Abstract modelling of the nonlinear mechanical response of joints in fiber reinforced composite assemblies

在航空航天和汽车工业中，用于大型组件数值分析的模型可能包含数百甚至数千个接头。这使得模型中接头的数值有效表示，同时保持足够的精度，这是一个至关重要的问题。特别是，这适用于与预测碰撞有关的模拟-为了这些目的，接头的非线性响应的精确表示是必不可少的。详细的建模方法具有真实的连接器属性和几何形状的表示，通常作为固体，允许关于非线性响应的可靠预测。然而，这通常导致非常高的自由度，即未知数的总数，即使对于单个连接器也是如此。这使得这种方法对于大型组件的分析是不可行的。作为详细建模的替代方案，不基于关节的实际几何形状或物理特性的忠实表示的策略。在这种情况下，关节被减少到少量节点，通常只有两个，这种方法被称为抽象建模，它通常与低自由度相关，因此适合于大型装配模型。有限元分析中的连接件抽象建模已经是工业应用中广泛使用的工具，然而，由于纤维增强复合材料的特殊性质，其应用范围越来越广，这一发展促进了现有方法的扩展。这些材料最突出的特性是它们的各向异性，这是现有方法无法处理的。此外，复合材料中各种接头的失效响应更为复杂，例如，层合板中的螺栓连接件可能以伪韧性或脆性的方式失效。这激发了本提议，该提议涉及考虑复合材料失效模式和各向异性的抽象接头建模策略的发展。

Models for the numerical analysis of large scale assemblies in the aerospace and automotive industry may comprise hundreds, or even thousands, of joints.This makes a numerically efficient representation of the joint in the model, while maintaining sufficient accuracy, an issue of paramount importance.In particular, this applies to simulations concerned with the prediction crash-worthiness or service life time of an assembly.For these purposes an accurate representation of the nonlinear response of joints is essential.Detailed modeling approaches featuring a faithful representation of the actual connector properties and geometry, typically as a solid, allow reliable predictions regarding the nonlinear response.However, this generally results on a very high degree of freedom, i.e. total number of unknowns, even for individual connectors.This renders this approach infeasible for the analysis of large scale assemblies.As an alternative to detailed modeling, strategies that is not based on a faithful representation of the actual geometry or physical properties of the joint.In that case the joint is reduced to a low number of nodes, typically only two, and their load-displacement relation is adjusted to be equivalent to the macroscopically apparent load-displacement relation of the actual joint.This approach is referred to as abstract modeling here, it is generally associated with a low degree of freedom and is, hence, suitable for large assembly models.Abstract modeling of joints in finite element analysis is already a wide spread tool in industrial applications, however, typically limited to linear analysis and isotropic materials.Recently, the application of fiber reinforced composites has become more wide spread.This development prompts an extension of existing methods, due to the particular properties of these materials. The most prominent property of these materials is their anisotropy which can not be handled by existing approachesFurthermore, the failure response of various joints in composites is more complex, e.g. bolt connectors in laminates may fail in a pseudo ductile or brittle manner.This motivates the present proposal which is concerned with the development of abstract joint-modeling strategies accounting for failure modes and anisotropy of composites.