Download for free
PDF version of the exercise book (on Research Gate)
Jupyter Notebook files (Python) with all the exercises on GitHub
This guide provides a solid introduction to the application of machine learning algorithms in various industrial and engineering contexts. It is designed to equip students with both the theoretical foundations and the practical skills required to implement control solutions, deploy forecasting techniques, perform clustering, classification, optimization, and decision-making.
The book begins with an overview of the fundamental concepts of control theory and machine learning, and their role in industrial applications. It explores key topics such as open-loop and closed-loop control systems, data acquisition, and data preprocessing.
It then continues with a curated set of exercises aimed at supporting students in applying machine learning techniques across different domains. The guide includes exercises on modeling various dynamic systems either from first principles or using historical data, as well as the application of algorithms for:
Forecasting
Techniques for building/training models for time series forecasting and interdependent variables, including:
- Mean value
- Weighted mean
- Moving average
- Single exponential smoothing
- Double exponential smoothing (Holt’s method)
- Triple exponential smoothing (Holt–Winters method)
- Regression method (correlation analysis)
- Recurrent Neural Networks (RNN / LSTM)
Clustering
Clustering techniques enable the identification of natural groupings and structures within datasets.
- K-means
Classification
The book examines algorithms for classifying states and objects using artificial intelligence methods (through artificial neural networks), including:
- Support Vector Machine (SVM)
- Multilayer Perceptron (MLP)
- Convolutional Neural Network (CNN)
Optimisation
The book also provides an example exercise for solving a linear optimization problem, including:
- The Traveling Salesman Problem
- The General Routing Problem
Decision-Making
An essential component of any intelligent control system is the decision-making module, especially when decisions must be made under uncertainty. The guide presents an example from the field of:
- Fuzzy Logic