Transformer upper layer oil temperature prediction method, system and apparatus based on big data

Abstract

The invention discloses a transformer upper layer oil temperature prediction method based on big data. The method includes steps: obtaining detection data of a plurality of preset times, wherein the detection data includes oil temperature data, temperature data and main transformer load data; regarding the detection data of the same time as a group of sample data, substituting each group of sampledata into a pre-established regression prediction model, and determining a coefficient corresponding to the regression prediction model; and obtaining target data of a current time, substituting thetarget data into the regression prediction model which determines the coefficient, and determining the transformer upper layer oil temperature of the current time, wherein the target data includes other various kinds of data except for the oil temperature data. By adopting the method, the transformer upper layer oil temperature of the temperature of a transformer winding can be accurately reflected, and the condition of false alarm caused by abnormal data is reduced. The invention also provides a transformer upper layer oil temperature prediction system and apparatus based on big data, and thecorresponding technical effects are achieved.

Description

technical field [0001] The invention relates to the technical field of power grids, in particular to a method, system and equipment for predicting oil temperature in the upper layer of a transformer based on big data. Background technique [0002] The transformer is the core equipment of the substation, and the operation status of the transformer needs to be controlled in time so that the transformer can operate safely. The winding temperature of the transformer is an important indicator that affects the life of the transformer. In some occasions, the temperature of the winding of the transformer is directly measured to control the start and stop of the cooler. However, since the winding temperature is not convenient for direct measurement, the application of this method is very limited. In other occasions, the average temperature of the transformer winding is obtained for corresponding control, but because the average temperature is used, the accuracy is low, especially in ...

AI Technical Summary

Problems solved by technology

For example, a problem with the corresponding sensor causes the detected oil temperature data of the upper layer of the transformer to be much higher than the real oil temperature data, resulting in a large number of false alarm information

In many occasions, it is necessary to continuously detect the upper oil temperature of the transformer throughout the day, and the number of transformers may be large, so a large amount of detection data will be generated, that is to say, under the background of this big data, false alarms Occurrences can be very high

In addition, it takes a certain amount of time for the upper oil temperature to reflect the temperature of the transformer winding, that is, the change of the upper oil temperature cannot realize the synchronous change with the winding temperature.

Especially in the case of rapid load increase, it takes a long time for the oil temperature on the upper layer of the transformer to reflect the temperature of the transformer winding, which will cause the cooler to not start in time and reduce the service life of the transformer

Images