Frequency-independent-element cross-coupled-secondaries model for multiwinding transformers

Abstract

For high-frequency transformers with any number of windings, an equivalent circuit is presented that accurately predicts the effects of ac winding resistance and leakage inductance over a wide range of frequencies. This model is suitable for circuit simulation with nonsinusoidal excitation and nonlinear loads.

Parameter extraction for the extended cantilever model of magnetic component windings

“…The model for a magnetic component consists of a core model and a winding model [2][3][4][5][6][7][8][9][10][11][12]. This correspondence deals with the latter.…”

“…A winding model is generally a set of parameter-dependent equations or a parameter-dependent equivalent circuit that relates the currents and voltages at the winding terminals. The model parameters are usually extracted by performing open-and short-circuit measurements, followed by solution of coupled nonlinear equations to extract model parameters from the data [6]. The extended cantilever model (ECM) introduced in [12] is unique in that, in principle, each parameter can be obtained from one open-circuit-voltage or one short-circuit-current measurement.…”

“…It has received a great deal of attention in recent years due to its unique power in handling problems with both structural and parametric uncertainties and/or changes and in decomposing a complex problem into simpler subproblems. Its great recent success ranges from target tracking to fault detection and isolation, and from biomedical signal processing to process control (see, e.g., [6]). In his Bode Lecture given at the 1995 IEEE Conference on Decision and Control, Kumpati S. Narendra advocated this powerful, robust, and adaptive method.…”

“…Many applications of this MM estimator can be found in the literature under various names, such as the "multiple model adaptive estimator," "parallel processing algorithm," "partitioned filter," "self-tuning estimator," "filter bank," and "Gaussian sum adaptive filter" [4] (see [6] for the long list of references therein). These names suggest the structure, features, and capability of this "first-generation" MM estimator, to be referred to Manuscript received November 16, 1998; revised July 3, 1999; released for publication October 6, 1999. …”