Design Electrical Machines in Minutes, not Weeks.
Motor Agent is an agentic AI-assisted, data-driven motor design system developed within the Power Electronics, Machines and Control (PEMC) Research Institute at the University of Nottingham.
Motor agent
We believe electrical machine design is a slow, iterative and expertise-driven loop of “spec → design → simulate → revise”, and that agentic AI systems can compress this cycle by guiding engineers with a pre-developed motor design database, finite-element solvers, and structured reasoning.
Current Development Problems
- Expert-dependent
- Iterative and manual
- Repetitive across similar motor families
- Poorly standardised for manufacturing scalability
Lack of Cohesion
Engineers frequently restart early-stage designs when specifications are not met, wasting valuable time. Manufacturing teams require designs aligned with production lines, but early design rarely incorporates these constraints systematically.
Increasing Demands
Meanwhile, global demand for high-performance electrical machines is accelerating rapidly, motor Agent addresses the bottleneck at the early design stage.
Our Solution
Motor Agent enables engineers to input natural-language specifications and receive concept-level motor designs within minutes.
It recommends motor topology and key parameters, reduces repeated manual iteration with automatic operation, integrates solver validation, links designs to a structured database and supports modular and standardised motor families.
This shifts the process from trial-and-error expert manual iteration to data-guided agentic AI assisted inverse design.
Motor Agent Team
Yiwei Wang
Entrepreneurial Lead
PhD Candidate, PEMC Research Institute
University of Nottingham
Dr. Sarah Newman
Technology Transfer Officer
Licensing/ Commercialisation Executive
University of Nottingham
Prof. Tao Yang
Principal Scientific Advisor
Prof. of Aerospace Electrical Systems
University of Nottingham