Neural Network Method for Helicopters Turboshaft Engines Dynamic Efficiency Monitoring

The work is devoted to the development of a neural network method for monitoring the dynamics of changes in the efficiency of the engine compressor, which characterizes the efficiency of air compression in the compressor and, as a result, ultimately determines the fuel consumption, which affects the helicopter flight. The use of the neural network ensemble “RBF – Perceptron – Neuro-Fuzzy Classifier” is proposed, which allows to determine the current engine status with an accuracy of 99.62 % and quickly decide on the possibility of carrying out (continuing) the flight. The results of a comparative analysis with other neural networks architectures showed that the use of the neural network ensemble “RBF – Perceptron –Neuro-Fuzzy Classifier” provides a decrease in errors of the first and second kind from 1.23 to 5.96 times.