1. komponenta

Lecture type	Total
Lectures	15
Laboratory exercises	30

* Load is given in academic hour (1 academic hour = 45 minutes)

Course objective is to teach students the performance and operation of systems that apply artificial intelligence in chemical engineering and process industry.

Course content:
Overview of artificial intelligence methods. Historical development. Examples of artificial intelligence method applications. Intelligent systems in the industry.
Artificial neural networks. Biological and artificial neuron. Neural network topology. Multilayer perceptron neural networks. Types of neural networks. Learning algorithms. Back propagation algorithm.
Evaluation of the model. Sensitivity analysis. Examples of the application of neural networks in chemical engineering.
Development of a neural network model from experimental data.
Fuzzy logic. Fuzzy sets and membership functions. Fuzzy rules and relationships. Determination of the structure and parameters of a fuzzy model. Process design and control using fuzzy logic. Hybrid systems.
Introduction to machine learning. Machine learning methods. Supervised learning. Regression and classification. Developing a machine learning model. Unsupervised learning. Techniques of unsupervised learning.
Preparation and data pre-processing. Methods for data dimension reducing.
Algorithms for clustering data. Examples.
Support vector machine method. Linear and nonlinear support vector methods. Optimizing the use of support vector methods. Kernel function. Examples.
Methods based on decision trees. Basic learning algorithm for decision tree. Example.
Optimization algorithms inspired by nature. Evolutionary algorithms. Genetic algorithms. Basic concepts (individual, population, generation). Selection, crossing and mutation operators. Fitness function and scaling of fitness function. Basic steps in genetic algorithm optimization. An example of optimization (parameter estimation) using genetic algorithms in chemical engineering. Method of pattern search. Method of simulated annealing.

1. Identify the engineering problems suitable for solving with artificial intelligence methods;
2. Select, adapt and apply the artificial intelligence method to solve a typical engineering problem;
3. Develop and apply neural network models in chemical engineering;
4. Apply methods and techniques for data processing and visualization;
5. Interpret the results and statistical criteria for model evaluation;
6. Apply machine learning methods for prediction, classification and diagnostics;
7. Apply genetic algorithms for process and model optimization.

1. Machine Learning, Mitchell, T. M., McGrowHill, 1997.
2. The elements of statistical learning: Data mining, inference, and prediction, Hastie, T., Tibshirani, R., Friedman, J.H., Springer, New York, 2009.

Izborna grupa - Regular modul - Chemical Engineering in Environmental Protection
Izborna grupa - Regular modul - Chemical Process Engineering
Izborna grupa - Regular modul - Chemical Technologies and Products