A method and system for predicting the oil temperature range of the top layer of a power transformer

Abstract

The invention discloses a method and system for predicting the oil temperature interval of the top layer of a power transformer. Based on the kernel limit learning machine and the Bootstrap method, the original training set data is obtained, and the sub-training set is generated through the Bootstrap method; multiple kernel limits are trained using the sub-training set data. The top-level oil temperature prediction model of the learning machine; the original training set is predicted by multiple kernel limit learning models, and the training samples of the noise prediction kernel limit learning machine are generated according to the prediction results, and the noise prediction kernel limit learning machine is trained; multiple kernel limit The top oil temperature prediction model of the learning machine predicts the verification set, and uses the noise prediction kernel extreme learning machine to predict the observation noise variance of the top oil temperature; according to the variance of the top oil temperature prediction results and the predicted observation noise based on multiple kernel extreme learning The variance is calculated to obtain the prediction interval of the top oil temperature. The invention can obtain a clear and reliable oil temperature prediction interval at the top layer of the transformer at a certain confidence level.

Description

technical field [0001] The invention relates to a method and system for predicting the oil temperature range of the top layer of a power transformer. Background technique [0002] The safety and aging rate of the oil-paper insulation system of an oil-immersed power transformer are mainly affected by its internal temperature rise, and the top layer oil temperature is one of the important thermal variables describing the internal temperature rise state of the transformer. During the operation of the transformer load, it must be ensured that it does not exceed the limit value. According to the load curve of the transformer and the ambient temperature conditions, the top oil temperature can be accurately and reliably predicted in the future. On the one hand, it can make full use of its load capacity under the premise of ensuring the safety and reliability of the transformer. On the other hand, it can prevent the transformer from appearing. Overheating faults are significant. ...

AI Technical Summary

Problems solved by technology

[0003] At present, the prediction methods for the oil temperature on the top layer of the transformer are all point prediction methods, that is, the specific prediction value at a certain moment is given, and the overall error level is only used as the performance index of the prediction accuracy, without considering the uncertainty and prediction of the prediction model itself. Positive and negative characteristics of error

Images