Sealing Ring Reliability Assessment for Hydraulic Systems Based on BP Neural Networks

In this paper, reliability assessment of sealing ring in hydraulic systems is carried out based on the Back Propagation (BP) neural networks. An O-ring Finite Element Model (FEM) is established based on ANSYS software, by which the calculated maximum contact stress is taken as failure criteria. The diameter, oil pressure, the amount of compression, and modulus of elasticity are treated as random variables to take into account the parameters variations for studying the effect of parameter uncertainty on the sealing performance. Simulation data that generated from the FEM model is trained by using the artificial neural network toolbox to fit the relationship between the input data and output response. The trained neural network model is finally combined with the importance sampling method, to study the effect of the parameter variations on the reliability of the seal.