CN111696003B - Intelligent identification and early warning method for abnormal period of line loss rate of distribution room under drive of mass data - Google Patents
Intelligent identification and early warning method for abnormal period of line loss rate of distribution room under drive of mass data Download PDFInfo
- Publication number
- CN111696003B CN111696003B CN202010533154.2A CN202010533154A CN111696003B CN 111696003 B CN111696003 B CN 111696003B CN 202010533154 A CN202010533154 A CN 202010533154A CN 111696003 B CN111696003 B CN 111696003B
- Authority
- CN
- China
- Prior art keywords
- loss rate
- line loss
- abnormal
- user
- time interval
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 247
- 238000000034 method Methods 0.000 title claims abstract description 51
- 230000005611 electricity Effects 0.000 claims description 45
- 230000008569 process Effects 0.000 claims description 26
- 230000005856 abnormality Effects 0.000 claims description 25
- 238000003064 k means clustering Methods 0.000 claims description 4
- 230000002349 favourable effect Effects 0.000 abstract description 4
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses an intelligent identification and early warning method for a line loss rate abnormal period under the drive of mass data. The method can well solve the problems of identifying abnormal time periods of the line loss rate of the distribution area and performing abnormal early warning under mass data, improves the accuracy and efficiency of judging the abnormal line loss rate, and is favorable for improving the practical problems that the abnormal line loss rate of the distribution area cannot be judged timely and is difficult to investigate. The method can automatically early warn and quickly position the line loss rate abnormity of the distribution area, improve the accuracy and efficiency of judging the line loss rate abnormity, and is favorable for improving the practical problems that the line loss rate abnormity is not judged timely and is difficult to investigate in the distribution area.
Description
Technical Field
The invention relates to an intelligent identification and early warning method for abnormal periods of line loss rate of a platform area driven by mass data, belonging to the technology for judging abnormal periods of line loss rate.
Background
The effective reduction of the power loss is a long-term target in the line loss management work of the power enterprises, and the line loss management is the key point in the operation management of the power enterprises. The analysis of the line loss abnormity is the core of the line loss management work, and the analysis of the line loss data is used for timely grasping the operation conditions of all parts in the power grid and finding out corresponding problems, so that the abnormity reason can be timely found out, the faults of the power grid can be eliminated, and the overall management level of a power supply enterprise can be effectively improved. When the line loss rate abnormality analysis is performed, the existing line loss rate abnormality period judging method is single in condition, only shows the line loss rate, cannot specify a specific time period of the line loss rate abnormality, cannot specify a high line loss rate or a negative line loss rate, lacks of judging the duration of the line loss rate, lacks of analyzing the line loss rate by combining the line loss rate and the power consumption of a user for discussion, and greatly influences the work of the line loss rate abnormality analysis.
The judgment work of the abnormal time interval of the line loss of the transformer area can help personnel to accurately find out corresponding abnormal line loss rate data, so that the efficiency is improved, a scientific basis is provided for line loss management, and finally the abnormal place of the line loss of the transformer area can be found in time. Because the electric quantity loss in the electric energy transmission process of the power grid cannot be avoided, the line loss reasons are numerous, the line loss quantity changes complicatedly and cannot be determined in time, and the data quantity of the line loss rate of the transformer area is various, the line loss rate of the transformer area needs to be judged in abnormal time periods before the reason for judging the abnormal line loss rate is judged.
Disclosure of Invention
The purpose of the invention is as follows: in order to overcome the defects in the prior art, the invention provides an intelligent identification and early warning method for abnormal periods of the line loss rate of a distribution room under the drive of mass data, and the line loss rate data of the distribution room are obtained; establishing a membership function of the line loss rate according to historical data of the line loss rate; classifying the line loss rate through a membership function, and determining time periods with different line loss rate anomalies; calculating the duration of the abnormal line loss rate; determining the abnormal time interval of the electricity consumption of the user in the abnormal time interval of the line loss rate; calculating the contact ratio of the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the corresponding user; establishing a membership function about the contact ratio of two abnormal time periods in the fuzzy rule according to expert experience; identifying the relationship between the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user by utilizing the membership function, and finding out an abnormal associated user; and giving early warning information of the abnormal line loss rate, and displaying the abnormal time period of the line loss rate of the transformer area, the corresponding duration and the associated user information influencing the abnormal line loss rate.
The technical scheme is as follows: in order to achieve the purpose, the invention adopts the technical scheme that:
a mass data driven intelligent identification and early warning method for abnormal period of line loss rate of a platform area comprises the following steps:
s1, obtaining historical data of line loss rate of the transformer area;
s2, establishing a membership function in the fuzzy rule according to the historical data of the line loss rate of the transformer area;
s3, classifying the line loss rate through the membership function obtained in the step S2, and determining each abnormal time period of the line loss rate;
s4, calculating the abnormal duration of the line loss rate;
s5, calculating the coincidence degree of the abnormal time periods of the electricity consumption of the users corresponding to the abnormal time periods of the line loss rate;
s6, establishing a membership function about the coincidence degree of the line loss rate abnormal time period and the user power consumption abnormal time period in the fuzzy rule according to expert experience;
s7, analyzing the relation between the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user by using the membership function obtained in the step S6, and finding out an abnormal associated user;
and S8, outputting line loss rate abnormity early warning information, and displaying the line loss rate abnormity time period, the line loss rate abnormity duration and abnormity associated users.
Specifically, in step S2, when the membership function in the fuzzy rule is established according to the line loss rate historical data of the distribution room: inputting historical data of line loss rate of a transformer area, and preparing for subsequent data processing; selecting a curve shape as the curve shape of the membership function based on expert experience; then, a k-means clustering method is selected to cluster the historical data of the line loss rate of the transformer area, and a clustering center of the normal line loss rate is found out, wherein the clustering center of the normal line loss rate is represented as delta P%zAnd the clustering center of the abnormal line loss rate is expressed as delta P%y,ΔP%yIs delta P%zTwice of; and finally, establishing a membership function in the fuzzy rule based on the national standard of the normal line loss rate, namely the industry management distribution line loss standard and the distribution line loss rate standard after rural power grid transformation:
wherein: mu.sfIs the membership degree of the negative line loss; mu.szThe membership degree of the normal line loss rate; mu.sgThe membership degree of the high linear loss rate; x is the line loss rate; delta P%aThe corresponding line loss rate of the point with the membership degree of 1 in the membership degree function representing the normal line loss rate takes the value of delta P percentz;ΔP%bAnd (3) the corresponding line loss rate of a point with the membership degree of 1 in the membership degree function representing the high line loss rate.
Specifically, in step S3, the step of classifying the line loss rate by the membership function obtained in step S2 and determining each abnormal period of the line loss rate includes the following steps:
s31, deriving historical data of line loss rate of the station area, and dividing the judged line loss rate time interval into time sets { T } in units of days1,T2,…,Tn,…,TN},TnThe period represents a line loss rate period of the nth day;
s32, initializing a normal line loss rate time setNegative line loss rate time setAnd high-loss-rate time set
S33, initializing n to 1;
s34, pair TnThe line loss rate delta P% of the time interval is taken as x to be substituted into the membership degreeThe function is classified:
the following conditions are: if corresponding mu in the membership functionf<μzAnd mug<μzThen, determine TnThe time interval is a normal time interval, the corresponding line loss rate is a normal line loss rate, and T is calculatednClassifying into alpha;
case two: if corresponding mu in the membership functionfIf greater than 0, then T is judgednThe time interval is an abnormal time interval, the corresponding line loss rate is a negative line loss rate, and T is calculatednClassifying into beta;
case (c): if corresponding mu in the membership functiong>μzThen, determine TnThe time interval is an abnormal time interval, the corresponding line loss rate is a high line loss rate, and T is calculatednClassifying into gamma;
s35, n ═ n + 1: if N is more than or equal to N +1, entering step 36; otherwise, go to step S34;
s36, obtaining the time set α ═ T of normal line loss rateiT, time set of negative line loss ratejAnd a high-loss-rate time set γ ═ Tk},Ti≠Tj≠Tk;TiDenotes the ith normal line loss rate period, TjIndicates the jth negative line loss rate abnormal period, TkRepresenting a kth high-line-loss-rate anomaly period;
s37, judging the classification result obtained in the step S36: if both β and γ are empty sets, the line loss rate of the station area is normal within the determined line loss rate period, and the step S8 is entered; otherwise, the line loss rate of the station area is abnormal in the judged line loss rate time period, and beta and gamma are output.
Specifically, in step S4, when calculating the duration of the line loss rate anomaly, the historical data of the line loss rate of the station area is derived, and the determined line loss rate period is divided into time sets { T } in units of days1,T2,…,Tn,…,TN},TnThe period represents a line loss rate period of the nth day; then the following steps are carried out:
s41, initializing the negative line loss rate abnormality duration as a-0, initializing the high line loss rate abnormality duration as b-0, c-4, and n-1;
S42、if TnIf the time slot line loss rate is the normal line loss rate, if c is c-1, the process proceeds to step S47; otherwise, go to step S43;
s43, if TnIf the line loss rate in the time interval is the negative line loss rate, the step S44 is entered; if TnIf the time interval line loss rate is the high line loss rate, the step S45 is entered;
s44, if b is equal to 0, a is equal to a +1, and the process proceeds to step S47; otherwise, go to step S46;
s45, if a is equal to 0, b is equal to b +1, and the process proceeds to step S47; otherwise, go to step S46;
s46, c is 0, and n is n-1, and the process proceeds to step S47;
s47, where n is n +1, the process proceeds to step S48;
s48, if c is equal to 0, the process proceeds to step S49; otherwise, go to step S42;
s49, determining that the line loss rate abnormality duration is t days, and t is max { a, b }, and proceeding to step S410;
s410, if N is larger than or equal to N +1, the step S5 is executed; otherwise, the process proceeds to step S42.
Specifically, in step S5, the method calculates the overlap ratio of the abnormal electricity consumption periods of the user corresponding to the abnormal electricity consumption periods of the line loss rates, divides all the abnormal electricity consumption periods of the line loss rates into continuous sub-time periods, determines the abnormal electricity consumption periods of the user in each sub-time period, and calculates the overlap ratio of the abnormal electricity consumption periods of the user corresponding to the abnormal electricity consumption periods of the line loss rates, and includes the following steps:
s51, according to the abnormal duration time of the line loss rate and the time set beta of the negative line loss rate ═ T1,T2,…,Tj,…,TJAnd a high-loss-rate time set γ ═ T1,T2,…,Tk,…,TKAnd re-dividing the abnormal time period of the line loss rate:
s511, time set β ═ T for negative line loss rate1,T2,…,Tj,…,TJThe repartitioning comprises the following steps:
s5112, mixing TjDue to XhPerforming the following steps;
s5113 according to TjLine loss rate abnormal duration t corresponding to time intervaljWill TjAfter a period of time (t)j-1) all time periods are ascribed to XhPerforming the following steps;
S5114、j=j+tj,h=h+1;
s5115, if J is larger than J, the process goes to step S512; otherwise, go to step S5112;
s512, for the high-loss-rate time set gamma ═ T1,T2,…,Tk,…,TKThe repartitioning comprises the following steps:
s5122, mixing TkDue to XLPerforming the following steps;
s5123 according to TkLine loss rate abnormal duration t corresponding to time intervalkWill TkAfter a period of time (t)k-1) all time periods are ascribed to XLPerforming the following steps;
S5124、k=k+tk,l=l+1;
s5125, if K is larger than K, the process goes to step S52; otherwise, go to step S5122;
s52, obtaining an abnormal time period time set { X1,X2,…,Xl,…,XL};
S53, initializing h to 1;
s54, introducing XhElectricity consumption of M users over a period of time, using ZhmIndicates that user m is at XhEstablishing a corresponding user electricity consumption data set { Z ] according to the electricity consumption in a time intervalh1,,Zh2,…,Zhm,…,ZhM};
S55, initializing m to 1;
s56, counting that the user m is XhThe number of days of abnormal electricity usage e (e.g., the number of days of abnormal electricity usage is always 0 or too high);
s57, judging the relation between the electricity consumption of the user and the line loss rate of the distribution room by using the Pearson correlation coefficient:
wherein: r represents the Pearson correlation coefficient between X and Y, X representing user m at XhElectricity consumption in time interval ZhmY represents XhLine loss rate of time slot and station area, N represents XhThe line loss rate abnormal duration corresponding to the time interval; the larger the Pearson correlation coefficient r is, the larger X ishThe stronger the correlation between the line loss rate of the time interval station area and the electricity consumption of the user m is, and when r is more than or equal to 0.8, judging that the electricity consumption of the user m is abnormal within the abnormal duration time of the line loss rate;
s58, calculating XhThe abnormal duration time of the user electricity consumption of the user m in the time interval is d-e + N; calculating XhThe coincidence degree of the time interval and the abnormal time interval of the user electricity consumption of the user m is
S59、m=m+1;
S510, if M is larger than M, the step S511 is carried out; otherwise, go to step S56;
S511、h=h+1;
s512, if H is larger than H, the step S513 is executed; otherwise, go to step S54;
s513, an overlap ratio data set S ═ S of the abnormal time interval of the loss rate of the output line and the abnormal time interval of the electricity consumption of the userhm}。
Specifically, in step S6, when a membership function regarding the overlap ratio between the abnormal time period of the line loss rate and the abnormal time period of the power consumption of the user in the fuzzy rule is established according to expert experience: firstly, selecting a curve shape as the curve shape of a membership function based on expert experience; then, the coincidence degree of the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user is converted into a fuzzy subset { NB ZO PB }, and a membership function table based on the coincidence degree (namely the failure reason) of the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user is established; and finally, establishing a membership function based on a membership function table:
wherein: mu.sN(NB) represents membership degree with small overlap ratio between the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user; mu.sZ(ZO) represents a degree of membership moderate in the degree of coincidence of abnormal periods; mu.sP(PB) represents a membership degree at which the coincidence degree of the abnormal period is large; and x is the coincidence degree of the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user.
Specifically, in the step S7, the relationship between the abnormal period of the line loss rate and the abnormal period of the power consumption of the user is analyzed by using the membership function obtained in the step S6, so as to find out the abnormal associated user, which includes the following steps:
s71, deriving an abnormal time period time set { X1,X2,…,Xh,…,XHH is initialized to 1;
s72, deriving coincidence degree data set S ═ S of abnormal time interval of line loss rate and abnormal time interval of power consumption of userhmInitializing m to 1;
s73, for XhAnd classifying the line loss rate of the time interval:
the following conditions are: if corresponding mu in the membership functionN≥μZAnd muN≥μPThen, determine XhThe coincidence degree of the two types of abnormity in the time period is very small, the correlation between the line loss rate abnormity and a user m is very small, and the user m is a three-level abnormity associated user;
case two: if corresponding mu in the membership functionZ≥μNAnd muZ≥μPThen, determine XhThe coincidence of the two types of abnormity in the time period is medium, the line loss rate abnormity is partially related to a user m, and the user m is a secondary abnormity related user;
case (c): if corresponding mu in the membership functionP>μZThen, determine XhThe superposition of two types of abnormity in time interval is large, the correlation between the line loss rate abnormity and a user m is large, and the user m is a first-level abnormity associated user;
S74、m=m+1;
s75, if M is larger than M, the step goes to S76; otherwise, go to step S73;
S76、h=h+ 1
s77, if H is more than H, the step goes to S78; otherwise, go to step S72;
s78, if there is one-stage abnormal associated user, outputting all the one-stage abnormal associated users; if no primary abnormal associated user exists but a secondary abnormal associated user exists, outputting a set number of secondary abnormal associated users; and if the primary abnormal associated user and the secondary abnormal associated user do not exist, outputting the result.
The invention starts from the practical application of intelligent identification and early warning analysis of the abnormal stage of the line loss rate of the transformer area, analyzes the line loss rate data of the transformer area, calculates the duration of the abnormal line loss rate, and simultaneously establishes a membership function by considering the characteristics of the overlap ratio of the power consumption of users and the line loss rate of the transformer area, thereby finding out abnormal associated users and giving early warning information of the abnormal line loss rate. The method can well solve the problems of identifying abnormal time periods of the line loss rate of the distribution area and performing abnormal early warning under mass data, improves the accuracy and efficiency of judging the abnormal line loss rate, and is favorable for improving the practical problems that the abnormal line loss rate of the distribution area cannot be judged timely and is difficult to investigate.
Has the advantages that: the method can perform automatic early warning and rapid positioning on the abnormal line loss rate of the distribution area, improve the accuracy and efficiency of judging the abnormal line loss rate, and is favorable for improving the practical problems that the abnormal line loss rate in the distribution area cannot be judged timely and is difficult to investigate; according to the method, the line loss rate is fuzzified, a fuzzy membership function adaptive to the line loss rate of the transformer area is established, the abnormal time period of the line loss rate of the transformer area is automatically identified, the property of the abnormal line loss rate is contrasted and analyzed from the time perspective, and the relation between the abnormal time period of the power consumption of a user and the abnormal time period of the line loss rate is judged, so that the early warning of the abnormal line loss rate of the transformer area is finally realized.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention;
FIG. 2(a) is a flow chart of establishing a membership function in a fuzzy rule according to line loss rate historical data of a distribution room;
FIG. 2(b) is a graph of membership functions with respect to line loss rates obtained based on FIG. 2 (a);
FIG. 3 is a graph of membership functions for time overlap for periods of abnormal line loss and for periods of abnormal power usage by a user;
FIG. 4 is a flowchart illustrating the analysis of each abnormal line loss rate period;
fig. 5 is a flowchart of an analysis of abnormal line loss rate duration.
Detailed Description
The present invention will be further described with reference to the accompanying drawings.
As shown in fig. 1, the invention provides an intelligent identification and early warning method for abnormal line loss rate time period of a platform area driven by mass data, which comprises the following steps:
step one, obtaining historical data of line loss rate of a transformer area.
And step two, establishing a membership function in the fuzzy rule according to the historical data of the line loss rate of the transformer area.
As shown in fig. 2(a) and 2(b), the line loss rate history data of the station area is input first, and preparation is made for the subsequent data processing; then selecting a curve shape as the curve shape of the membership function based on expert experience, wherein the example selects a trapezoid as the curve shape of the membership function; then, a k-means clustering method is selected to cluster the historical data of the line loss rate of the transformer area, and a clustering center of the normal line loss rate is found out, wherein the clustering center of the normal line loss rate is represented as delta P%zAnd the clustering center of the abnormal line loss rate is expressed as delta P%y,ΔP%yIs delta P%zTwice of; finally, the national standard based on the normal line loss rate, namely industry managementThe distribution line loss standard and the distribution line loss rate standard after rural power grid transformation establish membership function in the fuzzy rule:
wherein: mu.sfIs the membership degree of the negative line loss; mu.szThe membership degree of the normal line loss rate; mu.sgThe membership degree of the high linear loss rate; x is the line loss rate; delta P%aThe corresponding line loss rate of the point with the membership degree of 1 in the membership degree function representing the normal line loss rate takes the value of delta P percentz;ΔP%bAnd (3) the corresponding line loss rate of a point with the membership degree of 1 in the membership degree function representing the high line loss rate.
And step three, classifying the line loss rate through the membership function obtained in the step two, and determining each abnormal time period of the line loss rate.
As shown in fig. 4, the method comprises the following steps:
s31, deriving historical data of line loss rate of the station area, and dividing the judged line loss rate time interval into time sets { T } in units of days1,T2,…,Tn,…,TN},TnThe period represents a line loss rate period of the nth day;
s32, initializing a normal line loss rate time setNegative line loss rate time setAnd high-loss-rate time set
S33, initializing n to 1;
s34, pair TnAnd (3) substituting the line loss rate delta P% of the time interval as x into a membership function for classification:
the following conditions are: if corresponding mu in the membership functionf<μzAnd mug<μzThen, determine TnThe time interval is a normal time interval, the corresponding line loss rate is a normal line loss rate, and T is calculatednClassifying into alpha;
case two: if corresponding mu in the membership functionfIf greater than 0, then T is judgednThe time interval is an abnormal time interval, the corresponding line loss rate is a negative line loss rate, and T is calculatednClassifying into beta;
case (c): if corresponding mu in the membership functiong>μzThen, determine TnThe time interval is an abnormal time interval, the corresponding line loss rate is a high line loss rate, and T is calculatednClassifying into gamma;
s35, n ═ n + 1: if N is more than or equal to N +1, entering step 36; otherwise, go to step S34;
s36, obtaining the time set α ═ T of normal line loss rateiT, time set of negative line loss ratejAnd a high-loss-rate time set γ ═ Tk},Ti≠Tj≠Tk;TiDenotes the ith normal line loss rate period, TjIndicates the jth negative line loss rate abnormal period, TkRepresenting a kth high-line-loss-rate anomaly period;
s37, judging the classification result obtained in the step S36: if both β and γ are empty sets, the line loss rate of the station area is normal within the determined line loss rate period, and the step S8 is entered; otherwise, the line loss rate of the station area is abnormal in the judged line loss rate time period, and beta and gamma are output.
And step four, calculating the abnormal duration of the line loss rate.
As shown in the figure5, deriving the historical data of line loss rate of the station area, and dividing the determined line loss rate period into time sets { T } in units of days1,T2,…,Tn,…,TN},TnThe period represents a line loss rate period of the nth day; then the following steps are carried out:
s41, initializing the negative line loss rate abnormality duration as a-0, initializing the high line loss rate abnormality duration as b-0, c-4, and n-1;
s42, if TnIf the time slot line loss rate is the normal line loss rate, if c is c-1, the process proceeds to step S47; otherwise, go to step S43;
s43, if TnIf the line loss rate in the time interval is the negative line loss rate, the step S44 is entered; if TnIf the time interval line loss rate is the high line loss rate, the step S45 is entered;
s44, if b is equal to 0, a is equal to a +1, and the process proceeds to step S47; otherwise, go to step S46;
s45, if a is equal to 0, b is equal to b +1, and the process proceeds to step S47; otherwise, go to step S46;
s46, c is 0, and n is n-1, and the process proceeds to step S47;
s47, where n is n +1, the process proceeds to step S48;
s48, if c is equal to 0, the process proceeds to step S49; otherwise, go to step S42;
s49, determining that the line loss rate abnormality duration is t days, and t is max { a, b }, and proceeding to step S410;
s410, if N is larger than or equal to N +1, the step S5 is executed; otherwise, the process proceeds to step S42.
And step five, calculating the contact ratio of the abnormal time periods of the electricity consumption of the users corresponding to the abnormal time periods of the line loss rate.
Dividing all the abnormal time periods of the line loss rate into continuous sub-time periods, determining the abnormal time period of the user power consumption in each sub-time period, and calculating the contact ratio of the abnormal time periods of the user power consumption corresponding to the abnormal time periods of the line loss rate, wherein the method comprises the following steps:
s51, according to the abnormal duration time of the line loss rate and the time set beta of the negative line loss rate ═ T1,T2,…,Tj,…,TJAnd a high-loss-rate time set γ ═ T1,T2,…,Tk,…,TKAnd re-dividing the abnormal time period of the line loss rate:
s511, time set β ═ T for negative line loss rate1,T2,…,Tj,…,TJThe repartitioning comprises the following steps:
s5112, mixing TjDue to XhPerforming the following steps;
s5113 according to TjLine loss rate abnormal duration t corresponding to time intervaljWill TjAfter a period of time (t)j-1) all time periods are ascribed to XhPerforming the following steps;
S5114、j=j+tj,h=h+1;
s5115, if J is larger than J, the process goes to step S512; otherwise, go to step S5112;
s512, for the high-loss-rate time set gamma ═ T1,T2,…,Tk,…,TKThe repartitioning comprises the following steps:
s5122, mixing TkDue to XLPerforming the following steps;
s5123 according to TkLine loss rate abnormal duration t corresponding to time intervalkWill TkAfter a period of time (t)k-1) all time periods are ascribed to XLPerforming the following steps;
S5124、k=k+tk,l=l+1;
s5125, if K is larger than K, the process goes to step S52; otherwise, go to step S5122;
s52, obtaining an abnormal time period time set { X1,X2,…,Xl,…,XL};
S53, initializing h to 1;
s54, introducing XhElectricity consumption of M users over a period of time, using ZhmIndicates that user m is at XhEstablishing a corresponding user electricity consumption data set { Z ] according to the electricity consumption in a time intervalh1,,Zh2,…,Zhm,…,ZhM};
S55, initializing m to 1;
s56, counting that the user m is XhThe number of days of abnormal electricity usage e (e.g., the number of days of abnormal electricity usage is always 0 or too high);
s57, judging the relation between the electricity consumption of the user and the line loss rate of the distribution room by using the Pearson correlation coefficient:
wherein: r represents the Pearson correlation coefficient between X and Y, X representing user m at XhElectricity consumption in time interval ZhmY represents XhLine loss rate of time slot and station area, N represents XhThe line loss rate abnormal duration corresponding to the time interval; the larger the Pearson correlation coefficient r is, the larger X ishThe stronger the correlation between the line loss rate of the time interval station area and the electricity consumption of the user m is, and when r is more than or equal to 0.8, judging that the electricity consumption of the user m is abnormal within the abnormal duration time of the line loss rate;
s58, calculating XhThe abnormal duration time of the user electricity consumption of the user m in the time interval is d-e + N; calculating XhThe coincidence degree of the time interval and the abnormal time interval of the user electricity consumption of the user m is
S59、m=m+1;
S510, if M is larger than M, the step S511 is carried out; otherwise, go to step S56;
S511、h=h+1;
s512, if H is larger than H, the step S513 is executed; otherwise, go to step S54;
s513, an overlap ratio data set S ═ S of the abnormal time interval of the loss rate of the output line and the abnormal time interval of the electricity consumption of the userhm}。
And step six, establishing a membership function about the coincidence degree of the line loss rate abnormal time period and the user power consumption abnormal time period in the fuzzy rule according to expert experience.
When a membership function about the coincidence degree of the line loss rate abnormal time period and the user power consumption abnormal time period in the fuzzy rule is established according to expert experience: firstly, selecting a curve shape as the curve shape of a membership function based on expert experience, and selecting a triangle as the curve shape of the membership function in the example; then, the coincidence degree of the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user is converted into a fuzzy subset { NB ZO PB }, and a membership function table based on the coincidence degree (namely, the failure reason) of the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user is established, as shown in FIG. 3 and Table 1:
TABLE 1 membership function Table
Note: the data in the table are the membership levels corresponding to different time overlap ratios, for example, when the linguistic variable is ZO, the membership level of ZO at a quantization level of 25 is 0.45, and so on.
And finally, establishing a membership function based on a membership function table:
wherein: mu.sN(NB) represents membership degree with small overlap ratio between the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user; mu.sZ(ZO) represents a degree of membership moderate in the degree of coincidence of abnormal periods; mu.sP(PB) representsMembership degree with high coincidence degree in abnormal time period; and x is the coincidence degree of the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user.
And seventhly, analyzing the relation between the abnormal period of the line loss rate and the abnormal period of the power consumption of the user by using the membership function obtained in the sixth step, and finding out the abnormal associated user.
S71, deriving an abnormal time period time set { X1,X2,…,Xh,…,XHH is initialized to 1;
s72, deriving coincidence degree data set S ═ S of abnormal time interval of line loss rate and abnormal time interval of power consumption of userhmInitializing m to 1;
s73, for XhAnd classifying the line loss rate of the time interval:
the following conditions are: if corresponding mu in the membership functionN≥μZAnd muN≥μPThen, determine XhThe coincidence degree of the two types of abnormity in the time period is very small, the correlation between the line loss rate abnormity and a user m is very small, and the user m is a three-level abnormity associated user;
case two: if corresponding mu in the membership functionZ≥μNAnd muZ≥μPThen, determine XhThe coincidence of the two types of abnormity in the time period is medium, the line loss rate abnormity is partially related to a user m, and the user m is a secondary abnormity related user;
case (c): if corresponding mu in the membership functionP>μZThen, determine XhThe superposition of two types of abnormity in time interval is large, the correlation between the line loss rate abnormity and a user m is large, and the user m is a first-level abnormity associated user;
S74、m=m+1;
s75, if M is larger than M, the step goes to S76; otherwise, go to step S73;
S76、h=h+ 1
s77, if H is more than H, the step goes to S78; otherwise, go to step S72;
s78, if there is one-stage abnormal associated user, outputting all the one-stage abnormal associated users; if no primary abnormal associated user exists but a secondary abnormal associated user exists, outputting a set number of secondary abnormal associated users; and if the primary abnormal associated user and the secondary abnormal associated user do not exist, outputting the result.
And step eight, outputting line loss rate abnormity early warning information, and displaying the line loss rate abnormity time period, the line loss rate abnormity duration and abnormity associated users.
The invention is further described below with reference to a specific embodiment.
It is known that the line loss rate data and the power consumption data of each user in the west tower substation 04 in the year 1 month to 2019 month 2 are recorded in units of days, and the line loss rate and the power consumption data of each user in the year 1 month and february in 2019 are shown in table one.
Firstly, introducing 2018 station area line loss rate data, carrying out clustering analysis on the station area line loss rate by using a k-means clustering method, determining the number of clusters to be 3, and carrying out clustering by using a sps software to obtain three clustering centers.
Table 2 cluster centers:
note: the table intermediate data is data of three cluster centers of line loss rates, for example, the line loss rate of the cluster center with the number of 1 is 2.08%, 373 days are in the station area line loss rate data, and so on.
The clustering center with the number of 1 can be judged to be the clustering center with the normal line loss rate, namely delta P percent according to the number of caseszIs 2.08%, and Δ P%y4.16%, and 10% line loss rate standard because the plateau region is 10kv plateau region, and 2.08% or more than 0.5 times 10% to reduce error according to delta P%zAnd. delta. P%yA membership function is established with respect to the line loss rate.
The membership function can be used for quickly judging the line loss rate conditions of the transformer area in 1 month and february in 2019, judging that the line loss rates of January and February are abnormal, and judging the time period of the abnormal line loss rate and outputting related data of the abnormal line loss rate time period.
Table 3 line loss rate abnormal period:
note: the data in the table indicates the time of the line loss rate abnormality and the corresponding line loss rate, for example, the line loss rate abnormality of 1 month and 7 days in 2019, the abnormal line loss rate is 5.42%, the high line loss rate, the line loss rate abnormality of 1 month and 8 days in 2019, the abnormal line loss rate is 3.7%, the high line loss rate, and the like.
Then, the duration of the abnormal line loss rate is calculated, and the duration of the line loss rate in different periods can be calculated.
Table 4 abnormal line loss rate duration:
note: the data in the middle of the table are the time and duration of the line loss rate abnormality, for example, the line loss rate abnormality of 1 month and 8 days in 2019, the abnormal line loss rate is 3.7%, the duration of the line loss rate abnormality is 10 days, the line loss rate abnormality of 1 month and 14 days in 2019 is 3.85%, the duration of the line loss rate abnormality is 10 days, and so on.
And then judging the abnormal condition of the line loss rate of the platform zone in the month 1 of 2019. And importing user electricity consumption data, analyzing abnormal time periods of the user electricity consumption, and judging the contact ratio of the abnormal time periods of the user electricity consumption and the abnormal time periods of the line loss rate. The line loss rate abnormal period may be divided into 2 segments.
TABLE 5 segmentation of the line loss rate anomaly periods
First, a first abnormal time interval is analyzed, and it can be found that the electricity consumption of the user in the time interval of the 1 month in 2019 is not 0, and e is 0. The Pearson coefficient of the electricity consumption of the user A in the period and the line loss rate in the period reaches a strong correlation range, is larger than 0.8, and N is 10, so that the time length d of the abnormal electricity consumption period of the user is 10 days.
And analyzing the relationship between the two abnormal time periods by using the membership function of the abnormal time period of the line loss rate and the abnormal time period of the power consumption of the user, and judging that the probability of the abnormal line loss rate of the time period related to the user A is extremely high, wherein the abnormal power consumption of the user is the reason of the abnormal line loss rate of the time period.
And analyzing the next abnormal time interval according to the same method, and finding that the probability of the line loss rate abnormality in the third time interval related to the user A is extremely high, and the abnormal power consumption of the user is the reason of the line loss rate abnormality in the time interval.
Table 6 shows the coincidence degree of the calculated time interval and the abnormal electricity consumption time interval of the A user
Note: the data in the middle of the table are the time of the line loss rate abnormality, the abnormal line loss rate and the corresponding duration, for example, 1, 7 days to 21 days in 2019, the abnormal line loss rate is high line loss, the abnormal duration of the line loss rate is 10 days, the user causing the line loss rate abnormality is a user, and so on.
TABLE 7 COMPARATIVE TABLE FOR USER POWER CONSUMPTION
The scheme is through fuzzifying the line loss rate, a fuzzy membership function adaptive to the line loss rate of the distribution room is established, abnormal time intervals of the line loss rate of the distribution room are automatically identified, the properties of the abnormal line loss rate are contrasted and analyzed from the time angle, and the abnormal line loss rate early warning of the distribution room is finally realized by judging the relation between the abnormal time intervals of the power consumption of a user and the abnormal time intervals of the line loss rate.
The above description is only of the preferred embodiments of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the invention and these are intended to be within the scope of the invention.
Claims (1)
1. A mass data driven intelligent identification and early warning method for abnormal period of line loss rate of a platform area is characterized by comprising the following steps: the method comprises the following steps:
s1, obtaining historical data of line loss rate of the transformer area;
s2, establishing a membership function in the fuzzy rule according to the historical data of the line loss rate of the transformer area: firstly, inputting historical data of line loss rate of a transformer area; selecting a curve shape as the curve shape of the membership function based on expert experience; then, a k-means clustering method is selected to cluster the historical data of the line loss rate of the transformer area, and a clustering center of the normal line loss rate is found out, wherein the clustering center of the normal line loss rate is represented as delta P%zAnd the clustering center of the abnormal line loss rate is expressed as delta P%y,ΔP%yIs delta P%zTwice of; and finally, establishing a membership function in the fuzzy rule based on the national standard of the normal line loss rate:
wherein: mu.sfIs the membership degree of the negative line loss; mu.szThe membership degree of the normal line loss rate; mu.sgThe membership degree of the high linear loss rate; x is the line loss rate; delta P%aThe corresponding line loss rate of the point with the membership degree of 1 in the membership degree function representing the normal line loss rate takes the value of delta P percentz;ΔP%bRepresenting the corresponding line loss rate of a point with the membership degree of 1 in the membership degree function of the high line loss rate;
s3, classifying the line loss rate through the membership function obtained in the step S2, and determining each abnormal time period of the line loss rate, comprising the following steps:
s31, deriving historical data of line loss rate of the station area, and dividing the judged line loss rate time interval into time sets { T } in units of days1,T2,…,Tn,…,TN},TnThe period represents a line loss rate period of the nth day;
s32, initializing a normal line loss rate time setNegative line loss rate time setAnd high-loss-rate time set
S33, initializing n to 1;
s34, pair TnAnd (3) substituting the line loss rate delta P% of the time interval as x into a membership function for classification:
the following conditions are: if at the degree of membershipCorresponding mu in functionf<μzAnd mug<μzThen, determine TnThe time interval is a normal time interval, the corresponding line loss rate is a normal line loss rate, and T is calculatednClassifying into alpha;
case two: if corresponding mu in the membership functionfIf greater than 0, then T is judgednThe time interval is an abnormal time interval, the corresponding line loss rate is a negative line loss rate, and T is calculatednClassifying into beta;
case (c): if corresponding mu in the membership functiong>μzThen, determine TnThe time interval is an abnormal time interval, the corresponding line loss rate is a high line loss rate, and T is calculatednClassifying into gamma;
s35, n ═ n + 1: if N is more than or equal to N +1, entering step 36; otherwise, go to step S34;
s36, obtaining the time set α ═ T of normal line loss rateiT, time set of negative line loss ratejAnd a high-loss-rate time set γ ═ Tk},Ti≠Tj≠Tk;TiDenotes the ith normal line loss rate period, TjIndicates the jth negative line loss rate abnormal period, TkRepresenting a kth high-line-loss-rate anomaly period;
s37, judging the classification result obtained in the step S36: if both β and γ are empty sets, the line loss rate of the station area is normal within the determined line loss rate period, and the step S8 is entered; otherwise, the line loss rate of the station area is abnormal in the judged line loss rate time period, and beta and gamma are output;
s4, calculating the abnormal duration of the line loss rate, firstly deriving the historical data of the line loss rate of the station area, and dividing the judged line loss rate time interval into time sets { T } by taking the day as a unit1,T2,…,Tn,…,TN},TnThe period represents a line loss rate period of the nth day; then the following steps are carried out:
s41, initializing the negative line loss rate abnormality duration as a-0, initializing the high line loss rate abnormality duration as b-0, c-4, and n-1;
s42, if TnIf the time slot line loss rate is the normal line loss rate, if c is c-1, the process proceeds to step S47; otherwise, go to step S43;
s43, if TnIf the line loss rate in the time interval is the negative line loss rate, the step S44 is entered; if TnIf the time interval line loss rate is the high line loss rate, the step S45 is entered;
s44, if b is equal to 0, a is equal to a +1, and the process proceeds to step S47; otherwise, go to step S46;
s45, if a is equal to 0, b is equal to b +1, and the process proceeds to step S47; otherwise, go to step S46;
s46, c is 0, and n is n-1, and the process proceeds to step S47;
s47, where n is n +1, the process proceeds to step S48;
s48, if c is equal to 0, the process proceeds to step S49; otherwise, go to step S42;
s49, determining that the line loss rate abnormality duration is t days, and t is max { a, b }, and proceeding to step S410;
s410, if N is larger than or equal to N +1, the step S5 is executed; otherwise, go to step S42;
s5, calculating the contact ratio of the abnormal periods of the user power consumption corresponding to the abnormal periods of the line loss rate, dividing all the abnormal periods of the line loss rate into continuous sub-time periods, determining the abnormal time periods of the user power consumption in each sub-time period, and calculating the contact ratio of the abnormal periods of the user power consumption corresponding to the abnormal periods of the line loss rate, wherein the method comprises the following steps:
s51, according to the abnormal duration time of the line loss rate and the time set beta of the negative line loss rate ═ T1,T2,…,Tj,…,TJAnd a high-loss-rate time set γ ═ T1,T2,…,Tk,…,TKAnd re-dividing the abnormal time period of the line loss rate:
s511, time set β ═ T for negative line loss rate1,T2,…,Tj,…,TJThe repartitioning comprises the following steps:
s5112, mixing TjDue to XhPerforming the following steps;
s5113 according to TjCorresponding to time intervalLine loss rate anomaly duration tjWill TjAfter a period of time (t)j-1) all time periods are ascribed to XhPerforming the following steps;
S5114、j=j+tj,h=h+1;
s5115, if J is larger than J, the process goes to step S512; otherwise, go to step S5112;
s512, for the high-loss-rate time set gamma ═ T1,T2,…,Tk,…,TKThe repartitioning comprises the following steps:
s5122, mixing TkDue to XLPerforming the following steps;
s5123 according to TkLine loss rate abnormal duration t corresponding to time intervalkWill TkAfter a period of time (t)k-1) all time periods are ascribed to XLPerforming the following steps;
S5124、k=k+tk,l=l+1;
s5125, if K is larger than K, the process goes to step S52; otherwise, go to step S5122;
s52, obtaining an abnormal time period time set { X1,X2,…,Xl,…,XL};
S53, initializing h to 1;
s54, introducing XhElectricity consumption of M users over a period of time, using ZhmIndicates that user m is at XhEstablishing a corresponding user electricity consumption data set { Z ] according to the electricity consumption in a time intervalh1,,Zh2,…,Zhm,…,ZhM};
S55, initializing m to 1;
s56, counting that the user m is XhThe number of days e when the electricity consumption is abnormal in a time period;
s57, judging the relation between the electricity consumption of the user and the line loss rate of the distribution room by using the Pearson correlation coefficient:
wherein: r represents the Pearson correlation coefficient between X and Y, X representing user m at XhElectricity consumption in time interval ZhmY represents XhLine loss rate of time slot and station area, N represents XhThe line loss rate abnormal duration corresponding to the time interval; the larger the Pearson correlation coefficient r is, the larger X ishThe stronger the correlation between the line loss rate of the time interval station area and the electricity consumption of the user m is, and when r is more than or equal to 0.8, judging that the electricity consumption of the user m is abnormal within the abnormal duration time of the line loss rate;
s58, calculating XhThe abnormal duration time of the user electricity consumption of the user m in the time interval is d-e + N; calculating XhThe coincidence degree of the time interval and the abnormal time interval of the user electricity consumption of the user m is
S59、m=m+1;
S510, if M is larger than M, the step S511 is carried out; otherwise, go to step S56;
S511、h=h+1;
s512, if H is larger than H, the step S513 is executed; otherwise, go to step S54;
s513, an overlap ratio data set S ═ S of the abnormal time interval of the loss rate of the output line and the abnormal time interval of the electricity consumption of the userhm};
S6, establishing a membership function of the coincidence degree of the line loss rate abnormal time period and the user power consumption abnormal time period in the fuzzy rule according to expert experience: firstly, selecting a curve shape as the curve shape of a membership function based on expert experience; then, the coincidence degree of the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user is converted into a fuzzy subset { NB ZO PB }, and a membership function table based on the coincidence degree of the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user is established; and finally, establishing a membership function based on a membership function table:
wherein: mu.sN(NB) represents membership degree with small overlap ratio between the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user; mu.sZ(ZO) represents a degree of membership moderate in the degree of coincidence of abnormal periods; mu.sP(PB) represents a membership degree at which the coincidence degree of the abnormal period is large; x is the coincidence degree of the abnormal time interval of the line loss rate and the abnormal time interval of the power consumption of the user;
s7, analyzing the relationship between the abnormal period of the line loss rate and the abnormal period of the power consumption of the user by using the membership function obtained in the step S6, and finding out the abnormal associated user, wherein the method comprises the following steps:
s71, deriving an abnormal time period time set { X1,X2,…,Xh,…,XHH is initialized to 1;
s72, deriving coincidence degree data set S ═ S of abnormal time interval of line loss rate and abnormal time interval of power consumption of userhmInitializing m to 1;
s73, for XhAnd classifying the line loss rate of the time interval:
the following conditions are: if corresponding mu in the membership functionN≥μZAnd muN≥μPThen, determine XhThe coincidence degree of the two types of abnormity in the time period is very small, the correlation between the line loss rate abnormity and a user m is very small, and the user m is a three-level abnormity associated user;
case two: if corresponding mu in the membership functionZ≥μNAnd muZ≥μPThen, determine XhThe coincidence of the two types of abnormity in the time period is medium, the line loss rate abnormity is partially related to a user m, and the user m is a secondary abnormity related user;
case (c): if corresponding mu in the membership functionP>μZThen, determine XhThe superposition of two types of abnormity in time interval is large, the correlation between the line loss rate abnormity and a user m is large, and the user m is a first-level abnormity associated user;
S74、m=m+1;
s75, if M is larger than M, the step goes to S76; otherwise, go to step S73;
S76、h=h+1
s77, if H is more than H, the step goes to S78; otherwise, go to step S72;
s78, if there is one-stage abnormal associated user, outputting all the one-stage abnormal associated users; if no primary abnormal associated user exists but a secondary abnormal associated user exists, outputting a set number of secondary abnormal associated users; if no primary abnormal associated user or secondary abnormal associated user exists, no output is performed;
and S8, outputting line loss rate abnormity early warning information, and displaying the line loss rate abnormity time period, the line loss rate abnormity duration and abnormity associated users.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010533154.2A CN111696003B (en) | 2020-06-12 | 2020-06-12 | Intelligent identification and early warning method for abnormal period of line loss rate of distribution room under drive of mass data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010533154.2A CN111696003B (en) | 2020-06-12 | 2020-06-12 | Intelligent identification and early warning method for abnormal period of line loss rate of distribution room under drive of mass data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111696003A CN111696003A (en) | 2020-09-22 |
CN111696003B true CN111696003B (en) | 2021-03-16 |
Family
ID=72480591
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010533154.2A Active CN111696003B (en) | 2020-06-12 | 2020-06-12 | Intelligent identification and early warning method for abnormal period of line loss rate of distribution room under drive of mass data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111696003B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106503919A (en) * | 2016-11-03 | 2017-03-15 | 国家电网公司 | A kind of power distribution network evaluation methodology based on power supply zone characteristic |
WO2018197904A1 (en) * | 2017-04-28 | 2018-11-01 | Gb Gas Holdings Limited | Method and system for detecting anomalies in energy consumption |
CN110739686A (en) * | 2019-10-15 | 2020-01-31 | 福建网能科技开发有限责任公司 | distribution room line loss management method and system based on summary chart anomaly analysis |
-
2020
- 2020-06-12 CN CN202010533154.2A patent/CN111696003B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106503919A (en) * | 2016-11-03 | 2017-03-15 | 国家电网公司 | A kind of power distribution network evaluation methodology based on power supply zone characteristic |
WO2018197904A1 (en) * | 2017-04-28 | 2018-11-01 | Gb Gas Holdings Limited | Method and system for detecting anomalies in energy consumption |
CN110739686A (en) * | 2019-10-15 | 2020-01-31 | 福建网能科技开发有限责任公司 | distribution room line loss management method and system based on summary chart anomaly analysis |
Non-Patent Citations (1)
Title |
---|
基于kmeans聚类算法的低压台区线损异常辨别方法;陈洪涛 等;《南方电网技术》;20190228;第13卷(第2期);第2-6页 * |
Also Published As
Publication number | Publication date |
---|---|
CN111696003A (en) | 2020-09-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108985380B (en) | Point switch fault identification method based on cluster integration | |
CN104573000B (en) | Automatic call answering arrangement and method based on sequence study | |
CN102819772B (en) | Power matching network builds material requirements Forecasting Methodology and device | |
CN109272138B (en) | Sorting equipment cell number optimization method and device | |
CN110610121B (en) | Small-scale source load power abnormal data identification and restoration method based on curve clustering | |
CN108241925A (en) | A kind of discrete manufacture mechanical product quality source tracing method based on outlier detection | |
CN112231971B (en) | Blast furnace fault diagnosis method based on relative integral trend diffusion fault sample generation | |
CN113743538A (en) | Intelligent building energy consumption prediction method, equipment and medium based on IPSO-BP neural network | |
CN110880044B (en) | Markov chain-based load prediction method | |
CN116128544A (en) | Active auditing method and system for electric power marketing abnormal business data | |
CN109615115B (en) | Task reliability-oriented integrated production task scheduling method | |
CN113327047B (en) | Power marketing service channel decision method and system based on fuzzy comprehensive model | |
CN111696003B (en) | Intelligent identification and early warning method for abnormal period of line loss rate of distribution room under drive of mass data | |
CN116757562B (en) | Intelligent manufacturing scheduling method with self-learning capability | |
CN113112177A (en) | Transformer area line loss processing method and system based on mixed indexes | |
CN109508820A (en) | Campus electricity demand forecasting modeling method based on differentiation modeling | |
CN107808245A (en) | Based on the network scheduler system for improving traditional decision-tree | |
CN114578015B (en) | Intelligent quality detection method for soft magnetic ferrite | |
CN116883184A (en) | Financial tax intelligent analysis method based on big data | |
CN114118592B (en) | Smart power grids power consumption end short-term energy consumption prediction system | |
CN115936389A (en) | Big data technology-based method for matching evaluation experts with evaluation materials | |
CN116090702A (en) | ERP data intelligent supervision system and method based on Internet of things | |
CN115392663A (en) | Data acquisition and processing method based on big data | |
CN115186900A (en) | Dynamic blast furnace gas production prediction method and system suitable for multiple working condition types | |
CN105260948B (en) | A kind of water-supply systems daily planning scheduling decision method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |