Struct GpuEulerGrid 

pub struct GpuEulerGrid {
    pub nx: usize,
    pub ny: usize,
    pub nz: usize,
    pub dx: f64,
    pub rho: Vec<f64>,
    pub u: Vec<f64>,
    pub v: Vec<f64>,
    pub w: Vec<f64>,
    pub p: Vec<f64>,
}

A 3-D staggered MAC grid for Eulerian fluid simulation.

Velocity components are face-centred; pressure and density are cell-centred.

Fields

nx: usize

Number of cells in x.

ny: usize

Number of cells in y.

nz: usize

Number of cells in z.

dx: f64

Grid spacing (m).

rho: Vec<f64>

Cell-centred density (kg/m³).

u: Vec<f64>

x-face velocity (m/s), size (nx+1)·ny·nz.

v: Vec<f64>

y-face velocity (m/s), size nx·(ny+1)·nz.

w: Vec<f64>

z-face velocity (m/s), size nx·ny·(nz+1).

p: Vec<f64>

Cell-centred pressure (Pa).

Implementations

impl GpuEulerGrid

pub fn new(nx: usize, ny: usize, nz: usize, dx: f64) -> Self

Create a new GpuEulerGrid with nx × ny × nz cells and spacing dx.

pub fn index(&self, i: usize, j: usize, k: usize) -> usize

Flat cell index for cell (i, j, k).

pub fn cell_count(&self) -> usize

Total number of cells.

pub fn max_velocity(&self) -> f64

Maximum speed over all velocity components.

pub fn cfl_timestep(&self, cfl: f64) -> f64

CFL-safe time step: cfl · dx / max_velocity (or dx if velocity is zero).

pub fn gpu_apply_gravity(&mut self, gravity: f64, dt: f64)

Apply gravitational acceleration to the v (y) velocity field.

pub fn gpu_compute_divergence(&self) -> Vec<f64>

Compute the divergence field ∇·u at every cell.

pub fn divergence_at(&self, i: usize, j: usize, k: usize) -> f64

Local divergence at cell (i, j, k).

pub fn vorticity_at(&self, i: usize, j: usize, k: usize) -> [f64; 3]

Local vorticity (curl of velocity) at cell (i, j, k).

Returns [ωx, ωy, ωz] using central differences.

pub fn gpu_enforce_solid_bc(&mut self)

Zero normal velocities at domain boundaries (solid wall BC).

pub fn gpu_pressure_projection(&mut self, max_iter: usize, tol: f64) -> f64

Jacobi pressure projection to enforce incompressibility.

Iterates up to max_iter times; returns the final L-inf residual.

pub fn gpu_update_velocity_from_pressure(&mut self, dt: f64)

Update velocity from pressure gradient: u -= dt · ∇p.

pub fn gpu_advect_semi_lagrange(&mut self, dt: f64)

Semi-Lagrangian advection of velocity (first-order back-trace).

pub fn gpu_vorticity_confinement(&mut self, epsilon: f64)

Vorticity confinement: adds a force proportional to epsilon toward regions of high vorticity to counteract numerical diffusion.

pub fn compute_stats(&self) -> FluidSimStats

Compute simulation diagnostics.

Trait Implementations

impl Clone for GpuEulerGrid

fn clone(&self) -> GpuEulerGrid

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GpuEulerGrid

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

Formats the value using the given formatter.