Search

The purpose of this paper is synthesis of intellectual automatic control system (ACS) of the water level of drum boiler unit based on the theory of neural networks. We develop a mathematical model of the research object. Analysis of the transients quality under different operating boiler shows that the traditional ACS has significant drawbacks associated with significant regulation time and the first rejection. To improve the traditional ACS and its efficiency, we propose to use a neural network approach to control. We synthesize a neural network controller (NNC) trained to determine the optimal control actions to compensate for the deep disturbances caused by peaks and adjusting modes of power. To test the proposed solutions in the program MatLab (Simulink, NT) computer experiments are conducted to develop and to test a neural network control system. Analysis of the transients quality when conducting an experiment to determine the robustness and stability of the NNC in the deep impact of external and internal disturbances, advantages of adaptive neural network system compared to the traditional combined one, requires a period of adaptation. [ABSTRACT FROM AUTHOR]