Elastoplastic Damage Analysis And Compressive Strength Prediction Of Laminates With Fiber Wrinkle Defects

Fiber wrinkle is a typical type of manufacturing defect in fiber-reinforced composite material and structures,which can result in severe degradation of strength and stiffness of composite laminates,especially for compressive strength.Therefore,the analysis of compressive failure and strength prediction are greatly important for laminates with fiber wrinkle defects.To investigate the effect of fiber wrinkle on the damage evolution and compressive strength,numerical analysis was conducted to simulate the progressive failure behavior of fiber-reinforced composite laminates with wrinkle defects subjected to axial compression based on the finite element method(FEM)and back propagation neural network(BPNN)in this paper.The contributions and conclusions in this paper are as follows:(1)A three-dimensional(3D)elastoplastic damage model with four damage modes(i.e.,fiber tensile failure,matrix failure,fiber kinking/splitting,and delamination)was developed based on the La RC05 criterion with considering the elastoplastic effect of the matrix.In the matrix analysis,a modified algorithm of searching for the fracture angle was proposed by combining the golden section search method and an inverse interpolation.The efficiency of the approach is 10 times higher than that of the traversal method.(2)Numerical simulations were performed to investigate the damage evolution of multi-directional laminates under axial compression based on the developed 3D elastoplastic damage model.The results show that the 3D elastoplastic damage model can not only reproduce the nonlinearity of the laminate during the evolution of the damage,but also provide more accurate damage location and compressive strength predictions than the previous model.The compressive strength of laminates with wrinkle defects has a nearly linear decrease with the increase of the maximum misalignment angle.For wrinkle level 11.4°,the compressive strength reduces to 68% of the pristine strength.The dominant failure mode of laminates changes from fiber kinking to delamination failure.For laminates with low wrinkle levels,the plastic effect of matrix has a significant effect on the damage evolution and compressive strength,which cannot be ignored.(3)Numerical analysis of wrinkled unidirectional laminates(UD)under off-axis compression was carried out based on the developed 3D elastoplastic damage model.The results show that the compressive strength decreases nonlinearly with the increase of the maximum misalignment angle,though the stiffness of the laminate has little change with the wrinkle level.The dominant failure mode of UD laminates change from fiber kinking to delamination failure with the increase wrinkle levels.The off-axis compressive strength of UD laminates with same wrinkle level varies sinusoidal with the increase of the orientation angle,and the minimum strength occurs at the case of orientation angle 45°.The waviness plays a great role on the amplitude and distribution of shear stresses,which results in increase contribution of shear stress to the off-axis compressive failure of laminates.(4)The BPNN with three hidden layers was constructed to predict the compressive strength of laminates containing embedded fiber wrinkles by using the numerical results.The best number of the neurons,i.e.12,was determined based on the golden section method by comparing the prediction results and the assessment indicators of different cases.The results show that,the error of the strengths of the laminates with the maximum wrinkle angles of 5.6°,9.9° and 11.4° predicted by the developed BP neural networks are 3.4%,4.6%,and-0.01%,respectively.The comparison of the prediction with the test results shows that,the developed BPNN has high accuracy and robustness.