The Effect of Distributed Generation Plants’ Prognostic Controllers on Power Quality in Power Supply Systems of Non-traction Consumers

Abstract

AbstractIn power supply systems powered from mainline railroads’ traction substations, the problem of ensuring the electric power quality is acute. One of the effective ways to solve it is the use of smart grid technologies, an important segment of which are automatically controlled distributed generation plants. The article presents the results of computer simulation of transients during an increase in traction load, which leads to deterioration of power quality in non-traction consumers’ power supply systems. The purpose of the study was to determine the efficiency of using distributed generation plants to improve the power quality in power supply systems of on-site facilities powered from mainline railroads’ traction substations. Particular attention was paid to the setting and algorithms of the automatic controllers of the distributed generation plant synchronous generator and their influence on the indicators of the transient process and the power quality in the power supply system. The studies were conducted in the MATLAB development and simulation environment. Their results indicated that the presence of distributed generation plants allows to improve the power quality factors at the busbars of non-traction consumer; concordant setting of controllers provides the possibility for additional improvement of these indicators. The use of prognostic controllers ensures maintaining of power quality factors at a constant level when changing the generator load. The use of direct current link and prognostic controllers almost completely eliminates voltage asymmetry and harmonic distortions, significantly reduces the oscillability and value of voltage overshoot and generator rotor speed of the distributed generation plant.KeywordsRailroad power supply systemsDistributed generation plantsSynchronous generatorsAutomatic voltage controllerAutomatic speed controllerPrognostic controllersPower qualityModeling