With the development of emerging technologies in the new era,higher requirements are put forward for the fault di-agnosis technology of substation equipment.Once equipment failure occurs in the substation,if there is no advance warning and treatment,it will have a huge impact.The active learning random forest model proposed in this paper extracts a data sample set from the original substation equipment data,performs feature extraction,and uses it to construct an adaptive random tree.Ac-cording to the corresponding weight of each random tree,the classification of known faults and the detection of unknown faults are completed through the proposed voting mechanism.The method proposed in this paper can automatically identify new faults,adjust the model automatically,and introduce an adaptive voting mechanism to improve the classification accuracy of the model.The analysis results verify the effectiveness of the method.In addition,compared with the traditional random forest model,its classification accuracy,misrecognition rate and rejection rate are all excellent.
维普
万方数据