Investigation of a DC OCT signal processing method based on an improved Kalman algorithm

Abstract

In this paper, we propose an improved Kalman filtering algorithm, to solve problems associated with processing of direct current (DC) signals in optical current transformers (OCTs). The algorithm was realized as follows. Based on the characteristics of the OCT, mixed noise was included with the parameters and state information of the measurement model, to ensure that it could be processed. State predictions and update equations for the improved Kalman algorithm were obtained, according to the minimum mean squared error criterion, using the established model. Finally, real-time noise estimation and correction was achieved by designing a recursive noise filter. Results from simulations and experiments demonstrate that the improved signal processing algorithm is effective in filtering mixed noise from DC signals measured by an OCT. Furthermore, the proposed algorithm is able to achieve real-time noise correction during the filtering process.