LAUSR

Because the current method of attack detection of current flow microgrid is slow, it leads to the increase of detection time. Therefore, a two-stage segmentation model based attack detection method for DC microgrid network is proposed. The structure of DC microgrid is analyzed, the model of DC microgrid network is constructed, and the sparse region of the micro grid network is divided. According to the results of regional division, a two-level and multi-segment model is constructed, which is composed of the above-level central neural network and the lower level is edge end neural network. The parallel computing structure of the model is set up to get the attack detection results of DC microgrid network. The experimental results show that the method has high recall and precision, low error detection rate and short detection time, which can realize the rapid and accurate detection of network attacks.