CN116298698A - Method and system for detecting data abnormality of voltage monitoring points of power distribution network - Google Patents

Abstract

The invention relates to the technical field of data anomaly detection, in particular to a method and a system for detecting data anomalies of power distribution network voltage monitoring points. The method comprises the following steps: acquiring voltage data of voltage monitoring points of the power distribution network according to a preset sampling frequency; eliminating zero value data and null value data in the voltage data set to obtain a first data set to be detected; performing out-of-limit detection on the first data set to be detected to obtain an out-of-limit value data set; performing first-level detection on the threshold-crossing data set according to a quartile data interval criterion, and removing data with abnormal detection results in the first-level detection to obtain a second data set to be detected; performing second-level detection on a second data set to be detected according to a median deviation method to obtain a first abnormal data set; according to the method, the interference items can be removed in a two-layer identification mode, so that the identification result is more accurate, and the problem that the result of the existing data anomaly detection method for the power distribution network voltage monitoring points is inaccurate is effectively solved.

Description

Method and system for detecting data abnormality of voltage monitoring points of power distribution network

Technical Field

The invention relates to the technical field of data anomaly detection, in particular to data anomaly detection of a power distribution network voltage monitoring point.

Background

The power distribution network voltage monitoring point refers to a node for monitoring the voltage value and checking the voltage quality of the power distribution network, and generally represents the general level of the power distribution network voltage; with the continuous improvement of the digital and intelligent level of the power grid, the data of the bus bars, the feeder lines, the distribution transformers and the user voltages of the transformer substations monitored at the voltage monitoring points of the power distribution network become more and more abundant.

However, in a real situation, when the data is transmitted to an automatic measurement and background management system, the voltage monitoring terminals can cause abnormal conditions such as missing values, null values, outliers and the like of the finally collected voltage data due to the influences of equipment or network faults, abnormal events and human factors; therefore, abnormal value detection of the voltage monitoring points of the power distribution network becomes an indispensable link.

At present, although the abnormal value detection methods of the voltage monitoring points of the power distribution network are various, due to the fact that the abnormal value is more in variety, and when different abnormal values are detected, small abnormal data can not be effectively detected due to the fact that large abnormal value data exist, large errors exist in detection results generally, and accurate and effective judgment basis is difficult to provide for occurrence of abnormal conditions of the voltage monitoring points.

Disclosure of Invention

The invention provides a method and a system for detecting data anomalies of power distribution network voltage monitoring points, which are used for solving the problem that the results of the existing method for detecting the data anomalies of the power distribution network voltage monitoring points are inaccurate.

The invention provides a data anomaly detection method for a voltage monitoring point of a power distribution network, which is characterized by comprising the following steps:

acquiring voltage data of voltage monitoring points of the power distribution network according to a preset sampling frequency to obtain a voltage data set;

eliminating zero value data and null value data in the voltage data set to obtain a first data set to be detected;

performing out-of-limit detection on the first data set to be detected to obtain an out-of-limit value data set;

performing first-level detection on the threshold-crossing value data set according to a quartile data interval criterion, and removing data with abnormal detection results in the first-level detection to obtain a second data set to be detected;

and carrying out second-level detection on the second data set to be detected according to a median deviation method to obtain a first abnormal data set.

Specifically, the threshold crossing detection is performed on the first to-be-detected data set to obtain a threshold crossing data set, which specifically includes:

acquiring a preset upper limit percentage of voltage management, a preset lower limit percentage of voltage management and a voltage class of a power distribution network;

calculating to obtain a voltage management upper limit according to the voltage management upper limit percentage and the power distribution network voltage level;

calculating to obtain a lower voltage management limit according to the lower voltage management limit percentage and the voltage class of the power distribution network;

and judging the data which is larger than the upper voltage management limit or smaller than the lower voltage management limit in the first data set to be detected as the out-of-limit value data, and collecting all the out-of-limit value data to obtain the out-of-limit value data set.

Specifically, the first level detection is performed on the threshold-crossing data set according to a quartile data interval criterion, specifically:

acquiring a preset upper limit abnormal coefficient and a preset lower limit abnormal coefficient;

quartering the data in the threshold crossing data set, and selecting first quartile data and third quartile data;

determining upper limit abnormal data according to the first quartile data, the third quartile data and the upper limit abnormal coefficient;

determining lower limit anomaly data according to the first quartile data, the third quartile data and the lower limit anomaly coefficient;

and marking the upper limit abnormal data and the lower limit abnormal data as abnormal data of a detection result.

Specifically, the second level detection is performed on the second data set to be detected according to the median deviation method, and specifically includes the following steps:

s1: acquiring a preset first abnormality judgment coefficient, a preset second abnormality judgment coefficient and a preset proportionality factor constant;

s2: starting from the first data of the data to be detected, setting a sliding window with a preset data length;

s3: sequentially arranging the data in the sliding window from small to large, and determining the median of the arranged data in the sliding window;

s4: calculating absolute intermediate difference data according to the intermediate number of the data in the sliding window and the data in the sliding window;

s5: arranging the absolute median difference data from small to large, and determining the median of the arranged absolute median difference data;

s6: determining window abnormal data in the sliding window according to the median of the data in the sliding window, the first abnormal judgment coefficient, the second abnormal judgment coefficient, the scale factor constant and the median of the absolute median difference data;

s7: and sliding the sliding window by a preset data length and repeating the steps S3-S6 until the detection of all data in the second data set to be detected is completed.

Specifically, a method for detecting abnormal data of a voltage monitoring point of a power distribution network is characterized in that the abnormal data of the window in the sliding window is determined according to the median of the data in the sliding window, the first abnormal judgment coefficient, the second abnormal judgment coefficient, the proportionality factor constant and the median of the absolute median difference data, specifically:

determining first window abnormal data in the sliding window according to the median of the data in the sliding window, the first abnormal judgment coefficient, the scale factor constant and the median of the absolute median difference data;

determining second window abnormal data in the sliding window according to the median of the data in the sliding window, the second abnormal judgment coefficient, the scale factor constant and the median of the absolute median difference data;

and determining the first window abnormal data and the second window abnormal data as the window abnormal data.

Another aspect of the present invention provides a system for detecting data anomalies at voltage monitoring points of a power distribution network, including: sampling module, first data set acquisition module, detection module that crosses limit, first level detection module and second level detection module, wherein:

the sampling module is used for acquiring voltage data of voltage monitoring points of the power distribution network according to a preset sampling frequency to obtain a voltage data set;

the first data set to be detected acquisition module is connected with the sampling module and is used for eliminating zero value data and null value data in the voltage data set to obtain a first data set to be detected;

the out-of-limit detection module is connected with the first to-be-detected data set acquisition module, and out-of-limit detection is carried out on the first to-be-detected data set to obtain an out-of-limit value data set;

the first level detection module is connected with the out-of-limit detection module and is used for carrying out first level detection on the out-of-limit value data set according to a quartile data interval criterion, and eliminating data with abnormal detection results in the first level detection to obtain a second data set to be detected;

the second level detection module is connected with the first level detection module and is used for carrying out second level detection on the second data set to be detected according to a median deviation method to obtain a first abnormal data set.

Specifically, the out-of-limit detection module includes: the device comprises a first acquisition unit, an upper limit calculation unit, a lower limit calculation unit and a first judgment unit;

the first acquisition unit is used for acquiring a preset upper limit percentage of voltage management, a preset lower limit percentage of voltage management and a voltage class of the power distribution network;

the upper limit calculation unit is connected with the first acquisition unit and is used for calculating the upper limit of the voltage management according to the upper limit percentage of the voltage management and the voltage class of the power distribution network;

the lower limit calculation unit is connected with the first acquisition unit and is used for calculating a voltage management lower limit according to the voltage management lower limit percentage and the power distribution network voltage level;

the first judging unit is respectively connected with the upper limit calculating unit and the lower limit calculating unit, and is used for judging the data which is larger than the upper limit of the voltage management or smaller than the lower limit of the voltage management in the first data set to be detected as the out-of-limit value data, and acquiring the out-of-limit value data set from all the out-of-limit value data sets.

Specifically, the first level detection module includes: an abnormal coefficient acquisition unit, a quartile selection unit, an out-of-limit abnormal data determination unit and a marking unit;

the anomaly coefficient setting unit is used for acquiring an anomaly coefficient beyond the upper limit and an anomaly coefficient beyond the lower limit;

the quartile selection unit is used for quartering the data in the threshold value crossing data set and selecting first quartile data and third quartile data;

the out-of-limit abnormal data determining unit is connected with the abnormal coefficient setting unit and the quartile selecting unit, and is used for determining upper-limit abnormal data according to the first quartile data, the third quartile data and the upper-limit abnormal coefficient, and is also used for determining lower-limit abnormal data according to the first quartile data, the third quartile data and the lower-limit abnormal coefficient;

the marking unit is connected with the out-of-limit abnormal data determining unit and is used for marking the out-of-limit abnormal data and the out-of-limit abnormal data as abnormal data of a detection result.

Specifically, the second-level detection module includes: the device comprises a second acquisition unit, a sliding window setting unit, a first median determining unit, an absolute median calculating unit, a second median determining unit, an abnormal data determining unit and a sliding unit;

the second acquisition unit is used for acquiring a preset first abnormality judgment coefficient, the second abnormality judgment coefficient and a preset scale factor constant;

the sliding window setting unit is used for setting a sliding window with preset data length from the first data of the data to be detected;

the first median determining unit is connected with the sliding window setting unit and is used for sequentially arranging the data in the sliding window from small to large and determining the median of the arranged data in the sliding window;

the absolute median difference calculating unit is connected with the first median determining unit and is used for calculating absolute median difference data according to the median of the data in the sliding window and the data in the sliding window;

the second median determining unit is connected with the absolute median calculating unit and is used for arranging the absolute median data from small to large and determining the median of the arranged absolute median data;

the abnormal data determining unit is respectively connected with the first median determining unit, the coefficient and constant adding unit and the second median determining unit and is used for determining window abnormal data in the sliding window according to the median of the data in the sliding window, the first abnormal judgment coefficient, the second abnormal judgment coefficient, the scale factor constant and the median of the absolute median difference data;

the sliding unit is connected with the abnormal data determining unit and is used for sliding the sliding window by a preset data length and repeating the steps S3-S6 until detection of all data in the second data set to be detected is completed.

Specifically, the abnormal data determining unit includes: the system comprises a first abnormal data determining subunit, a second abnormal data determining subunit and a summarizing subunit, wherein:

the first abnormal data determining subunit is configured to determine first window abnormal data in the sliding window according to a median of data in the sliding window, the first abnormal judgment coefficient, the scale factor constant, and a median of the absolute median difference data;

the second abnormal data determining subunit is configured to determine second window abnormal data in the sliding window according to a median of data in the sliding window, the second abnormal judgment coefficient, the scale factor constant, and a median of the absolute median difference data;

the summarizing subunit is configured to determine the first window abnormal data and the second window abnormal data as the window abnormal data.

The method for detecting the data abnormality of the power distribution network voltage monitoring points has the beneficial effects that the method for detecting the data abnormality of the power distribution network voltage monitoring points provided by the embodiment of the invention comprises the following steps: acquiring voltage data of voltage monitoring points of the power distribution network according to a preset sampling frequency to obtain a voltage data set; eliminating zero value data and null value data in the voltage data set to obtain a first data set to be detected; performing out-of-limit detection on the first data set to be detected to obtain an out-of-limit value data set; performing first-level detection on the threshold-crossing value data set according to a quartile data interval criterion, and removing data with abnormal detection results in the first-level detection to obtain a second data set to be detected; and carrying out second-level detection on the second data set to be detected according to a median deviation method to obtain a first abnormal data set.

According to the data anomaly detection method for the power distribution network voltage monitoring points, interference items such as zero values and null value anomaly data can be removed, out-of-limit value data sets are obtained, first-level detection is carried out on the out-of-limit value data sets, when the first-level detection is carried out, the characteristic that discrete values are easily distinguished by using a quartile data interval criterion is utilized, anomaly data with larger dispersion are screened out firstly, contrast between second anomaly data with smaller dispersion and normal data which are more difficult to distinguish from normal values is increased, interference items are further reduced, and then biased data are detected from normal data one by one through a median deviation method, so that a first anomaly data set is obtained; different identification modes are adopted for different abnormal value types through a two-layer identification mode, abnormal value data of smaller abnormal values are effectively detected, detection accuracy is further improved, and the problem that the detection result of the existing method for detecting abnormal data of the power distribution network voltage monitoring points is inaccurate is effectively solved.

Drawings

In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.

Fig. 1 is a flowchart of a method for detecting data anomalies at voltage monitoring points of a power distribution network.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

The invention provides a method for detecting data anomalies of power distribution network voltage monitoring points, referring to fig. 1, fig. 1 is a flow chart of a method for detecting data anomalies of power distribution network voltage monitoring points, which specifically comprises the following steps:

acquiring voltage data of voltage monitoring points of power distribution network according to preset sampling frequency

Obtaining a voltage data set->

Eliminating zero value data and null value data in the voltage data set to obtain a first data set to be detected;

out-of-limit detection is carried out on the first data set to be detected to obtain an out-of-limit value data set omega ^yx ；

For threshold-crossing data sets omega according to the quartile data spacing criterion (IQR) ^yx Performing first-level detection, and removing data with abnormal detection results in the first-level detection to obtain a second data set to be detected;

performing second-level detection on the second data set to be detected according to a median deviation method to obtain a first abnormal data set omega ^anomaly_TL 。

In the present embodiment, the data of the detection result abnormality in the first level detection is the highly abnormal data Ω ^anomaly-H First abnormal data set omega ^anomaly_TL In order to obtain middle-low abnormal data, the height abnormal data is larger than the middle-low abnormal data in the abnormal value, so that when detection is carried out, the characteristic that discrete values are easily distinguished by using a quartile data interval criterion is firstly utilized, the height abnormal value is screened out from out-of-limit data, the difference between the rest middle-low abnormal data and normal data is more obvious, the middle-low abnormal data is easily distinguished from the normal data, and then the middle-low abnormal data and the normal data are distinguished by using a data statistics-based median deviation method, so that the voltage monitoring of the current power distribution network is effectively solvedAnd the result of the data anomaly detection method of the measuring point is inaccurate.

In a specific embodiment of the present invention, based on the foregoing embodiment, the out-of-limit detection is performed on the first to-be-detected data set to obtain an out-of-limit value data set Ω ^yx The formula is as follows:

wherein: x is X ₁ To preset the upper voltage limit, X ₂ A preset voltage lower limit;

in a more specific embodiment:

wherein, the liquid crystal display device comprises a liquid crystal display device,

the conventional rated voltage levels of the power distribution network are represented, wherein the conventional rated voltage levels comprise a lower management limit coefficient percentage and an upper management limit coefficient of 110kV, 35kV, 10kV, 380V/220V, and lambda% and gamma% respectively correspond to the voltage levels.

In a particular embodiment of the invention, a voltage dataset is determined

Voltage data in (a)

Whether or not to meet->

If yes, will->

Recorded as zero value data omega ^zero ；

Determining a voltage dataset

Voltage data>

Whether or not to meet->

If yes, will->

Outputting the null data omega one by one ^null 。

In a more specific embodiment of the present invention, based on the previous embodiment, the threshold-crossing data set Ω is compared according to a quartile data spacing criterion ^yx Performing first-level detection, specifically:

acquiring a preset upper limit height anomaly coefficient b _up And a preset lower limit height abnormality coefficient b _low ；

The threshold value data set omega will be exceeded ^yx Quarter of the data in the data block, and selecting the first quartile data Q ₁ And third quartile data Q ₃ ；

According to the first quartile data Q ₁ Third quartile data Q ₃ Higher limit height anomaly coefficient b _up A first preset threshold X ₁ Determining an upper limit height outlier;

according to the first quartile data Q ₁ Third quartile data Q ₃ Higher limit height anomaly coefficient b _low A second preset threshold X ₂ Determining a lower limit height anomaly value according to the following formula:

the upper limit height abnormal value and the lower limit height abnormal value are marked as data of abnormal detection results.

In a more specific embodiment of the present invention, based on the foregoing embodiment, the second level detection is performed on the second data set to be detected according to a median deviation method, and specifically includes the following steps:

s1: acquiring a preset first abnormality determination coefficient beta ₁ Second abnormality determination coefficient beta ₂ A preset proportionality factor constant K;

s2: from the data to be detected Ω ^T The first data of the preset data length W is set, and the sliding window W with the preset data length W is set as follows:

wherein: w-k represents the first bit of the data, and w+k represents the last bit of the data;

s3: sequentially arranging the data in the sliding window from small to large to obtain ordered data

Determining data within a sliding window

Is>

Wherein: media () is the median;

s4: based on data within sliding window

Is>

Data in sliding window +.>

Calculating absolute medium bit difference data, wherein the formula is as follows;

wherein DeltaY is absolute medium bit difference data, Y _W Is the last bit data of the absolute medium bit difference data deltay,

for data in sliding window->

First data of->

For data +.>

Last data of (2);

wherein: w (W) _p Representing the location of the window in which it is located;

s5: arranging the data in the absolute medium-bit difference data set delta Y from small to large to obtain an arranged absolute medium-bit difference data set

Determining the median of the absolute median difference dataset deltay:

s6: based on the median of the data in the sliding window

First abnormality determination coefficient beta ₁ Second abnormality determination coefficient beta ₂ A scale factor constant K, a median of the absolute median difference data +.>

Determining window abnormality data within a sliding window, i.e. when +.>

When the method is used, the following steps are:

wherein: isoutlier () is outlier detection, β=β ₁ Or β=β ₂ ；

S7: sliding the sliding window by a preset data length w and repeating the steps S3-S6 until the data omega to be detected is completed ^T All data of the data set.

In a more specific embodiment of the present invention, the first anomaly determination coefficient β is based on the median of the data in the sliding window ₁ The second abnormality determination coefficient beta ₂ A median of the scale factor constant K and the absolute median difference data

The window abnormal data in the sliding window is determined, specifically:

based on the median of the data in the sliding window

The first abnormality determination coefficient beta ₁ The median of the scale factor constant K and the absolute median difference data +.>

Determining first window anomaly data within the sliding window;

based on the median of the data in the sliding window

The second abnormality determination coefficient beta ₂ The scaling factor constant KThe median of the absolute median difference data +.>

Determining second window anomaly data within the sliding window;

and determining the first window abnormal data and the second window abnormal data as the window abnormal data.

In another specific embodiment of the present invention, the first abnormality determination coefficient β ₁ And the second abnormality determination coefficient beta ₂ The intervals are preset.

In the specific implementation process, the first abnormal data omega is obtained by setting different abnormal judgment coefficients ^anomaly_TL Distinguishing into moderately abnormal data sets omega ^anomaly_T And a low anomaly dataset Ω ^anomaly_L When the first abnormality determination coefficient beta ₁ Is larger than the second abnormality determination coefficient beta ₂ In the time, the coefficient beta is judged according to the first abnormality ₁ The obtained first window abnormal data set of all windows is a moderate abnormal data set omega ^anomaly_T Determining the coefficient beta according to the first abnormality ₂ The obtained second window abnormal data set of all windows is a low-level abnormal data set omega ^anomaly_L 。

In a more specific embodiment of the present invention, based on the foregoing embodiment, the first abnormality determination coefficient β ₁ Is 3, the second abnormality determination coefficient beta ₂ 2.

The invention also provides a data anomaly detection system of the voltage monitoring point of the double-layer progressive power distribution network, which comprises the following steps: sampling module, first data set acquisition module, detection module that crosses limit, first level detection module and second level detection module, wherein:

the sampling module is used for acquiring voltage data of voltage monitoring points of the power distribution network according to a preset sampling frequency to obtain a voltage data set;

the first data set to be detected acquisition module is connected with the sampling module and is used for eliminating zero value data and null value data in the voltage data set to obtain a first data set to be detected;

the out-of-limit detection module is connected with the first to-be-detected data set acquisition module, and out-of-limit detection is carried out on the first to-be-detected data set to obtain an out-of-limit value data set;

the first level detection module is connected with the out-of-limit detection module and is used for carrying out first level detection on the out-of-limit value data set according to a quartile data interval criterion, and eliminating data with abnormal detection results in the first level detection to obtain a second data set to be detected;

the second-level detection module is connected with the first-level detection module and is used for carrying out second-level detection on the second data set to be detected according to a median deviation method to obtain a first abnormal data set.

In a more specific embodiment of the present invention, on the basis of the foregoing embodiment, the out-of-limit detection module includes: the device comprises a first acquisition unit, an upper limit calculation unit, a lower limit calculation unit and a first judgment unit;

the first acquisition unit is used for acquiring a preset upper limit percentage of voltage management, a preset lower limit percentage of voltage management and a voltage class of the power distribution network;

the upper limit calculation unit is connected with the first acquisition unit and is used for calculating the upper limit of the voltage management according to the upper limit percentage of the voltage management and the voltage class of the power distribution network;

the lower limit calculation unit is connected with the first acquisition unit and is used for calculating a voltage management lower limit according to the voltage management lower limit percentage and the power distribution network voltage level;

the first judging unit is respectively connected with the upper limit calculating unit and the lower limit calculating unit, and is used for judging the data which is larger than the upper limit of the voltage management or smaller than the lower limit of the voltage management in the first data set to be detected as the out-of-limit value data, and acquiring the out-of-limit value data set from all the out-of-limit value data sets.

In a more specific embodiment of the present invention, based on the foregoing embodiment, the first level detection module includes: an abnormal coefficient acquisition unit, a quartile selection unit, an out-of-limit abnormal data determination unit and a marking unit;

the anomaly coefficient setting unit is used for acquiring an anomaly coefficient beyond the upper limit and an anomaly coefficient beyond the lower limit;

the quartile selection unit is used for quartering the data in the threshold value crossing data set and selecting first quartile data and third quartile data;

the out-of-limit abnormal data determining unit is connected with the abnormal coefficient setting unit and the quartile selecting unit, and is used for determining upper-limit abnormal data according to the first quartile data, the third quartile data and the upper-limit abnormal coefficient, and is also used for determining lower-limit abnormal data according to the first quartile data, the third quartile data and the lower-limit abnormal coefficient;

the marking unit is connected with the out-of-limit abnormal data determining unit and is used for marking the out-of-limit abnormal data and the out-of-limit abnormal data as abnormal data of a detection result.

In a more specific embodiment of the present invention, based on the foregoing embodiment, the second level detection module includes: the device comprises a second acquisition unit, a sliding window setting unit, a first median determining unit, an absolute median calculating unit, a second median determining unit, an abnormal data determining unit and a sliding unit;

the second acquisition unit is used for acquiring a preset first abnormality judgment coefficient, the second abnormality judgment coefficient and a preset scale factor constant;

the sliding window setting unit is used for setting a sliding window with preset data length from the first data of the data to be detected;

the first median determining unit is connected with the sliding window setting unit and is used for sequentially arranging the data in the sliding window from small to large and determining the median of the arranged data in the sliding window;

the absolute median difference calculating unit is connected with the first median determining unit and is used for calculating absolute median difference data according to the median of the data in the sliding window and the data in the sliding window;

the second median determining unit is connected with the absolute median calculating unit and is used for arranging the absolute median data from small to large and determining the median of the arranged absolute median data;

the abnormal data determining unit is respectively connected with the first median determining unit, the coefficient and constant adding unit and the second median determining unit and is used for determining window abnormal data in the sliding window according to the median of the data in the sliding window, the first abnormal judgment coefficient, the second abnormal judgment coefficient, the scale factor constant and the median of the absolute median difference data;

the sliding unit is connected with the abnormal data determining unit and is used for sliding the sliding window by a preset data length and repeating the steps S3-S6 until detection of all data in the second data set to be detected is completed.

In a more specific embodiment of the present invention, on the basis of the foregoing embodiment, the abnormal data determining unit includes: the system comprises a first abnormal data determining subunit, a second abnormal data determining subunit and a summarizing subunit, wherein:

the first abnormal data determining subunit is configured to determine first window abnormal data in the sliding window according to a median of data in the sliding window, the first abnormal judgment coefficient, the scale factor constant, and a median of the absolute median difference data;

the second abnormal data determining subunit is configured to determine second window abnormal data in the sliding window according to a median of data in the sliding window, the second abnormal judgment coefficient, the scale factor constant, and a median of the absolute median difference data;

the summarizing subunit is configured to determine the first window abnormal data and the second window abnormal data as the window abnormal data.

The invention also provides a specific embodiment:

in an embodiment, the voltage data of the grid voltage monitoring points are shown in table 1;

table 1:24 hours voltage data (unit: V)

From Table 1, it can be seen that there is a zero value in the voltage data

Null value->

According to the quartile data spacing criterion of the first level, the data is {236.55,236.92,237.37,238.3,246.46,526.33} after the out-of-limit data of which the size is more than 220× (1+7%) -235V is ordered;

wherein, the liquid crystal display device comprises a liquid crystal display device,

detecting the data as highly abnormal data according to the IQR;

thereby, the data omega to be detected of the second level is obtained ^T As shown in table 2:

table 2:

228.07	228.24	229.18	226.85	227.91	226.93
						228.43	228.69	226.93	227.44	226.57	226.91
226.36	246.46	237.48	238.3	236.55	227.44
						236.92	268.82	237.37

for simplicity of illustrating the present invention, the sliding window size is taken as w=2k+1=11, and the half window size k=5;

then Ω ^T The data will be detected by two W windows as to whether the anomaly is detected, the data in window 1

For a pair of

The medium elements are ordered from small to large, and the ordered data are shown in table 3:

table 3:

226.57	226.85	226.93	226.93	227.44	227.91
						228.07	228.24	228.43	228.69	229.18

calculation of

The median of (2) is as follows:

calculating absolute medium bit difference data delta Y:

arranging the data in the absolute medium bit difference data delta Y from small to large to obtain the ordered absolute medium bit difference data delta Y ^p ＝{0,0.16,0.33,0.47,0.52,0.78,0.98,0.98,1.06,1.27,1.34}；

Next, the median of the absolute median difference data Δy is calculated as follows:

taking the first abnormality judgment coefficient β1=3, the judgment threshold value of the abnormality of the voltage data in the window W1 is:

it can be seen that {0.16,0.33,1.27,1.06,0,0.98,0.52,0.78,0.98,0.47,1.34} in window W1 is less than 3.4692, i.e., there is no voltage anomaly in window W1.

Similarly, the abnormality detection is performed on the voltage data in the window W2, and the process and the result are as follows:

the voltage data within window W2 is as in table 4:

table 4:

226.91	226.36	246.46	237.48	238.3	236.55
						227.44	236.92	268.82	237.37

the voltage data within window W2 is ordered from small to large as in table 5:

table 5:

226.36	226.91	227.44	236.55	236.92	237.37
						237.48	238.3	246.46	268.82

median of data within window W2: 237.145V;

absolute mid-range data Δy within window W2 is shown in table 6:

table 6:

10.235	10.785	9.315	0.335	1.155	0.595
						9.705	0.225	31.675	0.225

median of absolute median difference data within window W2:

voltage data abnormality discrimination threshold within window W2:

/>

available, 268.82 in window W2 is exception data;

outputting an abnormal detection result of voltage data of the voltage monitoring point:

from this point, the abnormality detection ends.

The terms "first," "second," "third," "fourth," and the like in the description of the present application and in the above-described figures, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that embodiments of the present application described herein may be capable of operation in sequences other than those illustrated or described herein, for example. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

It should be understood that, in the present application, "and/or" is used to describe association relationships of association objects, three relationships may exist, for example, "a and/or B" may denote: only a, only B and both a and B are present, wherein a, B may be singular or plural. The character "/" generally indicates that the context-dependent object is an "or" relationship.

In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The above embodiments are merely for illustrating the technical solution of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.

Claims (10)

1. The data anomaly detection method for the voltage monitoring points of the power distribution network is characterized by comprising the following steps of:

acquiring voltage data of voltage monitoring points of the power distribution network according to a preset sampling frequency to obtain a voltage data set;

eliminating zero value data and null value data in the voltage data set to obtain a first data set to be detected;

performing out-of-limit detection on the first data set to be detected to obtain an out-of-limit value data set;

performing first-level detection on the threshold-crossing value data set according to a quartile data interval criterion, and removing data with abnormal detection results in the first-level detection to obtain a second data set to be detected;

and carrying out second-level detection on the second data set to be detected according to a median deviation method to obtain a first abnormal data set.

2. The method for detecting abnormal data of voltage monitoring points of a power distribution network according to claim 1, wherein the out-of-limit detection is performed on the first data set to be detected to obtain an out-of-limit data set, specifically:

acquiring a preset upper limit percentage of voltage management, a preset lower limit percentage of voltage management and a voltage class of a power distribution network;

calculating to obtain a voltage management upper limit according to the voltage management upper limit percentage and the power distribution network voltage level;

calculating to obtain a lower voltage management limit according to the lower voltage management limit percentage and the voltage class of the power distribution network;

and judging the data which is larger than the upper voltage management limit or smaller than the lower voltage management limit in the first data set to be detected as the out-of-limit value data, and collecting all the out-of-limit value data to obtain the out-of-limit value data set.

3. The method for detecting abnormal data of voltage monitoring points of a power distribution network according to claim 1, wherein the first level detection is performed on the out-of-limit value data set according to a quartile data interval criterion, specifically:

acquiring a preset upper limit abnormal coefficient and a preset lower limit abnormal coefficient;

quartering the data in the threshold crossing data set, and selecting first quartile data and third quartile data;

determining upper limit abnormal data according to the first quartile data, the third quartile data and the upper limit abnormal coefficient;

determining lower limit anomaly data according to the first quartile data, the third quartile data and the lower limit anomaly coefficient;

and marking the upper limit abnormal data and the lower limit abnormal data as abnormal data of a detection result.

4. The method for detecting abnormal data of voltage monitoring points of power distribution network according to claim 1, wherein the second level detection is performed on the second data set to be detected according to a median deviation method, and specifically comprises the following steps:

s1: acquiring a preset first abnormality judgment coefficient, a preset second abnormality judgment coefficient and a preset proportionality factor constant;

s2: starting from the first data of the data to be detected, setting a sliding window with a preset data length;

s3: sequentially arranging the data in the sliding window from small to large, and determining the median of the arranged data in the sliding window;

s4: calculating absolute intermediate difference data according to the intermediate number of the data in the sliding window and the data in the sliding window;

s5: arranging the absolute median difference data from small to large, and determining the median of the arranged absolute median difference data;

s6: determining window abnormal data in the sliding window according to the median of the data in the sliding window, the first abnormal judgment coefficient, the second abnormal judgment coefficient, the scale factor constant and the median of the absolute median difference data;

s7: and sliding the sliding window by a preset data length and repeating the steps S3-S6 until the detection of all data in the second data set to be detected is completed.

5. The method for detecting abnormal data at voltage monitoring points of power distribution network according to claim 4, wherein the determining the abnormal data of the window in the sliding window according to the median of the data in the sliding window, the first abnormal judgment coefficient, the second abnormal judgment coefficient, the scale factor constant and the median of the absolute median difference data comprises:

determining first window abnormal data in the sliding window according to the median of the data in the sliding window, the first abnormal judgment coefficient, the scale factor constant and the median of the absolute median difference data;

determining second window abnormal data in the sliding window according to the median of the data in the sliding window, the second abnormal judgment coefficient, the scale factor constant and the median of the absolute median difference data;

and determining the first window abnormal data and the second window abnormal data as the window abnormal data.

6. The utility model provides a data anomaly detection system of distribution network voltage monitoring point which characterized in that includes: sampling module, first data set acquisition module, detection module that crosses limit, first level detection module and second level detection module, wherein:

the sampling module is used for acquiring voltage data of voltage monitoring points of the power distribution network according to a preset sampling frequency to obtain a voltage data set;

the first data set to be detected acquisition module is connected with the sampling module and is used for eliminating zero value data and null value data in the voltage data set to obtain a first data set to be detected;

the out-of-limit detection module is connected with the first to-be-detected data set acquisition module, and out-of-limit detection is carried out on the first to-be-detected data set to obtain an out-of-limit value data set;

the first level detection module is connected with the out-of-limit detection module and is used for carrying out first level detection on the out-of-limit value data set according to a quartile data interval criterion, and eliminating data with abnormal detection results in the first level detection to obtain a second data set to be detected;

the second level detection module is connected with the first level detection module and is used for carrying out second level detection on the second data set to be detected according to a median deviation method to obtain a first abnormal data set.

7. The system for detecting abnormal data at voltage monitoring points of a power distribution network according to claim 6, wherein said out-of-limit detection module comprises: the device comprises a first acquisition unit, an upper limit calculation unit, a lower limit calculation unit and a first judgment unit;

the first acquisition unit is used for acquiring a preset upper limit percentage of voltage management, a preset lower limit percentage of voltage management and a voltage class of the power distribution network;

the upper limit calculation unit is connected with the first acquisition unit and is used for calculating the upper limit of the voltage management according to the upper limit percentage of the voltage management and the voltage class of the power distribution network;

the lower limit calculation unit is connected with the first acquisition unit and is used for calculating a voltage management lower limit according to the voltage management lower limit percentage and the power distribution network voltage level;

the first judging unit is respectively connected with the upper limit calculating unit and the lower limit calculating unit, and is used for judging the data which is larger than the upper limit of the voltage management or smaller than the lower limit of the voltage management in the first data set to be detected as the out-of-limit value data, and acquiring the out-of-limit value data set from all the out-of-limit value data sets.

8. The system for detecting abnormal data at voltage monitoring points of a power distribution network of claim 6, wherein said first level detection module comprises: an abnormal coefficient acquisition unit, a quartile selection unit, an out-of-limit abnormal data determination unit and a marking unit;

the anomaly coefficient setting unit is used for acquiring an anomaly coefficient beyond the upper limit and an anomaly coefficient beyond the lower limit;

the quartile selection unit is used for quartering the data in the threshold value crossing data set and selecting first quartile data and third quartile data;

the out-of-limit abnormal data determining unit is connected with the abnormal coefficient setting unit and the quartile selecting unit, and is used for determining upper-limit abnormal data according to the first quartile data, the third quartile data and the upper-limit abnormal coefficient, and is also used for determining lower-limit abnormal data according to the first quartile data, the third quartile data and the lower-limit abnormal coefficient;

the marking unit is connected with the out-of-limit abnormal data determining unit and is used for marking the out-of-limit abnormal data and the out-of-limit abnormal data as abnormal data of a detection result.

9. The system for detecting abnormal data at voltage monitoring points of a power distribution network of claim 6, wherein said second level detection module comprises: the device comprises a second acquisition unit, a sliding window setting unit, a first median determining unit, an absolute median calculating unit, a second median determining unit, an abnormal data determining unit and a sliding unit;

the second acquisition unit is used for acquiring a preset first abnormality judgment coefficient, the second abnormality judgment coefficient and a preset scale factor constant;

the sliding window setting unit is used for setting a sliding window with preset data length from the first data of the data to be detected;

the first median determining unit is connected with the sliding window setting unit and is used for sequentially arranging the data in the sliding window from small to large and determining the median of the arranged data in the sliding window;

the absolute median difference calculating unit is connected with the first median determining unit and is used for calculating absolute median difference data according to the median of the data in the sliding window and the data in the sliding window;

the second median determining unit is connected with the absolute median calculating unit and is used for arranging the absolute median data from small to large and determining the median of the arranged absolute median data;

the abnormal data determining unit is respectively connected with the first median determining unit, the coefficient and constant adding unit and the second median determining unit and is used for determining window abnormal data in the sliding window according to the median of the data in the sliding window, the first abnormal judgment coefficient, the second abnormal judgment coefficient, the scale factor constant and the median of the absolute median difference data;

the sliding unit is connected with the abnormal data determining unit and is used for sliding the sliding window by a preset data length and repeating the steps S3-S6 until detection of all data in the second data set to be detected is completed.

10. The system for detecting abnormal data at voltage monitoring points of power distribution network according to claim 9, wherein said abnormal data determining unit comprises: the system comprises a first abnormal data determining subunit, a second abnormal data determining subunit and a summarizing subunit, wherein:

the first abnormal data determining subunit is configured to determine first window abnormal data in the sliding window according to a median of data in the sliding window, the first abnormal judgment coefficient, the scale factor constant, and a median of the absolute median difference data;

the second abnormal data determining subunit is configured to determine second window abnormal data in the sliding window according to a median of data in the sliding window, the second abnormal judgment coefficient, the scale factor constant, and a median of the absolute median difference data;

the summarizing subunit is configured to determine the first window abnormal data and the second window abnormal data as the window abnormal data.

Images