Struct ArcLength 

pub struct ArcLength<NId = NodeId> {

    pub initial_radius: f64,
    pub min_radius: f64,
    pub max_radius: f64,
    pub target_iterations: usize,
    pub max_retries: usize,
    pub direction: ArcDirection,
    pub scales: ArcScales,
    pub predictor: ArcPredictor,
    pub seed: Option<ArcSeed<NId>>,
    pub arc_tolerance: f64,
    pub correction_tolerance: Option<f64>,
    pub backtracking: Backtracking,
    pub stop: ArcStopCriteria<NId>,
}

Arc-length continuation settings (Integrator::ArcLength). Build with ArcLength::fixed or ArcLength::adaptive and adjust fields from there; every field is validated when the first step runs.

Fields

initial_radius: f64

Radius of the first step, in the dimensionless metric.

min_radius: f64

Cutbacks never go below this radius.

max_radius: f64

Growth never goes above this radius. Setting all three radii equal gives a fixed-radius run with no retries or growth.

target_iterations: usize

Corrector solves the adaptive radius aims for (docs/arclength.md §7: s_next = s * clamp(sqrt(target / N), 0.5, 1.5)).

max_retries: usize

Cutbacks allowed per step before the step fails.

direction: ArcDirection

Initial direction of travel in lambda.

scales: ArcScales

How the metric is scaled.

predictor: ArcPredictor

Predictor used after the first step.

seed: Option<ArcSeed<NId>>

Orients the first tangent when it is singular (or overrides the default direction-based orientation when given).

arc_tolerance: f64

Acceptance tolerance on the constraint, |g| / s^2.

correction_tolerance: Option<f64>

Optional acceptance tolerance on the last undamped coupled correction, relative to the radius: ||delta||_M <= tol * s.

backtracking: Backtracking

stop: ArcStopCriteria<NId>

Implementations

impl<NId> ArcLength<NId>

pub fn fixed(radius: f64, scales: ArcScales) -> Self

Fixed radius radius with every other setting at its default.

pub fn adaptive( initial_radius: f64, min_radius: f64, max_radius: f64, scales: ArcScales, ) -> Self

Adaptive radius in [min_radius, max_radius], starting at initial_radius, with every other setting at its default.

pub fn validate_with( &self, algorithm: &Algorithm, test: &ConvergenceTest, ) -> Result<(), AnalysisError>

Checks this configuration together with the algorithm and convergence test it will run under — the same checks every arc step makes, exposed so wire decoders can reject bad input up front. Arc length needs Newton { tangent: Current, line_search: None } (its coupled backtracking replaces line search) and a force test (NormUnbalance or Combined); displacement/work tests alone don’t certify equilibrium.

pub fn validate(&self) -> Result<(), AnalysisError>

Range checks on this configuration alone.

Trait Implementations

impl<NId: Clone> Clone for ArcLength<NId>

fn clone(&self) -> ArcLength<NId>

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl<NId: Debug> Debug for ArcLength<NId>

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<NId: PartialEq> PartialEq for ArcLength<NId>

fn eq(&self, other: &ArcLength<NId>) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl<NId: PartialEq> StructuralPartialEq for ArcLength<NId>