Neural Network Method for Helicopters Turboshaft Engines Working Process Parameters Identification at Flight Modes

This work is devoted to the development of a neural network method for helicopters turboshaft engines working process thermo-gas-dynamic parameters identification. The developed method is based on a neural network - a three-layer perceptron of architecture 7-53-36, the optimal training method for which - back propagation algorithm. To solve the applied problem of identification of helicopter turboshaft engines (using the example of an aircraft engine TV3-117). The error of identification of thermo-gas-dynamic parameters of the working process of turboshaft helicopter using the developed three-layer perceptron does not exceed 0.68 %. For the classical method (thermo-gas-dynamic model of helicopter turboshaft engine) it is about 2 % in the considered range of changes in the engine operating mode.