Classification of Unstructured Power Grid Data Based on Self-organizing Map Neural Network

In order to provide an effective means for power enterprises to manage non-structural data, a classification and grading method for non-structural data of electrified wire netting based on a self-organizing mapping neural network is studied. This method designs the technical architecture of grid non-structural data acquisition, uses it to obtain grid non-structural data, and then uses the information entropy suppression method to eliminate the redundant data in the grid non-structural data. Based on the grid non-structural data without redundant data, the machine reading understanding model is used to segment the grid non-structural data. Then, the segmentation results are taken as the input of a self-organizing map neural network, and the classification results of non-structural data of electrified wire netting are output through a self-organizing map neural network. Then, the classification results of non-structural data of electrified wire netting are obtained using the data classification method based on immune features. The experimental results show that this method can effectively collect the non-structural data of electrified wire netting, remove the redundancy, and classify and grade them, which has a good application effect.