Developing dynamic machine learning surrogate models of physics-based industrial process simulation models

Computer Science

• Simulation Mode 100%
• Case Study 33%
• Machine Learning 33%
• Root Mean Squared Error 19%
• Artificial Neural Network 9%
• Testing 9%
• Workflows 9%
• Nonlinearities 4%
• Optimization 4%
• Model Generation 4%
• In-Process 4%
• Calculation Process 4%
• Modeling and Simulation 4%
• Expected Result 4%
• Dynamic Behavior 4%
• Accuracy Requirement 4%
• Neural Network Model 4%
• Accuracy 4%
• Simulation 4%
• Application 4%
• Motivation 4%
• Open Source 4%
• Prevents 4%
• Normalized Model 4%
• Validation Dataset 4%
• Generalization Ability 4%

Engineering

• Models 95%
• Root-Mean-Squared Error 19%
• Libraries 14%
• Dataset 9%
• Literature Study 9%
• Recurrent 9%
• Good Level 9%
• Artificial Neural Network 9%
• Surrogate Model 9%
• Testing 9%
• Research 9%
• Error Value 9%
• Neural Network Architecture 9%
• Optimization 4%
• Simulation Software 4%
• Technical Research Centre 4%
• Dynamic Behavior 4%
• Unit Operations 4%
• Nonlinearity 4%
• Long Short-Term Memory 4%
• Applications 4%
• Simulation 4%
• Water 4%
• Accuracy 4%
• Testing Dataset 4%