The article presents a computer-aided method of diagnostics of gas turbine blades with use of artificial neural networks. The subject of presentation is the developed neural network, with help of which – on the basis of features of blade surface images – realised is determination of their condition (operable element – inoperable element). Basing on conclusions formulated on the basis of microstructure examinations and concerning evaluation of state of overheating (blades suitable and not suitable for further operation), as patterns assumed were surface images representing blades in various states (neural pattern classification). Additionally, combining and segregating (according to their applicability for the network teaching process) image parameters, acquired from histograms as well as from matrix of events, automated and increased was the credibility (computer aiding) of decision process. The application of artificial neural network enables better representation of complex relations between blade image and its condition, than in the case of subjective methods used currently by diagnosticians.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.