CN112783939A - Low-voltage distribution network running state evaluation method based on data mining - Google Patents

Low-voltage distribution network running state evaluation method based on data mining Download PDF

Info

Publication number
CN112783939A
CN112783939A CN202011635023.1A CN202011635023A CN112783939A CN 112783939 A CN112783939 A CN 112783939A CN 202011635023 A CN202011635023 A CN 202011635023A CN 112783939 A CN112783939 A CN 112783939A
Authority
CN
China
Prior art keywords
distribution network
evaluation
term
low
index
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.)
Pending
Application number
CN202011635023.1A
Other languages
Chinese (zh)
Inventor
江霖
曹安瑛
万新宇
陈卓航
李爱平
李文晖
裴星宇
刘尧
黄培专
付博
侯成
郑卫文
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Power Grid Co Ltd
Zhuhai Power Supply Bureau of Guangdong Power Grid Co Ltd
Original Assignee
Guangdong Power Grid Co Ltd
Zhuhai Power Supply Bureau of Guangdong Power Grid Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangdong Power Grid Co Ltd, Zhuhai Power Supply Bureau of Guangdong Power Grid Co Ltd filed Critical Guangdong Power Grid Co Ltd
Priority to CN202011635023.1A priority Critical patent/CN112783939A/en
Publication of CN112783939A publication Critical patent/CN112783939A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2465—Query processing support for facilitating data mining operations in structured databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • 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
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • 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

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Databases & Information Systems (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • General Business, Economics & Management (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • Quality & Reliability (AREA)
  • Fuzzy Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Operations Research (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Game Theory and Decision Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Evolutionary Biology (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention provides a low-voltage distribution network running state evaluation method based on data mining, which comprises the following steps of: s1, establishing a low-voltage distribution network loop impedance model according to the collected and processed low-voltage distribution network data; s2, evaluating and analyzing the short-term running state of the low-voltage distribution network according to the loop impedance model; s3, evaluating and analyzing the medium-long-term running state of the low-voltage distribution network according to the loop impedance model; and S4, multi-index evaluation based on the improved radar map. The method realizes the evaluation of the running state of the low-voltage distribution network side, applies the big data technology to the distribution network data, can efficiently and deeply dig out useful values and assist the power grid personnel to make running decisions; the operation effect of the power distribution network is comprehensively evaluated, the implicit rule among the relevant data of the power distribution network is excavated, and visual display is realized.

Description

Low-voltage distribution network running state evaluation method based on data mining
Technical Field
The invention relates to the technical field of operation and maintenance of low-voltage power distribution networks, in particular to a low-voltage power distribution network operation state evaluation method based on data mining.
Background
The power distribution network is one of the important links of power transmission, and the low voltage distribution network is used as a ring connecting the power distribution network and a user side, plays a key role in power services such as electric energy transmission, user response and the like, and has important influence on a main network and the user side in the operation state. The running state of the low-voltage distribution network directly affects the running state of the power grid and the overall economic benefit, so the evaluation of the running state of the low-voltage distribution network is particularly important. The existing state evaluation technology is mainly carried out on medium and high voltage power grids, and common methods comprise a multi-source information fusion method, a fuzzy comprehensive evaluation method, a credibility theory method, a subjective Bayes method, a subjective risk analysis method, a multi-level analysis method and the like.
The operation state evaluation of the low-voltage distribution network is different from that of a medium-high voltage distribution network, the number of user sides of the low-voltage distribution network is large, the structure of the power supply network is complex, the quality of power supply lines is uneven, and devices and lines are frequently abnormal, so that graphs and models of the low-voltage distribution network are difficult to effectively draw and measure, a measurement and acquisition device is not configured on the existing low-voltage distribution line, and an intelligent electric meter is only configured on the user side. Therefore, the evaluation mode of the running state of the low-voltage distribution network is different from that of the medium-high voltage distribution network at present, and the evaluation method of the state of the low-voltage distribution network is less. For example, chinese patent CN111061821A discloses a method and a system for checking topology of a low voltage distribution network based on an improved k-value clustering algorithm; the method has extremely high similarity to the clustered intra-cluster curves, and has a good clustering effect for the data sets with obviously different inter-cluster curves, and the algorithm has a simple structure and is easy to implement. However, due to different initial clustering center selections of different clustering algorithms, the operation state evaluation result is largely related to the initial clustering center point selection, the power distribution network state evaluation result is not unique, and the conditions of non-unique low-voltage power distribution network state evaluation result and low accuracy can occur; when the state of the low-voltage distribution network is evaluated, only qualitative analysis can be performed, and quantitative analysis cannot be performed; and the operation state evaluation result cannot be visually displayed, so that the data is not convenient to understand.
Disclosure of Invention
The invention provides a low-voltage distribution network operation state evaluation method based on data mining, aiming at solving the problems that the operation state evaluation result is related to the selection of the initial clustering center point to a great extent due to different initial clustering center selections of different clustering algorithms, the power distribution network state evaluation result is not unique, the low-voltage distribution network state evaluation result is not unique and the accuracy is not high in the prior art. The method and the device realize the evaluation of the running state of the low-voltage distribution network side, comprehensively evaluate the running effect of the distribution network, dig out the implicit rules among the relevant data of the distribution network and realize visual display.
In order to solve the technical problems, the invention adopts the technical scheme that: a low-voltage distribution network operation state evaluation method based on data mining comprises the following steps:
s1, establishing a low-voltage distribution network loop impedance model according to the collected and processed low-voltage distribution network data;
s2, evaluating and analyzing the short-term running state of the low-voltage distribution network according to the loop impedance model;
s3, evaluating and analyzing the medium-long-term running state of the low-voltage distribution network according to the loop impedance model;
s4, multi-index evaluation based on the improved radar map: selecting evaluation indexes according to the short-term operation state evaluation analysis result and the medium-term operation state evaluation analysis result, calculating the optimal combination weight of each evaluation index of the low-voltage distribution network by using a minimum deviation combination weight method, selecting a new feature vector to construct an evaluation function, and comprehensively evaluating the operation state of the low-voltage distribution system through the evaluation function.
Further, the S1 specifically includes: the loop impedance model of the low-voltage distribution network defines the loop impedance of the intelligent ammeter, and the loop impedance is the sum of the impedances in a loop formed by an upstream live wire, a zero line, a T-connection line and a distribution transformer; the loop impedance is approximately represented by the change rate of the voltage and the current measured by the intelligent ammeter, and is specifically described as follows:
Figure BDA0002876033750000021
Ubi＝Uenb-(RsIsb+Udb+Ufb)-(RlibIlib+RzibIzib)
Uai＝Uena-(RsIsa+Uda+Ufa)-(RliaIlia+RziaIzia)
Figure BDA0002876033750000022
in the above formulaiIs t1,t2Voltage U of two-time intelligent electric meter iiAnd current IiA ratio of rates of change; i isai、IbiAre each t1,t2The current value of the intelligent ammeter i at two moments; u shapeai、UbiAre each t1,t2The voltage value of the intelligent ammeter i at two moments; u shapeda、UdbAre each t1,t2The voltage value of each branch on an upstream fire line of the intelligent ammeter i at two moments; u shapeena，UenbAre each t1,t2The two-time distribution and transformation of the equivalent power supply voltage value of the upstream transmission and distribution network; u shapefa、UfbAre each t1,t2Voltage values of all branches on an upstream zero line of the intelligent ammeter i at two moments; u shapedThe voltage value of each branch on an upstream fire line of the intelligent ammeter i is obtained; u shapefThe voltage value of each branch on an upstream zero line of the intelligent ammeter i is obtained;
the following relation exists among the upstream currents of the intelligent ammeter i:
Figure BDA0002876033750000031
Iithe unit is A, and the current value of the intelligent ammeter i is shown as the current value of the intelligent ammeter i; e is 0,1,2 … …, i-1, and the intelligent electric meter i is at t1,t2The loop impedance at both times is:
Figure BDA0002876033750000032
in the formula, Rawi、RbwiAre respectively an intelligent electric meter i at t1,t2Loop impedance values at two moments are in omega; according to the normal condition, the loop impedance value can not be suddenly changed, so that the method comprises the following steps:
Figure BDA0002876033750000033
Rwithe unit is omega, and the loop impedance value of the intelligent ammeter i is shown in the specification;
when int1,t2Voltage value U of two-time distribution and transformation upstream transmission and distribution network equivalent power supplyena，UenbWhen acquisition is difficult or the transformation is small, the above formula can be simplified as follows:
Figure BDA0002876033750000034
preferably, the short-term operation state evaluation analysis is to evaluate and analyze the power failure, fault, power stealing, line breaking and power restoration conditions of the low-voltage distribution network for 15min to 24 h.
Further, the S2 specifically includes: establishing a short-term evaluation loop impedance array to record loop impedance information of all the intelligent electric meters in the downstream of the distribution transformer, wherein the specific description is as follows:
Figure BDA0002876033750000035
Figure BDA0002876033750000041
in the above formula, Ms is a short-term evaluation loop impedance array; the loop impedance per unit value of one intelligent ammeter at different sampling time points within 1 day is taken for each line, and the sampling time interval is generally 15 min; h is the number of sampling points per day, and is generally 96; rwk is the per unit value of the loop impedance of the Kth sampling point of the intelligent electric meter; k ═ 1,2, … …, h; w is the number of the intelligent electric meter; w is 1,2, … …, m; m is the total number of the intelligent electric meters;
determining an operation state index corresponding to each data point according to the data in the Ms, wherein the operation state index comprises: the running state indexes of the system comprise comprehensive power failure X1, local power failure X2, fault power failure X3, power restoration X4, electricity stealing X6 and line breaking X7.
Further, the step S3 is to evaluate and analyze the number of times of power failure, power failure time, maximum load rate, power stealing capacity and line aging rate of the low-voltage distribution network from 1 month to 1 year.
Further, the S3 specifically includes: establishing a medium-long term operation state information array to record operation state information of all downstream intelligent electric meters of the distribution transformer in a period of time, wherein the specific description is as follows:
Figure BDA0002876033750000042
Mw1the power failure frequency information of the intelligent ammeter w in a period of time is expressed in units of times; mw2The unit is s, which is the total power failure time information of the intelligent electric meter w in a period of time; mw3The method comprises the following steps of obtaining maximum load rate information of an intelligent electric meter w in a period of time, namely the ratio of the maximum current value of the intelligent electric meter in a period of time to the rated current value of a downstream outlet line of the intelligent electric meter; mw4The unit of the electric quantity information is kWh, and the electric quantity information is lost due to electricity stealing of the intelligent electric meter w within a period of time; mw5The method comprises the following steps of providing upstream loop impedance aging information for a smart meter w in a period of time, wherein:
Figure BDA0002876033750000051
Iwmaxthe maximum current value of the intelligent ammeter in a period of time; i iswNThe rated current value is the downstream outlet line rated current value of the intelligent ammeter; rwmaxThe maximum loop impedance value of the intelligent ammeter in a period of time; rwminThe minimum loop impedance value of the intelligent ammeter in a period of time;
the evaluation indexes are specifically described as follows:
Figure BDA0002876033750000052
Figure BDA0002876033750000053
in the above formula, MAw1The power failure frequency information evaluation result of the intelligent electric meter w is obtained; mAw2Obtaining an evaluation result of the total power failure time information of the intelligent electric meter w; mAw3For intelligent electric meterThe maximum load rate information evaluation result of w; mAw4The electric quantity information evaluation result is the electricity stealing quantity information evaluation result of the intelligent electric meter w; mAw5Obtaining a loop impedance aging information evaluation result of the intelligent ammeter w; sw1The standard power failure times set for the intelligent electric meter w can be determined by the maintenance plan power failure times and the regional power supply reliability standard; sw2The standard power failure time set for the intelligent electric meter w is s and can be determined by the maintenance plan power failure time and the regional power supply reliability standard; sw3Setting a standard maximum load rate for the intelligent electric meter w; sw4The unit of the load electricity consumption measured by the intelligent electric meter w is kWh; sw5And setting a standard loop impedance aging rate for the intelligent electric meter w.
Further, the S4 includes the following steps:
s41, establishing an evaluation index system, and carrying out standardization processing on each index in the evaluation index system;
s42, replacing the original triangular area with the sector area, and calculating included angle values between the index shafts by using a minimum deviation combined weight method;
s43, drawing an improved radar map by taking the 95% probability large value of short-term and medium-term data as a data typical value, selecting the area and the perimeter of the improved radar map as characteristic values, constructing characteristic vectors according to the characteristic values, constructing an evaluation function according to the characteristic vectors, and carrying out quantitative scoring evaluation on the short-term and medium-term operation states of the low-voltage distribution network by using the evaluation function.
Further, the step S41 is specifically that, in the short-term operation state evaluation, an operation state index is selected: general power failure X1Local power failure X2Fault power failure X3Full-line complex electricity X4All the complex electricity X5Steal electricity X6And broken line X7(ii) a Selecting the operation state indexes of the intelligent electric meter w during the middle-long term operation state evaluation: power failure times information MAw1Total power off time information MAw2Maximum load factor information MAw3Information M of electric quantity of electricity stealingAw4And loop impedance aging information MAw5；
The method comprises the following steps of standardizing index data of short-term and medium-term running states of the low-voltage distribution network as follows:
Figure BDA0002876033750000061
Figure BDA0002876033750000062
in the above formula, Xk"is the value of the k-th item of data after being normalized; xkA measurement value representing the k-th index; xkmaxA maximum value representing the k-th index; mAwt"is the value after the t data is normalized; mAwtA measurement value representing the t-th index; mAwtmaxRepresents the maximum value of the t-th index.
Further, the S42 specifically includes: assuming that a decision maker determines the weight of each index by using q methods in common, wherein p methods are subjectively weighted and q-p methods are objectively weighted, the weight vectors corresponding to the short-term and medium-term indexes obtained by the jth method are:
uj＝(uj1,uj2,uj3,uj4,uj5,uj6,uj7),(j＝1,2,L,q)
dj＝(dj1,dj2,dj3,dj4,dj5),(j＝1,2,L,q)
where Σ ujk＝1,(k＝1,2L,7)，∑djt＝1,(t＝1,2L,5)；
The weighting coefficients of the jth weighting method of the short-term and medium-term evaluation indexes are respectively set as alphaj、λjThen, the weight vectors of different weighting methods can be recorded as:
Figure BDA0002876033750000063
introducing a deviation function to ensure that the weight deviation of various weighting methods takes a minimum value, and constructing an optimization model as follows:
Figure BDA0002876033750000064
Figure BDA0002876033750000071
in the formula, J represents a deviation function corresponding to a short-term operation state of the power distribution network, and L represents a deviation function corresponding to a medium-term operation state and a long-term operation state of the power distribution network;
and simultaneously, calculating the minimum deviation combination weight vector W (W is the minimum deviation combination weight vector W of each index of the power distribution network in short term and medium term by constructing a corresponding Lagrange function and according to the extreme value existing necessary conditions and the Cramer rule1,w2,w3,w4,w5,w6,w7)、N＝(n1,n2,n3,n4,n5)；
State evaluation index { X) of short-term operation of low-voltage distribution network1,X2,X3,X4,X5,X6,X7The corresponding weight vector is W ═ W (W)1,w2,w3,w4,w5,w6,w7) Then, the central angle vector θ of the sector area corresponding to the 7-term index is 2 pi W (θ)1,θ2,θ3,θ4,θ5,θ6,θ7) (ii) a Wherein the k-th index corresponds to the sector central angle thetak＝2πwkTaking a radial as a reference axis of a first index along the vertical direction with the center of a circle as a starting point, and taking a first normalized index value X on the reference axis1' is a radius, in theta1＝2πw1A sector area is formed in the anticlockwise direction of the circle center, namely the representative area of the first index, and the representative area of each index is formed in sequence in the same way;
state evaluation index { M) for long-term operation in low-voltage distribution networkAwt1,MAwt2,MAwt3,MAwt4,MAwt5The corresponding weight vector is N ═ N (N)1,n2,n3,n4,n5) Fan corresponding to 5 indexesThe shape region center angle vector θ '2 pi N ═ θ'1,θ'2,θ'3,θ'4,θ'5) (ii) a Wherein the t index corresponds to a fan-shaped central angle theta't＝2πwtTaking a radial as a reference axis of the first index along the vertical direction with the center of circle as a starting point, and taking a first normalized index value M on the reference axisAw1Is a radius of theta'1＝2πw1A sector area is made in the anticlockwise direction as the circle center, namely a representative area of the first index; similarly, the representative region of each index is made in turn.
Further, the S43 specifically includes: using the 95% probability large value of the short-term evaluation analysis data and the medium-term evaluation analysis data as a data typical value to draw an improved radar map, and selecting the area and the perimeter of the improved radar map as characteristic values; are respectively marked as Si、Li(i is 1,2), i is the number of improved radar maps, and the area evaluation value SiThe perimeter estimate L is the sum of all sector areas in the ith radar chartiThe sum of the arc lengths of all sectors in the ith radar chart is as follows:
Figure BDA0002876033750000072
Figure BDA0002876033750000081
in the formula, k is the total number of short-term state evaluation indexes and the total number of medium-term and long-term state evaluation indexes; according to SiAnd LiConfigurable feature vector D ═ Di1,Di2]；
Figure BDA0002876033750000082
In the formula, Di1Area evaluation value S representing ith radar chartiThe ratio to the maximum area; di2Indicates the ith circumference evaluation value LiThe ratio to the circumference under the same area; and Di1、Di2∈(0，1](ii) a Taking a feature vector Di1、Di2As an evaluation function, i.e. there are:
Figure BDA0002876033750000083
and carrying out quantitative scoring evaluation on the short-term running state and the medium-term running state of the low-voltage distribution network by using an evaluation function.
Compared with the prior art, the beneficial effects are:
according to the low-voltage distribution network simplified graph, a low-voltage loop impedance model is established; then, based on a loop impedance model, establishing a short-term evaluation loop impedance array for short-term operation state evaluation and a medium-term and long-term operation state information array for medium-term and long-term operation state evaluation; and finally, the improved radar maps in the short-term running state and the medium-term running state are obtained respectively through the comprehensive evaluation mode of the improved radar maps, multi-dimensional data are displayed visually and comprehensively, and the overall evaluation of the low-voltage distribution network is completed. The method and the device realize the evaluation of the running state of the low-voltage distribution network side, apply the big data technology to the distribution network data, can efficiently and deeply dig out useful values and assist the power network personnel to make running decisions. The operation effect of the power distribution network is comprehensively evaluated, the implicit rule among the relevant data of the power distribution network is excavated, and visual display is realized.
Drawings
Fig. 1 is a schematic block diagram of the present invention.
Fig. 2 is a simplified diagram of a low voltage distribution network according to the present invention.
FIG. 3 is an exemplary axial angle for short term operation in accordance with the present invention.
FIG. 4 is an angle between index axes in the middle and long term operation states of the present invention.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent; for the purpose of better illustrating the embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted. The positional relationships depicted in the drawings are for illustrative purposes only and are not to be construed as limiting the present patent.
The embodiment provides a low-voltage distribution network operation state evaluation method based on data mining, as shown in fig. 1, including the following steps:
s1, establishing a low-voltage distribution network loop impedance model according to the collected and processed low-voltage distribution network data;
s2, evaluating and analyzing the short-term running state of the low-voltage distribution network according to the loop impedance model;
s3, evaluating and analyzing the medium-long-term running state of the low-voltage distribution network according to the loop impedance model;
s4, multi-index evaluation based on the improved radar map: selecting evaluation indexes according to the short-term operation state evaluation analysis result and the medium-term operation state evaluation analysis result, calculating the optimal combination weight of each evaluation index of the low-voltage distribution network by using a minimum deviation combination weight method, selecting a new feature vector to construct an evaluation function, and comprehensively evaluating the operation state of the low-voltage distribution system through the evaluation function.
The specific evaluation method comprises the following steps:
s1, establishing a low-voltage distribution network loop impedance model
The loop impedance model of the low-voltage distribution network defines loop impedance of the intelligent electric meter, and the loop impedance is the sum of impedances in a loop formed by an upstream live wire, a zero line, a T-connection line and a distribution transformer. Meanwhile, the low-voltage distribution network loop impedance model provides a basis for subsequently establishing a short-term evaluation loop impedance array and a medium-and-long-term operation state information array. Since the impedance parameters of the line are fixed, the impedance value of the loop does not change abruptly under normal operating conditions. If the impedance of a certain power supply loop changes suddenly, the loop is abnormal, and the running state of the low-voltage distribution network is changed. The current intelligent electric meter has a remote meter reading function, and can acquire information such as system voltage, current, power and electric quantity in real time. Fig. 2 is a simplified diagram of a low-voltage distribution network, and it can be seen that a distribution transformer, a zero line, a T-junction line, an intelligent electric meter and a live line form an upstream power supply loop of the intelligent electric meter.
The loop impedance can be approximately represented by the change rate of the voltage and the current measured by the intelligent ammeter, and is specifically described as follows:
Figure BDA0002876033750000091
Ubi＝Uenb-(RsIsb+Udb+Ufb)-(RlibIlib+RzibIzib)
Uai＝Uena-(RsIsa+Uda+Ufa)-(RliaIlia+RziaIzia)
Figure BDA0002876033750000101
in the above formula, λiIs t1,t2Voltage U of two-time intelligent electric meter iiAnd current IiA ratio of rates of change; i isai、IbiAre each t1,t2The current value of the intelligent ammeter i at two moments; u shapeai、UbiAre each t1,t2The voltage value of the intelligent ammeter i at two moments; u shapeda、UdbAre each t1,t2The voltage value of each branch on an upstream fire line of the intelligent ammeter i at two moments; u shapeena，UenbAre each t1,t2The two-time distribution and transformation of the equivalent power supply voltage value of the upstream transmission and distribution network; u shapefa、UfbAre each t1,t2Voltage values of all branches on an upstream zero line of the intelligent ammeter i at two moments; u shapedThe voltage value of each branch on an upstream fire line of the intelligent ammeter i is obtained; u shapefAnd the voltage value of each branch on the upstream zero line of the intelligent electric meter i is obtained.
The following relation exists among the upstream currents of the intelligent ammeter i:
Figure BDA0002876033750000102
Iithe unit is A, and the current value of the intelligent ammeter i is shown as the current value of the intelligent ammeter i; e is 0,1,2 … …, i-1, and the intelligent electric meter i is at t1,t2The loop impedance at both times is:
Figure BDA0002876033750000103
in the formula, Rawi、RbwiAre respectively an intelligent electric meter i at t1,t2The loop impedance values at both times are in units of omega.
Because the loop impedance value can not suddenly change under the normal condition, so have:
Figure BDA0002876033750000104
Rwiand the unit is omega, and the loop impedance value of the intelligent ammeter i is shown.
When at t1,t2Voltage value U of two-time distribution and transformation upstream transmission and distribution network equivalent power supplyena，UenbWhen acquisition is difficult or the transformation is small, the above formula can be simplified as follows:
Figure BDA0002876033750000105
s2, evaluating and analyzing the short-term operation state of the low-voltage distribution network according to the loop impedance model
The short-term operation state evaluation mainly evaluates and analyzes conditions of power failure, faults, electricity stealing, line breaking, power restoration and the like of the low-voltage distribution network from 15min to 24 h. Establishing a short-term evaluation loop impedance array to record loop impedance information of all the intelligent electric meters in the downstream of the distribution transformer, wherein the specific description is as follows:
Figure BDA0002876033750000111
Figure BDA0002876033750000112
Msevaluating a loop impedance array for a short term; the loop impedance per unit value of one intelligent ammeter at different sampling time points within 1 day is taken for each line, and the sampling time interval is generally 15 min; h is the number of sampling points per day, and is generally 96; rwkThe per unit value of the loop impedance of the Kth sampling point of the intelligent ammeter is obtained; k ═ 1,2, … …, h; w is the number of the intelligent electric meter; w is 1,2, … …, m; and m is the total number of the intelligent electric meters. Determining an operation state index corresponding to each data point according to the data in the Ms, specifically:
if M issAll the K-th line data are-1, and all the K-1-th line data are non-1 (the 0-th line data are defined to be non-1), it means that the distribution transformation has a total power failure X at the time of the K-th data point1；
If M issWhere column K is-1 and all column K-1 data is non-1 (column 0 data is defined as non-1), it indicates that the distribution transformation has a local power failure X at the time of the K-th data point2(ii) a If the outage area is not contained, then the outage event is a fault outage X3；
If M issThe Kth line of data is all-1, and the K +1 th line of data is all non-1, which means that the distribution transformation changes from full-line power failure to full-line power restoration X at the K-th data point4；
If M issThe data part of the K-th line is-1, and the data of the K + 1-th lines are all non-1, which means that the distribution transformation changes from partial power failure to full power restoration X at the time of the K-th data point5；
If the loop impedance per unit value RwkAnd a preceding plurality of data Rw,k-1,Rw,k-2,……Rw,k-p+1When the unit value is-4, the fact that the downstream load voltage of the intelligent ammeter is changed and the current is unchanged in the duration is shown, the situation can be that the current transformer of the intelligent ammeter is in short circuit through fixed impedance or the current coil of the intelligent ammeter is in short circuit through fixed impedance, and the fact that electricity stealing X is possible is shown6A situation occurs;
if the loop impedance of the intelligent ammeter exceeds the alarm limit value and exceeds the alarm number standard value SATime, represents the broken line X caused by line aging7The situation exists in an upstream line of the intelligent electric meter and needs to be overhauled in time; the concrete description is as follows:
Figure BDA0002876033750000121
in the formula, LAThe alarm limit value for line impedance aging can be generally determined by 5-10 times of the impedance of a normal loop; a. thekThe loop impedance exceeds an alarm limit number, wherein K is 1,2, … … h;
if the loop impedance per unit value RwkThe conditions of 1) -6) are not met, and the condition indicates that the low-voltage distribution network corresponding to the intelligent electric meter is in a normal operation state.
S3, evaluating and analyzing the medium-long term running state of the low-voltage distribution network according to the loop impedance model
The evaluation of the medium-long term operation state mainly comprises the evaluation and analysis of the power failure times, power failure time, maximum load rate, electricity stealing capacity, line aging rate and the like of the low-voltage distribution network from 1 month to 1 year. Establishing a medium-long term operation state information array to record operation state information of all downstream intelligent electric meters of the distribution transformer in a period of time, wherein the specific description is as follows:
Figure BDA0002876033750000122
Mw1the power failure frequency information of the intelligent ammeter w in a period of time is expressed in units of times; mw2The unit is s, which is the total power failure time information of the intelligent electric meter w in a period of time; mw3The method comprises the following steps of obtaining maximum load rate information of an intelligent electric meter w in a period of time, namely the ratio of the maximum current value of the intelligent electric meter in a period of time to the rated current value of a downstream outlet line of the intelligent electric meter; mw4The unit of the electric quantity information is kWh, and the electric quantity information is lost due to electricity stealing of the intelligent electric meter w within a period of time; mw5The method comprises the following steps of providing upstream loop impedance aging information for a smart meter w in a period of time, wherein:
Figure BDA0002876033750000123
Iwmaxthe maximum current value of the intelligent ammeter in a period of time; i iswNThe rated current value is the downstream outlet line rated current value of the intelligent ammeter; rwmaxThe maximum loop impedance value of the intelligent ammeter in a period of time; rwminThe minimum loop impedance value of the intelligent electric meter in a period of time.
The evaluation indexes are specifically described as follows:
Figure BDA0002876033750000131
Figure BDA0002876033750000132
MAw1the power failure frequency information evaluation result of the intelligent electric meter w is obtained; mAw2Obtaining an evaluation result of the total power failure time information of the intelligent electric meter w; mAw3The maximum load rate information evaluation result of the intelligent electric meter w is obtained; mAw4The electric quantity information evaluation result is the electricity stealing quantity information evaluation result of the intelligent electric meter w; mAw5And obtaining the loop impedance aging information evaluation result of the intelligent electric meter w. Sw1The standard power failure times set for the intelligent electric meter w can be determined by the maintenance plan power failure times and the regional power supply reliability standard; sw2The standard power failure time set for the intelligent electric meter w is s and can be determined by the maintenance plan power failure time and the regional power supply reliability standard; sw3Setting a standard maximum load rate for the intelligent electric meter w; sw4The unit of the load electricity consumption measured by the intelligent electric meter w is kWh; sw5And setting a standard loop impedance aging rate for the intelligent electric meter w.
S4, multi-index evaluation method based on improved radar map
The multi-index evaluation method based on the improved radar map utilizes a minimum deviation combination weight method to calculate the optimal combination weight of each evaluation index of low pressure. Compared with the traditional radar mapping method, the improved radar mapping method further improves the determination of the included angle value between the index axes and the selection of the characteristic vector. Firstly, a sector area is adopted to replace the original triangular area, and the included angle value between the index axes is calculated by utilizing a minimum deviation combined weight method. And then selecting a new feature vector to construct an evaluation function, and comprehensively evaluating the running state of the low-voltage distribution system. Compared with the traditional radar map, the improved radar map method solves the problems of information sharing among indexes and non-unique evaluation result in the traditional radar map method, and enables the evaluation result to be more accurate and reasonable
S41, establishing an evaluation index system, and standardizing each index in the evaluation index system
When the running state of the power distribution system is comprehensively evaluated, the selection of the evaluation index is very important. According to the actual operation condition of the power distribution system, selecting a comprehensive power failure X in short-term operation state evaluation1Local power failure X2Fault power failure X3X for recovery of electricity4X for avoiding or indicating fraudulent use of electricity6And broken wire X7Waiting for important operation state indexes; selecting power failure frequency information M of intelligent electric meter w during medium and long-term running state evaluationAw1Total power off time information MAw2Maximum load factor information MAw3Information M of electric quantity of electricity stealingAw4Loop impedance aging information MAw5And waiting for important operation state indexes.
Before comprehensive evaluation of each index, in order to eliminate the influence of different dimensions or different orders of magnitude among different indexes, each index needs to be standardized. The data of the short-term and medium-term running states of the low-voltage distribution network can be standardized as follows:
Figure BDA0002876033750000141
Figure BDA0002876033750000142
in the above formula, Xk"is the value of the k-th item of data after being normalized; xkA measurement value representing the k-th index; xkmaxA maximum value representing the k-th index; mAwt"is the value after the t data is normalized; mAwtA measurement value representing the t-th index; mAwtmaxRepresents the maximum value of the t-th index. .
S42, replacing the original triangular area with the sector area, and calculating included angle values between the index shafts by using a minimum deviation combined weight method;
the minimum deviation combined weight method combines the main weight and the objective weight according to the principle of minimum deviation; the specific calculation steps are as follows:
assuming that a decision maker determines the weight of each index by using q methods in common, wherein p methods are subjectively weighted and q-p methods are objectively weighted, the weight vectors corresponding to the short-term and medium-term indexes obtained by the jth method are:
uj＝(uj1,uj2,uj3,uj4,uj5,uj6,uj7),(j＝1,2,L,q)
dj＝(dj1,dj2,dj3,dj4,dj5),(j＝1,2,L,q)
where Σ ujk＝1,(k＝1,2L,7)，∑djt＝1,(t＝1,2L,5)。
The weighting coefficients of the jth weighting method of the short-term and medium-term evaluation indexes are respectively set as alphaj、λjThen, the weight vectors of different weighting methods can be recorded as:
Figure BDA0002876033750000143
Figure BDA0002876033750000151
in order to comprehensively consider the subjective opinion of a decision maker and the objectivity of decision, a deviation function is introduced, so that the weight deviation of various weighting methods takes a minimum value, and an optimization model is constructed as follows:
Figure BDA0002876033750000152
Figure BDA0002876033750000153
in the formula, J represents a deviation function corresponding to a short-term operation state of the power distribution network, and L represents a deviation function corresponding to a medium-term operation state and a long-term operation state of the power distribution network;
meanwhile, by constructing a corresponding Lagrange function and according to the extreme value, the minimum deviation combination weight vector W (W is the minimum deviation combination weight vector W) of each index of the power distribution network in short term and medium term can be calculated according to the existence of necessary conditions and the Cramer rule of extreme values1,w2,w3,w4,w5,w6,w7)、N＝(n1,n2,n3,n4,n5)。
As shown in FIG. 3, the state evaluation index { X ] of the short-term operation of the low-voltage distribution network1,X2,X3,X4,X5,X6,X7The corresponding weight vector is W ═ W (W)1,w2,w3,w4,w5,w6,w7) Then, the central angle vector θ of the sector area corresponding to the 7-term index is 2 pi W (θ)1,θ2,θ3,θ4,θ5,θ6,θ7) (ii) a Wherein the k-th index corresponds to the sector central angle thetak＝2πwkTaking a radial as a reference axis of a first index along the vertical direction with the center of a circle as a starting point, and taking a first normalized index value X on the reference axis1' is a radius, in theta1＝2πw1And (4) making a sector area in the anticlockwise direction of the center of the circle, namely making the representative area of the first index, and sequentially making the representative area of each index in the same way.
Similarly, as shown in fig. 4, the state evaluation index { M) of long-term operation in the low-voltage distribution networkAwt1,MAwt2,MAwt3,MAwt4,MAwt5Couple (c)The weight vector should be N ═ N (N)1,n2,n3,n4,n5) Then, the fan-shaped area central angle vector θ 'corresponding to the 5-item index is 2 pi N (θ'1,θ'2,θ'3,θ'4,θ'5). Wherein the t index corresponds to a fan-shaped central angle theta't＝2πwtTaking a radial as a reference axis of the first index along the vertical direction with the center of circle as a starting point, and taking a first normalized index value M on the reference axisAw1Is a radius of theta'1＝2πw1A sector area is made in the anticlockwise direction as the circle center, namely a representative area of the first index; similarly, the representative region of each index is made in turn.
S43, drawing an improved radar map by taking the 95% probability large value of short-term and medium-term data as a data typical value, selecting the area and the perimeter of the improved radar map as characteristic values, constructing characteristic vectors according to the characteristic values, constructing an evaluation function according to the characteristic vectors, and carrying out quantitative scoring evaluation on the short-term and medium-term operation states of the low-voltage distribution network by using the evaluation function.
The key of comprehensively evaluating the running state of the low-voltage power distribution system by utilizing an improved radar map method is the construction of an evaluation function, a 95% probability large value of short-term and medium-term data is taken as a data typical value to draw an improved radar map, and the area and the perimeter of the improved radar map are selected as characteristic values; are respectively marked as Si、Li(i is 1,2), i is the number of improved radar maps, and the area evaluation value SiThe perimeter estimate L is the sum of all sector areas in the ith radar chartiThe sum of the arc lengths of all sectors in the ith radar chart is as follows:
Figure BDA0002876033750000161
Figure BDA0002876033750000162
and k is the total number of the short-term state evaluation indexes and the total number of the medium-term and long-term state evaluation indexes. According to SiAnd LiConfigurable feature vector D ═ Di1,Di2]；
Figure BDA0002876033750000163
The definition of the evaluation vector is known, Di1Area evaluation value S representing ith radar chartiThe ratio to the maximum area; di2Indicates the ith circumference evaluation value LiThe ratio to the circumference under the same area; and Di1、Di2∈(0，1]. Taking a feature vector Di1、Di2As an evaluation function, i.e. there are:
Figure BDA0002876033750000164
and quantitative scoring evaluation can be respectively carried out on the short-term running state and the medium-term running state of the low-voltage distribution network by utilizing the finally obtained evaluation function.
In this embodiment, based on the CRISP-DM model principle, the external data of the power distribution network, such as the operation data and meteorological data inside the low-voltage power distribution network, is collected from the production scheduling management system, the power marketing system, the equipment asset management system, the power distribution automation system, and the power consumption collection system. The low-voltage distribution network has abundant data sources, and can be divided into internal data and external data of the distribution network according to whether the data are generated by the distribution network. Internal operation data can be collected from different informatization systems, and the internal operation data comprises a production scheduling management system, an electric power marketing system, an equipment asset management system, a power distribution automation system, an electricity utilization collection system and the like. Fault power failure data, unplanned power failure data and planned power failure data can be extracted from the production scheduling management system; extracting terminal online rate, remote signaling data, remote control data and feeder automation data from a distribution network automation system; the slave power consumption acquisition system can acquire data such as heavy overload, light and no load of distribution transformer, low voltage of user, overvoltage of user and the like of the power distribution network; and the external data of the power distribution network mainly comprises meteorological data, power distribution corridor data, regional economic development data, traffic line data and the like. Because the transmission height of the power distribution network is low, the population of the area is generally dense, the traffic condition is complex, the positions of the power distribution equipment are scattered, and although the external data are not generated by the power distribution network, the external data are closely related to the safe operation of the power distribution network. The collected data is then pre-processed, including data cleansing and data merging. The data cleaning needs to be started from the aspects of name normalization, time field normalization and logic judgment, related data unit normalization, sampling data processing and null value data processing. After data is cleaned, data needs to be merged for convenience of data mining. After the data is merged, the new data table contains specified fields such as line name, start time, end time, running index, etc.
When the running state of the low-voltage distribution network is evaluated, a low-voltage loop impedance model is established according to the simplified graph of the low-voltage distribution network. Then, establishing a short-term evaluation loop impedance array for short-term running state evaluation based on a loop impedance model, wherein the short-term evaluation loop impedance array is mainly used for evaluating and analyzing conditions of faults, power failure, electricity stealing, line breaking, power restoration and the like of the low-voltage distribution network for 15min to 24 h; establishing a medium-long term operation state information array for medium-long term operation state evaluation, wherein the medium-long term operation state evaluation is mainly used for evaluating and analyzing the power failure times, power failure time, maximum load rate, electricity stealing capacity, line aging rate and the like of the low-voltage distribution network for 1 month to 1 year; and finally, the improved radar maps in the short-term running state and the medium-term running state are obtained respectively through the comprehensive evaluation mode of the improved radar maps, multi-dimensional data are displayed visually and comprehensively, and the overall evaluation of the low-voltage distribution network is completed. According to the method and the device, the running state of the low-voltage power distribution network side is evaluated, the big data technology is applied to the power distribution network data, the useful value can be efficiently and deeply excavated, and the power grid personnel are assisted to make running decisions. The operation effect of the power distribution network is comprehensively evaluated, the implicit rule among the relevant data of the power distribution network is excavated, and visual display is realized.
It should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.

Claims (10)

1. A low-voltage distribution network operation state evaluation method based on data mining is characterized by comprising the following steps:
s1, establishing a low-voltage distribution network loop impedance model according to the collected and processed low-voltage distribution network data;
s2, evaluating and analyzing the short-term running state of the low-voltage distribution network according to the loop impedance model;
s3, evaluating and analyzing the medium-long-term running state of the low-voltage distribution network according to the loop impedance model;
s4, multi-index evaluation based on the improved radar map: selecting evaluation indexes according to the short-term operation state evaluation analysis result and the medium-term operation state evaluation analysis result, calculating the optimal combination weight of each evaluation index of the low-voltage distribution network by using a minimum deviation combination weight method, selecting a new feature vector to construct an evaluation function, and comprehensively evaluating the operation state of the low-voltage distribution system through the evaluation function.
2. The method for evaluating the operation state of the low-voltage distribution network based on data mining according to claim 1, wherein the step S1 is specifically as follows: the loop impedance model of the low-voltage distribution network defines the loop impedance of the intelligent ammeter, and the loop impedance is the sum of the impedances in a loop formed by an upstream live wire, a zero line, a T-connection line and a distribution transformer; the loop impedance is approximately represented by the change rate of the voltage and the current measured by the intelligent ammeter, and is specifically described as follows:
Figure FDA0002876033740000011
Ubi＝Uenb-(RsIsb+Udb+Ufb)-(RlibIlib+RzibIzib)
Uai＝Uena-(RsIsa+Uda+Ufa)-(RliaIlia+RziaIzia)
Figure FDA0002876033740000012
in the above formulaiIs t1,t2Voltage U of two-time intelligent electric meter iiAnd current IiA ratio of rates of change; i isai、IbiAre each t1,t2The current value of the intelligent ammeter i at two moments; u shapeai、UbiAre each t1,t2The voltage value of the intelligent ammeter i at two moments; u shapeda、UdbAre each t1,t2The voltage value of each branch on an upstream fire line of the intelligent ammeter i at two moments; u shapeena，UenbAre each t1,t2The two-time distribution and transformation of the equivalent power supply voltage value of the upstream transmission and distribution network; u shapefa、UfbAre each t1,t2Voltage values of all branches on an upstream zero line of the intelligent ammeter i at two moments; u shapedThe voltage value of each branch on an upstream fire line of the intelligent ammeter i is obtained; u shapefThe voltage value of each branch on an upstream zero line of the intelligent ammeter i is obtained;
the following relation exists among the upstream currents of the intelligent ammeter i:
Figure FDA0002876033740000021
Iithe unit is A, and the current value of the intelligent ammeter i is shown as the current value of the intelligent ammeter i; e is 0,1,2 … …, i-1, and the intelligent electric meter i is at t1,t2The loop impedance at both times is:
Figure FDA0002876033740000022
in the formula, Rawi、RbwiAre respectively an intelligent electric meter i at t1,t2Loop impedance values at two moments are in omega;
according to the normal condition, the loop impedance value can not be suddenly changed, so that the method comprises the following steps:
Figure FDA0002876033740000023
Rwithe unit is omega, and the loop impedance value of the intelligent ammeter i is shown in the specification;
when at t1,t2Voltage value U of two-time distribution and transformation upstream transmission and distribution network equivalent power supplyena，UenbWhen acquisition is difficult or the transformation is small, the above formula can be simplified as follows:
Figure FDA0002876033740000024
3. the data mining-based low-voltage distribution network operation state evaluation method according to claim 1, wherein the short-term operation state evaluation analysis is an evaluation analysis of power failure, fault, power stealing, line breaking and power restoration conditions of the low-voltage distribution network for 15min to 24 h.
4. The method for evaluating the operation state of the low-voltage distribution network based on the data mining as claimed in claim 3, wherein the step S2 is specifically as follows: establishing a short-term evaluation loop impedance array to record loop impedance information of all the intelligent electric meters in the downstream of the distribution transformer, wherein the specific description is as follows:
Figure FDA0002876033740000025
Figure FDA0002876033740000031
in the above formula, Ms is a short-term evaluation loop impedance array; the loop impedance per unit value of one intelligent ammeter at different sampling time points within 1 day is taken for each line, and the sampling time interval is generally 15 min; h is the number of sampling points per day, and is generally 96; rwk is the per unit value of the loop impedance of the Kth sampling point of the intelligent electric meter; k ═ 1,2, … …, h; w is the number of the intelligent electric meter; w is 1,2, … …, m; m is the total number of the intelligent electric meters;
determining an operation state index corresponding to each data point according to the data in the Ms, wherein the operation state index comprises: the running state indexes of the system comprise comprehensive power failure X1, local power failure X2, fault power failure X3, power restoration X4, electricity stealing X6 and line breaking X7.
5. The method for evaluating the operation state of the low-voltage distribution network based on the data mining as claimed in claim 4, wherein the step S3 is implemented by evaluating and analyzing the power failure times, power failure time, maximum load rate, power stealing capacity and line aging rate of the low-voltage distribution network from 1 month to 1 year.
6. The method for evaluating the operation state of the low-voltage distribution network based on the data mining as claimed in claim 4, wherein the step S3 is specifically as follows: establishing a medium-long term operation state information array to record operation state information of all downstream intelligent electric meters of the distribution transformer in a period of time, wherein the specific description is as follows:
Figure FDA0002876033740000032
Mw1the power failure frequency information of the intelligent ammeter w in a period of time is expressed in units of times; mw2The unit is s, which is the total power failure time information of the intelligent electric meter w in a period of time; mw3The method comprises the following steps of obtaining maximum load rate information of an intelligent electric meter w in a period of time, namely the ratio of the maximum current value of the intelligent electric meter in a period of time to the rated current value of a downstream outlet line of the intelligent electric meter; mw4Is an intelligent ammeter w is in oneThe unit of the electric quantity information lost due to electricity stealing in a period of time is kWh; mw5The method comprises the following steps of providing upstream loop impedance aging information for a smart meter w in a period of time, wherein:
Figure FDA0002876033740000041
Iwmaxthe maximum current value of the intelligent ammeter in a period of time; i iswNThe rated current value is the downstream outlet line rated current value of the intelligent ammeter; rwmaxThe maximum loop impedance value of the intelligent ammeter in a period of time; rwminThe minimum loop impedance value of the intelligent ammeter in a period of time;
the evaluation indexes are specifically described as follows:
Figure FDA0002876033740000042
Figure FDA0002876033740000043
in the above formula, MAw1The power failure frequency information evaluation result of the intelligent electric meter w is obtained; mAw2Obtaining an evaluation result of the total power failure time information of the intelligent electric meter w; mAw3The maximum load rate information evaluation result of the intelligent electric meter w is obtained; mAw4The electric quantity information evaluation result is the electricity stealing quantity information evaluation result of the intelligent electric meter w; mAw5Obtaining a loop impedance aging information evaluation result of the intelligent ammeter w; sw1The standard power failure times set for the intelligent electric meter w can be determined by the maintenance plan power failure times and the regional power supply reliability standard; sw2The standard power failure time set for the intelligent electric meter w is s and can be determined by the maintenance plan power failure time and the regional power supply reliability standard; sw3Setting a standard maximum load rate for the intelligent electric meter w; sw4The unit of the load electricity consumption measured by the intelligent electric meter w is kWh; sw5And setting a standard loop impedance aging rate for the intelligent electric meter w.
7. The method for evaluating the operation state of the low-voltage distribution network based on the data mining as claimed in claim 6, wherein the step of S4 comprises the steps of:
s41, establishing an evaluation index system, and carrying out standardization processing on each index in the evaluation index system;
s42, replacing the original triangular area with the sector area, and calculating included angle values between the index shafts by using a minimum deviation combined weight method;
s43, drawing an improved radar map by taking the 95% probability large value of short-term and medium-term data as a data typical value, selecting the area and the perimeter of the improved radar map as characteristic values, constructing characteristic vectors according to the characteristic values, constructing an evaluation function according to the characteristic vectors, and carrying out quantitative scoring evaluation on the short-term and medium-term operation states of the low-voltage distribution network by using the evaluation function.
8. The method for evaluating the operation state of the low-voltage distribution network based on the data mining as claimed in claim 7, wherein S41 is specifically that the operation state indexes are selected during the short-term operation state evaluation: general power failure X1Local power failure X2Fault power failure X3Full-line complex electricity X4All the complex electricity X5Steal electricity X6And broken line X7(ii) a Selecting the operation state indexes of the intelligent electric meter w during the middle-long term operation state evaluation: power failure times information MAw1Total power off time information MAw2Maximum load factor information MAw3Information M of electric quantity of electricity stealingAw4And loop impedance aging information MAw5；
The method comprises the following steps of standardizing index data of short-term and medium-term running states of the low-voltage distribution network as follows:
Figure FDA0002876033740000051
Figure FDA0002876033740000052
in the above formula, Xk"is the value of the k-th item of data after being normalized; xkA measurement value representing the k-th index; xkmaxA maximum value representing the k-th index; mAwt"is the value after the t data is normalized; mAwtA measurement value representing the t-th index; mAwtmaxRepresents the maximum value of the t-th index.
9. The method for evaluating the operation state of the low-voltage distribution network based on the data mining as claimed in claim 8, wherein the step S42 is specifically as follows: assuming that a decision maker determines the weight of each index by using q methods in common, wherein p methods are subjectively weighted and q-p methods are objectively weighted, the weight vectors corresponding to the short-term and medium-term indexes obtained by the jth method are:
uj＝(uj1,uj2,uj3,uj4,uj5,uj6,uj7),(j＝1,2,L,q)
dj＝(dj1,dj2,dj3,dj4,dj5),(j＝1,2,L,q)
where Σ ujk＝1,(k＝1,2L,7)，∑djt＝1,(t＝1,2L,5)；
The weighting coefficients of the jth weighting method of the short-term and medium-term evaluation indexes are respectively set as alphaj、λjThen, the weight vectors of different weighting methods can be recorded as:
Figure FDA0002876033740000053
introducing a deviation function to ensure that the weight deviation of various weighting methods takes a minimum value, and constructing an optimization model as follows:
Figure FDA0002876033740000061
Figure FDA0002876033740000062
in the formula, J represents a deviation function corresponding to a short-term operation state of the power distribution network, and L represents a deviation function corresponding to a medium-term operation state and a long-term operation state of the power distribution network;
and simultaneously, calculating the minimum deviation combination weight vector W (W is the minimum deviation combination weight vector W of each index of the power distribution network in short term and medium term by constructing a corresponding Lagrange function and according to the extreme value existing necessary conditions and the Cramer rule1,w2,w3,w4,w5,w6,w7)、N＝(n1,n2,n3,n4,n5)；
State evaluation index { X) of short-term operation of low-voltage distribution network1,X2,X3,X4,X5,X6,X7The corresponding weight vector is W ═ W (W)1,w2,w3,w4,w5,w6,w7) Then, the central angle vector θ of the sector area corresponding to the 7-term index is 2 pi W (θ)1,θ2,θ3,θ4,θ5,θ6,θ7) (ii) a Wherein the k-th index corresponds to the sector central angle thetak＝2πwkTaking a radial as a reference axis of a first index along the vertical direction with the center of a circle as a starting point, and taking a first normalized index value X on the reference axis1' is a radius, in theta1＝2πw1A sector area is formed in the anticlockwise direction of the circle center, namely the representative area of the first index, and the representative area of each index is formed in sequence in the same way;
state evaluation index { M) for long-term operation in low-voltage distribution networkAwt1,MAwt2,MAwt3,MAwt4,MAwt5The corresponding weight vector is N ═ N (N)1,n2,n3,n4,n5) Then, the fan-shaped area central angle vector θ 'corresponding to the 5-item index is 2 pi N (θ'1,θ'2,θ'3,θ'4,θ'5) (ii) a Wherein the t index corresponds to a fan-shaped central angle theta't＝2πwtTaking a radial as a reference axis of the first index along the vertical direction with the center of circle as a starting point, and taking a first normalized index value M on the reference axisAw1Is a radius of theta'1＝2πw1A sector area is made in the anticlockwise direction as the circle center, namely a representative area of the first index; similarly, the representative region of each index is made in turn.
10. The method for evaluating the operation state of the low-voltage distribution network based on the data mining as claimed in claim 9, wherein the step S43 is specifically as follows: using the 95% probability large value of the short-term evaluation analysis data and the medium-term evaluation analysis data as a data typical value to draw an improved radar map, and selecting the area and the perimeter of the improved radar map as characteristic values; are respectively marked as Si、Li(i is 1,2), i is the number of improved radar maps, and the area evaluation value SiThe perimeter estimate L is the sum of all sector areas in the ith radar chartiThe sum of the arc lengths of all sectors in the ith radar chart is as follows:
Figure FDA0002876033740000071
Figure FDA0002876033740000072
in the formula, k is the total number of short-term state evaluation indexes and the total number of medium-term and long-term state evaluation indexes; according to SiAnd LiConfigurable feature vector D ═ Di1,Di2]；
Figure FDA0002876033740000073
In the formula, Di1Area evaluation value S representing ith radar chartiThe ratio to the maximum area; di2Indicates the ith circumference evaluation value LiRatio to circumference of the same area(ii) a And Di1、Di2∈(0，1](ii) a Taking a feature vector Di1、Di2As an evaluation function, i.e. there are:
Figure FDA0002876033740000074
and carrying out quantitative scoring evaluation on the short-term running state and the medium-term running state of the low-voltage distribution network by using an evaluation function.
CN202011635023.1A 2020-12-31 2020-12-31 Low-voltage distribution network running state evaluation method based on data mining Pending CN112783939A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011635023.1A CN112783939A (en) 2020-12-31 2020-12-31 Low-voltage distribution network running state evaluation method based on data mining

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011635023.1A CN112783939A (en) 2020-12-31 2020-12-31 Low-voltage distribution network running state evaluation method based on data mining

Publications (1)

Publication Number Publication Date
CN112783939A true CN112783939A (en) 2021-05-11

Family

ID=75754913

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011635023.1A Pending CN112783939A (en) 2020-12-31 2020-12-31 Low-voltage distribution network running state evaluation method based on data mining

Country Status (1)

Country Link
CN (1) CN112783939A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113270944A (en) * 2021-06-08 2021-08-17 国网甘肃省电力公司天水供电公司 Low-voltage power distribution user side state evaluation method
CN114757001A (en) * 2022-03-14 2022-07-15 国网福建省电力有限公司 Active power distribution network line transformation relation identification method and terminal

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107064758A (en) * 2017-04-18 2017-08-18 江西明正变电设备有限公司 A kind of dry-type transformer on-Line Monitor Device and monitoring method
CN107623319A (en) * 2017-08-17 2018-01-23 广东电网有限责任公司惠州供电局 A kind of power network critical circuits discrimination method based on more evaluation indexes
CN107832930A (en) * 2017-10-27 2018-03-23 国网山东省电力公司菏泽供电公司 A kind of active distribution network operation situation appraisal procedure based on improvement radar map
CN110795692A (en) * 2019-09-16 2020-02-14 浙江工业大学 Active power distribution network operation state evaluation method
CN112101785A (en) * 2020-09-15 2020-12-18 国网河南省电力公司经济技术研究院 Method for analyzing comprehensive benefits of power and communication sharing iron tower

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107064758A (en) * 2017-04-18 2017-08-18 江西明正变电设备有限公司 A kind of dry-type transformer on-Line Monitor Device and monitoring method
CN107623319A (en) * 2017-08-17 2018-01-23 广东电网有限责任公司惠州供电局 A kind of power network critical circuits discrimination method based on more evaluation indexes
CN107832930A (en) * 2017-10-27 2018-03-23 国网山东省电力公司菏泽供电公司 A kind of active distribution network operation situation appraisal procedure based on improvement radar map
CN110795692A (en) * 2019-09-16 2020-02-14 浙江工业大学 Active power distribution network operation state evaluation method
CN112101785A (en) * 2020-09-15 2020-12-18 国网河南省电力公司经济技术研究院 Method for analyzing comprehensive benefits of power and communication sharing iron tower

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
何山 等: "基于数据挖掘的低压配电网运行状态评估方法", 广东电力, vol. 32, no. 5, 25 May 2019 (2019-05-25), pages 80 - 86 *
殷自力和张伟: "基于回路阻抗模型的城乡低压配电网络运行状态评估", 农业工程学报, vol. 34, no. 22, 23 November 2018 (2018-11-23), pages 162 - 167 *
赵会茹 等: "基于组合赋权及扇形雷达图法的电网公司月度运营预警模型", 中国电力, vol. 49, no. 9, 5 September 2016 (2016-09-05), pages 119 - 124 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113270944A (en) * 2021-06-08 2021-08-17 国网甘肃省电力公司天水供电公司 Low-voltage power distribution user side state evaluation method
CN114757001A (en) * 2022-03-14 2022-07-15 国网福建省电力有限公司 Active power distribution network line transformation relation identification method and terminal

Similar Documents

Publication Publication Date Title
CN108959424B (en) Operation method of urban electricity utilization map for monitoring load of power system
CN111429027B (en) Regional power transmission network operation multidimensional analysis method based on big data
CN105512448B (en) A kind of appraisal procedure of power distribution network health index
CN110097297A (en) A kind of various dimensions stealing situation Intellisense method, system, equipment and medium
CN109146093A (en) A kind of electric power equipment on-site exploration method based on study
CN108074021A (en) A kind of power distribution network Risk Identification system and method
CN109491339B (en) Big data-based substation equipment running state early warning system
CN112783939A (en) Low-voltage distribution network running state evaluation method based on data mining
CN103810533A (en) Cloud-model-based power distribution network fault risk identification method
CN110095661B (en) Distribution transformer high-voltage side open-phase fault first-aid repair method
CN110826228B (en) Regional power grid operation quality limit evaluation method
CN109711664B (en) Power transmission and transformation equipment health assessment system based on big data
Shin et al. Spatiotemporal load-analysis model for electric power distribution facilities using consumer meter-reading data
CN117113157B (en) Platform district power consumption fault detection system based on artificial intelligence
CN114814402A (en) Abnormal electricity utilization analysis method based on integrated electricity quantity and line loss system big data
CN106340963A (en) Electric power distribution room intelligent commission maintenance monitoring system
CN116298675B (en) Intelligent algorithm-based power distribution network line loss anomaly detection method and system
CN110674240B (en) GIS-based distributed multistage intelligent fault diagnosis system for power equipment
Lin et al. Voltage sag severity analysis based on improved FP-Growth algorithm and AHP algorithm
Sachan et al. Multiple correspondence analysis to study failures in a diverse population of a cable
Liu et al. Research on online status evaluation technology for main equipment of power transmission and transformation based on digital twin
CN115146715A (en) Power utilization potential safety hazard diagnosis method, device, equipment and storage medium
Zhang et al. Research on intelligent operation and maintenance technology of primary equipment in substation
Han et al. Application of artificial intelligence technology in power grid enterprises based on enterprise architecture method
Dehbozorgi et al. Association rule mining application to diagnose smart distribution power system outage root cause

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