Fatigue behaviour of 3-D steel structural specimens at low temperatures was investigated in this paper. Simulation model adopting backpropagation neural network approach is used for the analysis. The updated Lagrange formulation of motion combined with pseudo-force technique in the FETM-wave approach was used for the treatment of non-linear problems. Each step of iteration approaches the solution of linear problem and the feasibility of parallel processsing combined with backpropagation neural network is established. Numerical and experimental analysis is submitted in order to demonstrate the efficiency of the procedures suggested.