Time Series Prediction Method of Dissolved Gas Content in Transformer Oilased on ARIMA Model
[1]Jia Rubin,Gao Jinfeng.Time Series Prediction Method of Dissolved Gas Content in Transformer Oilased on ARIMA Model[J].Journal of Zhengzhou University (Engineering Science),2020,41(02):67-72.[doi:10.13705/j.issn.1671-6833.2020.03.010]
Copy
Journal of Zhengzhou University (Engineering Science)[ISSN
1671-6833/CN
41-1339/T] Volume:
41
Number of periods:
2020 02
Page number:
67-72
Column:
Public date:
2020-05-31
- Title:
- Time Series Prediction Method of Dissolved Gas Content in Transformer Oilased on ARIMA Model
- Keywords:
- Transformer oil; gas content; sequentially; ARIMA model; predict
- CLC:
- -
- DOI:
- 10.13705/j.issn.1671-6833.2020.03.010
- Abstract:
- The dissolved gas content in transformer oil is an important index to measure the operation status of transformers. The differential autoregressive moving average model (ARIMA) is used to predict the gas content in transformer oil. This method uses the time corresponding to the gas content value as an index to input the prediction model through python programming. The original non-stationary time series is converted into a stationary time series by means of difference processing, and then several sets of models are obtained by using the autocorrelation function and partial autocorrelation function parameter selection principles, and are used in the process of optimizing several sets of models. A set of optimal models were obtained by Chichi information, Bayesian information, and Hannan-Quine criteria. Finally, the residuals of the optimal models were tested by correlation testing methods, and the gas content was predicted using the models that met the residual requirements. Experiments show that the proposed prediction method has high prediction accuracy, which can provide a valuable reference for rationally arranging the condition-based maintenance of transformers.
- References:
- -
- Similar References:
Memo
-
Last Update:
2020-05-30