The thesis introduces possible solutions for reducing harmful emissions produced in the industry and transport sectors. The thesis describes the system for reducing nitrogen oxides (NOx) emissions in exhaust air, which comes from the dryers and ovens in a factory of automotive catalytic converters.
The exhaust air from the dryers and ovens contains NOx emissions, which must be reduced before the air can be released in to the atmosphere. The system for NOx emissions reduction efficiently operates only in predetermined temperature and air flow ranges of the exhaust air. Due to those conditions, exhaust air from the dryers and ovens must be prepared in advance by controlling the fan speed and fresh and exhaust air dampers positions. At the same time must be ensured that operating conditions in the dryers and ovens are within defined ranges. Operation outside of the defined ranges can result in lower product quality or even in the production shutdown.
In the second part of the thesis, a control system analysis of the exhaust air preparation proposed by the costumer is given. We think the proposed control system is not optimal and could be improved. An example of improved control system of the exhaust air preparation is given.
To design the improved control system, mathematical models must have been obtained. The exhaust air preparation process was identified using identification method, with help of Takagi-Sugeno type fuzzy logic. Brief description of the fuzzy logic is also presented.
In the last part of the thesis, obtained fuzzy models are used to demonstrate the improved control system for the exhaust air preparation. The control system uses proportional-integral (PI) controllers with added dead zones on error signals and manipulated values. Process value signals filtering and the supervisory logic of the control system are also added to ensure the correct operation of the control system.
The control system and the supervisory logic are designed as the concept and are tested with simulation. The purpose of the thesis is to show that better and more optimal control systems can be used for the exhaust air preparation control in comparison with the solution proposed by the costumer. With minor changes of the controller parameters and supervisory logic settings, presented control system can be ready for the implementation on the real process.