Oil and Gas Petro chemical Pulp and Paper Water and wastewater Metal Industries

Applications
Data pre- ML model AI algorithm Optimizer DSS
Data
processing development development
Modelling &
APC
Simulation

Dr. Senthilmurugan Subbiah, Department of Chemical Engineering, IITG.

Application of Al and ML in chemical engineering

Introduction
Objective of course

• To introduce the Machine learning model development lifecycle

• To introduce Artificial intelligence and ML
• Demonstrate the Application of AI& ML chemical engineering
• Decision support system
• Process control & real-time optimization
• Modelling and simulation
• Industrial example
• Oil & Gas
• Petro Chemical
• Pulp and paper
• Water and wastewater
• Metal Industries
January 15, 2024 | Slide 2
Syllabus
Application of Al and ML in chemical engineering (2-0-2-6)

• Introduction to AI and ML • Introduction to software tools used in AI

• Types of learning problems
& ML
• Python scikit-learn
• Supervised, Unsupervised,
• (https://scikit-learn.org/stable/)
Semi-supervised
• MATLAB Statistics and Machine
• Regression and Classification
Learning Toolbox
• Overview of optimization (https://in.mathworks.com/product
techniques s/statistics.html);
• Application in chemical • Complete machine learning
engineering process demo – Simple
regression/classification example
in Python/MATLAB

January 15, 2024 | Slide 3

Syllabus

• Overview of Machine learning process • Data preprocessing:

lifecycle • Data visualization (line plot, scatter
• Data collection, plot, histogram)
• Preprocessing, • Outlier detection (z-score, Inter-

• Exploratory Data Analysis (EDA) Quartile Range (IQR))

• Model development (training, • Smoothing techniques (moving

validation), average, exponential average,

• Deployment
Savitzky–Golay (SG) filter)
• Data scaling (Standardization,
• Maintenance
Normalization)
• Process lifecycle in MATLAB, Python

January 15, 2024 | Slide 4

Syllabus

• Dimensionality reduction • Model selection

• PCA : Principle Component Analysis
• ICA : Integrated Component Analysis • Hybrid Cross-Validation methods
• t-SNE : t- distributed Stochastic • Leave-one-out cross-
Neighbour Embedding
• UMAP : Uniform Manifold Approximation validation (LpOCV)
Projection • K-fold
• Feature extraction, selection
• Model Evaluation & identification • Time-series cv
• Performance metrics • Nested cv
• Regression - MAE, MSE, RMSE
• Classification - Precision, Recall, F1 • Hyperparameter Tuning
score, Receiver operating characteristic
(RUC) curve,
• Residual analysis

January 15, 2024 | Slide 5

Syllabus

• Model development: • Regression

• Classification • Linear regression
• Logistic regression • simple

• Naïve bayes classifier • multiple

• K-nearest neighbors • Kernel

• Support vector machines • Regression analysis

• Decision trees • Decision tree-based regression

• Random forests models

• Boosting Algorithms

January 15, 2024 | Slide 6

Syllabus

• Time series forecasting –

• Autoregressive (AR) model
• Neural Network:
• Moving average (MA) model • Introduction (Perceptron,
• Autoregressive moving average (ARMA) model
• Autoregressive integrated moving average
multilayer perceptron)
(ARIMA) model
• Backpropagation
• Seasonal autoregressive integrated moving
average (SARIMA) model • Extreme Learning Machines (ELM)
• Vector autoregressive (VAR) model
• Vector error correction (VECM) model • Convolutional Neural Network
• Autoregressive Integrated Moving Average (CNN)
eXogenous Variable Models (ARIMAX)
• Seasonal Auto-Regressive Integrated Moving • Recurrent Neural Network (RNN);
Average with eXogenous factors (SARIMAX)
• Vector Autoregressive Moving Average • Long Short Term Memory
(VARMA)
• Vector Autoregression Moving-Average with (LSTM)
Exogenous Regressors (VARMAX)

January 15, 2024 | Slide 7

Syllabus

• Unsupervised Techniques: • Use cases in Chemical engineering:

• K-means • decision support system,

• Hierarchical and spectral • process control & real time

clustering, optimization
• Density-based spatial clustering of • modeling and simulation
applications with noise (DBSCAN)
• Affinity propagation

• Apriori algorithm

January 15, 2024 | Slide 8

References:

1. Hastie, T., Tibshirani, R., Friedman, J.H., The Elements of Statistical Learning Data
Mining, Inference, and Prediction, Second Edition, 2009
2. Abu-Mostafa, Y.S., Magdon-Ismail, M., Hsuan-Tein, L., Learning from Data.
AMLBook, 2012
3. Bishop, C., Pattern Recognition and Machine Learning. Springer-Verlag, 2006
4. Gareth, J., Witten, D., Hastie. T., Tibshirani, R., An Introduction to Statistical
Learning with Applications in R, Springer-Verlag, 2013
5. Müller, A. C., Gudio, S., Introduction to Machine Learning with Python, O'Reilly
Media, Inc., 2016
6. Shalev-Shwartz, S. and Ben-David, S., Understanding Machine Learning: From
Theory to Algorithms, Cambridge University Press., 2014

January 15, 2024 | Slide 9

Evaluation criteria
Mid Semester

Evaluation Marks
Mid Semester 20
End Semester 30
Assignment 10
Quiz 10
Project 30
Attendance As per institute norms

January 15, 2024 | Slide 10

TA Details

S.No TA Name Email ID Phone

1 B. Sai Mukesh [email protected] 7075570330
Reddy
2 Bijoyendra Sharma [email protected] 7577814920

January 15, 2024 | Slide 11

Class Schedule
Monday (4:00 pm – 5:00 pm) Tuesday (2:00 pm – 4:00 pm) Wednesday (4:00 pm – 5:00 pm)

1 2 3
January

8 9 10

15 Holiday 16 17

22 23 24

February 5 6 7

12 13 14

19 20 21

26 Mid term 27 Mid term 28 Mid term

January 15, 2024 | Slide 12

 Monday (4:00 pm – 5:00 pm) Tuesday (2:00 pm – 4:00 pm) Wednesday (4:00 pm – 5:00 pm)

March 4 5 6

11 12 13

18 19 20

25 Holiday 26 27

April 1 2 3

8 9 10

15 Holiday 16 17

22 23 24

29 End sem 30 End sem

January 15, 2024 | Slide 13
 Monday (4:00 pm – 5:00 pm) Tuesday (2:00 pm – 4:00 pm) Wednesday (4:00 pm – 5:00 pm)

May 1 End sem

6 7 8

13 14 15

20 21 22

January 15, 2024 | Slide 14