Yunpeng Gao

Fast S-Transform for Time-Varying Voltage Flicker Analysis

Voltage flicker analysis is one of the elementary tools of power quality assessment in power systems. A new algorithm, fast S-transform (FST) is proposed and is used to analyze the time-varying voltage flicker in this paper. A voltage flicker waveform envelope, which captures the main flicker characteristic, is obtained by Teager-Kaiser energy operator instantaneously. FST achieves much lower computational complexity in comparison to the S-transform. FST is used to extract both the time and the frequency domain information of each flicker component. The proposed algorithm can be easily implemented with hardware multipliers, making the method a good choice for real-time applications for the time-frequency analysis of flicker. The implementation of the proposed algorithm in the dual-core processor-digital signal processor and advanced RISC machines-based voltage flickermeter is also introduced. The simulation and application results validate the accuracy and efficiency of the proposed algorithm.

Dielectric loss factor measurement based on nuttall self-convolution window phase difference correction

Some windows which have small side-lobe level and quick fall off rate have been used to eliminate the measurement errors of dielectric loss factor tanδ caused by non-synchronized sampling and non-integral period truncation conditions. A new kind of window called Nuttall Self-Convolution Window (D-Nuttall Window for short) developed by convolving Nuttall Window is presented in this paper. D-Nuttall Window time domain and frequency domain functions are built, an approach for tanδ measurement based on D-Nuttall Window phase difference correction is proposed. The formulas of fundamental initial phase and tanδ are deduced, and the hardware implementation for the measurement system based on this approach is described. D-Nuttall Window with simple structure has smaller side-lobe level and quicker fall off rate than traditional combined cosine windows, and it is unnecessary to solve the high-order equations for spectrum phase difference correction algorithm, which is easy for embedded system realization. The simulation and experiment test results show that D-Nuttall Window function can well suppress spectral leakage, and the influences on tanδ measurement caused by harmonic disturbance, fundamental frequency fluctuation, change in sampling time length and white noise are all restricted by the approach presented in this paper. This approach can be designed and realized easily and has precise and steady measurement results, which can be used in tanδ offline measurement and online monitoring.