Method and device for predicting oil temperature of transformer top layer based on error prediction and correction

Abstract

The invention discloses a method and device for predicting oil temperature at the top layer of a transformer based on error prediction correction, belonging to the field of transformer on-line monitoring. The method includes: obtaining a prediction data set, which includes the load current and ambient temperature of the transformer; The load current of the transformer is the input quantity, and the Susa thermal circuit model is used to predict the top oil temperature of the transformer, and the predicted value of the top oil temperature of the Susa thermal circuit model is obtained; the predicted value of the top oil temperature of the Susa thermal circuit model and the predicted data set are normalized processing; using the normalized Susa thermal path model top oil temperature prediction value and prediction data set as input, using the GA-KELM model for regression prediction, and obtaining the regression prediction error; denormalizing the regression prediction error ; Subtract the regression prediction error after denormalization from the predicted value of the top oil temperature of the Susa thermal circuit model before normalization to obtain the corrected top oil temperature prediction value of the transformer. The invention can accurately predict the oil temperature on the top layer of the transformer.

Description

technical field [0001] The invention relates to the field of on-line monitoring of transformers, in particular to a method and device for predicting oil temperature at the top of a transformer based on error prediction and correction. Background technique [0002] The dynamic load capacity and insulation aging speed of power transformers mainly depend on their thermal characteristics. The top layer oil temperature is an important index to measure the thermal characteristics of the transformer, and it is also one of the important monitoring quantities during the operation of the transformer. Accurate and reliable prediction of the top layer oil temperature is of great significance for the reasonable guidance and arrangement of the dynamic load of the transformer and the prevention of thermal failure of the transformer. [0003] At present, there are many prediction methods for transformer top oil temperature, including two typical models, semi-physical models based on heat tr...

AI Technical Summary

Problems solved by technology

The former has the problems of oversimplification of the model, inaccurate parameter calculation, and great influence of environmental factors, and the model has systematic errors; the latter model has the problem of unclear physical meaning, and its generalization needs to be studied

[0004] The Susa thermal circuit model is a typical semi-physical model based on the principle of thermoelectric analogy in heat transfer. It considers the influence of oil viscosity on thermal resistance and oil time constant. It has a clear physical meaning, but it is generally targeted.

Images