Introduction to the FEA Solver

The Pipe Stress FEA Engine is a specialized, 3D structural analysis tool built explicitly for piping and pipeline systems.

At its core, the engine utilizes a Direct Stiffness Method matrix formulation. It relates the global nodal displacements $\{d\}$ to the applied nodal forces $\{F\}$ through the global stiffness matrix $[K]$:

$\{F\} = [K]\{d\}$

Unlike generalized structural FEA tools (which often require manual meshing and complex boundary definitions), this engine is purposefully heavily automated. It abstracts away the complex discretization of piping components—such as elbows, tees, and buried soil-spring interactions—allowing the engineer to focus on the piping layout rather than the mesh topology.

The Three-Phase Architecture

To handle the combination of static loads, non-linear boundaries, and dynamic code-compliance checks, the engine executes every analysis through a strict three-phase sequence.

Phase 1: Preprocessing & Matrix Assembly

Before solving, the engine interprets the physical piping geometry and prepares it for matrix math:

• Discretization: Elbows are sliced into faceted segments to capture true flexibility. Buried pipes are automatically sub-meshed based on their characteristic diameter to apply distributed soil springs.
• Property Assignment: Pipe section properties, insulation mass, fluid mass, and ASME B31.3 Appendix D flexibility factors and Stress Intensification Factors (SIFs) are calculated.
• Matrix Building: The $12 \times 12$ local stiffness matrices for each element are assembled, rotated into the global coordinate system, and injected into the Global Stiffness Matrix $[K]$.

Phase 2: State Resolution (The Sequencer)

Because piping systems experience different non-linear behaviors depending on the load case, the engine acts as a "traffic cop" to route the matrices through the appropriate mathematical solver:

1. Standard Non-Linear States (SUS, OPE): The engine iterates a Penalty Method for gap closures, one-way lift-off and limit stops, combined with an elastic-Coulomb stick-slip friction model in which each sticking support contributes a real tangential spring to the stiffness matrix (see Boundary Conditions).
2. Expansion Range (EXP): For nonlinear models the displacement-stress range is computed as the algebraic difference of fully converged states (operating − sustained, enveloped over hot/cold candidates), using the classical L1 − L2 combination (see Static & Thermal Loads).
3. Occasional States (Wind, Seismic): The engine superimposes the environmental loads on top of the deformed Operating (OPE) state. The operating contact state is frozen, dynamic snubbers are locked using rigid penalty stiffnesses, and friction is omitted from the overlay.
4. Modal Analysis: The solver bypasses the static force vector, builds a Consistent Mass Matrix $[M]$, and performs a Generalized Eigenvalue Extraction to find the fundamental frequencies, normalized mode shapes, and per-direction mass participation.

Phase 3: Code Compliance & Post-Processing

Once the global displacements $\{d\}$ are solved and the system has converged, the engine translates those movements back into localized forces.

• Element Forces: Displacements are rotated back to local axes to calculate the in-plane, out-of-plane, and torsional bending moments acting on every component.
• Stress Evaluation: The localized forces are routed through the selected Design Code (e.g., ASME B31.3, ASME B31.1, or AS 4041) to calculate the Sustained ($S_L$) and Expansion ($S_E$) stresses.
• Compliance Checks: Calculated stresses are compared against the material's allowable limits, triggering B31 "Liberal Allowable" logic where applicable.
• Flange Leakage: Local bending moments and axial forces are extracted to evaluate equivalent pressures for flange leakage assessments.

Automation Philosophy

The engine is designed to eliminate "setup fatigue." For example, when you define a pipeline as "Buried," you do not need to manually calculate the soil spring rates or determine the mesh density. The Preprocessor handles the discretization and ALA/ASCE spring calculations automatically during Phase 1.