CN112070352A - Large user drop factor analysis method based on improved principal component-gray correlation - Google Patents

Large user drop factor analysis method based on improved principal component-gray correlation Download PDF

Info

Publication number
CN112070352A
CN112070352A CN202010771384.2A CN202010771384A CN112070352A CN 112070352 A CN112070352 A CN 112070352A CN 202010771384 A CN202010771384 A CN 202010771384A CN 112070352 A CN112070352 A CN 112070352A
Authority
CN
China
Prior art keywords
power grid
factor index
index
grid factor
principal component
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.)
Granted
Application number
CN202010771384.2A
Other languages
Chinese (zh)
Other versions
CN112070352B (en
Inventor
魏聪
杨东俊
赵红生
周辉
吴军
王博
乔立
徐小琴
熊志
刘巨
徐敬友
陈�峰
贺继锋
唐靖
雷何
杨洁
郑旭
熊炜
张东寅
阮博
任羽纶
王佳
关钦月
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.)
State Grid Corp of China SGCC
Wuhan University WHU
Economic and Technological Research Institute of State Grid Hubei Electric Power Co Ltd
Original Assignee
State Grid Corp of China SGCC
Wuhan University WHU
Economic and Technological Research Institute of State Grid Hubei Electric Power 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 State Grid Corp of China SGCC, Wuhan University WHU, Economic and Technological Research Institute of State Grid Hubei Electric Power Co Ltd filed Critical State Grid Corp of China SGCC
Priority to CN202010771384.2A priority Critical patent/CN112070352B/en
Publication of CN112070352A publication Critical patent/CN112070352A/en
Application granted granted Critical
Publication of CN112070352B publication Critical patent/CN112070352B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • 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
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06315—Needs-based resource requirements planning or analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • 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
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • 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)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Marketing (AREA)
  • Mathematical Physics (AREA)
  • Development Economics (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Quality & Reliability (AREA)
  • Mathematical Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Pure & Applied Mathematics (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Primary Health Care (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Water Supply & Treatment (AREA)
  • Algebra (AREA)
  • Public Health (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

A large user drop factor analysis method based on improved principal component-gray correlation is characterized in that the drop number of typical industrial large users in an area and power grid factor indexes influencing the drop of the typical industrial large users are determined, the information content of each power grid factor index is determined by the improved principal component analysis method, the correlation value between the drop number of the typical industrial large users and each power grid factor index is calculated by adopting a gray correlation model based on time effect, the influence weighted value and the influence degree ratio of each power grid factor index are calculated by adopting a combined weighting method according to the information content and the correlation value of each power grid factor index, and the power grid factor indexes are graded according to the influence degree ratio. The design can improve the stability and the safety of the system operation after the large user accesses the power grid.

Description

Large user drop factor analysis method based on improved principal component-gray correlation
Technical Field
The invention belongs to the field of power grid planning, and particularly relates to a large user drop factor analysis method based on improved principal component-gray correlation.
Background
In recent years, the burden reduction policy of continuous enterprises in China is that large and medium-sized enterprises are continuously formed and grown, the occupation ratio in regional economy is increased day by day, and most of the enterprises have the characteristics of relatively advanced technology, large electricity utilization potential and high proportion of electricity charge to cost. Therefore, large users with electricity consumption of 315kVA or more or power supply voltage of 10kV or more in the year become main power supply service objects of the power grid.
At present, the site of the large user has a series of influences on the stability, economy and safety of a regional power grid, such as voltage and power out-of-limit, interval utilization rate reduction and the like. Meanwhile, the current power grid level has an important influence on the site selection of a large user in the near term, and poor energy consumption experience directly causes economic loss of the user, so that the power grid loses large customer resources, and the power grid is not favorable for attracting the site of the large user. Therefore, the power grid factors influencing the large user drop are effectively analyzed, the matching degree of the load access of the large power user and the electric energy resource is improved, the power supply network actively adapts to the access of the large user in the planning, and the method is an effective way for solving the problem of connection between the large user planning and the power grid planning.
Disclosure of Invention
The invention aims to overcome the problems in the prior art, and provides a large user drop factor analysis method based on the improved principal component-gray correlation, which can improve the matching degree of the load access of a large power user and electric energy resources, and improve the stability and safety of system operation after the large user accesses a power grid.
In order to achieve the above purpose, the invention provides the following technical scheme:
the method for analyzing the large user drop factor based on the improved principal component-gray correlation sequentially comprises the following steps of:
step A, collecting the number of fallen households of each typical industry large user in an area, and determining power grid factor indexes influencing the fallen households according to the load characteristics of each typical industry large user;
b, determining the information content of each power grid factor index by adopting an improved principal component analysis method, and calculating the association value between the number of the large users in each typical industry and each power grid factor index by adopting a grey association model based on a time effect;
step C, calculating the influence weight value and the influence degree ratio of each power grid factor index by adopting a combined weighting method according to the information content and the relevance value of each power grid factor index;
and D, grading the factor indexes of each power grid according to the influence degree ratio.
In the step B, the step of determining the information content of each power grid factor index by adopting an improved principal component analysis method sequentially comprises the following steps:
1. the data is subjected to completion and standardization processing to form a standardized matrix X ═ Xij)a×bWherein, a is the number of the grid factor indexes, b is the total annual number of the selected data, i is 1, 2.
2. Calculating the correlation coefficient matrix R ═ R (R) of the normalized matrixij)a×b；
3. Solving a characteristic root of the correlation coefficient matrix;
4. calculating the accumulated contribution rate rho of each principal component according to the characteristic root, and outputting the principal components with the accumulated contribution rates rho being more than 75% and the grading coefficients thereof;
5. calculating the information content of each power grid factor index by the following formula
Figure BDA0002616748910000021
Figure BDA0002616748910000022
Figure BDA0002616748910000023
In the above formula, m is the number of output principal components, n is 1, 2.. multidot.m,
Figure BDA0002616748910000024
for the grid factor index XiThe total amount of information contained in the message,
Figure BDA0002616748910000025
as a main component FnFactor index X of medium power gridiThe amount of information contained in the information,
Figure BDA0002616748910000026
as a main component FnThe scoring coefficient of (1).
In the step B, the step of calculating the correlation values between the number of the large users in the typical industry and the power grid factor indexes by adopting the grey correlation model based on the time effect sequentially comprises the following steps:
s1, constructing an analysis matrix [ Y, X ]:
Figure BDA0002616748910000027
in the above formula, T is the total annual number of data selected, y0(t) isThe number of typical industrial large users in the T year, T1, 2i(t) is a data value of an ith grid factor index in a t year, wherein i is 1, 2.
S2, carrying out normalization processing on the analysis matrix [ Y, X ] through the following formula to form an initial value matrix:
Figure BDA0002616748910000031
Figure BDA0002616748910000032
s3, firstly, the initial value matrix is based on delta0i(t)＝|y'0(t)-x'i(t) | forms a difference matrix Δ0iThen selects a difference matrix delta0iMaximum value delta of medium pole rangemaxTo a minimum value Δmin；
S4, calculating the correlation coefficient lambda between the typical large-user number of the industry and the grid factor index in the t-th year through the following formula0i(t) forming a correlation coefficient matrix
Figure BDA0002616748910000033
Figure BDA0002616748910000034
In the above formula, β is a resolution coefficient;
s5, adopting an equal ratio series model to endow a weight value W to the time effect:
Figure BDA0002616748910000035
in the above formula, TmaxIs the latest year of the selected data, a1＝1,q＝0.8；
S6, calculating the number of typical large users and electricity of the production industry through the following formulasGrey relevance value r of network factor indexj0：
Figure BDA0002616748910000036
In step C, the weight value of each power grid factor index is influenced
Figure BDA0002616748910000037
And influence degree ratio
Figure BDA0002616748910000038
The formula is adopted to calculate the following formula:
Figure BDA0002616748910000041
Figure BDA0002616748910000042
in the above formula, the first and second carbon atoms are,
Figure BDA0002616748910000043
for the grid factor index XiThe amount of information contained in the information,
Figure BDA0002616748910000044
for the grid factor index XiK is the grid factor index number.
In step D, the grade division standard is:
if the influence degree of a certain power grid factor index accounts for more than or equal to 15%, judging that the power grid factor index is a class of influence index;
if the influence degree of a certain power grid factor index accounts for less than 15% and more than 10%, judging that the power grid factor index is a second-class influence index;
and if the influence degree of a certain power grid factor index accounts for less than or equal to 10%, judging that the power grid factor index is three types of influence indexes.
In the step A, the typical industries comprise medicine manufacturing industry, traffic manufacturing industry, electronic manufacturing industry, environmental protection industry and new energy industry, and the power grid factor indexes comprise voltage qualification rate, power supply reliability rate indexes, regional total power consumption, regional total power generation, average total industrial value per ten thousand yuan power consumption, transformation capacity and industrial power consumption price.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention is based on the improved principal component-grey correlation large user drop factor analysis method, firstly determining the drop number of each typical industry large user in the area and the power grid factor index influencing the drop, then determining the information content of each power grid factor index by the improved principal component analysis method, calculating the correlation value between the drop number of each typical industry large user and each power grid factor index by adopting a grey correlation model based on time effect, then calculating the influence weight value and the influence degree ratio of each power grid factor index by adopting a combined weighting method according to the information content and the correlation value of each power grid factor index, and finally grading each power grid factor index according to the influence degree ratio, so that the main power influence index can be determined according to the type of the drop industry in the power grid planning process, and planning the power grid according to the main power influence indexes, thereby improving the matching degree of the load access of the large power user and the electric energy resource, reducing the influence of the site of the large power user on the regional power grid, and improving the stability and the safety of the system operation after the large power user accesses the power grid. Therefore, the invention can improve the stability and the safety of the system operation after the large user accesses the power grid.
2. According to the method for analyzing the major user drop factor based on the improved principal component-gray correlation, the power grid factor indexes comprise a voltage qualification rate, a power supply reliability rate index, the total regional power consumption, the total regional power generation capacity, the average total industrial value power consumption per ten thousand yuan, the transformation capacity and the industrial power consumption price. Therefore, the analysis result of the invention has better rationality and accuracy.
Drawings
Fig. 1 is a distribution diagram of influence degree ratio of each grid factor index of each typical industry obtained in embodiment 1 of the present invention.
Detailed Description
The present invention will be further described with reference to the following embodiments.
The method for analyzing the large user drop factor based on the improved principal component-gray correlation sequentially comprises the following steps of:
step A, collecting the number of fallen households of each typical industry large user in an area, and determining power grid factor indexes influencing the fallen households according to the load characteristics of each typical industry large user;
b, determining the information content of each power grid factor index by adopting an improved principal component analysis method, and calculating the association value between the number of the large users in each typical industry and each power grid factor index by adopting a grey association model based on a time effect;
step C, calculating the influence weight value and the influence degree ratio of each power grid factor index by adopting a combined weighting method according to the information content and the relevance value of each power grid factor index;
and D, grading the factor indexes of each power grid according to the influence degree ratio.
In the step B, the step of determining the information content of each power grid factor index by adopting an improved principal component analysis method sequentially comprises the following steps:
1. the data is subjected to completion and standardization processing to form a standardized matrix X ═ Xij)a×bWherein, a is the number of the grid factor indexes, b is the total annual number of the selected data, i is 1, 2.
2. Standardization of calculationCorrelation coefficient matrix R ═ R (R) of matrixij)a×b；
3. Solving a characteristic root of the correlation coefficient matrix;
4. calculating the accumulated contribution rate rho of each principal component according to the characteristic root, and outputting the principal components with the accumulated contribution rates rho being more than 75% and the grading coefficients thereof;
5. calculating the information content phi contained in each power grid factor index by the following formulaXi：
Figure BDA0002616748910000061
Figure BDA0002616748910000062
In the above formula, m is the number of output principal components, n is 1, 2.. multidot.m,
Figure BDA0002616748910000063
for the grid factor index XiThe total amount of information contained in the message,
Figure BDA0002616748910000064
as a main component FnFactor index X of medium power gridiThe amount of information contained in the information,
Figure BDA0002616748910000065
as a main component FnThe scoring coefficient of (1).
In the step B, the step of calculating the correlation values between the number of the large users in the typical industry and the power grid factor indexes by adopting the grey correlation model based on the time effect sequentially comprises the following steps:
s1, constructing an analysis matrix [ Y, X ]:
Figure BDA0002616748910000066
in the above formula, T is the total annual number of data selected, y0(t) is a typical industryThe number of large users in the T year, T1, 2, T, j is the number of grid factor indexes, xi(t) is a data value of an ith grid factor index in a t year, wherein i is 1, 2.
S2, carrying out normalization processing on the analysis matrix [ Y, X ] through the following formula to form an initial value matrix:
Figure BDA0002616748910000067
Figure BDA0002616748910000068
s3, firstly, the initial value matrix is based on delta0i(t)＝|y'0(t)-x'i(t) | forms a difference matrix Δ0iThen selects a difference matrix delta0iMaximum value delta of medium pole rangemaxTo a minimum value Δmin；
S4, calculating the correlation coefficient lambda between the typical large-user number of the industry and the grid factor index in the t-th year through the following formula0i(t) forming a correlation coefficient matrix
Figure BDA0002616748910000071
Figure BDA0002616748910000072
In the above formula, β is a resolution coefficient;
s5, adopting an equal ratio series model to endow a weight value W to the time effect:
Figure BDA0002616748910000073
in the above formula, TmaxIs the latest year of the selected data, a1＝1,q＝0.8；
S6, calculating the number of typical large users and the power grid factor through the following formulaTarget Grey relevance value rj0：
Figure BDA0002616748910000074
In step C, the weight value of each power grid factor index is influenced
Figure BDA0002616748910000075
And influence degree ratio
Figure BDA0002616748910000076
The formula is adopted to calculate the following formula:
Figure BDA0002616748910000077
Figure BDA0002616748910000078
in the above formula, the first and second carbon atoms are,
Figure BDA0002616748910000079
for the grid factor index XiThe amount of information contained in the information,
Figure BDA00026167489100000710
for the grid factor index XiK is the grid factor index number.
In step D, the grade division standard is:
if the influence degree of a certain power grid factor index accounts for more than or equal to 15%, judging that the power grid factor index is a class of influence index;
if the influence degree of a certain power grid factor index accounts for less than 15% and more than 10%, judging that the power grid factor index is a second-class influence index;
and if the influence degree of a certain power grid factor index accounts for less than or equal to 10%, judging that the power grid factor index is three types of influence indexes.
In the step A, the typical industries comprise medicine manufacturing industry, traffic manufacturing industry, electronic manufacturing industry, environmental protection industry and new energy industry, and the power grid factor indexes comprise voltage qualification rate, power supply reliability rate indexes, regional total power consumption, regional total power generation, average total industrial value per ten thousand yuan power consumption, transformation capacity and industrial power consumption price.
The principle of the invention is illustrated as follows:
the invention provides a large user drop factor analysis method based on improved principal component-gray correlation. In view of the difference of different types of industrial users on the electric energy demand and the different main aspects influenced by the power grid factors, the invention selects four typical novel local industries to respectively develop research, constructs a more comprehensive large-user household influence factor system from the power grid perspective, integrates the influences of the three aspects of the electric energy quality (voltage qualification rate and power supply reliability rate index), the power supply capacity (total power consumption of the area, total power generation of the area, average total industrial value power consumption per ten thousand yuan, power transformation capacity) and the power consumption price (industrial power consumption price), determines the interpretation capability of the selected indexes in the influence of the power grid factors on the household falling of the large-user based on an improved principal component analysis method, determines the number of principal components and the contained information content of each index, and simultaneously improves a gray correlation model by considering the influence of time effect, the method comprises the steps of researching the relevance between each influence factor and the typical four-class large-user house-drop situation, then calculating the influence weight value and the influence proportion of each index to different industries by integrating the information content of the index and the relevance, and finally determining the main electric power influence factors aiming at different industries through an influence index grade division table, so that an electric power company can take targeted measures in the active large-user power grid access planning, and the improvement of a power grid is realized to enhance the adaptation degree of large-user access.
Cumulative contribution ratio ρ: the SPSS software is used for analyzing and calculating the accumulated contribution rate rho of each main component, if rho is larger than 75%, the power grid factor index has strong interpretative ability, and the influence of the power grid factor index on the large user drop can be researched through the main components.
Score coefficient of principal component: the positive and negative scoring coefficients of the power grid factor indexes in the main components can judge whether the indexes can promote the drop of large users. If the scoring coefficient of the index is positive, the positive influence on the large user to select the area users is larger if the index value is larger; if the score coefficient of the index is negative, the larger the index value is, the larger the negative influence on the users in the area selected by the large users is.
In the invention, the grey correlation model based on the time effect is improved based on the time effect, and considering that the time efficiency of data acquisition and utilization is met, the grey correlation model based on the time effect is in accordance with the principle of 'big-end-up and small-end-up', the analysis value and the time efficiency of data closer to the research stage are higher, and the grey correlation model is in accordance with the current development trend, so that the grey correlation model is preliminarily improved, and the analysis result is more in accordance with the reality.
Grading: according to the invention, each power grid factor index is divided into a first-class influence index, a second-class influence index and a third-class influence index through a specific grade division standard, and the first-class influence index has the largest influence on a fallen household, so that the improvement of the factor needs to be considered emphatically in user access planning; the influence degree of the second kind of influence indexes is second to that of the first kind of indexes, and the second kind of influence indexes can be incorporated into power grid planning on the premise that the first kind of indexes are complete; for the three types of influence indexes, the influence degree of the indexes is not obvious, and the electric power company can be improved by other measures such as operation mode adjustment after a user leaves a house under the condition of limited fund and time period.
Example 1:
referring to fig. 1, a method for analyzing a large user drop factor based on an improved principal component-gray correlation takes a certain region in China as a research object, and sequentially comprises the following steps:
step 1, collecting the number of fallen households from 2015 to 2017 of typical industry large users in the area by referring to data such as relevant statistical yearbooks and the like, analyzing the load characteristics of the typical industry large users, and determining the power grid factor indexes influencing the fallen households, wherein the typical industry is medicine manufacturing industry, traffic manufacturing industry, electronic industryManufacturing industry, environmental protection and new energy industry, the power grid factor index comprises voltage qualification rate X1Power supply reliability index X2General power consumption X in region3Total generated energy X in region4Average total industrial production value of each ten thousand yuan of power consumption X5Variable capacitance X6Industrial electricity price X7；
Step 2, determining the information content of each power grid factor index by adopting an improved principal component analysis method, and calculating the association value between the number of the large users in each typical industry and each power grid factor index by adopting a grey association model based on a time effect, wherein,
the specific steps of determining the information content of each power grid factor index by adopting an improved principal component analysis method are as follows:
1. the data is subjected to completion and standardization processing to form a standardized matrix X ═ Xij)a×bWherein, a is the number of the grid factor indexes, b is the total annual number of the selected data, i is 1, 2.
2. Calculating the correlation coefficient matrix R ═ R (R) of the normalized matrixij)a×b；
3. Solving a characteristic root of the correlation coefficient matrix;
4. using SPSS software, the cumulative contribution ratio ρ of each principal component was calculated from the feature root analysis (see table 1 for results):
table 1 statistics table of variation numbers of power grid factor index specification
Figure BDA0002616748910000101
As can be seen from the above table, the principal component F1 can explain 92.673% of the influence reasons of the grid factor indexes on the large user drop, which far exceeds the contribution requirement of 75%, and the influence of the grid factor on the large user drop can be studied through the principal component, so that the principal component F1 and the scoring coefficient thereof are output (see table 2);
TABLE 2 scoring coefficients for principal component F1
Principal component F1
Percent of pass of voltage .142
Reliability of power supply .153
Total electricity consumption of area .153
Total power generation in area .153
Average total industrial production per ten thousand yuan power consumption -.141
Price of electricity for industry -.142
Variable capacitance .154
5. Calculating the information content of each power grid factor index by the following formula
Figure BDA0002616748910000102
As shown in table 3:
Figure BDA0002616748910000103
in the above formula, the first and second carbon atoms are,
Figure BDA0002616748910000104
for the grid factor index XiThe total amount of information contained in the message,
Figure BDA0002616748910000105
as a main component FnFactor index X of medium power gridiThe amount of information contained in the information,
Figure BDA0002616748910000106
as a main component FnThe scoring coefficient of (a);
the specific steps of calculating the association value between the number of the large users in the typical industry and each power grid factor index by adopting the grey association model based on the time effect are as follows:
s1, constructing an analysis matrix [ Y, X ]:
Figure BDA0002616748910000111
in the above formula, T is the total annual number of data selected, y0(T) is the number of typical large industrial users in the T year, T is 1,2i(t) is a data value of an ith grid factor index in a t year, wherein i is 1, 2.
S2, carrying out normalization processing on the analysis matrix [ Y, X ] through the following formula to form an initial value matrix:
Figure BDA0002616748910000112
Figure BDA0002616748910000113
s3, firstly, the initial value matrix is based on delta0i(t)＝|y'0(t)-x'i(t) | forms a difference matrix Δ0iThen selects a difference matrix delta0iMaximum value delta of medium pole rangemaxTo a minimum value Δmin；
S4, calculating the correlation coefficient lambda between the typical large-user number of the industry and the grid factor index in the t-th year through the following formula0i(t) forming a correlation coefficient matrix
Figure BDA0002616748910000114
Figure BDA0002616748910000115
In the above formula, β is a resolution coefficient;
s5, adopting an equal ratio series model to endow a weight value W to the time effect:
Figure BDA0002616748910000116
in the above formula, TmaxIs the latest year of the selected data, a1＝1,q＝0.8；
S6, calculating the grey correlation value r of the typical large-user number of the production and the power grid factor index through the following formulaj0：
Figure BDA0002616748910000121
The grey correlation values obtained in this example are shown in table 3:
TABLE 3 information content and relevance value of each grid factor index
X1 X2 X3 X4 X5 X6 X7
Information quantity value 0.142 0.153 0.153 0.153 0.141 0.154 0.142
[Y1,x]Relevance value 0.777 0.777 0.637 0.649 0.725 0.592 0.418
[Y2,x]Relevance value 0.654 0.654 0.733 0.73 0.621 0.735 0.432
[Y3,x]Relevance value 0.695 0.695 0.768 0.786 0.649 0.705 0.425
[Y4,x]Relevance value 0.813 0.813 0.633 0.645 0.719 0.587 0.414
In the above table, Y1-medicine manufacturing, Y2-traffic manufacturing, Y3-electronics manufacturing, Y4-environmental protection and new energy industry;
step 3, according to the information content and the relevance value contained in each power grid factor index,
Figure BDA0002616748910000122
calculating the influence weight value of each power grid factor index by adopting a combined weighting method
Figure BDA0002616748910000123
And the ratio of the influence degrees, and the specific results are shown in fig. 1:
Figure BDA0002616748910000124
Figure BDA0002616748910000125
in the above formula, the first and second carbon atoms are,
Figure BDA0002616748910000126
for the grid factor index XiThe amount of information contained in the information,
Figure BDA0002616748910000127
for the grid factor index XiK is the number of the indexes of the power grid factors;
and 4, grading the power grid factor indexes according to the influence degree ratio, wherein the grading standard is as follows:
if the influence degree of a certain power grid factor index accounts for more than or equal to 15%, judging that the power grid factor index is a class of influence index;
if the influence degree of a certain power grid factor index accounts for less than 15% and more than 10%, judging that the power grid factor index is a second-class influence index;
and if the influence degree of a certain power grid factor index accounts for less than or equal to 10%, judging that the power grid factor index is three types of influence indexes.
The results of the partitioning for this example are shown in table 4:
table 4 power grid factor index grade division results
Index of one kind Index of class two Index of three categories
Pharmaceutical manufacturing industry X1,X2,X5 X3,X4,X6 X7
Traffic manufacturing industry X3,X4,X6 X1,X2,X5 X7
Electronic manufacturing industry X3,X4,X5,X2 X1,X5 X7
Environmental protection and new energy industry X1,X2 X3,X4,X6 X7
As can be seen from table 4, for the large users in the pharmaceutical manufacturing industry, the environmental protection industry and the new energy industry, when the power company plans to access the power grid for the two types of users, it is necessary to improve the voltage qualification rate and the power supply reliability of the area, and the pharmaceutical manufacturing industry needs to improve the energy utilization efficiency to reduce the average total power consumption per ten thousand yuan of the total industrial value, and for the traffic manufacturing industry and the electronic manufacturing industry, the influence degree of the total power consumption and the power generation amount of the area on the power grid is significant, so that when the power company actively adapts to the access of the two types of users to the power grid, the power supply capacity of the power grid needs to be improved to ensure that the users can supply power sufficiently when the installation capacity is large, and the power transformation capacity of the electronic manufacturing industry needs to be improved. The industrial electricity price belongs to three indexes for the four types of industries, so the influence of the electricity price on the large-user household location selection process is relatively small, and the influence of the electricity price on the large-user can be temporarily not considered on the premise that the electric power company meets the first-class indexes of the users. And for the second type of indexes of different industrial large users, the requirements of the users can be responded as much as possible in power grid planning under the conditions of perfecting the first type of indexes and having sufficient capital time.

Claims (6)

1. A large user drop factor analysis method based on improved principal component-gray correlation is characterized by comprising the following steps:
the method comprises the following steps in sequence:
step A, collecting the number of fallen households of each typical industry large user in an area, and determining power grid factor indexes influencing the fallen households according to the load characteristics of each typical industry large user;
b, determining the information content of each power grid factor index by adopting an improved principal component analysis method, and calculating the association value between the number of the large users in each typical industry and each power grid factor index by adopting a grey association model based on a time effect;
step C, calculating the influence weight value and the influence degree ratio of each power grid factor index by adopting a combined weighting method according to the information content and the relevance value of each power grid factor index;
and D, grading the factor indexes of each power grid according to the influence degree ratio.
2. The method of analyzing big user drop factor based on improved principal component-gray correlation of claim 1, wherein:
in the step B, the step of determining the information content of each power grid factor index by adopting an improved principal component analysis method sequentially comprises the following steps:
1. the data is subjected to completion and standardization processing to form a standardized matrix X ═ Xij)a×bWherein, a is the number of the grid factor indexes, b is the total annual number of the selected data, i is 1, 2.
2. Calculating the correlation coefficient matrix R ═ R (R) of the normalized matrixij)a×b；
3. Solving a characteristic root of the correlation coefficient matrix;
4. calculating the accumulated contribution rate rho of each principal component according to the characteristic root, and outputting the principal components with the accumulated contribution rates rho being more than 75% and the grading coefficients thereof;
5. calculating the information content of each power grid factor index by the following formula
Figure FDA0002616748900000011
Figure FDA0002616748900000012
Figure FDA0002616748900000013
In the above formula, m is the number of output principal components, n is 1, 2.. multidot.m,
Figure FDA0002616748900000014
for the grid factor index XiThe total amount of information contained in the message,
Figure FDA0002616748900000021
as a main component FnFactor index X of medium power gridiThe amount of information contained in the information,
Figure FDA0002616748900000022
as a main component FnThe scoring coefficient of (1).
3. The method for analyzing the factor of falling account of large users based on the improved principal component-gray correlation as claimed in claim 1 or 2, wherein:
in the step B, the step of calculating the correlation values between the number of the large users in the typical industry and the power grid factor indexes by adopting the grey correlation model based on the time effect sequentially comprises the following steps:
s1, constructing an analysis matrix [ Y, X ]:
Figure FDA0002616748900000023
in the above formula, T is the total annual number of data selected, y0(T) is the number of typical large industrial users in the T year, T is 1,2i(t) is a data value of an ith grid factor index in a t year, wherein i is 1, 2.
S2, carrying out normalization processing on the analysis matrix [ Y, X ] through the following formula to form an initial value matrix:
Figure FDA0002616748900000024
Figure FDA0002616748900000025
s3, firstly, the initial value matrix is based on delta0i(t)＝|y′0(t)-x′i(t) | forms a difference matrix Δ0iThen selects a difference matrix delta0iMaximum value delta of medium pole rangemaxTo a minimum value Δmin；
S4, calculating the correlation coefficient lambda between the typical large-user number of the industry and the grid factor index in the t-th year through the following formula0i(t) forming a correlation coefficient matrix
Figure FDA0002616748900000026
Figure FDA0002616748900000031
In the above formula, β is a resolution coefficient;
s5, adopting an equal ratio series model to endow a weight value W to the time effect:
Figure FDA0002616748900000032
in the above formula, TmaxIs the latest year of the selected data, a1＝1,q＝0.8；
S6, calculating the grey correlation value r of the typical large-user number of the production and the power grid factor index through the following formulaj0：
Figure FDA0002616748900000033
4. The method for analyzing the factor of falling account of large users based on the improved principal component-gray correlation as claimed in claim 1 or 2, wherein:
in step C, the weight value of each power grid factor index is influenced
Figure FDA0002616748900000034
And influence degree ratio
Figure FDA0002616748900000035
The formula is adopted to calculate the following formula:
Figure FDA0002616748900000036
Figure FDA0002616748900000037
in the above formula, the first and second carbon atoms are,
Figure FDA0002616748900000038
for the grid factor index XiThe amount of information contained in the information,
Figure FDA0002616748900000039
for the grid factor index XiK is the grid factor index number.
5. The method for analyzing the factor of falling account of large users based on the improved principal component-gray correlation as claimed in claim 1 or 2, wherein:
in step D, the grade division standard is:
if the influence degree of a certain power grid factor index accounts for more than or equal to 15%, judging that the power grid factor index is a class of influence index;
if the influence degree of a certain power grid factor index accounts for less than 15% and more than 10%, judging that the power grid factor index is a second-class influence index;
and if the influence degree of a certain power grid factor index accounts for less than or equal to 10%, judging that the power grid factor index is three types of influence indexes.
6. The method for analyzing the factor of falling account of large users based on the improved principal component-gray correlation as claimed in claim 1 or 2, wherein:
in the step A, the typical industries comprise medicine manufacturing industry, traffic manufacturing industry, electronic manufacturing industry, environmental protection industry and new energy industry, and the power grid factor indexes comprise voltage qualification rate, power supply reliability rate indexes, regional total power consumption, regional total power generation, average total industrial value per ten thousand yuan power consumption, transformation capacity and industrial power consumption price.
CN202010771384.2A 2020-08-04 2020-08-04 Large user drop factor analysis method based on improved principal component-gray correlation Active CN112070352B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010771384.2A CN112070352B (en) 2020-08-04 2020-08-04 Large user drop factor analysis method based on improved principal component-gray correlation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010771384.2A CN112070352B (en) 2020-08-04 2020-08-04 Large user drop factor analysis method based on improved principal component-gray correlation

Publications (2)

Publication Number Publication Date
CN112070352A true CN112070352A (en) 2020-12-11
CN112070352B CN112070352B (en) 2024-05-10

Family

ID=73656877

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010771384.2A Active CN112070352B (en) 2020-08-04 2020-08-04 Large user drop factor analysis method based on improved principal component-gray correlation

Country Status (1)

Country Link
CN (1) CN112070352B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102999792A (en) * 2012-12-20 2013-03-27 诸暨市供电局 Method for comprehensive evaluation of power distribution network optimization allocation
CN104700325A (en) * 2015-03-30 2015-06-10 国家电网公司 Power distribution network stability evaluation method
CN105427053A (en) * 2015-12-07 2016-03-23 广东电网有限责任公司江门供电局 Relative influence analysis model applied to evaluation of distribution network construction and renovation schemes and power supply quality indexes
US20170308934A1 (en) * 2016-04-22 2017-10-26 Economy Research Institute of State Grid Zhejiang Electric Power Management method of power engineering cost
CN108985964A (en) * 2018-06-11 2018-12-11 昆明理工大学 A kind of method of part throttle characteristics Quantitative Analysis of Influence Factors
CN109035730A (en) * 2018-07-16 2018-12-18 河海大学 It is a kind of to consider that the concrete dam that Service Environment influences damages dynamic warning method
CN110503462A (en) * 2019-07-18 2019-11-26 上海交通大学 Power grid investment measuring and calculating method, system and medium based on grey correlation degree analysis

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102999792A (en) * 2012-12-20 2013-03-27 诸暨市供电局 Method for comprehensive evaluation of power distribution network optimization allocation
CN104700325A (en) * 2015-03-30 2015-06-10 国家电网公司 Power distribution network stability evaluation method
CN105427053A (en) * 2015-12-07 2016-03-23 广东电网有限责任公司江门供电局 Relative influence analysis model applied to evaluation of distribution network construction and renovation schemes and power supply quality indexes
US20170308934A1 (en) * 2016-04-22 2017-10-26 Economy Research Institute of State Grid Zhejiang Electric Power Management method of power engineering cost
CN108985964A (en) * 2018-06-11 2018-12-11 昆明理工大学 A kind of method of part throttle characteristics Quantitative Analysis of Influence Factors
CN109035730A (en) * 2018-07-16 2018-12-18 河海大学 It is a kind of to consider that the concrete dam that Service Environment influences damages dynamic warning method
CN110503462A (en) * 2019-07-18 2019-11-26 上海交通大学 Power grid investment measuring and calculating method, system and medium based on grey correlation degree analysis

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
任志涛: "基于熵值-主成分分析的环境治理公众参与水平评价研究", 《环境保护科学》, pages 3 *
孙艺新;秦超;张玮;闫庆友;谭忠富;: "基于故障树方法的供电可靠性灰色关联分析", 中国电力, no. 05, 5 May 2016 (2016-05-05) *
李娜;佟春生;: "基于DPSIR模型的城市配电网规划评价指标体系研究", 电气工程学报, no. 02, 25 February 2016 (2016-02-25) *
范丽彬: "大工业用户用电最大需量影响因素分析与预测模型研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》, pages 9 - 19 *
雷庆生;鄢晶;柴继勇;吕国峰;殷奕恒;王可;李璐瑶;胡钋;: "县域配电网规划差异化评价体系与评分方法", 电力科学与技术学报, no. 04, 28 December 2017 (2017-12-28) *

Also Published As

Publication number Publication date
CN112070352B (en) 2024-05-10

Similar Documents

Publication Publication Date Title
CN108280479A (en) A kind of power grid user sorting technique based on Load characteristics index weighted cluster algorithm
CN107944630A (en) A kind of seasonality tou power price optimization formulating method
CN110111003A (en) A kind of new energy typical scene construction method based on improvement FCM clustering algorithm
CN110009145B (en) Power load prediction method and device based on time-frequency transformation feature extraction and autoregressive trend prediction technology
CN110109971A (en) A kind of low-voltage platform area user power utilization Load Characteristic Analysis method
CN105225021B (en) The optimum choice method of power distribution network project yet to be built
CN112785119B (en) Distribution network voltage out-of-limit reason analysis method based on clustering and hierarchical analysis algorithm
CN110807598A (en) User load regulation and control value evaluation method participating in orderly power utilization
CN108428055A (en) A kind of load characteristics clustering method considering load vertical characteristics
CN111724049B (en) Research and judgment method for potential electric power energy efficiency service clients
CN111126499A (en) Secondary clustering-based power consumption behavior pattern classification method
CN116701965A (en) BIRCH clustering algorithm-based panoramic carbon representation method for enterprise users
CN111798333A (en) Energy utilization evaluation and electricity utilization safety analysis method and system
CN111523819A (en) Energy-saving potential evaluation method considering distributed power supply output uncertainty
CN113450031B (en) Method and device for selecting intelligent energy consumption service potential transformer area of residents
CN114912825A (en) Carbon neutralization comprehensive evaluation method based on sparse logarithm principal component analysis
CN112734274B (en) Low-carbon power grid operation leading influence factor mining and comprehensive evaluation method
CN112070352A (en) Large user drop factor analysis method based on improved principal component-gray correlation
CN112288496A (en) Load classification calculation method and tracking analysis method for power industry
CN116629632A (en) Comprehensive evaluation method and device for power quality of power distribution network and storage medium
CN116882776A (en) Method and device for setting energy-saving strategy in steel production, electronic equipment and storage medium
CN114860808A (en) Power distribution network equipment abnormal event correlation analysis method based on big data
CN114169790A (en) Comprehensive evaluation method for ordered power utilization potential of user by considering relevance among indexes
CN106886851A (en) Electric energy meter standby redundancy Forecasting Methodology based on electric energy meter repair based on condition of component
Soile et al. Resource abundance and energy intensity: A cross country analysis

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