空间矢量法自适应检测基波正负序及谐波分量

为了满足系统各分量的实时检测,结合空间矢量概念与自适应干扰对消原理提出了适用于三相系统的基波正、负序分量和谐波分量的自适应检测方法,并详细地分析了检测电路的频率特性。结果表明,所提出的自适应检测方法对于基波检测回路相当于一个理想的二阶带通滤波器,对于谐波检测回路相当于一个理想的二阶陷波器。当被检测信号发生严重畸变时,为克服高次特征谐波对检测精度的影响,采用了在积分器前串接低通滤波器的方法来减小谐波引起的权值波动。用软件PSCAD/EMTDC仿真带不对称负荷的三相电压型逆变器的结果证明了所提检测方案正确有效。

Adaptive Detecting Method for Fundamental Positive Sequence, Negative Sequence Components and Harmonic Components Based on Space Vector

For all kinds of FACTS equipments based on Voltage Source Converter and PWM converters, in order to decrease the over-current and the ripple of the voltage on dc side generated by unbalanced voltage of AC system, it is usually necessary to detect the negative and positive sequency components of the ac voltage and current in real time. And in order to economically and efficiently compensate the negative sequence and fundamental reactive components and harmonics in power systems, it also needs real-time detections of each component. In this paper, based on the concept of space vector and the theory of adaptive interference canceling, an adaptive detecting method for the fundamental positive sequence component, fundamental negative sequence component and harmonic component is proposed. The proposed detecting circuit is a continuously regulated closed loop system and its operating characteristics are independent of the parameters of the circuit elements provided that the signal does not exceed its linear dynamic range. In addition, the spectrum characteristic of the detecting circuit has been analyzed in detail, and the results show that with the proposed adaptive detecting method, the fundamental component-detecting loop is equivalent to an ideal band-pass filter, while the harmonic detecting loop is equivalent to an ideal notch-filter. When the detected signal is seriously distorted, low pass filters are adopted in the detecting circuit to reduce the ripple of weight value, so that higher detecting precise is reached. Finally, with PSCAD/EMTDC, simulations on three-phase voltage source inverter with unbalanced load prove that the proposed method can overcomes the limitations of the conventional detecting methods and has a better performance.