On the basis of the analysis of the advantages and limitations of two different ways to predict the condition of power transformer, an approach to the prediction of characteristic parameters based on the discrete gray model is introduced into the reliability prediction of power transformer, as a solution to avoid the possible instability of the traditional gray model prediction. And according to the data characteristics of the characteristic parameters of power transformer, data preprocessing of strong randomicity sequences, non-equidistance data equalized in the time is also described to make the discrete gray model suitable for original data prediction and to improve the prediction accuracy of characteristic parameters, and the method of continuous parameter-prediction is also proposed in the paper, so that all the obstacles from the theoretical analysis to the engineering application of condition prediction of power transformer by predicting the condition parameters values with discrete gray model are removed, and the right prediction of power transformer condition are achieved.
[1]
Hv Power,et al.
APPLICATION OF GREY VERHULST MODEL IN MIDDLE AND LONG TERM LOAD FORECASTING
,
2003
.
[2]
Liu Si-feng,et al.
Research on Extension of Discrete Grey Model and Its Optimize Formula
,
2006
.
[3]
Huang Dao,et al.
Diagnosis and Prediction of Chemical Equipment Based on Grey Theory
,
2006
.
[4]
Liu Si-feng,et al.
Discrete GM(1,1) and Mechanism of Grey Forecasting Model
,
2005
.
[5]
Liu Jian-xin.
Transformer fault prediction using new grey prediction model
,
2007
.
[6]
Yang Xiang,et al.
Application of combined optimum grey model to mid and long term load forecasting
,
2007
.
[7]
Zhang Xiaoqi.
Prediction model for dissolved gas in transformer oil based on improved grey theory
,
2008
.