Browser-native execution: Run thermodynamic simulations directly in the browser
TabularBackend: Pre-computed fluid tables for fast property lookups (100x faster than direct EOS calls)
Zero server dependency: No backend required - runs entirely client-side
JSON serialization: All results are JSON-serializable for easy integration

Installation

npm install @entropyk/wasm

Quick Start

import init, {
    WasmSystem,
    WasmCompressor,
    WasmCondenser,
    WasmEvaporator,
    WasmExpansionValve,
    WasmFallbackConfig
} from '@entropyk/wasm';

// Initialize the WASM module
await init();

// Create components
// WasmCompressor(fluid, speed_rpm, displacement_m3_per_rev, efficiency, m1..m10)
const compressor = new WasmCompressor(
    "R134a",    // fluid
    2900.0,     // speed_rpm
    0.0001,     // displacement_m3_per_rev
    0.85,       // mechanical_efficiency
    1.0, 0.0, 0.0, 0.0, 0.0,  // AHRI 540 coefficients m1-m5
    0.0, 0.0, 0.0, 0.0, 0.0   // AHRI 540 coefficients m6-m10
);

// WasmCondenser(ua) — thermal conductance in W/K
const condenser = new WasmCondenser(1000.0);

// WasmEvaporator(ua) — thermal conductance in W/K
const evaporator = new WasmEvaporator(800.0);

// WasmExpansionValve(fluid, capacity_kW) — pass 0 for no capacity limit
const valve = new WasmExpansionValve("R134a", 0.0);

// Create system and add components
const system = new WasmSystem();
const n0 = system.add_component(compressor.into_component());
const n1 = system.add_component(condenser.into_component());
const n2 = system.add_component(valve.into_component());
const n3 = system.add_component(evaporator.into_component());

// Connect components in a cycle
system.add_edge(n0, n1)?; // compressor → condenser
system.add_edge(n1, n2)?; // condenser → valve
system.add_edge(n2, n3)?; // valve → evaporator
system.add_edge(n3, n0)?; // evaporator → compressor

// Finalize topology
system.finalize()?;

// Configure solver
const config = new WasmFallbackConfig();

// Solve
const result = system.solve(config);

console.log(result.toJson());
// {
//   "converged": true,
//   "iterations": 12,
//   "final_residual": 1e-8,
//   "status": "Converged"
// }

// Get thermodynamic state at a node
const state = system.get_node_result(0);
console.log(state.toJson());
// {
//   "pressure": { "pascals": 1000000.0 },
//   "temperature": { "kelvin": 310.5 },
//   "enthalpy": { "joules_per_kg": 420000.0 },
//   "mass_flow": { "kg_per_s": 0.0 }
// }

API Reference

Core Types

WasmPressure — Pressure in Pascals (constructor validates non-negative)
- new(pascals) — from Pascals
- from_bar(bar) — from bar
WasmTemperature — Temperature in Kelvin (constructor validates non-negative)
- new(kelvin) — from Kelvin
- from_celsius(celsius) — from Celsius
WasmEnthalpy — Enthalpy in J/kg (constructor rejects NaN)
- new(joules_per_kg) — from J/kg
- from_kj_per_kg(kj_per_kg) — from kJ/kg
WasmMassFlow — Mass flow in kg/s (constructor rejects NaN)
- new(kg_per_s) — from kg/s

All types have a toJson() method returning a JSON string.

Components

WasmCompressor — AHRI 540 compressor model
- new(fluid, speed_rpm, displacement, efficiency, m1..m10) — all 10 AHRI coefficients required
WasmCondenser — Heat rejection heat exchanger
- new(ua) — thermal conductance in W/K (must be positive)
WasmEvaporator — Heat absorption heat exchanger
- new(ua) — thermal conductance in W/K (must be positive)
WasmExpansionValve — Isenthalpic expansion device
- new(fluid, capacity_kW) — pass 0 for no capacity limit
WasmPipe — Transport pipe
- new(fluid, length, diameter, density, viscosity) — all parameters required

Each component has an into_component() method that converts it to a generic WasmComponent for adding to the system.

Solver

WasmSystem — Thermodynamic system container
- add_component(component) → node index
- add_edge(from_idx, to_idx) — connect two component nodes
- finalize() — finalize topology before solving
- solve(config) — solve with fallback strategy
- solve_newton(config) — solve with Newton-Raphson only
- get_node_result(idx) — get thermodynamic state at a node after solving
WasmNewtonConfig — Newton-Raphson solver configuration
- set_max_iterations(n), set_tolerance(tol)
WasmPicardConfig — Sequential substitution solver configuration
- set_max_iterations(n), set_relaxation_factor(omega) — omega clamped to (0, 1]
WasmFallbackConfig — Intelligent fallback solver configuration
- set_fallback_enabled(bool), set_return_to_newton_threshold(f64), set_max_fallback_switches(usize)
WasmConvergedState — Solver result with converged, iterations, final_residual, status

Fluid Management

list_available_fluids() — returns array of available fluid names
load_fluid_table(json_string) — load a custom fluid table from JSON

Build Requirements

Rust 1.70+
wasm-pack: cargo install wasm-pack
wasm32 target: rustup target add wasm32-unknown-unknown

Building from Source

# Clone the repository
git clone https://github.com/entropyk/entropyk.git
cd entropyk/bindings/wasm

# Build for browsers
wasm-pack build --target web

# Build for Node.js
wasm-pack build --target nodejs

# Run Rust WASM tests
wasm-pack test --node

TypeScript Support

TypeScript definitions are auto-generated by wasm-pack build in the pkg/ directory. After building, import the generated .d.ts file:

import init, { WasmSystem } from './pkg/entropyk_wasm';

Performance

Operation	Target	Typical
Simple cycle solve	< 100ms	30-50ms
Property query	< 1μs	~0.5μs
Cold start	< 500ms	200-300ms

Limitations

CoolProp unavailable: The WASM build uses TabularBackend with pre-computed tables. CoolProp C++ cannot compile to WebAssembly.
Limited fluid library: By default, only R134a is embedded. Additional fluids can be loaded from JSON tables generated by the entropyk CLI.
Custom fluid tables: load_fluid_table validates the table but registration in the global backend requires additional infrastructure (planned).

Browser Compatibility

Chrome/Edge 80+
Firefox 75+
Safari 14+
Node.js 14+

License

MIT OR Apache-2.0

README.md

Entropyk WebAssembly Bindings

Features