Deep Transfer Learning-Based Fault Diagnosis for Gearbox under Complex Working Conditions

Table 2

Fault diagnosis process of the proposed algorithm.

Step	Description
1	Collect the signal under single working condition as training data set ; collect the signal under complex working conditions as testing data set .
2	Extract feature data set of time domain and frequency domain in Table 1 as preparation.
3	Calculate sensitive factor in (2) to keep these features in which the value of is large. These parameters constitute sensitivity parameter set as input data.
4	Let in (12); train network to gain suitable parameter set and the source features.
5	Assign suitable values in (12) to validate the network by target data set until minimizing the cost function in (12) by comparing the distance between the target features and source features, using from step 4 as initial parameters.
6	After step 5 is done, record the parameters and features of testing.
7	Send the features into classifier to gain the fault types.