No abstract

To obtain more comprehensive customer requirements and accurately determine the importance ratings of customer requirements in automotive clutch Quality Function Deployment (QFD), Affinity Diagram, expert evaluation method and rough Analytic Hierarchy Process (RAHP) were introduced in this paper. Firstly, fourteen requirements were identified for automotive clutch by using Affinity Diagram. Then rough numbers and rough boundary intervals were utilized to characterize vagueness and uncertainty of customer requirements, and constructed rough group decision matrices and rough pairwise comparison matrices. Finally, the importance ratings of customer requirements were determined by solving eigenvalues and eigenvectors of rough pairwise comparison matrices.

To further investigate the fracture response in propellant grain, numerical methodology is proposed to cope with crack propagation simulation especially for the mixed mode condition. The numerical discrete scheme of the propellant linear viscoelastic constitutive model is proposed, which provides a key means for the simulation of crack propagation. In order to simulate the cohesive traction distribution on the new crack surface, the extrinsic Park-Paulino-Roesler (PPR) cohesive zone model (CZM) is introduced. To let the crack propagate along any direction determined, element splitting technique and its corresponding topological operations are proposed step by step. Then, computational simulation implementation process is explained in greater detail. Typical fracture problem, single edge-notched tension test (SENT) is solved to demonstrate the efficiency and accuracy of the proposed method. In addition, double edge-notched tension test (DENT) as well as plate tension test with a slant crack is conducted to show the special fracture characters in viscoelastic solid propellant, like time dependence. Computational results reveal that the method proposed can be utilized in further fracture investigation in solid propellant combined with the experimental findings.

No abstract