基于RLS的自适应滤波算法研究

Adaptive Filtering Algorithm Based on RLS Algorithm

为抑制谐振接地系统对地电容中的工频信号和其他谐波成分,提出了一种基于RLS算法的自适应滤波算法。该算法通过给定n-1次迭代滤波器抽头权向量最小二乘估计,依据新到达的数据计算n次迭代权向量最新估计。当滤波器输入信号变化时可保持最佳输出信号,在一定程度上补偿了滤波器元件参数变化带来的运算误差。对其滤波理论及算法进行仿真研究。结果表明:该算法提高了滤波的稳定性、快速性和精确性,优于其他自适应算法。

To suppress the power frequency signal and other harmonic components in earth capacitance of resonant grounded system, an adaptive filtering algorithm is proposed based on RLS （recursive least square） algorithm. The algorithm given by n-1 iterations filter tap weight vector least squares estimation, based on the data of new arrivals n iterations of the latest estimated vector. When the filter input signal changes, it can remain the best output signal, which compensates the computing error caused by change of filter element parameters to a certain extent. Its filtering theory and algorithm are simulated. The study results show that the stability, speed and accuracy are improved, and this algorithm is superior to other adaptive algorithm.