Nonlinear load modeling for analysis of non-sinusoidal conditions in electrical networks based on measurements of harmonic parameters

Non-sinusoidal conditions in electrical networks need to be calculated for their control and development of technical measures to maintain harmonic parameters according to the requirements of regulatory documents. These calculations are impossible without electrical network and nonlinear load models that adequately reflect them in computational programs. Nonlinear load models have been developed for a long time. Some studies present general modeling principles and models of various nonlinear devices. Others consider some nonlinear devices as equivalent nonlinear loads connected to low and medium voltage networks. A whole host of high-power nonlinear electrical equipment is connected to high voltage networks. Modeling nonlinear loads connected to these networks is a problem. Research of measured parameters of harmonic conditions in electrical networks has shown that they are random values. The probabilistic nature is determined by the network configuration, a range of network components, the number of nonlinear loads, wave and frequency properties of the network, harmonic source phase currents, voltage at terminals of nonlinear electrical equipment, changes in operating conditions and load power, and many other factors. Nonlinear loads can only be modeled based on the measurements of parameters of harmonic conditions due to their unpredictability. The paper presents an overview of existing methods for modeling nonlinear loads, a methodological approach to modeling nonlinear loads based on measured parameters, an algorithm for modeling harmonics of active and reactive currents, a computational program algorithm designed to identify distribution functions of measured current harmonics, and modeling results for current harmonics of railway transformers supplying power to electric locomotives.