Open AccessVirtual NOx-emission sensors for robust aero engine automatic control system
Author(s) -
T. A. Kuznetcova,
В. Г. Августинович
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/862/5/052014
Subject(s) - nox , combustor , combustion , probability density function , artificial neural network , combustion chamber , control theory (sociology) , computer science , automotive engineering , mechanics , engineering , mathematics , chemistry , physics , artificial intelligence , statistics , control (management) , organic chemistry
Modern robust control systems use built-in mathematical models for estimation of unmeasured by the direct methods parameters such as NOx emission. The two models of NOx emissions virtual sensor built into the smart controller of an aero engine low-emission combustion chamber are proposed in this study. A stochastic nonlinear mathematical model is based on the Zeldovich equation. It applies the superposition principle of NOx production in diffusion and homogeneous flames. Probability density distribution functions of the air-fuel mixture concentration in these flames take into account both of a spatial non-uniformity of the mixture composition and a harmonic component of the acoustic waves generated by the heat release. The NOx generation rate is averaged over the fuel flow through diffusion and homogeneous flames based on the probability density distribution function. An exponentiality of a decrease in the generation rate along the length of the combustor space is proposed. The concept of integral relations models has been developed with the use of numerical modeling of spatial and temporal non-uniformities of the air-fuel mixture concentration (4D-metamodeling) and available experimental data. Another virtual sensor model is based on the neural network. A well-known approach of NOx emissions prediction used in monitoring systems of gas turbine power plants is applied. The example of a neural network and results of its training on a real combustion chamber is presented. It is shown that the two or three-layer neural network having 20… 30 neurons provides an acceptable error (not exceeding 10%) of the NOx emission display and can be used as a virtual emission sensor in an engine control system. The normalized level of NOx emission per take-off and landing cycle is considered as a target function of the automatic control of low-emission combustion. The estimation of the level of NOx emission by a built-in virtual sensor is proposed for robust aero engine automatic control.