Short-term Prediction of Oil Temperature of a Transformer during Summer and Winter by Self-organizing Map

H. Du, M. Inui, M. Ohki, and M. Ohkita (Japan)


SelfOrganizing Map (SOM), Transformer, Oil Temperature, Prediction Accuracy, Forecast Atmospheric Temperature


Prediction of oil temperature of a transformer has been practiced using a conventional method based on explicit numerical calculations. When the technique is applied, sometimes there is a limitation in the prediction accuracy due to assumption made on the characteristics of the transformer. This paper considers an application of the Self-Organizing Map (SOM), which is an effective technique for classification of multi-dimensional data, to the prediction of the oil temperature of a transformer. The SOMs are constructed by atmospheric temperatures and oil temperature during three months from June to August and other three months from January to March. The short-term prediction of the oil temperature of transformers can be realized by the SOMs based on the maximum and minimum values of the forecast atmospheric temperature, and the prediction accuracy is higher than that obtained using the conventional method.

Important Links:

Go Back