| Download | - View final version: Machine learning enhanced design and knowledge discovery for multi-junction photonic power converters (PDF, 3.0 MiB)
- View supplementary information: Machine learning enhanced design and knowledge discovery for multi-junction photonic power converters (PDF, 64.6 MiB)
|
|---|
| DOI | Resolve DOI: https://doi.org/10.1038/s41598-025-16408-4 |
|---|
| Author | Search for: Hunter, Robert F. H.; Search for: Forcade, Gavin P.; Search for: Grinberg, Yuri1ORCID identifier: https://orcid.org/0000-0003-3349-1590; Search for: Wilson, D. Paige; Search for: Beattie, Meghan N.; Search for: Valdivia, Christopher E.; Search for: de Lafontaine, Mathieu; Search for: St-Arnaud, Louis-Philippe; Search for: Helmers, Henning; Search for: Höhn, Oliver; Search for: Lackner, David; Search for: Pellegrino, Carmine; Search for: Krich, Jacob J.; Search for: Walker, Alexandre W.2ORCID identifier: https://orcid.org/0000-0002-1791-2140; Search for: Hinzer, Karin |
|---|
| Affiliation | - National Research Council of Canada. Digital Technologies
- National Research Council of Canada. Quantum and Nanotechnologies
|
|---|
| Funder | Search for: National Research Council Canada; Search for: National Sciences and Engineering Research Council of Canada; Search for: The Canadian Foundation for Innovation; Search for: The Government of Ontario; Search for: The German Federal Ministry of Education and Research; Search for: ERC grant PHASE |
|---|
| Format | Text, Article |
|---|
| Subject | machine learning; dimensionality reduction; design discovery; optimization acceleration; knowledge discovery; multi-junction photonic power converters |
|---|
| Abstract | Machine learning is proving to be a revolutionary tool across many disciplines, including optoelectronic device design. In this report, we compare classical and machine learning enhanced design optimization methodologies. We investigate, as an example case, the design of the complex structures of tenjunction InP lattice matched photonic power converters with In₀.₅₃Ga₀.₄₇As absorbers optimized for operation at 1550 nm. We find that the implicit pattern recognition capabilities of dimensionality reduction using principal component analysis accelerate design discovery, optimization, and the understanding of complex optical phenomena in the simulated devices. The dimensionality reduction approach offers over twenty times as many optimal designs with greater variability and with a 15% reduction in computational cost compared to a classical optimization method. Furthermore, we find that the representation of the reduced dimensionality subspace offers an intuitive interpretation of optical phenomena expected to occur in this design problem. This method is general and offers the potential for knowledge discovery, expanded design perspective, and optimization acceleration in conjunction with a significant reduction in computational expense in systems which can be numerically modeled. |
|---|
| Publication date | 2025-09-26 |
|---|
| Publisher | Springer Nature |
|---|
| Licence | |
|---|
| In | |
|---|
| Language | English |
|---|
| Peer reviewed | Yes |
|---|
| Export citation | Export as RIS |
|---|
| Report a correction | Report a correction (opens in a new tab) |
|---|
| Record identifier | 0a590f7b-cb4f-4bd3-bbb5-7ac3afd1d19a |
|---|
| Record created | 2025-10-08 |
|---|
| Record modified | 2025-11-03 |
|---|
