Syllabus

Introduction to mathematical modeling and dynamic simulation
Conservation principles, Definition of mathematical model and different types of models and necessity of them, transfer function models, Time-Domain, Laplace Domain, -Z-domain transformation, Dynamic simulation and solvers.
Development of mathematical models using conservation principle and their responses
I. Distillation columns,
II. Isothermal and non- isothermal reactors
III. Heat exchanger
Linearization and state space models
Process Identification
Time series data and their attributes, auto correlation, partial correlation, cross correlation, Stationary and non-stationary data, sampling continuous signal, reconstruction of continuous signals from their discrete time values, conversion of continuous to discrete Time models. AR, ARMA, ARMAX, NRMA, NRMAX, output error models for correlating time series data which will be useful for control and monitoring.
Machine Learning
Introduction to Machine Learning, Chemmometrics, and Artificial Intelligence (AI). supervised learning, unsupervised learning, and reinforcement learning and its application for Chemical Proses Industries in Specific, Data acquisition and preprocessing, attributes of data. Dimensionality Reduction and Regression and classification Algorithms, k- Nearest Neighbors (kNN), K-Means Algorithms.

Exploration of MATLAB BASED TOOLBOX
System identification toolbox, Differential equation solver, Machine Learning toolbox.

1. The students will be able to develop mathematical models of processes related to mass, heat and fluid transfer, separation and reaction using conservation principles. 2. The students will be able to analyze the dynamics of the aforesaid processes performing dynamic simulation using the MATLAB/SIMULINK environment which would help them to monitor an ongoing process and design appropriate controllers for that process to ensure product quality and process safety.

The students will be able to develop input-output models like ARX, AR, ARMA, NRMA etc. using plant/experimental time series data, which might help them in successful monitoring and control of a specific process

The students will be able to utilize various machine learning algorithms including AI in solving regression, classification, parameter estimation problems in developing process monitoring tools and controlling of a process

System identification toolbox, Differential equation solver, and machine-learning toolbox

1. The students will be able to Develop mathematical models for various chemical processes with the help of first principles. They will find the numerical solution of the proposed model using available numerical methods and MATLAB as simulation Platform.
2. The students will be able to Identify the process (transfer function) from plant data (discrete time series signals, their collection, pre-processing).
3. The students will be able to implement Machine learning algorithms and its superset Artificial Intelligence (AI) in various chemical processes (as predictive models/classifier etc.)
4. The students will be able to explore and utilize MATLAB BASED TOOLBOX: System identification toolbox, Differential equation solver, and machine-learning toolbox

W. L. Luyben, Process Modelling, Simulation and Control for Chemical Engineers, McGraw Hill , Second Edition, McGraw-Hill, 1996

B. Roffel, B. Betlem, Process Dynamics & Control: Modeling for control and prediction, John Wiley & Sons Ltd , 2006.

B. W. Bequette, Process Control: Modeling, Design, and Simulation (International Series in the Physical and Chemical Engineering Sciences), Prentice-Hall India , 2nd Edition, 2021

Madhusree Kundu, Palash Kundu, Seshu Kumar Damarla, A Chemometric Approach to Monitoring: Product Quality Assessment, Process Fault Detection and Miscellaneous Applications, CRC Press, Taylor & Francis Group , 2017

Applied Time-Series Analysis - NPTEL Online Course by Prof. Arun Tangirala

Statistics and Machine Learning Toolbox™ of MathWorks