DeepM&Mnet for hypersonics: Predicting the coupled flow and finite-rate chemistry behind a normal shock using neural-network approximation of operators.
Zhiping MaoLu LuOlaf MarxenTamer A. ZakiGeorge Em KarniadakisPublished in: J. Comput. Phys. (2021)
Keyphrases
- neural network
- discrete random variables
- artificial neural networks
- neural network model
- pattern recognition
- back propagation
- fuzzy logic
- activation function
- continuous functions
- approximation error
- flow field
- approximation algorithms
- morphological operators
- genetic algorithm
- feedforward neural networks
- backpropagation neural network
- neural network is trained
- bp neural network
- associative memory
- self organizing maps
- error bounds
- feed forward
- np hard
- image reconstruction from projections
- totally ordered
- knn
- laplacian operator
- flow patterns
- neural nets
- relative error
- recurrent neural networks
- fuzzy neural network
- finite number
- hidden layer
- multi layer