Unsupervised integrated method and device of instability detection and fault module positioning

An unsupervised integrated method of instability detection and fault module positioning, comprising following steps:

• acquiring a topological structure of a direct current (DC) microgrid and collecting electrical data of each node in the topological structure, intercepting the electrical data having the same data length at different time points, to construct a corresponding first enhancement dataset and a corresponding second enhancement dataset;
• encoding a fault type of the nodes in an instable status of the DC microgrid and a correlation among the nodes based on the topological structure, to construct a fault type pool;
• constructing a corresponding classification network comprising a data enhancement module, a feature extraction module, a target classification module, and a label mapping module based on a twinborn network framework;
• the data enhancement module being configured to enhance input electrical data and intercept the input electrical data with a preset data length, to output enhancement data;
• the feature extraction module being configured to extract a time-series feature of the enhancement data;
• the target classification module being configured to perform classified calculation on the time-series feature, to output a corresponding classified time-series feature vector;
• and
• the label mapping module being configured to input the classified time-series feature vector to the fault type pool to construct a confusion matrix and perform label matching, to output an optimal matching result;
• training the classification network using the first enhancement dataset and the second enhancement dataset, to obtain a detection model for instability detection and fault module positioning;
• and
• inputting the electrical data of the DC microgrid to be detected to the detection model, to output whether the DC microgrid to be detected has system stability and a corresponding fault type.