About the Book
This book provides an application-focused exposition of modern ML tools that have proven useful in process industry and hands-on illustrations on how to develop ML-based solutions for process monitoring, predictive maintenance, fault diagnosis, inferential modeling, dimensionality reduction, and process control. This book considers unique characteristics of industrial process data and uses real data from industrial systems for illustrations. With the focus on practical implementation and minimal programming or ML prerequisites, the book covers the gap in available ML resources for industrial practitioners. The authors of this book have drawn from their years of experience in developing data-driven industrial solutions to provide a guided tour along the wide range of available ML methods and declutter the world of machine learning. The readers will find all the resources they need to deal with high-dimensional, correlated, noisy, corrupted, multimode, and nonlinear process data.
The book has been divided into four parts. Part 1 provides a perspective on the importance of ML in process systems engineering and lays down the basic foundations of ML. Part 2 provides in-detail presentation of classical ML techniques and has been written keeping in mind the various characteristics of industrial process systems. Part 3 is focused on artificial neural networks and deep learning. Part 4 covers the important topic of deploying ML solutions over web and shows how to build a production-ready process monitoring web application.
Broadly, the book covers the following:
- Varied applications of ML in process industry
- Fundamentals of machine learning workflow
- Practical methodologies for pre-processing industrial data
- Classical ML methods and their application for process monitoring, fault diagnosis, and soft sensing
- Deep learning and its application for predictive maintenance
- Reinforcement learning and its application for process control
- Deployment of ML solution over web
If you do not have a PayPal account, you can purchase the book at Google Play.
About the Authors
Ankur Kumar holds a PhD degree (2016) in Process Systems Engineering from the University of Texas at Austin and a bachelor’s degree (2012) in Chemical Engineering from the Indian Institute of Technology Bombay. He currently works at Linde in the Advanced Digital Technologies & Systems Group in Linde’s Center of Excellence, where he has developed several in-house machine learning-based monitoring and process control solutions for Linde’s hydrogen and air-separation plants. Ankur’s tools have won several awards both within and outside Linde. Most recently, one of his tools, PlantWatch (a plantwide fault detection and diagnosis tool), received the 2021 Industry 4.0 Award by the Confederation of Industry of the Czech Republic. Ankur has authored or co-authored more than 10 peer-reviewed journal papers (in the areas of data-driven process modeling and optimization), is a frequent reviewer for many top-ranked Journals, and has served as Session Chair at several international conferences. Ankur also served as an Associate Editor of the Journal of Process Control from 2019 to 2021.
Jesus Flores-Cerrillo is currently an Associate Director - R&D at Linde and manages the Advanced Digital Technologies & Systems Group in Linde’s Center of Excellence. He has over 20 years of experience in the development and implementation of monitoring technologies and advanced process control & optimization solutions. Jesus holds a PhD degree in Chemical Engineering from McMaster University and has authored or co-authored more than 40 peer-reviewed journal papers in the areas of multivariate statistics and advanced process control among others. His team develops and implements novel plant monitoring, machine learning, IIOT solutions to improve the efficiency and reliability of Linde’s processes. Jesus’s team received the Artificial Intelligence and Advanced Analytics Leadership 2020 award from the National Association of Manufacturers’ Manufacturing Leadership Council.