基于卡尔曼滤波和加窗插值谐波分析法的介损测量方法

Dielectric Loss Measurement Based on Kalman Filter and Interpolating Windowed Fast Fourier Transform Algorithm

加窗插值的谐波分析法可减轻非同步采样对介质损耗角测量的影响,是一种非常有效的介质损耗角计算方法。但在有噪声干扰、特别是在信噪比较低的情况下,用这种方法进行介质损耗角测量时产生的误差足以掩盖真实值。为此作者提出先将信号进行卡尔曼滤波处理,再使用加窗插值的谐波分析法计算的介质损耗角测量方法。仿真结果表明,与原有算法相比,所提出的方法能明显提高测量准确度。

As a very effective computing method for dielectric loss angle, the windowed interpolation FFT algorithm can mitigate the affect of non-synchronous sampling on the measurement of dielectric loss angle. However, in the environment of noise interference, especially under lower singal to noise ratio （SNR） the measuring error produced during the measurement of dielectric loss angle by this method would be so high that the ture value of dielectric loss angle might be concealed. For this reason, the authors propose that at first the measured signal is filtered by Kalman filter, then dielectric loss angle is calculated by windowed interpolation FFT algorithm. Simulation results show that comparing with original windowed interpolation FFT algorithm, the proposed method can improve measuring accuracy evidently.