High-Performance Computing and Artificial Intelligence in Process Engineering.

Format:

Book

Author/Creator:

Li, Mingheng.

Contributor:

Heng, Yi.

Series:

IOP Ebooks Series

Language:

English

Physical Description:

1 online resource (335 pages)

Edition:

1st ed.

Place of Publication:

Bristol : Institute of Physics Publishing, 2025.

Summary:

High-performance computing (HPC) and artificial intelligence (AI) boost production efficiency, lower costs, and enhance system stability in process engineering, fostering industrial intelligence and smart manufacturing.

Contents:

Intro

Acknowledgements

Editor biographies

Mingheng Li

Yi Heng

List of contributors

Chapter Artificial intelligence and the future of process engineering

1.1 Introduction

1.2 Types of neural networks

1.3 Applications in chemical and process engineering

1.3.1 Transport phenomena

1.3.2 Kinetics and reactor design

1.3.3 Thermodynamics

1.3.4 Process control

1.3.5 Sample study for applying neural networks in transport phenomena

1.4 Conclusions

Bibliography

Chapter Machine learning in optimal control and process modeling

2.1 Introduction to machine learning

2.1.1 Supervised learning

2.1.2 Unsupervised learning

2.1.3 Reinforcement learning

2.2 Reinforcement learning for optimal control

2.2.1 Review of reinforcement learning

2.2.2 Class of nonlinear systems and problem formulation

2.2.3 Stabilization and safety via CLBF

2.2.4 Control Lyapunov-barrier function-based RL

2.2.5 Application to a chemical process example

2.3 Supervised learning in modeling nonlinear systems

2.3.1 Data-driven modeling using feedforward neural network

2.3.2 Application of NN model-based RL

2.3.3 Application of NN model-based MPC

2.4 Conclusion

Chapter Graph-based control invariant set approximation and its applications

3.1 Introduction

3.2 Preliminaries

3.2.1 Notation

3.2.2 System description

3.3 CIS approximation

3.3.1 Set invariance conditions for autonomous systems

3.3.2 Robust control invariance conditions and approximation

3.3.3 Application to a chemical process

3.4 CIS in economic MPC through zone tracking

3.4.1 MPC with zone tracking

3.4.2 Economic MPC through zone tracking

3.4.3 Application to the CSTR

3.5 RCIS in RL

3.5.1 Safe RL

3.5.2 Application to the CSTR

3.6 Conclusion

Bibliography.

Chapter Machine learning-based multiscale modeling and control of quantum dot manufacturing and their applications

4.1 Introduction, motivation, and literature review

4.1.1 Motivation

4.1.2 Literature review

4.1.3 Objectives and organization of this chapter

4.2 Multiscale modeling and control of tubular crystallizer for continuous QD manufacturing

4.2.1 Mathematical modeling of PFC

4.2.2 Experimental validation results

4.2.3 Optimal operation of PFC

4.3 Multiscale modeling of slug flow crystallizers for QD production

4.3.1 CFD-based multiscale modeling of SFC

4.3.2 Results and discussion

4.3.3 Multivariable optimal operation problem

4.4 Future directions

4.4.1 Transformer-enhanced hybrid modeling of QD systems

4.5 Conclusions

Chapter The rise of time-travelers: are transformer-based models the key to unlocking a new paradigm in surrogate modeling for dynamic systems?

5.1 Introduction

5.2 Time-series transformers

5.2.1 Operation of encoder-decoder transformers

5.2.2 TST architecture

5.3 Utilizing time-series transformers

5.3.1 CrystalGPT

5.3.2 TST-based hybrid modeling approaches

5.4 Insights and applications of transformer models in chemical systems

5.4.1 Advancements in multiscale modeling through TST models

5.4.2 Replacing existing data-driven system identification approaches

5.4.3 Harnessing transfer learning in chemical engineering: a new era

5.4.4 Integrating multiple data sources with transformer models

5.5 Conclusions

Chapter Optimization-based algorithms for solving inverse problems of parabolic PDEs

6.1 Introduction

6.1.1 Definition of the forward problem

6.1.2 Definition of the inverse problem

6.2 Fast and robust 3D IHTP solution strategies

6.2.1 Optimization-based conventional Tikhonov method for IHTP.

6.2.2 Bayesian optimization-based method

6.3 Applications and analysis

6.3.1 Chip heat dissipation

6.3.2 Pool boiling

6.4 Conclusions

References

Chapter Deep learning-based approach for solving forward and inverse partial differential equation problems

7.1 Introduction

7.1.1 Forward problems

7.1.2 Inverse problems

7.2 Deep-learning-based methods

7.2.1 Deep-learning-based methods for forward problems

7.2.2 Deep learning-based methods for inverse problems

7.3 Applications and analysis

7.3.1 Predicting the transport process in reverse osmosis desalination

7.3.2 Identification of highly transient surface heat flux

7.4 Conclusions

Chapter An active subspace based swarm intelligence method with its application in optimal design problem

8.1 Introduction

8.2 Modeling and methods

8.2.1 Active subspace method

8.2.2 Particle swarm optimization algorithm

8.2.3 Active subspace particle swarm optimization algorithm

8.3 Applications and analysis

8.3.1 Benchmark problem test for ASPSO

8.3.2 PDE constraints for multi-scale optimal design for RO seawater desalination

8.3.3 Simulation experiments for multi-scale optimal design for RO seawater desalination

8.4 Conclusions

Chapter Supercomputing and machine-learning-aided optimal design of high permeability seawater reverse osmosis membrane systems

9.1 Introduction

9.2 Potential evaluation of module and system design

9.3 Multiscale optimization design framework

9.3.1 Small-scale multi-physics modeling

9.3.2 Model identification with MLN

9.3.3 System-level modeling at the meter scale

9.3.4 Optimal design of the RO system

9.4 Results and discussion

9.4.1 Supercomputing-based machine-learning-driven model identification

9.4.2 Surrogate model evaluation.

9.4.3 Optimal design of high permeability SWRO systems

9.5 Conclusions

Chapter Supercomputing-based inverse identification of high-resolution atmospheric pollutant source intensity distributions

10.1 Introduction

10.2 Lagrangian models

10.3 Methods and theories

10.3.1 Forward simulation framework

10.3.2 High-throughput parallel inverse computing strategy

10.4 Applications and analysis

10.4.1 Data product

10.4.2 Case study: application to SO2 transport from volcanic eruptions

10.4.3 Case study: application to greenhouse gas CO2 transport from forest fires

10.5 Conclusions

Chapter Enhancing boiling heat transfer via model-based experimental analysis

11.1 Introduction

11.1.1 Pool boiling applications

11.1.2 Extensive investigations of pool boiling

11.1.3 Motivation

11.2 Modeling and methods

11.2.1 Fabrication of honeycomb porous structured surfaces

11.2.2 Reconstruction of geometric models for honeycomb surfaces

11.2.3 Numerical simulation

11.3 Applications and analysis

11.3.1 Experimental analysis of boiling heat transfer

11.3.2 Numerical analysis of boiling heat transfer

11.4 Conclusions

References.

Notes:

Description based on publisher supplied metadata and other sources.

ISBN:

9780750361743

0750361743

OCLC:

1513860265