Deep Learning Emulator Towards Both Forward and Adjoint Modes of Atmospheric Gas-Phase Chemical Process
Bewaard in:
| Gepubliceerd in: | Atmosphere vol. 16, no. 9 (2025), p. 1109-1126 |
|---|---|
| Hoofdauteur: | |
| Andere auteurs: | , , |
| Gepubliceerd in: |
MDPI AG
|
| Onderwerpen: | |
| Online toegang: | Citation/Abstract Full Text + Graphics Full Text - PDF |
| Tags: |
Geen labels, Wees de eerste die dit record labelt!
|
MARC
| LEADER | 00000nab a2200000uu 4500 | ||
|---|---|---|---|
| 001 | 3254466741 | ||
| 003 | UK-CbPIL | ||
| 022 | |a 2073-4433 | ||
| 024 | 7 | |a 10.3390/atmos16091109 |2 doi | |
| 035 | |a 3254466741 | ||
| 045 | 2 | |b d20250101 |b d20251231 | |
| 084 | |a 231428 |2 nlm | ||
| 100 | 1 | |a Liu, Yulong | |
| 245 | 1 | |a Deep Learning Emulator Towards Both Forward and Adjoint Modes of Atmospheric Gas-Phase Chemical Process | |
| 260 | |b MDPI AG |c 2025 | ||
| 513 | |a Journal Article | ||
| 520 | 3 | |a Gas-phase chemistry has been identified as a major computational bottleneck in both the forward and adjoint modes of chemical transport models (CTMs). Although previous studies have demonstrated the potential of deep learning models to simulate and accelerate this process, few studies have examined the applicability and performance of these models in adjoint sensitivity analysis. In this study, a deep learning emulator for gas-phase chemistry is developed and trained on a diverse set of forward-mode simulations from the Community Multiscale Air Quality (CMAQ) model. The emulator employs a residual neural network (ResNet) architecture referred to as FiLM-ResNet, which integrates Feature-wise Linear Modulation (FiLM) layers to explicitly account for photochemical and non-photochemical conditions. Validation within a single timestep indicates that the emulator accurately predicts concentration changes for 74% of gas-phase species with coefficient of determination (R2) exceeding 0.999. After embedding the emulator into the CTM, multi-timestep simulation over one week shows close agreement with the numerical model. For the adjoint mode, we compute the sensitivities of ozone (O3) with respect to O3, nitric oxide (NO), nitrogen dioxide (NO2), hydroxyl radical (OH) and isoprene (ISOP) using automatic differentiation, with the emulator-based adjoint results achieving a maximum R2 of 0.995 in single timestep evaluations compared to the numerical adjoint sensitivities. A 24 h adjoint simulation reveals that the emulator maintains spatially consistent adjoint sensitivity distributions compared to the numerical model across most grid cells. In terms of computational efficiency, the emulator achieves speed-ups of 80×–130× in the forward mode and 45×–102× in the adjoint mode, depending on whether inference is executed on Central Processing Unit (CPU) or Graphics Processing Unit (GPU). These findings demonstrate that, once the emulator is accurately trained to reproduce forward-mode gas-phase chemistry, it can be effectively applied in adjoint sensitivity analysis. This approach offers a promising alternative approach to numerical adjoint frameworks in CTMs. | |
| 651 | 4 | |a China | |
| 653 | |a Air quality | ||
| 653 | |a Nitric oxide | ||
| 653 | |a Datasets | ||
| 653 | |a Sensitivity analysis | ||
| 653 | |a Nitrogen dioxide | ||
| 653 | |a Artificial neural networks | ||
| 653 | |a Neural networks | ||
| 653 | |a Chemistry | ||
| 653 | |a Computer applications | ||
| 653 | |a Photochemicals | ||
| 653 | |a Distance learning | ||
| 653 | |a Simulation | ||
| 653 | |a Ozone | ||
| 653 | |a Emulators | ||
| 653 | |a Numerical models | ||
| 653 | |a Isoprene | ||
| 653 | |a Graphics processing units | ||
| 653 | |a Chemical transport | ||
| 653 | |a Chemical bonds | ||
| 653 | |a Hydroxyl radicals | ||
| 653 | |a Central processing units--CPUs | ||
| 653 | |a Accuracy | ||
| 653 | |a Deep learning | ||
| 653 | |a Atmospheric chemistry | ||
| 653 | |a Data assimilation | ||
| 653 | |a Photochemistry | ||
| 653 | |a Machine learning | ||
| 653 | |a Graphics | ||
| 653 | |a Gases | ||
| 653 | |a Mathematical models | ||
| 653 | |a Air quality models | ||
| 653 | |a Ordinary differential equations | ||
| 653 | |a Embedding | ||
| 700 | 1 | |a Liao Meicheng | |
| 700 | 1 | |a Liu, Jiacheng | |
| 700 | 1 | |a Cheng, Zhen | |
| 773 | 0 | |t Atmosphere |g vol. 16, no. 9 (2025), p. 1109-1126 | |
| 786 | 0 | |d ProQuest |t Publicly Available Content Database | |
| 856 | 4 | 1 | |3 Citation/Abstract |u https://www.proquest.com/docview/3254466741/abstract/embedded/L8HZQI7Z43R0LA5T?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text + Graphics |u https://www.proquest.com/docview/3254466741/fulltextwithgraphics/embedded/L8HZQI7Z43R0LA5T?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text - PDF |u https://www.proquest.com/docview/3254466741/fulltextPDF/embedded/L8HZQI7Z43R0LA5T?source=fedsrch |