Satellite fault diagnosis method based on AdaBoost algorithm

Satellite fault diagnosis method based on AdaBoost algorithm, belongs to the technical field of satellite fault diagnosis. For the present invention, a conventional method of fault diagnosis system of Naive Bayes smaller number of weak fault diagnostic capability status issue, satellite fault diagnosis method based on AdaBoost algorithm: Algorithm AdaBoost algorithm integrated reasonable application naive Bayes existing fault diagnosis system; after each training session, changing the weight of the sample, that is, change the value of the wrong type event features sub-sample to enhance the degree of concern fault diagnosis system of this feature; create new ones correspondence between the state and the event features matrix, re-training the classifier, the classifier is trained for the new integrated fault diagnosis classifier. The present invention is applicable to the field of satellite fault diagnosis.