End-to-end simulation
Placement → compaction → cure → residual stress → strength, in one twin.
Products
Plyoq Layup & Fiber Placement Voidra Defect Sensing & NDT Curevo Cure & Consolidation Strandix Laminate & Cure Digital Twin Roboply Robotics & Handling Yieldex Yield & Orchestration Request a pilotHit target strength before the cure — the composites digital twin.
A GPU-accelerated laminate-and-cure digital twin that simulates fiber placement, compaction, cure kinetics, exotherm, residual stress and structural strength — so a layup and cure recipe is validated and auto-optimized to hit target strength and yield before the cure.
Composite parts are qualified by building and testing them — an expensive, slow loop where a laminate’s true strength is only known after layup, cure and destructive coupon testing. Engineers cannot see, before committing carbon fiber and autoclave energy, whether a given layup and cure recipe will consolidate fully and hit target strength and yield. Process design stays conservative and iteration is measured in weeks and scrapped parts.
Hit target strength before the cure — the composites digital twin.— Strandix, part of the Curelum autonomy platform
No open-loop hoping. Strandix senses, reasons against the qualified envelope, acts within bounds, and logs every action.
The twin models fiber placement, steering and compaction to predict as-built ply geometry.
Cure kinetics, exotherm, consolidation and residual stress are simulated across the cycle.
Structural strength and allowables are predicted for the as-built, as-cured laminate.
The recipe is optimized in Omniverse for strength, yield, cycle time and energy — then pushed as control envelopes to Plyoq and Curevo.
Placement → compaction → cure → residual stress → strength, in one twin.
Optimizes for target strength, yield, cycle time and energy.
Generates rare porosity, wrinkle, FOD and cure-fault scenarios via Cosmos.
Omniverse + accelerated compute for fast, high-fidelity runs.
Validated envelopes flow directly to layup and cure control.
Model-validation records for NPI and airworthiness sign-off.
Every Strandix readout maps to a real measurement on the floor — tow geometry, cure trajectory, C-scan integrity, or predicted strength. The interface is precision test equipment: numbers you can certify, not a dashboard of vanity charts.
Strandix deploys on the factory edge and integrates with the systems you already run, under your security and export controls.
| Physics | Placement, compaction, cure kinetics, exotherm, residual stress, strength |
|---|---|
| Optimization | Strength, yield, cycle time, energy |
| Synthetic data | Rare porosity / wrinkle / FOD / cure-fault via Cosmos |
| Platform | Omniverse + accelerated compute |
| Output | Control envelopes for Plyoq & Curevo |
| Use | NPI, process design, qualification support |
Strandix runs in shadow, then assist, then bounded autonomy — grounded, auditable, and human-supervised at every airworthiness-critical checkpoint.
Strandix surrogates are validated against real parts, NDT and coupon data; predictions ship with confidence and are continuously corrected as ground truth arrives.
It reduces and focuses it. You iterate in simulation, then confirm with far fewer physical builds and coupons.
Optimized recipes are exported as bounded control envelopes consumed directly by Plyoq (layup) and Curevo (cure).
Start with one cell, a hard success metric, and a paid pilot. Prove ROI, then expand across the loop.