Physics-Informed Neural Network for Unreacted-Core Shrinking Model of Coal Gasification
Physics-informed neural network (PINN) has achieved success in many science and engineering disciplines by encoding physical laws into the loss functions of the neural network. Compared with traditional numerical method, PINN transformed the problem of solving differential equations into the optimiz...
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
AIDIC Servizi S.r.l.
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/2dd7470c37d64f4389bfa627e61f6f2b |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | Physics-informed neural network (PINN) has achieved success in many science and engineering disciplines by encoding physical laws into the loss functions of the neural network. Compared with traditional numerical method, PINN transformed the problem of solving differential equations into the optimization of loss functions by automatic differentiation. In this work, PINN is apply to solve the coal gasification chemical kinetic problems of gas-solid reactions in the form of the unreacted-core shrinking model with governing ordinary differential equations (ODEs). The results show that the prediction performance of the developed PINN is comparable with that of the widely-used Runge-Kutta method, and thus it opens the possibility for the application of deep learning to the modelling of complex chemical kinetics systems. |
|---|