Gas insulated switchgear (GIS) has been widely used in power grid due to its excellent performance

and it is necessary to assess its insulation condition to ensure safe and stable operation of power grid. However

the analysis of traditional GIS condition assessment model is only based on the partial discharge data collected from sensors

and the information recorded in numbers of historical cases is ignored

which makes it difficult to improve the accuracy of the model. Therefore

a GIS intelligence condition assessment model combined with historical cases is proposed in this paper. First

the partial discharge data of GIS over the past 5 days is analyzed by using the convolutional neural network

and the feature vectors of the partial discharge pattern are extracted. Then

the characters in the case text are converted into word embedding vectors by using the word embedding model

and the case feature vectors of the text are extracted by using the long short-term memory (LSTM) neural network. Finally

the pattern feature vector and the case feature vector are added

and LSTM is used for information analysis to eventually obtain the insulation condition of GIS. The verification on the actual samples collected by the power company is performed

showing tat the algorithm proposed in this paper can effectively assess the insulation conditon of GIS

and the effect is improved by 11.45% compared to the traditional models.