Power abnormal data detection method

Abstract

The invention discloses a power abnormal data detection method, relates to the field of detection, and aims to solve the problems of poor detection effect and high error rate of detecting abnormal data through fusion analysis of mass power data in the prior art. The method comprises the following steps: step 1, collecting operation data of a power system, setting the size of a sliding window and a sliding step length, and dividing the operation data of the power system to obtain a plurality of sliding windows; 2, comparing the operation data of the power system in every two adjacent sliding windows, and detecting an abnormal sliding window from the operation data; step 3, performing normalization processing on the operation data of the power system in the abnormal sliding window to obtain normalized data; and step 4, calculating the similarity between each normalized data and the normal power data by using a similarity detection method, and finding out the normalized data corresponding to the similarity exceeding a set similarity range from the normalized data as abnormal data. The method is used for detecting abnormal data.

Description

Technical Field

[0001] The invention relates to the detection field, and belongs to data detection. Background Art

[0002] Initially, the detection method for abnormal power data was manual detection, with specialized personnel checking and auditing the abnormal power data one by one to detect abnormal data in the data. However, with the gradual expansion of the scale of the power system, the amount of data generated during the operation of the power system is very large, and the workload of abnormal power data audit detection is large. Existing methods extract abnormal features of power data through fusion analysis of massive power data, and detect abnormal data through audit. However, the abnormal feature extraction effect of power data in the existing detection method is poor, which often leads to false detection in practice, and the hit rate is low, which cannot achieve the expected detection effect. Summary of the invention

[0003] The purpose of the present invention is to solve the problem that the existing abnormal data detection by fusion analysis of massive power data has poor detection effect and high error rate, and proposes a method for detecting abnormal power data.

[0004] A method for detecting abnormal power data, the method comprising the following contents:

[0005] Step 1: Collect power system operation data, set a sliding window size and a sliding step size to divide the power system operation data, and obtain a plurality of sliding windows;

[0006] Step 2: comparing the power system operation data in each two adjacent sliding windows to detect abnormal sliding windows;

[0007] Step 3: normalize the power system operation data within the abnormal sliding window to obtain normalized data;

[0008] Step 4: Calculate the similarity between each normalized data and the normal power data by using the similarity detection method, and find out the normalized data corresponding to the similarity exceeding the set similarity range as abnormal data.

[0009] Preferably, in step 1, the power system operation data in each two adjacent sliding windows are compared, specifically:

[0010] It is determined whether the average value of the power system operation data in each two adjacent sliding windows is less than a preset difference. If yes, it is determined to be a normal sliding window; if not, it is determined to be an abnormal sliding window.

[0011] Preferably, the normalized data y(x) * It is expressed as:

[0012]

[0013] In the formula, y(x) is the power data, y(x) min is the minimum value of a certain type of power data, y(x) max It is the maximum value of a certain type of power data.

[0014] Preferably, in step 4, the similarity r between each normalized data and the normal power data is expressed as:

[0015]

[0016] In the formula, ω s is the normalized data, Normal power data.

[0017] The beneficial effects of the present invention are:

[0018] The present invention uses a simple detection method to accurately and quickly detect abnormal data, thereby improving the operation stability of the power system and the power data transmission capability. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 The figure is a flow chart of a method for detecting abnormal power data. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0022] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0023] Example:

[0024] Combination Figure 1 The present embodiment describes a method for detecting abnormal power data, the method comprising the following contents:

[0025] Step 1: Collect power system operation data, set a sliding window size and a sliding step size to divide the power system operation data, and obtain a plurality of sliding windows;

[0026] Step 2: comparing the power system operation data in each two adjacent sliding windows to detect abnormal sliding windows;

[0027] Step 3: normalize the power system operation data within the abnormal sliding window to obtain normalized data;

[0028] Step 4: Calculate the similarity between each normalized data and the normal power data by using the similarity detection method, and find out the normalized data corresponding to the similarity exceeding the set similarity range as abnormal data.

[0029] Specifically, in order to facilitate the rapid collection of power system operation data, a collection model may be used.

[0030] The following further describes the comparison of the power system operation data in each two adjacent sliding windows in step 1, specifically:

[0031] It is determined whether the average value of the power system operation data in each two adjacent sliding windows is less than a preset difference. If yes, it is determined to be a normal sliding window; if not, it is determined to be an abnormal sliding window.

[0032] The normalized data y(x) is further introduced below: * :

[0033]

[0034] In the formula, y(x) is the power data, y(x) min is the minimum value of a certain type of power data, y(x) max It is the maximum value of a certain type of power data.

[0035] Specifically, considering that there are many types of collected power data, such as current, voltage, capacitance, power, temperature, etc., the power data attribute value has a greater impact on data clustering analysis. Therefore, the power data is normalized and the power data value is normalized to between 0 and 1.

[0036] The following is the similarity r between each normalized data and the normal power data:

[0037]

[0038] In the formula, ω s is the normalized data, Normal power data.

[0039] Specifically, Indicates power data under normal conditions. The larger the similarity value, the more similar the power data is to normal data. A similarity threshold is set according to the actual situation. If the value calculated by Formula 2 is less than or equal to the threshold, the audit result is that the power data is normal; if the value calculated by Formula 2 is greater than the threshold, the audit result is that the power data is abnormal. The normal data detected is represented by "0", and the abnormal data detected is represented by "1". The power data is classified according to the above rules to identify abnormal data.

[0040] Although the present invention is described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the present invention. It should therefore be understood that many modifications may be made to the exemplary embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the various dependent claims and features described herein may be combined in a manner different from that described in the original claims. It should also be understood that features described in conjunction with individual embodiments may be used in other described embodiments.

Claims

1. A method for detecting abnormal power data, characterized in that: The method comprises the following: Step 1: Collect power system operation data, set a sliding window size and a sliding step size to divide the power system operation data, and obtain a plurality of sliding windows; Step 2: comparing the power system operation data in each two adjacent sliding windows to detect abnormal sliding windows; Step 3: normalize the power system operation data within the abnormal sliding window to obtain normalized data; Step 4: Calculate the similarity between each normalized data and the normal power data by using the similarity detection method, and find out the normalized data corresponding to the similarity exceeding the set similarity range as abnormal data.

2. The power abnormality data detection method according to claim 1, characterized in that: In step 1, the power system operation data in each two adjacent sliding windows are compared, specifically: It is determined whether the average value of the power system operation data in each two adjacent sliding windows is less than a preset difference. If yes, it is determined to be a normal sliding window; if not, it is determined to be an abnormal sliding window.

3. The power abnormality data detection method according to claim 1, characterized in that: Normalized data y(x) * It is expressed as: In the formula, y(x) is the power data, y(x) min is the minimum value of a certain type of power data, y(x) max It is the maximum value of a certain type of power data.

4. The power abnormality data detection method according to claim 1, characterized in that: In step 4, the similarity r between each normalized data and the normal power data is expressed as: In the formula, ω s is the normalized data, Normal power data.

Images