Work

The paper is devoted to development of theoretical and practical control principles for multi-channel semiconductor converters using vibroacoustic signals on separate channels loads power. In the papers mathematical models are developed and analytical expressions that uniquely associate electrical signals with vibroacoustic are derived. The converters power transformer is the source of vibroacoustic signals due to sufficient transformer signal intensity and to the fact that it reflects the generalized channel parameters of multi-channel power supply system. The vibroacoustics signal model takes into account the core surface fluctuations due to phenomenon of magnetostriction and coil surface fluctuations as a result of interaction of windings under the influence of Lorentz force. The dependences of energy flows power passing through the transformer and its vibroacoustic signals on the main model parameters change were obtained using the model, the analysis of which shows a clear connection between electrical and vibroacoustic signals. The analytical expressions for calculation of energy flows power of input and output transformer windings directly on the core and windings surface fluctuations signals are derived on the model basis. Obtained analytical expressions differ in application areas for which they are best suited and for which the minimal calculation complexity is ensured. Also transfer functions are obtained in which the ampere-turns of primary and secondary windings and transformer core induction perform as output signals and the surface fluctuation functions as input signals. Suggested analytical expressions and transfer functions, compared to the capacities calculation method using windings voltages and currents, allow reducing the indirect power measurements error in proportion to the total number of primary and secondary windings with techniques fixed value method arithmetic operations non-exceedance that depends on the chosen set of analytical expressions for the particular transformer. The correctness of the proposed model and obtained analytical dependences is confirmed experimentally.