Sequential Deep Operator Networks (S-DeepONet) for predicting full-field solutions under time-dependent loads

Keyphrases

• Branched Network 20%
• Classical Formulation 20%
• Classical Neural Network 20%
• Computationally Expensive 20%
• Contour Plots 20%
• Deep Learning Operators 20%
• Deep Operator Network 100%
• DeepONet 100%
• Direct Finite Elements 20%
• Feedforward Neural Network 20%
• Finite Element Simulation 20%
• Forward Neural Network 20%
• Gated Recurrent Unit 40%
• Heat Transfer Problem 20%
• Infinite Dimension 20%
• Learning Model 20%
• Linear Solution 20%
• Loading History 20%
• Long Short-term Memory 40%
• Network Model 20%
• New Architecture 20%
• Nonlinear Solution 20%
• Operational Network 20%
• Order of Magnitude 20%
• Parametric Functions 20%
• Path Dependence 20%
• Prediction Accuracy 20%
• Prediction Error 20%
• R2 Value 20%
• Sequential Learning 20%
• Solution Operator 20%
• Superior Accuracy 20%
• Surrogate Model 20%
• Time-dependent Loading 20%
• Time-varying Load 100%
• Transient Heat Transfer 20%

Engineering

• Accurate Prediction 25%
• Contour Plot 25%
• Deep Learning 25%
• Deep Neural Network 25%
• Feedforward 25%
• Finite Element Simulation 25%
• Loading History 25%
• Long Short-Term Memory 50%
• Network Model 25%
• Network Operator 100%
• Prediction Error 25%
• Recurrent 50%
• Surrogate Model 25%
• Transients 25%

Chemical Engineering

• Deep Learning 50%
• Deep Neural Network 50%
• Feedforward Neural Network 50%
• Long Short-Term Memory 100%
• Neural Network 50%