Modeling method for prediction model of dissolved gas concentration in transformer oil

Abstract

The invention relates to a modeling method for a prediction model of dissolved gas concentration in transformer oil, which comprises the following steps of: selecting important factors which possibly have positive influence on prediction of dissolved gas in transformer oil, establishing a variable environment, and setting that prediction of each gas is positively influenced by all the other variables; using a long short term memory (LSTM) network prediction model for predicting and comparing errors respectively, performing verification hypothesis, extracting important factors having positive influence on gas variables to be predicted, establishing an LSTM prediction model fused with a time attention mechanism on the basis that the important factors are screened out, enhancing expression of key information in historical time series data of the important factors, taking the time series data corresponding to the important factors as the input of the LSTM fused with the time attention mechanism, and completing the modeling of the prediction model. According to the method, the problem of low precision of the method for predicting the concentration of the dissolved gas in the transformer oil based on the traditional correlation analysis method can be solved according to the acquired historical operation state data of the transformer.

Description

technical field [0001] The invention relates to the field of evaluating and predicting the operating state of power transformers, in particular to a modeling method for predicting the concentration of dissolved gas in transformer oil. Background technique [0002] As one of the most critical equipment in the power system, the power transformer has the function of transmitting and converting electric energy. It is very important to ensure its safe and stable operation for the entire power network. Relying on the traditional regular manual maintenance method to conduct health checks on transformer equipment will consume a lot of manpower, material resources and financial resources, and may also cause excessive or untimely maintenance, resulting in damage to the transformer and loss of life. Therefore, the future state of the transformer is predicted and implemented. The refined management of the whole life cycle is becoming a hot spot. [0003] Dissolved Gas Analysis (DGA) te...

Images