Research On Transformer Energy Efficiency Evaluation Under The Influence Of Multi Factor Power Quality

With the rapid development of domestic economy and the continuous growth of power grid capacity,people also put forward higher requirements for the stability of power.More and more researches on the influence of power quality factors on power equipment,especially transformers,are also made.This paper starts with the theoretical analysis of the influence of power quality factors on transformers,and summarizes the calculation mathematical models of transformer energy efficiency under the influence of harmonic,three-phase unbalance,voltage fluctuation and flicker factors at home and abroad.Through consulting the literature,it is found that most of the researches lack the support of field test data,only for theoretical analysis.In this paper,the modification of the transformer test data of a power quality factor test base is introduced to verify the mathematical model.It is worth mentioning that the corresponding literature is not found for the calculation mathematical model of transformer energy efficiency under the influence of multi power quality factors,Therefore,the additional loss caused by the transformer is linearly superimposed as the mathematical model of energy efficiency calculation under the influence of multi power quality factors.It is found that the error between the calculated loss data of the mathematical model of energy efficiency calculation and the field test data is large when the harmonic factors are affected by the comparison.After the correction of the harmonic loss coefficient,the error is maintained at about 5%.Based on the traditional mathematical calculation model of transformer energy efficiency under the influence of power quality factors,this paper proposes to apply intelligent algorithm to transformer evaluation under the influence of power quality factors.The two algorithms used in this paper are: CSO-BP(crisscross optimization back propagation)with fast and local best ability,The neural network algorithm with cross optimization and BDA(balanced distribution adaptation)algorithm with good migration learning ability and strong generalization ability.By comparing and analyzing the calculation results of the mathematical model of transformer energy efficiency under the same power quality factors,the error between the output loss results of CSO-BP neural network algorithm and BDA algorithm and the field test data,it can be concluded that CSO-BP neural network algorithm and BDA algorithm have good application effect in the evaluation of transformer energy efficiency under the influence of multi factor power quality,and the error of calculation results is relatively low.The results of CSO-BP neural network are better than that of single transformer,and BDA algorithm is more suitable for the loss calculation of multiple transformer data.The measurement data of transformer in a power station is replaced by two intelligent algorithms.The error between the calculation results and the field measurement data is small.The two intelligent algorithms are universal in the evaluation of transformer energy efficiency under the influence of multi factor power quality.