Parameter identification of the transformer winding based on least-squares method

Bayesian estimation of electrical transformer parameters

“…This can be seen by the equations (7) to (9). The system matrices are in a nonlinear relation with respect to the vector x, as can be seen by the equations (1) to (4). This is yet no problem, as the system is linear with respect to the input/output relation.…”

“…The estimation of transformer parameters has seen different developments over the past years [3], [4], [5], [6], [7], [8], [9], [10]. Figure 1 depicts the structure of a single core transformer.…”

“…In on-line methods the measurements are taken during the operation of the transformer where no special experiments can be performed. Several methods in this field recast the estimation problem into the form of an over determined equation system with respect to the parameters of interest [3], [4]. Then the Moore-Penrose inverse (pseudo inverse) is used to solve the equation system in a least squares sense.…”

The use of recursive least square to determine the model parameters of a transformer in different frequencies

“…The RLS can be applied in different frequencies providing (6) and 7. In (6) and 7:  represents the fundamental frequency of supply voltage, t  corresponds to the sample time. Thus, the RLS estimates the leakage and magnetization impedances for each frequency.…”

“…The model was found to be more accurate than existing model, which was based on magnetic theory, though at the expense of complexity of programming. A parameter identification method using the least squares based on orthogonal decomposition to identify the leakage inductances and resistances of transformer winding was made by [6]. The algorithm provided a good foundation for online identification of transformer parameters and offers great potential for practical application.…”