Douglas L. Jones

A Stiffness Degradation Model for Graphite/Epoxy Laminates

Stiffness reduction in composite laminates is an important measure of fa tigue damage. The determination of fatigue damage and the prediction of fatigue life can be made through the development of a stiffness degradation model. This paper proposes a stiffness degradation model that can be used to predict the statistical distribution of the residual stiffness of composite laminates subjected to fatigue cycling. Based on the pro posed model, two analytical methods are presented, which are capable of predicting the stiffness degradation of a particular composite specimen under cyclic loading. One method is based on the linear regression analysis and the other on the Bayesian approach. Experi ments have been performed on graphite/epoxy [90, +45, —45,0], laminates to generate sta tistically significant data for evaluating the proposed analytical models and for verifying the predicted results. It is shown that theoretical predictions for the stiffness degradation of an individual specimen and for the statistical distribution of the entire population corre late reasonably well with experimental results.

Statistical Fatigue of Graphite/Epoxy Angle-Ply Laminates in Shear1

A three-parameter fatigue and residual strength degradation model has been proposed to predict statistically the fatigue behavior of composite laminae under axial shear loadings. The fatigue behavior includes the fa tigue life and the fatigue damage expressed in terms of the residual strength degradation. An experimental test program using graphite/epoxy [±45°] 2s laminates has been conducted to generate statistically meaningful data in order to examine the validity of the theoretical model. It is shown that the correlation between the theoretical predictions and the test results on the statistical distributions of the fatigue life and the residual strength is excel lent. Test results on the shear modulus degradation are also presented and discussed in detail in order to provide insight for the establishment of a shear modulus degradation model.

A Stiffness-Based Statistical Model for Predicting the Fatigue Life of Graphite/Epoxy Laminates

A statistical model for predicting the fatigue life distribution of graphite/epoxy laminates under constant-amplitude tensile cyclic loadings has been proposed based on a stiffness degradation model and a failure stiffness criterion. The prediction of fatigue life requires base-line experimental data for the static ultimate strength and stiffness degradation. An experimental program has been conducted using graphite/epoxy [90, +45, −45,0] s laminates to generate statistically meaningful data for the verification of the proposed model

Development and Application of a Model Using Center of Gravity of Hysteresis Loops to Predict Fatigue Damage Accumulation in Fiber-reinforced Plastic Laminates

A new methodology for modeling fatigue damage accumulation in fiber-reinforced plastic (FRP) laminates is introduced. This new model, termed the CG method, or the CG Model, utilizes the center of gravity (CG) of hysteresis loops (HL). It is inspired by previous investigations into damage accumulation models based on HL, such as maximum displacement (the width of the HL), energy dissipation (the area of the HL), plastic strain (the width of the interior part of the HL), and stiffness reduction. The CG method combines all these models, relates them to each other, and presents very consistent results. The CG model is especially helpful in cases where the other models cannot be used. This study shows that using the CG model a set of straight lines is created that present the evolution of the positions of the CG of the HLs during fatigue tests. These ‘curves’ may then be employed to build the complete set of curves that, even for untested specimens at different stress levels, accurately predict fatigue damage accumulation in FRP laminates. Employing this method also provides more insight into the damage condition of the specimen than other models allow.

Computer-based Design Optimization IncludingCost And Environmental Factors

In the last 10 to 20 years computers clearly have become indispensable in helping to design higher quality products and reduce the time required to bring them to market. Computer-aided engineering (CAE) tools have generally been focused on improving the visualization of the design process, optimizing the virtual product, reducing the time required for prototyping, and producing engineering documentation. However, the success of commercial products ultimately depends upon the total life-cycle costs, including recycling and disposal of the components. These concerns have not yet been addressed to any significant degree in the CAE environment. A method is proposed by which the CAE environment can be used to facilitate the evaluation and analysis of all these different needs. This should lead to a design that is optimized not only for strength, material, and aesthetics, but also for cost and environmental factors that would produce a higher confidence level for success of the product.

Model-based Approaches For Robust ParameterDesign

Since Taguchi methods have been introduced in many major American industries, quality improvement experiments have emphasized optimizing the mean value of the performance characteristics, while placing secondary attention on the variance. However, some weaknesses in Taguchi methods have been identified, and some controversies still exist. As an alternative, several researchers have combined important aspects of the Taguchi methods with classical response surface methods. The focus in these enhancements has been placed on building models that involve direct interactions between the control factors (C*C), and establishing separate models for the mean and variance of the performance characteristics. In this paper, the responses of the mean and variance are derived according to the following scenarios: i) a model with control and external noise factors, ii) a model with control and internal noise factors, and iii) a model with control, external noise, and internal noise factors. Also, both the mean and variance can be obtained using combined and product arrays. Particularly, the model used is represented by linear main effects in the control and noise (N) factors, two-factor interactions and quadratic terms in the control factors, second-order C*N interactions, and possibly some third-order C*CxN interactions, even though these third-order interactions may be less important than the firstorder or second-order effects. In addition, these third-order terms should be included in modeling the performance characteristics when the product array is used for the experiments.