论文部分内容阅读
为了满足自适应滤波的准确性和时变噪声鲁棒性的要求,将具有递推特性的Kalman滤波理论引入到自适应滤波中,针对Kalman自适应滤波可能出现的由于计算舍入误差所引起的发散和时变噪声统计问题,提出了具有时变噪声统计的平方根Kalman自适应滤波。目前,应用平方根Kalman自适应滤波的自适应光学电流互感器(OCT)经过了严格的检测,稳态精度达到0.2级,暂态情况下的非周期分量电流的最大峰值瞬时值误差小于±1%。自适应OCT已经在河北省保定市某变电站挂网运行25个月。
In order to meet the requirements of adaptive filtering accuracy and time-varying noise robustness, the Kalman filter theory with recursive properties is introduced into adaptive filtering. For Kalman adaptive filtering, which may be caused by the calculation of rounding errors Divergence and time-varying noise statistics, a square root Kalman adaptive filtering with time-varying noise statistics is proposed. At present, the adaptive optical current transformer (OCT) applying square-root Kalman adaptive filtering has undergone rigorous testing and the steady-state accuracy reaches 0.2 level. The maximum peak instantaneous error of the non-periodic component current under transient conditions is less than ± 1% . Adaptive OCT has been in operation in a substation in Baoding, Hebei Province for 25 months.