Abstract:Peridynamics (PD) is widely used to simulate structural failure. However, PD models are time-consuming. To improve the computational efficiency, we developed an adaptive coupling model between PD and classical continuum mechanics (PD-CCM) based on the Morphing method [1], driven by the broken bond or strength criteria. We derived the dynamic equation of the coupled models from the Lagrangian equation and then the discretized finite element formulation. An adaptive coupling strategy was introduced by determining the key position using the broken bond or strength criteria. The PD subdomain was expanded by altering the value of the Morphing function around the key position. Additionally, the PD subdomain was meshed by discrete elements (DEs) (i.e., nodes were not shared between elements), allowing the crack to propagate freely along the boundary of the DE. The remaining subdomains were meshed by continuous elements (CEs). Following the PD subdomain expansion, the CEs were converted into DEs, and new nodes were inserted. The displacement vector and mass matrix were reconfigured to ensure calculation consistency throughout the solving process. Furthermore, the relationship between the expansion radius of the PD subdomain and the speed of crack propagation was also discussed. Finally, the effectiveness, efficiency, and accuracy of the proposed model were verified via three two-dimensional numerical examples.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is to improve computational efficiency while accurately simulating crack formation and propagation during the structural dynamic failure process. Specifically, the paper aims to overcome the time - consuming computational drawback of the Peridynamics (PD) model and solve the singularity problem encountered by the classical continuum mechanics (CCM) when dealing with discontinuous problems (such as cracks) by combining an adaptive coupling model of PD and CCM.
### Main Problems and Solutions
1. **High Computational Cost of the PD Model**:
- The PD model uses integral equations to describe the force equilibrium, avoiding derivative operations in traditional partial differential equations, so it can well simulate behaviors such as crack initiation, propagation, and branching.
- However, the numerical computational cost of the PD model is relatively high, especially when dealing with complex structures.
2. **Limitations of the CCM Model in Dealing with Discontinuous Problems**:
- The CCM assumes that the structure is continuous and the deformation is twice differentiable, which leads to singularity problems when dealing with discontinuous problems such as cracks.
- To solve these problems, researchers have proposed various crack - handling methods, such as the extended finite element method, the element erosion method, and the phase - field method, but these methods are still relatively complex when dealing with complex problems such as crack initiation, coalescence, and branching.
### Solution: Adaptive PD - CCM Coupling Model
To overcome the above problems, the paper proposes an adaptive PD - CCM coupling model based on the Morphing method. The main features of this model are as follows:
- **Adaptive Expansion of the PD Sub - domain**: By introducing broken bonds or strength criteria, key positions are determined and the PD sub - domain is dynamically expanded, thus ensuring that cracks can propagate freely where needed.
- **Reduction of Computational Cost**: The PD model is only used where necessary, and the CCM model is used for the rest, thus effectively reducing the computational cost.
- **Maintenance of Computational Consistency**: By adjusting the displacement vector and the mass matrix, computational consistency is ensured throughout the solution process.
### Formula Summary
- **Bond - based Constitutive Relationship in the PD Model**:
\[
\hat{f}(x, x', t)=\frac{1}{2} C(x, x')\cdot\eta(x, x', t)
\]
where \(\eta(x, x', t) = u(x', t)-u(x, t)\), and \(C(x, x')\) is the microscopic modulus tensor function.
- **Bond Elongation**:
\[
s=\frac{\|x' - x\|+\eta(x, x', t)}{\|x' - x\|}
\]
- **Bond Damage History Function**:
\[
\mu(x, t)=\begin{cases}
1 & \text{if } s(x, t)<s_{\text{crit}}\\
0 & \text{otherwise}
\end{cases}
\]
- **Lagrangian Equation**:
\[
\frac{d}{dt}\left(\frac{\partial L}{\partial\dot{u}}\right)-\frac{\partial L}{\partial u} = 0
\]
where \(L = T - U\), \(T\) is the total kinetic energy of the system, and \(U\) is the total potential energy of the system.
Through this adaptive coupling model, the paper has successfully improved the computational efficiency while accurately simulating the crack behavior during the structural dynamic failure process.