Learning-Based Soft Sensing and Predictions for Process Industries: Theory, Methodology and Applications covers prediction and soft sensing in industrial processes that are subject to specific challenges with AI-empowered learning algorithms. With the aid of a data-driven modeling strategy, the book explores the problems of industrial prediction and soft sensing and formulates a series of learning-based theory, methodologies, and applications. The book introduces the basics of prediction and soft sensing backgrounds, including different categories of prediction theory. Secondly, covers the foundations of machine learning methodologies, including supervised learning prediction, semi-supervised, and self-supervised prediction. Finally, the book examines novel learning-based models/architectures.
1. Chapter 1: Introduction to soft sensing - p. 3 2. Chapter 2: Introduction to data-driven prediction - p. 11 3. Chapter 3: Aluminum electrolysis industry - p. 21 4. Chapter 4: Paste-filling industry - p. 31 5. Part 2: Methodology - p. P2 6. Chapter 5: Machine learning basics for prediction and soft sensing - p. 41 7. Part 3: Applications - p. P3 8. Chapter 6: A novel semi-supervised soft sensor framework based on convolutional Laplacian ELM: CNN-LapsELM - p. 75 9. Chapter 7: A novel semi-supervised soft sensor framework based on stacked auto-encoder wavelet extreme learning machine: SAE-WELM - p. 87 10. Chapter 8: A novel soft senor based on Laplacian–Hessian semi-supervised hierarchical extreme learning machine: LHSS-HELM - p. 101 11. Chapter 9: A self-supervised prediction framework based on deep long short-term memory for aluminum electrolysis: SSDLSTM - p. 115 12. Chapter 10: A self-supervised prediction framework based on convolutional deep long short-term memory for aluminum temperature application: CNN-SSDLSTM - p. 127 13. Chapter 11: A novel probabilistic prediction framework based on Bayesian machine learning: BLSTM - p. 143 14. Chapter 12: Direct data-driven quantile regressor forecastor for underflow concentration soft sensing: DDQRF - p. 163 15. Chapter 13: A novel key-quality prediction framework for industrial deep cone thickener: DualLSTM - p. 177 16. Chapter 14: A deeply efficient long short-term memory framework for underflow concentration prediction: DE-LSTM - p. 191 17. Subject index - p. 203
Learning-Based Soft Sensing and Predictions for Process Industries: Theory, Methodology and Applications covers prediction and soft sensing in industrial processes that are subject to specific challenges with AI-empowered learning algorithms. With the aid of a data-driven modeling strategy, the book explores the problems of industrial prediction and soft sensing and formulates a series of learning-based theory, methodologies, and applications. The book introduces the basics of prediction and soft sensing backgrounds, including different categories of prediction theory. Secondly, covers the foundations of machine learning methodologies, including supervised learning prediction, semi-supervised, and self-supervised prediction. Finally, the book examines novel learning-based models/architectures.
1. Chapter 1: Introduction to soft sensing - p. 3 2. Chapter 2: Introduction to data-driven prediction - p. 11 3. Chapter 3: Aluminum electrolysis industry - p. 21 4. Chapter 4: Paste-filling industry - p. 31 5. Part 2: Methodology - p. P2 6. Chapter 5: Machine learning basics for prediction and soft sensing - p. 41 7. Part 3: Applications - p. P3 8. Chapter 6: A novel semi-supervised soft sensor framework based on convolutional Laplacian ELM: CNN-LapsELM - p. 75 9. Chapter 7: A novel semi-supervised soft sensor framework based on stacked auto-encoder wavelet extreme learning machine: SAE-WELM - p. 87 10. Chapter 8: A novel soft senor based on Laplacian–Hessian semi-supervised hierarchical extreme learning machine: LHSS-HELM - p. 101 11. Chapter 9: A self-supervised prediction framework based on deep long short-term memory for aluminum electrolysis: SSDLSTM - p. 115 12. Chapter 10: A self-supervised prediction framework based on convolutional deep long short-term memory for aluminum temperature application: CNN-SSDLSTM - p. 127 13. Chapter 11: A novel probabilistic prediction framework based on Bayesian machine learning: BLSTM - p. 143 14. Chapter 12: Direct data-driven quantile regressor forecastor for underflow concentration soft sensing: DDQRF - p. 163 15. Chapter 13: A novel key-quality prediction framework for industrial deep cone thickener: DualLSTM - p. 177 16. Chapter 14: A deeply efficient long short-term memory framework for underflow concentration prediction: DE-LSTM - p. 191 17. Subject index - p. 203
