Neural network control for thermal power plant boiler

In today's conditions where the development levels of the countries are determined with the energy scales consumed, not only energy production but also of which process and method being pick gained importance, these importance increasing day by day. In this context, it is of great importance to ensure the sustainability of the system by reducing fuel costs from an economic point of view and this improvment provides the establishment of a more sustainable world order by consuming less resources with changing environmental conditions and the perspectives of societies. Although the consumption of coal, which increased with the industrial revolution, decreased with the discovery of other energy sources, espacialy coal fueled power plants still placed top of list in our country. Although human beings have turned to renewable energy resources to ensure environmental sustainability, today's technologies have not yet reached the level of technology that can meet all the energy they need only with these resources. while studies on resources are in progress on the other hand, it is important to always keep improving the existing production processes. Improvements in the processes of thermal power plants, which are widely used in electrical energy production activities in our country and which already hold the base load in many countries of the world, are important. Improvements in the control process can create a more economical, more efficient and environmentally friendly structure. The control parameters determined during the design phase of thermal power plants deterioration due to changes in mechanical equipment over time caused problems such as,dynamic system structure with changing environmental conditions from the design stage, main input changing due to lack of fuel homogeneity, factors resulting from waste generated during the combustion process, Human (operator) behaviors Control by learning systems that can adapt to changing conditions to cope with disruptive factors is the focus of studies on the above-mentioned benefits. In the scope of this thesis, ANN-based modeling, ANN-based adaptive boiler controller design and simulation of a 157 MW fossil fuel thermal power plant in MATLAB 2018 B have been made. The results of modeling and control design are examined in terms of applicability and utility.