An Efficient Algorithm for Fluid Force and its Jacobian Matrix in Journal Bearing

Abstract

Based on the theory of variational inequality, a rapid efficient algorithm for fluid force and its Jacobian matrix in journal bearing is presented in this paper. Primarily, to solve the fluid force is transformed to solve a set of linear algebraic equations with tri-diagonal coefficient matrices. Meanwhile, an amendatory direct-method is proposed to solve the united equations about fluid forces and their Jacobian matrices, rapidly and synchronously. The Reynolds boundary condition has to be satisfied automatically during the process. Secondly, the coefficient matrices, which are involved in the previous process, can be decomposed to an assembly of a part of relative with journal motion and a part of invariable matrix, which can be prepared in advance and be referred to later repeatedly. Through these measures, many redundant operations are avoided. The numerical examples show that, under the accuracy guaranteed, the algorithm in this paper can reduce computational time remarkably, which reveals that the current method has a good operational characteristic and practicability.