The Reliability Analysis Of Single-Beam Bridge Crane Grider Based On Non-Intrusive Stochastic Finite Element

Crane plays a vital role in the process of modern industrialization,and its structural reliability and safety are also widely concerned by scholars.As an important part of the metal bridge structures,the crane girder plays the role of devices supporting and loads passing in the working process.So,the structure reliability of the crane girders largely affect the reliability of the whole system.At present,a lot of theory and application results have been obtained in the structural reliability study of cranes.However,the traditional method s of structural reliability are mainly based on explicit functions.For crane structure becoming large,complex and nonlinear increasingly,it is not easy to get exact explicit functions.Therefore,by using the finite element method,the stochastic finite element method developed by the theory of structural reliability stochastic analysis has important engineering value and practical significance.A non-intrusive stochastic finite element method based on the Chebyshev polynomial stochastic expansion is proposed by considering the parame ter uncertainty of medium and high nonlinear complex structures.The method effectively decouples the structural finite element analysis with reliability analysis.The Latin hypercube sampling points are selected as the collocation points to fit the Chebyshev orthogonal polynomial stochastic expansion model.The model avoids the pathological phenomena of the normal equation in the high order expansion and explicitly expresses the structural explicit performance function instead of the implicit one.Considering all the reliability factors of single-beam bridge crane box girder,the finite element models under strength?horizontal static stiffness?vertical static stiffness and dynamic stiffness working conditions are established.The effect of collocation points number on the calculated reliability is explored after picking out random variables by sensitivity analysis,and it is hypothesized that the number of collocation points required by the method is approximately twice that of the unknown coefficients.The reliability under the four working conditions can be acquired by the method based on the Chebyshev polynomial stochastic expansion.The results indicate that the reliability calculated by using 3 order Chebyshev polynomial stochastic expansion is slightly more accurate than by using the 2 order one,and both calculated results meet the demands.By balancing accuracy and efficiency,the proposed method in this paper can effectively evaluate the reliability of complex structures which leads to good engineering applicability.