Abstract
In this paper, we propose an optimization for deploying symbolic
regression-based energy management algorithms on embedded hardware.
We investigate two strategies: parameter reduction using Spearman
correlation analysis, and the use of fixed-point arithmetic to replace
floating-point operations. Both approaches are evaluated on an ARM
Cortex-A55 platform using real household power measurements. The results
show significant reductions in computational effort with minimal
loss in control performance, enabling efficient embedded deployment of
symbolic regression energy management systems.
regression-based energy management algorithms on embedded hardware.
We investigate two strategies: parameter reduction using Spearman
correlation analysis, and the use of fixed-point arithmetic to replace
floating-point operations. Both approaches are evaluated on an ARM
Cortex-A55 platform using real household power measurements. The results
show significant reductions in computational effort with minimal
loss in control performance, enabling efficient embedded deployment of
symbolic regression energy management systems.
| Original language | English |
|---|---|
| Publication status | Accepted/In press - 2026 |
| Event | 20th International Conference on Computer Aided Systems Theory - Museo Elder de la Ciencia y la Tecnología, Las Palmas de Gran Canaria, Spain Duration: 23 Feb 2026 → 27 Feb 2026 https://eurocast2026.fulp.es/ |
Conference
| Conference | 20th International Conference on Computer Aided Systems Theory |
|---|---|
| Abbreviated title | Eurocast 2026 |
| Country/Territory | Spain |
| City | Las Palmas de Gran Canaria |
| Period | 23.02.2026 → 27.02.2026 |
| Internet address |
Keywords
- Symbolic Regression
- Energy Management System
- Optimization
- Embedded Systems