Analysis of the impact of non-sinusoidal voltage levels in the power grid of apartment buildings on the reliability and efficiency of transformers

Abstract

The effects of mass implementation of energy-saving lighting on the permissible level of loading of the supply transformer were analyzed. The influence of non-sinusoidal voltages and currents on transformer wear was analyzed. Distortions of sinusoidal shape of voltages and currents lead to additional power losses in transformer. They cause heating of the insulation and windings of the transformer, increased losses and accelerated wear and tear. Higher harmonic components of voltage and current in a 0.4 kV network are caused by the presence of various energy-saving lighting lamps with non-linear volt-ampere characteristic, which include compact fluorescent and LED light sources. As a result, the transformer designed to operate at a frequency of 50 Hz experiences an additional adverse effect in the form of power losses and accelerated wear of electrical insulation, which leads not only to the deterioration of its technical and economic performance characteristics, but also to a significant decrease in reliability associated with the influence of higher harmonic components of voltage and current on equipment failure. It is shown that the lack of power factor correction schemes for low-power compact fluorescent and LED lamps causes the growth of nonsinusoidal voltage and current as they spread, which in turn causes the need to adjust the limit load of transformers to ensure their rated thermal mode. One way to maintain operational reliability under such conditions is to limit the transformer load depending on the level of higher harmonics in the network, as well as to predict the residual life of the operating transformer. Measuring the values of current transformer load and current levels of higher harmonic components of voltage and current will allow not only to calculate an acceptable load factor of the transformer (for the condition of insulation wear), but also to estimate the current residual life of transformer equipment using a neural network model.