Foundations of Mathematical Description of Solid Mechanics Problems
Solid mechanics seeks to predict how solid bodies deform, carry stress, and remain stable when subjected to external loads. Achieving reliable predictions demands a rigorous mathematical framework that links the geometry of deformation, the material’s physical response, and the governing balance laws. The following exposition outlines the essential components of that framework and highlights how they interlock to form a complete description of solid‑mechanics problems.
The first step is to characterize the change of shape that a body undergoes. In continuum mechanics two reference frames are commonly employed:
- Lagrangian (material) description – quantities are referred to the undeformed configuration, denoted by coordinates X.
- Eulerian (spatial) description – quantities are expressed in the current, deformed configuration, with coordinates x.
For problems involving large deformations it is crucial to distinguish the reference configuration (the original shape) from the current configuration (the deformed shape). The mapping
[
\mathbf{x} = \boldsymbol{\chi}(\mathbf{X},t)
]
relates the two, and its gradient defines the deformation gradient
[
\mathbf{F} = \frac{\partial \mathbf{x}}{\partial \mathbf{X}} .
]
(\mathbf{F}) captures local stretch, rotation, and shear. From (\mathbf{F}) we construct strain measures that quantify how distances change:
Small‑strain (Cauchy) tensor – appropriate when displacements are infinitesimal:
[
\varepsilon_{ij}= \tfrac12!\left(\frac{\partial u_i}{\partial x_j}+\frac{\partial u_j}{\partial x_i}\right),
]
where (\mathbf{u}=\mathbf{x}-\mathbf{X}) is the displacement field.Finite‑strain tensors – for large deformations we employ, for example, the Green–Lagrange strain
[
\mathbf{E}= \tfrac12!\left(\mathbf{F}^{!T}\mathbf{F}-\mathbf{I}\right)
]
or the Almansi strain[
\mathbf{e}= \tfrac12!\left(\mathbf{I}-\mathbf{F}^{-T}\mathbf{F}^{-1}\right).
]
These measures retain the geometric non‑linearity that becomes significant when rotations or stretches are no longer negligible.
Physical Laws and Constitutive Relations
Once the kinematic description is in place, the next task is to relate stresses to the chosen strain measure. This constitutive relation encodes the material’s intrinsic behavior and is the bridge between geometry and equilibrium.
Linear Elasticity
For small strains and isotropic materials, Hooke’s law in its generalized form reads
[
\sigma_{ij}= \lambda,\varepsilon_{kk},\delta_{ij}+2\mu,\varepsilon_{ij},
]
where (\lambda) and (\mu) are the Lamé constants, (\delta_{ij}) the Kronecker delta, and (\boldsymbol{\sigma}) the Cauchy stress tensor. The linear relation is the foundation of countless engineering analyses because of its simplicity and the availability of closed‑form solutions for many canonical problems.
Non‑linear Material Models
When the stress–strain response departs from linearity, more sophisticated models are required:
Hyperelasticity – stresses derive from a strain‑energy density function (W(\mathbf{F})). The Cauchy stress follows from
[
\boldsymbol{\sigma}= \frac{1}{J},\mathbf{F},\frac{\partial W}{\partial \mathbf{F}}^{!T},
]
with (J=\det\mathbf{F}). Popular forms include the Neo‑Hookean and Mooney–Rivlin models for rubber‑like substances.Elasto‑plasticity – combines an elastic law with a plastic flow rule. The J2 plasticity model, for instance, introduces a yield function (f(\boldsymbol{\sigma})) and an associated flow direction, allowing permanent deformations once the material reaches its yield surface.
Viscoelasticity – captures time‑dependent behavior. The Maxwell and Kelvin‑Voigt models represent the stress as a combination of elastic springs and viscous dashpots, often expressed through convolution integrals
[
\boldsymbol{\sigma}(t)=\int_{0}^{t}\mathbf{C}(t-\tau),\dot{\boldsymbol{\varepsilon}}(\tau),d\tau,
]
where (\mathbf{C}) is a relaxation tensor.
Choosing the appropriate constitutive law is a decisive step; it must reflect the material’s microstructure, loading rate, temperature, and any other relevant physical influences.
Balance (Equilibrium) Equations
The governing equations of solid mechanics stem from Newton’s second law applied to an infinitesimal volume element. In the static case, the sum of internal and external forces must vanish, leading to the equilibrium equation
[
\nabla!\cdot!\boldsymbol{\sigma} + \mathbf{b}= \mathbf{0},
]
where (\mathbf{b}) denotes body forces per unit volume (e.g., gravity).
For dynamic problems, inertia cannot be ignored, and the equation becomes
[
\nabla!\cdot!\boldsymbol{\sigma} + \mathbf{b}= \rho,\mathbf{a},
]
with (\rho) the material density and (\mathbf{a}) the acceleration field. These partial differential equations (PDEs) are the strong form of the problem; they must be satisfied at every point inside the domain.
In addition to force balance, the angular momentum balance imposes the symmetry of the Cauchy stress tensor for non‑polar materials:
[
\boldsymbol{\sigma} = \boldsymbol{\sigma}^{!T}.
]
Together, the equilibrium equations and the constitutive relations constitute a closed system once appropriate boundary conditions are prescribed.
Boundary Conditions and Solution Strategies
A well‑posed solid‑mechanics problem requires the specification of boundary data on the domain’s perimeter (\partial\Omega). Three principal types are used:
Displacement (Dirichlet) conditions – prescribe the motion on a portion (\Gamma_u):
[
\mathbf{u} = \bar{\mathbf{u}}\quad \text{on }\Gamma_u .
]Traction (Neumann) conditions – prescribe the surface force on (\Gamma_t):
[
\mathbf{t}= \bar{\mathbf{t}} = \boldsymbol{\sigma}\cdot\mathbf{n}\quad \text{on }\Gamma_t ,
]
where (\mathbf{n}) is the outward unit normal.Mixed or contact conditions – combine displacement and traction constraints, or enforce non‑penetration and friction laws on interacting bodies.
From Strong to Weak Form
Analytical solutions of the strong form are rare, especially for complex geometries or non‑linear material behavior. The variational (weak) form provides a pathway to numerical approximation. By multiplying the equilibrium equation by a virtual displacement (\delta\mathbf{u}) and integrating over the domain, we obtain the principle of virtual work:
[
\int_{\Omega}\boldsymbol{\sigma}:\nabla!\delta\mathbf{u},d\Omega
= \int_{\Omega}\mathbf{b}\cdot\delta\mathbf{u},d\Omega
- \int_{\Gamma_t}\bar{\mathbf{t}}\cdot\delta\mathbf{u},d\Gamma .
]
If the material is elastic, the left‑hand side can be expressed in terms of strain energy, leading to the minimum potential energy principle. This weak formulation is the foundation of most computational methods.
Numerical Discretization
The finite element method (FEM) discretizes the domain into elements, approximates the displacement field with shape functions, and assembles a global system of algebraic equations. For linear elastic problems the resulting system is linear; for non‑linear constitutive laws or large deformations, iterative schemes such as Newton–Raphson are employed.
Other discretization techniques—finite differences, spectral methods, meshfree approaches—share the same underlying philosophy: replace the continuous PDEs with a solvable set of algebraic equations while preserving the essential physics encoded in the original formulation.
Concluding Remarks
The mathematical description of solid‑mechanics problems rests on a layered structure:
- Geometric kinematics define how a body moves and deforms.
- Constitutive relations translate strain into stress, embodying material characteristics.
- Balance equations enforce Newtonian mechanics at every point.
- Boundary conditions complete the problem statement, guaranteeing uniqueness of the solution.
A solid grasp of each layer—and, more importantly, of how they interact—is indispensable for both analytical insight and the development of robust numerical tools. By carefully selecting the appropriate strain measure, constitutive model, and discretization strategy, engineers and researchers can predict the response of structures ranging from micro‑scale devices to massive civil infrastructures with confidence and precision.