Prediction Model of Dissolved Gas in Transformer Oil Based on VMD‐SMA‐LSSVM

Dissolved gas analysis in oil is an effective method for early fault diagnosis of transformers. Predicting the concentration of future characteristic gases in the transformer can assist operation and maintenance personnel in judging the operation trend of the transformer and ensure stable operation. In order to improve the prediction accuracy of dissolved gas in transformer oil based on a small number of samples, this paper proposes a VMD‐SMA‐LSSVM combined prediction model by using variational modal decomposition and least square support vector machine optimized by slime mold algorithm. First, use variational modal decomposition to decompose the gas signal. For each subsequence, a combined algorithm based on slime mold optimization and least square support vector machine is used to predict separately. Then the prediction results of each sub‐sequence are superimposed and reconstructed to obtain the final prediction value. The research results show that the prediction results obtained based on this method have better prediction effects than other models of machine learning models, other decomposition methods and optimization methods. The proposed method has good fitting characteristics when predicting seven characteristic gases, which verifies the effectiveness. © 2022 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.