PIC2O-Sim: A physics-inspired causality-aware dynamic convolutional neural operator for ultra-fast photonic device time-domain simulation
Optical simulation plays an important role in photonic hardware design flow. The finite-difference time-domain (FDTD) method is widely adopted to solve time-domain Maxwell equations. However, FDTD is known for its prohibitive runtime cost as it iteratively solves Maxwell equations and takes minutes...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIP Publishing LLC
2025-03-01
|
| Series: | APL Photonics |
| Online Access: | http://dx.doi.org/10.1063/5.0242728 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850259361035714560 |
|---|---|
| author | Pingchuan Ma Haoyu Yang Zhengqi Gao Duane S. Boning Jiaqi Gu |
| author_facet | Pingchuan Ma Haoyu Yang Zhengqi Gao Duane S. Boning Jiaqi Gu |
| author_sort | Pingchuan Ma |
| collection | DOAJ |
| description | Optical simulation plays an important role in photonic hardware design flow. The finite-difference time-domain (FDTD) method is widely adopted to solve time-domain Maxwell equations. However, FDTD is known for its prohibitive runtime cost as it iteratively solves Maxwell equations and takes minutes to hours to simulate a single device. Recently, AI has been applied to realize orders-of-magnitude speedup in partial differential equation solving. However, AI-based FDTD solvers for photonic devices have not been clearly formulated. Directly applying off-the-shelf models to predict the optical field dynamics shows unsatisfying fidelity and efficiency since the model primitives are agnostic to the unique physical properties of Maxwell equations and lack algorithmic customization. In this work, we thoroughly investigate the synergy between neural operator designs and the physical property of Maxwell equations and introduce a physics-inspired AI-based FDTD prediction framework PIC2O-Sim. PIC2O-Sim features a causality-aware dynamic convolutional neural operator as its backbone model that honors the space–time causality constraints via careful receptive field configuration and explicitly captures the permittivity-dependent light propagation behavior via an efficient dynamic convolution operator. Meanwhile, we explore the trade-offs among prediction scalability, fidelity, and efficiency via a multi-stage partitioned time-bundling technique in autoregressive prediction. Multiple key techniques have been introduced to mitigate iterative error accumulation while maintaining efficiency advantages during autoregressive field prediction. Extensive evaluations on three challenging photonic device simulation tasks have shown the superiority of our PIC2O-Sim method, showing 51.2% lower roll-out prediction error, 23.5 times fewer parameters than state-of-the-art neural operators, providing 133–310× or 31–89× higher simulation speed than an open-source single-process or eight-process parallel FDTD numerical solver. |
| format | Article |
| id | doaj-art-b279e8b1779c4a0b98e88ed3eec74e44 |
| institution | OA Journals |
| issn | 2378-0967 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | AIP Publishing LLC |
| record_format | Article |
| series | APL Photonics |
| spelling | doaj-art-b279e8b1779c4a0b98e88ed3eec74e442025-08-20T01:55:52ZengAIP Publishing LLCAPL Photonics2378-09672025-03-01103036104036104-1710.1063/5.0242728PIC2O-Sim: A physics-inspired causality-aware dynamic convolutional neural operator for ultra-fast photonic device time-domain simulationPingchuan Ma0Haoyu Yang1Zhengqi Gao2Duane S. Boning3Jiaqi Gu4School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona 85281, USANVIDIA Inc., 11001 Lakeline Blvd. Suite #100 Bldg. 2, Austin, Texas 78717, USADepartment of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Boston, Massachusetts 02139, USADepartment of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Boston, Massachusetts 02139, USASchool of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona 85281, USAOptical simulation plays an important role in photonic hardware design flow. The finite-difference time-domain (FDTD) method is widely adopted to solve time-domain Maxwell equations. However, FDTD is known for its prohibitive runtime cost as it iteratively solves Maxwell equations and takes minutes to hours to simulate a single device. Recently, AI has been applied to realize orders-of-magnitude speedup in partial differential equation solving. However, AI-based FDTD solvers for photonic devices have not been clearly formulated. Directly applying off-the-shelf models to predict the optical field dynamics shows unsatisfying fidelity and efficiency since the model primitives are agnostic to the unique physical properties of Maxwell equations and lack algorithmic customization. In this work, we thoroughly investigate the synergy between neural operator designs and the physical property of Maxwell equations and introduce a physics-inspired AI-based FDTD prediction framework PIC2O-Sim. PIC2O-Sim features a causality-aware dynamic convolutional neural operator as its backbone model that honors the space–time causality constraints via careful receptive field configuration and explicitly captures the permittivity-dependent light propagation behavior via an efficient dynamic convolution operator. Meanwhile, we explore the trade-offs among prediction scalability, fidelity, and efficiency via a multi-stage partitioned time-bundling technique in autoregressive prediction. Multiple key techniques have been introduced to mitigate iterative error accumulation while maintaining efficiency advantages during autoregressive field prediction. Extensive evaluations on three challenging photonic device simulation tasks have shown the superiority of our PIC2O-Sim method, showing 51.2% lower roll-out prediction error, 23.5 times fewer parameters than state-of-the-art neural operators, providing 133–310× or 31–89× higher simulation speed than an open-source single-process or eight-process parallel FDTD numerical solver.http://dx.doi.org/10.1063/5.0242728 |
| spellingShingle | Pingchuan Ma Haoyu Yang Zhengqi Gao Duane S. Boning Jiaqi Gu PIC2O-Sim: A physics-inspired causality-aware dynamic convolutional neural operator for ultra-fast photonic device time-domain simulation APL Photonics |
| title | PIC2O-Sim: A physics-inspired causality-aware dynamic convolutional neural operator for ultra-fast photonic device time-domain simulation |
| title_full | PIC2O-Sim: A physics-inspired causality-aware dynamic convolutional neural operator for ultra-fast photonic device time-domain simulation |
| title_fullStr | PIC2O-Sim: A physics-inspired causality-aware dynamic convolutional neural operator for ultra-fast photonic device time-domain simulation |
| title_full_unstemmed | PIC2O-Sim: A physics-inspired causality-aware dynamic convolutional neural operator for ultra-fast photonic device time-domain simulation |
| title_short | PIC2O-Sim: A physics-inspired causality-aware dynamic convolutional neural operator for ultra-fast photonic device time-domain simulation |
| title_sort | pic2o sim a physics inspired causality aware dynamic convolutional neural operator for ultra fast photonic device time domain simulation |
| url | http://dx.doi.org/10.1063/5.0242728 |
| work_keys_str_mv | AT pingchuanma pic2osimaphysicsinspiredcausalityawaredynamicconvolutionalneuraloperatorforultrafastphotonicdevicetimedomainsimulation AT haoyuyang pic2osimaphysicsinspiredcausalityawaredynamicconvolutionalneuraloperatorforultrafastphotonicdevicetimedomainsimulation AT zhengqigao pic2osimaphysicsinspiredcausalityawaredynamicconvolutionalneuraloperatorforultrafastphotonicdevicetimedomainsimulation AT duanesboning pic2osimaphysicsinspiredcausalityawaredynamicconvolutionalneuraloperatorforultrafastphotonicdevicetimedomainsimulation AT jiaqigu pic2osimaphysicsinspiredcausalityawaredynamicconvolutionalneuraloperatorforultrafastphotonicdevicetimedomainsimulation |