Parametric diagnostics of the condition of a dual-flow double-shaft turbojet engine using neural network simulation of the operating process

The article presents a method of parametric diagnostics of the condition of a dual-flow double-shaft turbojet engine (DDTE). The method is based on the identification (determination) of the condition of the DDTE components (the compressor, combustion chamber, turbine) with application of a mathematical model of the operating process which is presented as an artificial neural network (ANN) model. This model describes the relation between the monitored parameters of the DDTE (the air temperatures
$\left(\mathrm{T}_{\text{lpc}}{ }^*, ~\mathrm{T}_{\text{hpc}}{ }^*\right)$
beyond the low pressure compressor (LPC) and the high pressure compressor (HPC), the pressure beyond the LPC (P
lpc
), the fuel consumption rate (G
f
), the gas temperatures
$\left(\mathrm{T}_{\text{hpt}}{ }^*, ~\mathrm{T}_{\text{lpt}}{ }^*\right)$
beyond the high pressure turbine (HPT) and the low pressure turbine (LPT)) and the parameters of the condition of its components (the efficiencies of the LPC and the HPC
$\left(\eta_{\text{lpc}}{ }^*, \eta_{\text{hpc}}{ }^*\right)$
, the stagnation pressure recovery factor in the combustion chamber
$\left(\sigma_{\text{cc}}\right)$
, the efficiencies of the HPT and the LPT
$\left.\left(\eta_{\text{hpt}}{ }^*, \eta_{\text{lpt}}{ }^*\right)\right)$
. The parameters of the condition of the engine components
$\left(\eta_{\text{lpc}}{ }^*, \eta_{\text{hpc}}{ }^*, \sigma_{\text{cc}}, \eta_{\text{hpt}}{ }^*, \eta_{\text{lpt}}{ }^*\right)$
are the similarity criteria (integral criteria) which enable to identify the condition of the DDTE components to a high degree of reliability. Such analysis enables to detect defects at an early stage, even if the values of the monitored parameters
$\left(\mathrm{T}_{\text{lpc}}{ }^*\right$
.,
$\left.\mathrm{T}_{\text{hpc}}{ }^*, \mathrm{P}_{\text{lpc}}, \mathrm{G}_{\mathrm{f}}, \mathrm{T}_{\text{hpt}}{ }^*, ~\mathrm{T}_{\text{lpt}}{ }^*\right)$
are within the permissible limits. We provide the sequence of development of the ANN model diagnostics and the results of its performance resesarch depending on the various factors, including the software parametric setting.