Aerospace Instrument-Making Annotation << Back APPLICATION OF MACHINE LEARNING METHODS TO SELECT THE OPTIMAL BLADE CONFIGURATION FOR ENGINE DEVELOPMENT D.O. Pushkarev, D.Yu. Kiselev, Yu.V. Kiselev Using machine learning methods (radial-basic neural network), a neural network model for selecting the confi guration of gas turbine engine blades was developed to refine the engine during acceptance tests after repair. In the process of creating a neural network model, the thermogasdynamic parameters of the engine operation obtained during tests on stands were used. For verifi cation, we used data obtained from samples that were tested without the use of neural networks. The results of the developed method are compared with the statistical data for the past years. The results showed that the developed model gives high accuracy of results and improves the efficiency of engine diagnostics and engine development. Keywords: GTE, diagnostics, machine learning, neural network. DOI: 10.25791/aviakosmos.2.2023.1323 Pp. 30-38.