Courses – Fakultet tehničkih nauka u Novom Sadu

Subject: Methods of Optimization (17.E237)

Native organizations units: Sub-department for Automatic Control and Systems Engineering

Type of studies	Title
Undergraduate Academic Studies	Measurement-Information Technologies and Control Engineering (Godina: 4, Winter)

General information:

Category	Scientific-professional
Scientific or art field	Automatic Control and System Engineering
Interdisciplinary	No
ECTS	8

Educational goal:

Students learn about theoretical and practical bases of non-linear optimization of static and dynamic systems.

Educational outcome:

Students will lear to recognize, formulate and solve optimization problems, i.e. tasks involving identification of the best admissible solution of a given problem. Optimization problems are ubiquotious in engineering. Having in mind the existence of broad plaethora of optimization methods, the students will learn how to recognize the one best suited for the problem at hand, as well as how to implement the chosen method in the concrete situation.

Course content:

Formulation of optimization problem. Theoretical bases of static optimization. Analytical system determination, functions of one or more variables without constraints. Analytical determination of extremes, functions of one or more variables with constraints on the type of equality and inequality. Linear programming. Numerical solutions of one-dimensional problems. Numerical solutions of multi-dimensional problems with and without constraints. Fundamentals of variational calculus. Direct methods of variational calculus. Optimal control. Pontryagin`s maximum principle. Dynamic programming, linear regulators. Numerical methods of dynamic optimization. Modern optimization procedures: genetic algorithm, simulated annealing, PSO. Application of optimization procedures in training artificial neural networks and fuzzy logic systems. Examples of optimization of practical engineering problems.

Teaching methods:

Lectures, Numerical and calculation practice. Computer practice. Laboratory practice. Consultations. The examination is written and oral. The written part consists of at least four parts, in order to achieve a passing grade min 50 % each task must be completed successfully. The course material can be divided into two colloquia. The oral part of the examination is based on a list of examination questions. The colloquia, tests and examination are written. The written part is eliminating. The final grade is formed on the basis of colloquia, homework assignments, written and oral part of the examination.

Literature:

Authors	Title	Year	Publisher	Language
Vujanović, B., Spasić, D.	Metodi optimizacije	1998	Univerzitet u Novom Sadu, Novi Sad	Serbian language
Petrić, J., Zlobec, S.	Nelinearno programiranje	1983	Naučna knjiga, Beograd	Serbian language
Dimitri P. Bertsekas	Nonlinear Programming	2004	Athena Scientific	English

Knowledge evaluation:

Course activity	Pre-examination	Obligations	Number of points
Project	Yes	Yes	30.00
Practical part of the exam - tasks	No	Yes	40.00
Oral part of the exam	No	Yes	30.00
Coloquium exam	No	No	40.00

Lecturers:

Saradnik u nastavi Topalov Stefan

Teaching Associate

Computational classes

Saradnik u nastavi Golić Anastasija

Teaching Associate

Computational classes

prof. dr Jeličić Zoran

Full Professor

Lectures

Asistent Gavrilović Milan

Assistant - Master

Computational classes

prof. dr Rapaić Milan

Full Professor

Lectures

Asistent Dragutinović Đorđe

Assistant - Master

Practical classes

Saradnik u nastavi Zarić Dragan

Teaching Associate

Computational classes

Asistent Kičić Aleksandar

Assistant - Master

Practical classes