This article presents the principles of modeling three-phase asymmetrical currents in the form of secondary voltages of signal change sensors in the power supply system. The classical methods of signal transformation modeling do not provide the necessary accuracy due to the complexity in representing the interdependence between input and output quantities and parameters, as well as the difficulty in capturing the interchain dependencies and associated physical-technical effects. The analysis in this work focuses on the principles of signal transformation based on graph models, including the physical and technical effects of the three-phase primary current electromagnetic sensor that creates the interacting magnetic fields. The modifications to elements with nonlinear parameters and the values of the electric network currents in the power supply system during estimation have been considered as an object with concentrated parameters. In contrast to traditional single-phase current transformers, the complex transducing components, weight, and volume of the items are laborintensive when used in combined power measurement and control systems. They also do not provide the universality of the output value when interfacing with microprocessors and electronic technologies in intelligent systems.