
浏览全部资源
扫码关注微信
数据采样异常是引起电力系统误操作的主要原因,而在短数据窗下较难区分电气量的异常采样值和故障时的正常采样值。首先分析了电力系统正常运行及故障下电流、电压波形的特点,指出波形除若干间断点外在其他任意点处连续可导且导数同样分段连续,可利用这一特性判断采样值是否异常。因采样数据采样步长相对固定,波形在某点连续所要求的导数存在性,可通过与其前后相邻两点的差分获得。进一步在设定合适的差分误差门槛基础上,判断采样值是否异常,简称为采样值的3点连续有效判别法。给出了差分误差门槛的具体确定方法,分析了该方法应用中的几个具体问题,并用仿真验证了方法的可行性。
Abnormal sampling data is a main cause that results in mal-operation in power system. It is hard to distinguish abnormal sampling data from normal sampling data in fault circumstance under short sampling window. The character of electric current and voltage curves under normal and abnormal operation is firstly analyzed. It points out that the curves are continuously derivable except for several discontinuous points and the derivative is piecewise continuous. This character can be used to judge whether the sampling data is valid or not. As the sampling step is relatively fixed, the derivability can be judged by the backward and forward differences of continuous sampling data. Furthermore, sampling data are judged by setting reasonable differential error threshold. Then the method is briefly called sampling data judging method with three continuous valid data. The solving method for differential error threshold and analysis of several practical problems are given. Simulation results prove the feasibility of this method.
0
浏览量
2155
下载量
9
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621