Ultimate Bearing Capacity Prediction Of Static Pressure Pipe Pile Based On GA-SVM

Static pressure pipe pile is widely used due to its many advantages, how to make sure its bearing capacity should be the first problem in the engineering design. So, how to determine the actual ultimate bearing capacity accurately, simply and rapidly is a very tough question for technical personnel. Support vector machine theory seeks compromise between learning precision and the best learning ability on the basis of limited samples. It can not only guarantee the accuracy of the algorithm, but also ensure algorithm has better generalization ability. It is suitable for solving such complex issues as ultimate bearing capacity prediction of static pressure pipe pile.This paper elaborated the working mechanism of static pressure pipe pile and the influence factors of ultimate bearing capacity prediction; it introduced the basic concepts and principle of SVM and established the index system from three aspects: pile factors, Soil factors and construction parameters. Then it established the ultimate bearing capacity prediction model in which its parameters are optimized through the genetic algorithm. I write corresponding procedures using MATLAB 2010b tools. Finally, this paper established the ultimate bearing capacity prediction model based on GA-SVM, it also established the SVM prediction model, test results shows that GA-SVM prediction model has better prediction performance.