Transformer hot-spot temperature prediction method based on kernel principal component analysis and long-short term memory network

Abstract

The invention discloses a transformer hot-spot temperature prediction method based on kernel principal component analysis and a long-short term memory network, and the method comprises the steps: 1, obtaining parameters related to the transformer hot-spot temperature through a sensor, and constructing a feature set; 2, preprocessing the feature data, screening the preprocessed data by using kernel principal component analysis, calculating the contribution rate and the cumulative contribution rate of each feature value, screening the feature values of which the cumulative contribution rate of the principal component is greater than or equal to 90%, and constructing a new feature set; step 3, establishing a long and short-term memory network initialization model, inputting the screened feature data into the network model for training, and constructing a transformer hot-spot temperature prediction model based on the long and short-term memory network; and step 4, inputting effective test data after kernel principal component analysis processing into the constructed model, and outputting temperature of an output layer. The method can improve the data accuracy and calculation speed, can timely discover the temperature fault of the transformer, and guarantees the stable and reliable operation of a power system.

Description

technical field [0001] The invention belongs to the technical field of transformers, and relates to a transformer hot spot temperature prediction method based on kernel principal component analysis and long-short-term memory network. Background technique [0002] At present, the temperature monitoring of the transformer is an important means to measure the real-time operating conditions of the transformer and determine its safe operation. [0003] According to the transformer temperature, the corresponding fault state of the transformer can be judged. There are two basic forms of existing temperature measurement methods: the first type is contact measurement, and the contact measurement is based on the principle of "objects in the same equilibrium state have the same temperature". Basically, when measuring, the contacts of the meter must be in thermal equilibrium with the object to be measured, and the display results are accurate. The characteristic of the contact type is ...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the existing problem of difficult monitoring of transformer hot spot temperature, the present invention provides a transformer hot spot temperature prediction method based on kernel principal component analysis and long-short-term memory network

Images