A method of optimized neural network by L-M algorithm to transformer winding hot spot temperature forecasting

摘要

Transformers are essential device of the power system. Accurate computation of the hottest temperature (HST) of windings is of great meaning, for the HST is an fundamental parameter to guide the load operation mode and influence the life expansion of insulation. Based on the analysis of the heat transfer processes and the thermal characteristics inside transformers, influence on the oil-immersed transformers of factors like the sunshine, external wind speed etc. are taken into consideration. Experiment data and the neural network are used in modeling and protesting of the HST, and furthermore, investigations on optimization of the structure and algorithm of neutral network are conducted. Comparison among the measured value and calculation values using the recommended algorithm of IEC60076 and the neural network algorithm is made; it shows that the value computed with network algorithm approximates more to the measured value than the value computed with algorithm of IEC60076.