Prediction of the dissolved gas concentration in power transformer oil based on SARIMA model

Energy Reports

Guogang Zhang¹  Jiaxin Liu²  Zijian Zhao²  Yuanchen Zhong³  Chenchen Zhao⁴ 

[1] Corresponding author.;State Grid Liaoning Electric Power Company Limited Electric Power Research Institute, Shenyang 110006, China;State Grid Panjin Electric Power Supply Company, Panjin 124000, China;State Key Laboratory of Electrical Insulation and Power Equipment, Xi’an Jiaotong University, Xi’an 710049, China;

关键词: Power transformer;    Dissolved gas analysis;    Trend prediction;    SARIMA model;   

DOI  :  

来源: DOAJ

【 摘 要 】

As an important equipment in the power system, the operating state of the power transformers is related to the safe operation of the entire system. It is of great importance to determine the transformer state and take countermeasures in advance. Dissolved gas analysis (DGA) is one of the most common methods for state evaluation of the oil-immersed transformer. Therefore, a time series forecasting model based on Seasonal Autoregressive Integrated Moving Average (SARIMA) model is presented to forecast the dissolved gas concentration in transformer oil in this paper. Firstly, the influence of the period parameter on the prediction effect and the correlation between the ambient temperature and the gas concentration are analyzed. Then, comparing the prediction results of the autoregressive (AR) model, the long short-term memory (LSTM) model and the SARIMA model on the trend component of the gas, it is found that the SARIMA model has the most accurate and stable performance. And the accuracy of the prediction results is higher when the external factors related to the prediction data are considered in the SARIMA model. Finally, the reliability of the SARIMA model for the recent prediction on actual data is verified. The case studies show that the SARIMA model can be the suitable method for predicting the future trend of the dissolved gas concentration in power transformer oil.

【 授权许可】

Unknown