Automatic Control System of Low-Emission Combustion Chamber Based on Neural Network Emission Model

The paper aims to resolve the problem of reliability improvement of aircraft gas turbine engine's automatic control systems (ACS GTE) based on the use of the built-in mathematic model of the control object. The purpose of the study is the development of the on-board real-time monitoring and control of nitrogen and carbon oxide emissions in a modern GTE. The main purpose of the research is compliance with the emission standards while ensuring the stability of the combustion process under the influence of external and internal factors, based on the creation of robust control algorithms for low-emission combustion chambers (LECC). The mathematic model of the emission generation is designed for a work in conjunction with ACS GTE in a real environment and satisfy the requirements for compactness, speed and accuracy of engine parameters' identification in statics and dynamics in a wide range of operating modes, flight and engine conditions. The LECC's emission identification algorithms are based on the artificial neural networks. The data array of a full-scale experiment was obtained for studying the characteristics of emission during the industrial LECC operation. The neural network simulating LECC's emission was designed and trained on the basis of the obtained data. The MATLAB simulation results showed the high accuracy of the developed model. The different factors' importance for the model accuracy was studied. It turned out that the most important are the parameters of temperature and pressure. The obtained results can be used for the improvement of the reliability and environmental attractiveness of ACS GTE.