Research On Signal Processing Method Of Optical Current Transducer Based On Improved Kalman Filter Algorithm

Modern power system is developing rapidly in the direction of digital intellectualization of ultra-high voltage measurement, but the traditional electromagnetic transducer exposes some serious shortcomings like difficult to insulation, easy to magnetic saturation and explosive, so it’s unable to satisfy the modern electric power. With simple structure, no magnetic saturation, wide frequency response, offering analog to digital output and so on, optical current transducer(OCT) is tend to adapt to the rapid development in the field of smart power grid, and be an ideal equipment for measuring electric flow to replace traditional mutual inductor gradually. However, in the measuring process of OC T, photoelectric detector may produce random noise which affects the measurement accuracy, and the signal spectrum overlaps with noise. What’s more, it is difficult to obtain the noise statistical properties. Therefore, some research is made on OCT signal processing method, in order to improve the OCT’s output SNR and measuring accuracy.At first, this paper introduces the basic principle of OCT. Then selects an appropriate measurement structure, and do theoretical research on the noise source and characteristics OCT. On this basis, photoelectric detector noise is collected through the experiment, and is analyzed to obtain the signal-to-noise properties of OCT. Later, considering that low SNR and signal overlapping noise are both characters of OCT, two Kalman filtering algorithm is introduced into in the AC/DC signal processing of OCT.On the one hand,according to the characteristics of the DC OCT, improvement of the standard Kalman filtering algorithm is proposed. Establish the model of measuring state, blend in the process of mixed noise, and design noise recursive. After doing these, new comprehensive filtering processes are created. Writ e a simulation program to verify the effectiveness of the improved algorithm. The simulation results show that the improved algorithm can improve the measurement precision effectively and are more suitable for DC OCT signal processing.On the other hand,according to the characteristics of the AC OCT, this paper puts forward the improved central differential Kalman filtering algorithm(CDKF). Establish OCT AC measuring state space model. Improve the operation efficiency of CDKF in the aspects of sampling strategy, coloured observation and noise processing. Design Recursion to estimate the noise adaption. Write the Matlab simulation program. The extended Kalman algorithm, CDKF method and improved CDKF method are compared to each other on filtering effect. The results indicate that the improved CDKF algorithm is optimal on OCT communication signal processing, and can improve the AC OCT output SNR and measuring accuracy.Finally, design and build experiment platform for AC/DC measurement experiment of OCT. Measure the characteristic curve of the system, complete the correction of the system platform, and use the improved algorithm we design to compare the effect with Bessel filter. The experimental results show that the improved algorithm we propose has better filtering effect than Bessel filter, and it can effectively improve the OCT AC/DC measurement precision and signal-to-noise ratio.