Aerospace organisations must maximize the performance of components while minimizing weight and meeting strict certification requirements. This requires: innovation in materials design; optimization and control of processes; and robust testing programs. Alchemite can help all of these activities by ensuring that you get maximum return from the available data and by informing your next design, testing, or processing decision.
Case Study - Alloy Design
Challenge: Commercially-available superalloys are not always optimized for high-performance engineering applications, such as aerospace engines. But designing new alloys through empirical experimentation can take many years.
Solution: The Alchemite deep learning software can be used as a computational design tool, identifying which potential alloys have the greatest probability of meeting a design specification constrained by many different material properties. Alchemite can be trained using a combination of existing experimental data and computational thermodynamic predictions, despite the fact that combining these disparate sources resulted in many gaps in the data.
Outcome: Alchemite predicted that a new nickel-base alloy offered the optimal combination of properties for disc applications. Seven of these properties were experimentally verified, demonstrating better yield stress and oxidation resistance than commercially-available alternatives. Alchemite has also been used to design a nickel-base alloy for a combustor liner, and two Mo-based alloys for forging tools.
Alchemite for aerospace engineering
Aerospace scientists and engineers use Alchemite’s powerful deep learning methods to:
- Design new alloys, considering both material design and process parameters
- Run virtual experiments on candidate materials and processes
- Understand uncertainty and guide testing for certification programs
- Monitor outputs and optimize inputs for critical processes