Satellite Fault Diagnosis Using Support Vector Machines Based on a Hybrid Voting Mechanism

Algorithm 3

.

Input: sample set for training (including normal data and fault data): (),

is the eigenvalue, and y is the classifying label.

Output: combined decision value

(1) generate single model for each fault using the data of ;

(2) establish the weight for each model ;

(3) computer the decision functions of the data record using ;

(4) make the combined decision of the data record using combined classifiers:

  .