise angle of landslide model platform. Thanks to predictive control algorithm of least squares support vector machine (LS-SVM) model used in this paper, random real-time change of equivalent load applied on the hydraulic cylinder system effectively solved. The control system of the two-dimensional landslide model effectively eliminated crawling effect and pulse lifting in the condition of low speed and high load conditions of hydraulic system.

• Periodical: Applied Mechanics and Materials (Volumes 580-583)
• Pages: 947-953
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.580-583.947
• Online since: July 2014
• Authors: Zhong Tao Liu, Zheng Hua Feng*
• Keywords: Least Squares Support Vector Machine (Ls-SVM), Predictive Control, Two-Dimensional Landslide Model

References

[1] Li Zhongkui, Lu Darong, Zhong Shanyuan, etal. Development and application of new technology for 3D geo-mechanics model test of large underground houses[J]. Chinese Journal of Rock Mechanics and Engineering, 2003，22(9): 1430-1436.

[2] Cheng Shengguo, Luo Xianqi, Fang Kunhe. The test and design theory of the similar materials of soil-type landslide and the study on its assessment method[J]. Water Power, 2002, (4): 21–22.

[3] Luo Xianqi, Liu Defu, etal. Model test study on landslide under rainfall and reservoir water fluctuation. Chinese Journal of Rock Mechanics and Engineering, 2005, 24(14): 2476-2483.

[4] Zhang Hao-Ran, Han Zheng-Zhi, Li Chang-Gang. Support vector machine based unknown systems identification and control[J]. Journal of Shanghai Jiao Tong University, 2003, 37(1): 927-930.

[5] Alexei Pozdnoukhov, Mikhail Kanevski. Multi-scale support vector algorithms for hot spot detection and modelling[J], (2008).

DOI: 10.1007/s00477-007-0162-x

[6] Si Wu, Shun-Ichi Amari. Conformal Transformation of Kernel Functions: A Data-Dependent Way to Improve Support Vector Machine Classifiers[J], (2002).

[7] Lu Min, Zhang Zhanyu. Application of Support Vector Machine in Runoff Forecast[J], China Rural Water and Hydropower, 2006, (2): 47-49.

[8] Zhuo Sushuang, Su Guoshao, Yuan Qincheng. Long-term Forecast Method for Runoff Based on SVM[J]. Enterprise Science and Technology & Development, 2009, (18): 88-92.

[9] Qu Zhongyi, Chen Yaxin, Shi Haibin, etal. Structure and algorithm of BP Network for underground hydrology forecasting[J]. Journal of hydraulic, 2004, 35(2): 88-93.

[10] Liu Guodong, Ding Jing. Discussion on problems of BP neural networks applied to hydrological prediction[J]. Journal of hydraulic, 1999, 30(1): 65-70.

[11] Feng Guozhang. An analysis of effects of artificial neural network structures on precision of stream folw forecasting[J]. Journal of natural resources, 1998, 13(2): 169-174.

[12] Xing Zhenxiang. Application of modified genetic algorithm and back propagation artificial neural to water resource projects[D]. Northeast Agricultural University. (2004).

[13] Vapnik V N. The nature of statistical learning theory[M]. New York: Springer, (1995).

[14] Wang Yuhong, Huang Dexian, Gao Dongjie, Jin Yihui. Nonlinear Predictive Control Based on Support Vector Machine[J]. Information and Control, 2004, 33(2): 133-137.