Research On Assessment Of Sand Liquefaction Potential Based On Extenics And Support Vector Machine

China is among the countries where earthquakes take place most frequently in the world. 60 percent of the areas have high earthquake intensity, and sand liquefaction is the main reason that causes seismic disasters, so that the sand liquefaction is an important problem in earthquake engineering. Sand liquefaction comes through a complicate process, and it involves a great deal of influence factors, which have strong radomness and reciprocal nonlinearity. Assessment of sand liquefaction potential is the important link in the works of antiknock and disater-prevention, and is an important content in reconnaissance of geotechnical engineering, so evaluating sand liquefaction objuectively and exactly is the precondition of taking preventive measures on areas where liquefaction easily occurs. It also relates closely to the safety and economic benefits of the projects.The paper analyses and assesses traditional methods, and brings forward the theory ofextenics and support vector machine to the assessment of sand liquefaction potential, so as to find out a new path in the field of assessment of sand liquefaction potential.This paper has illustrated the essential theory of assessment method based on the extenics and realization technique of the method. According to that, a web assessment system based on the extenics theory has been developed with the web program technique, and this system was applied to the assessmento of sand liquefaction potential. Through detailed analyzing some influencing factiors of sand liquefaction, seven parameters were selected as assessment indexes. They are earthquake magnitude, peak ground surface acceleration, standard penetration value, specific penetration resistance, relative compaction, average particle diameter, and water table. The extension assessment model has been established that was applied to assessment of sand liquefaction potential. The model had been trained by the historical data and has been assessed in virtue of sand liquefaction from the historical data by simulating, and got a excellent result. After introduction of the support vector machine theory, this paper established the assessment model of sand liquefaction based on the support vector machine, and assessed in sand liquefaction from virtue data of some fields in home and overseas with the tool Libsvm, which was programmed based on the support vector machine theory. And got a good coherency between the assessement results and the actual liquefaction. The research expresses that the extension assessment method and the support vector machine are useful tool in the assessment and prediction of sand liquefaction potential.