Equivalent thermal model for main transformer in hot strip mill and optimization of load
暂无分享,去创建一个
The temperature of transformers is very important factor. If the temperature exceeds the rated temperature of the transformer, then its life decreases exponentially. This paper shows the relation between the sequence of slabs and the life loss of oil-immersed power transformer in steel works. Firstly, the equivalent thermal model for oil-immersed power transformer will be introduced. Using k-means clustering, current for slabs are classified into some current groups. To expect current groups of slabs that have not classified, a model for the system is needed. For this model, neural network with MLP was used. Finally, to know that the life loss of the transformer is influenced by different slab sequences, some different sequences were designed. The optimal sequence which minimizes the life loss of the transformer was obtained by means of Quasi-Newton optimization method. It is shown that proposed optimal sequence has the minimum loss of life among others.
[1] J. MacQueen. Some methods for classification and analysis of multivariate observations , 1967 .
[2] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .
[3] J. J. Moré,et al. Quasi-Newton Methods, Motivation and Theory , 1974 .