Newton迭代算法用于高精度故障数据分析

Precise Analysis Method for Fault Data Using the Newton Iterative Algorithm

为了提高采集电力故障的数字信号处理与计算的精度,解决系统采样数据同时受多个误差影响时难以得到准确结果的问题,分析了故障录波采样数据的特点,建立了故障信号在同时受衰减直流分量和非同步采样影响时的暂态信号模型;采用Newton迭代法求解该模型,得到误差参数的精确数值解,并求得故障信号的工频分量及各次谐波幅值和相位的高精度解。仿真实验表明,当较准确地估计迭代初始值时,该法收敛快、精度高,特别适合于要求对电力系统故障进行精确分析的场合。

With the widespread use of digital fault recorders （DFR） in power systems, the requirement of calculating and handling with the fault data becomes higher and higher. Existed fault data analyses methods always face to only one error factor, which make it difficult to get precise result when data sampled from the power system are simultaneously influenced by several error factors. Aiming at this problem, this paper analyzes the characteristics of the fault data from DFRs and discusses the phenomenon of spectrum leakage when using the digital Fourier algorithm. According to the typical expression of fault signal in power systems, a transient signal model represented by a set of nonlinear equations of fault data is built up in this paper, which involves the decaying DC component, asynchronous sampling parameter as well as the harmonics. Next, a method based on Newton iterative algorithm is illustrated, including how to get valid initial values estimation for iterative steps. Then, the method is used to get the numerical solutions of the nonlinear equations to acquire the accurate error parameters and the power system parameters. At last, some examples are given to demonstrate the feasibility and validity of the proposed method. As long as the initial values are estimated suitably, this method can get the power system parameters such as amplitudes, phases, work frequency and harmonics with high precision and convergence rate. With notable accurate parameter measurement, it makes a solid foundation of the power system＇s further supervision, measurement and control, and it is especially suitable for the occasions on which the accurate power system fault analysis is required.