International Journal of Aerospace Engineering / 2018 / Article / Tab 14

Research Article

Aircraft Gas Turbine Engine Health Monitoring System by Real Flight Data

Table 14

Details of the neural model.

Network parametersDetails

Architecture8 inputs, 1 outputs, and 4 hidden layers with 11, 8, 4, and 3 neurons in each layer (11-8-4-3)
SetTraining subset: 70% randomly selected recorded data (9045 patterns)
Validation subset: 15% randomly selected recorded data (1935 patterns)
Test subset: 15% randomly selected recorded data (1935 patterns)
ActivationFor hidden layers: tangent sigmoid tangent
FunctionFor output layer: linear
Training algorithmLevenberg-Marquardt
Performance function criteriaMinimum MSE
StoppingValidation stop
Criteria(Training is stopped when the validation error starts increasing)