Course detail

Optimization Methods and Artificial Intelligence in Computational Simulations

FSI-LOM Acad. year: 2025/2026 Winter semester

Optimization is a key element in the process of finding the best solution to a given engineering problem. Students will be introduced to the theoretical background of optimization and artificial intelligence techniques and then learn how to apply these principles to specific computational simulations. The course includes classical mathematical programming methods, gradient methods, heuristic approaches, the study of evolutionary algorithms, and artificial intelligence methods with a focus on artificial neural networks, both classical layered topologies and convolutional neural networks and feedback models.

Language of instruction

Czech

Number of ECTS credits

5

Supervisor

doc. Ing. Tomáš Mauder, Ph.D.

Department

Energy Institute

Entry knowledge

Advanced knowledge of computational modeling and simulation techniques. Basic knowledge of mathematical analysis, linear algebra, probability, statistics, and numerical methods within the scope of engineering degree requirements.

Rules for evaluation and completion of the course

Students will work on the assigned semester project. The presentation of the project will be followed by a professional discussion and classification.

Aims

Deepening theoretical and practical knowledge and skills in the field of optimization and AI methods and their implementation for computer simulation methods.

Deepening and expanding knowledge of programming in Python.

The study programmes with the given course

Programme N-SUE-P: Computational Simulations for Sustainable Energy, Master's, compulsory

Type of course unit

 

Lecture

26 hours, optionally

Syllabus


  1. Introduction to optimization, meaning of optimization, basic concepts

  2. Linear, non-linear, stochastic, integer programming

  3. Heuristic methods, gradient, evolutionary, swarm, human based methods

  4. Artificial intelligence – introduction, state space search

  5. Machine learning methods, basic tasks (regression problem, classification problem)

  6. Artificial neural networks – neuron, topology, activation functions

  7. Artificial neural networks – learning algorithm

  8. Convolutional neural networks

  9. Backpropagation neural networks

Computer-assisted exercise

26 hours, compulsory

Syllabus


  1. Linear programming in Python (Matlab)

  2. Heuristic methods in Python (Matlab)

  3. Artificial Intelligence in Python (Matlab)

  4. Combining optimization algorithms in Python with computational simulations

  5. Machine learning methods using Matlab (deep learning toolbox)

  6. Machine learning methods using the Tensorflow framework (Python)