Optimization and Control

The course subject is optimization. The candidates learn to formulate optimization problems and solve these through appropriate algorithms and software. Optimality conditions like the Karush-Kuhn-Tucker (KKT) conditions are discussed and conditions for global and local conditions are analyzed. Key optimization classes of problems including linear programming (LP), quadratic programming (QP) and nonlinear programming (NLP) are studied and applied in different settings. The course includes advanced control of dynamic systems with emphasis on Model Predictive Control (MPC).

Knowledge: - Ability to formulate appropriate engineering problems as a mathematical optimization problem. - Knowledge of typical engineering problems which are suitable for optimization. - Ability to analyze and solve an optimization problem; in particular linear programs (LP), quadratic programs (QP) and nonlinear programs (NP). - Ability to analyze and design optimal controllers; in particular Model Predictive Controllers (MPC). - Knowledge of optimization software. Skills: - Solve suitable optimization problems using Matlab. - Use optimization in controllers; in particular Model Predictive Control (MPC). - Complete a small optimization project. - Analyze a problem and contribute to innovative design solutions. General competence: - Communicate technical issues with specialists in cross-disciplinary teams and the general public. - Conscious attitude towards the use of optimization within engineering.

The course is given as a mixture of lectures, assignments, programming assessments and a laboratory project. Seven of the assignments and the laboratory report must be approved to enter the final exam.