Zazhitskiy O. Decision making of aircraft engine blades condition by neural network at the vibroacoustical monitoring

. The dissertation is devoted is devoted to problem solution of the aircraft engines blades vibroacoustical condition monitoring and diagnosis of the crack-like damages at the steady-state and non-steady-state modes. The neuron networks are used for decision mak-ing about blades condition monitoring. Classification of turbine blade condition for amplitude dimensionless characteristics of the vi-broacoustical signals was carried out using a Probability Neural Net-work at the steady-state and non-steady-state modes. Classification of turbine blade condition for bispectral characteristics of the vi-broacoustical signals was carried out using a Probability Neural Net-work and Adaptive Resonance Theory Network at the steady-state mode. The Neural Network classifiers are developed and optimized. The correct classification probability is evaluated and investigated. The classifier scheme, algorithms and software are developed.