Method and system for processing abnormal data of operation index data of power distribution network

[0027]Example one

[0028]This embodiment provides a method for processing abnormal data of distribution network operation index data, which includes the following steps:

[0029]Step 1: Obtain real-time operation index data of the distribution network;

[0030]There are n running index data, each data has m dimensions, S(t0) Collect data samples at the current time, and at different times tkThe collected distribution network operation index data sample is S(tk).

[0031]Step 2: Cluster the real-time operation index data at the current moment, and divide the clusters into normal clusters, suspected clusters, and abnormal clusters according to the cluster density;

[0032]The step 1 specifically includes:

[0033](1) Calculate the correlation coefficient between each data point in data set D and the remaining n-1 data;

[0034]Assumption Xi, Xj∈D, correlation coefficientThe correlation coefficient can be used to calculate the correlation between any two distribution network operation index data.

[0035]among

[0060]Example two

[0061]The purpose of this embodiment is to provide a data processing system for abnormal operation index data of a distribution network, including:

[0062]Data acquisition module to acquire real-time operation index data of the distribution network;

[0063]The data clustering module clusters the current real-time operation index data, and divides the clusters into normal clusters, suspected clusters, and abnormal clusters according to the cluster density;

[0064]Anomaly degree calculation module, which calculates the anomaly degree of data at the current moment according to Manhattan distance;

[0065]The abnormal data recognition module, according to the data belongs to the cluster and abnormality, to identify abnormal data.

[0066]Further, it also includes an abnormal data repair module, which repairs abnormal data according to normal data of neighbors.

[0067]Example three

[0068]The purpose of this embodiment is to provide an electronic device.

[0069]An electronic device includes a memory, a processor, and a computer program stored on the memory and capable of running on the processor. When the processor executes the program, the following steps are implemented, including:

[0070]Obtain real-time operation index data of the distribution network;

[0071]Cluster the current real-time operation index data, and divide the clusters into normal clusters, suspected clusters, and abnormal clusters according to the cluster density;

[0072]Calculate the anomaly degree of the data at the current moment based on Manhattan distance;

[0073]Identify abnormal data according to the cluster and degree of abnormality of the data.