The key to solve the problem of economic comparison and selection of gas insulated switchgear (GIS) equipment is to realize the accurate prediction of life cycle cost (LCC) data of substation GIS equipment. In accordance with the problem of few LCC data samples and abnormal data of existing substation GIS equipment

LCC data of similar substation GIS equipment is introduced and combined with self-attention (SA) mechanism to expand the sample scale

and then GRU model is used to predict abnormal time series data. A LCC data prediction method for substation GIS equipment based on SA-GRU model is proposed. Considering the differences of data characteristics between the similar data and the sample data

the self-attention mechanism algorithm is used to strengthen the data characteristics

and the weight of the influence degree between the similar data and the sample data is quantified

so that the data characteristics of the two are consistent. Then the GRU model is used to solve the problem of gradient disappearance explosion in training

and finally the LCC data in the whole life cycle are predicted. The LCC data of GIS equipment in a substation in a certain province in central China are taken as an example to verify the method. The proposed method demonstrates an improvement of 31.11% in accuracy based on the root mean square error compared to the Attention-GRU model

validating the accuracy of the method presented in this paper.