STOCHASTIC FINITE ELEMENT ANALYSIS FOR THE TEMPERATURE FIELD OF FROZEN SOIL ROADBED OF QINGHAI-TIBET RAILWAY

Based on the first order Taylor series expansion and finite element method, a stochastic finite element model is presented for the temperature field. The random temperature field of the frozen soil roadbed in the Bailu River section of Qinghai-Tibet Railway is calculated using the model. The simulation results show that the distribution of temperature-standard deviation is close to the temperature distribution in the roadbed, however, the standard deviation distribution inside the upper part of the roadbed presents a larger area than that of the corresponding temperature distribution. The effect of environmental temperature is remarkable at the top of the roadbed, and diminishes with depth. The values of standard deviation become uniform at the lower part of the roadbed. This paper also discusses advantages of stochastic finite element method.