CN114399323A - Time-of-use electricity price pricing method suitable for park comprehensive energy supply service - Google Patents

Time-of-use electricity price pricing method suitable for park comprehensive energy supply service Download PDF

Info

Publication number
CN114399323A
CN114399323A CN202111453603.3A CN202111453603A CN114399323A CN 114399323 A CN114399323 A CN 114399323A CN 202111453603 A CN202111453603 A CN 202111453603A CN 114399323 A CN114399323 A CN 114399323A
Authority
CN
China
Prior art keywords
time
electricity
park
matrix
load
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
CN202111453603.3A
Other languages
Chinese (zh)
Inventor
杨为城
吴芳蓉
周保中
陈耿伟
张继广
卢坤杰
李志钦
徐国平
朱烨扬
乐鹰
赵琦
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.)
Fujian Branch Of China Huadian Group Co ltd
Huadian Electric Power Research Institute Co Ltd
Original Assignee
Fujian Branch Of China Huadian Group Co ltd
Huadian Electric Power Research Institute 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 Fujian Branch Of China Huadian Group Co ltd, Huadian Electric Power Research Institute Co Ltd filed Critical Fujian Branch Of China Huadian Group Co ltd
Priority to CN202111453603.3A priority Critical patent/CN114399323A/en
Publication of CN114399323A publication Critical patent/CN114399323A/en
Pending legal-status Critical Current

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
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0206—Price or cost determination based on market factors
    • 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
    • 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
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • 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
    • G06F18/2323—Non-hierarchical techniques based on graph theory, e.g. minimum spanning trees [MST] or graph cuts
    • 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Accounting & Taxation (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Evolutionary Biology (AREA)
  • Computational Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Evolutionary Computation (AREA)
  • Marketing (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Business, Economics & Management (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Algebra (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Water Supply & Treatment (AREA)
  • Primary Health Care (AREA)
  • Public Health (AREA)
  • Tourism & Hospitality (AREA)
  • Software Systems (AREA)
  • Databases & Information Systems (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computing Systems (AREA)
  • Discrete Mathematics (AREA)

Abstract

The invention discloses a time-of-use electricity price pricing method suitable for park comprehensive energy supply service, which belongs to the field of comprehensive energy supply service of an electric power system and comprises the following steps: (1) carrying out clustering analysis on user groups; (2) determining time-sharing time intervals; (3) pricing time-of-use electricity price; (4) and (5) solving an algorithm. The implementation of the time-of-use electricity price can enable residents to enjoy the benefits on one hand, and is also beneficial to power plants, power grids and low-carbon economy on the other hand. The user changes the electricity utilization mode, so that the electricity is used more at the valley, and less at the peak, and the electricity charge expenditure can be reduced; the peak clipping, valley filling and load balancing effects brought by the implementation of time-of-use electricity prices can be beneficial to power generation enterprises and power grid enterprises; meanwhile, the improvement of the power generation efficiency can reduce carbon emission, protect the environment and benefit the society.

Description

Time-of-use electricity price pricing method suitable for park comprehensive energy supply service
Technical Field
The invention belongs to the field of comprehensive energy supply service of an electric power system, and particularly relates to a time-of-use electricity price pricing method suitable for comprehensive energy supply service of a park.
Background
The time-of-use electricity price is an electricity price mechanism which can effectively reflect the power supply cost difference of the power system at different periods, and is an important content of a demand side management project. Common forms include peak-valley time electricity price, seasonal electricity price, withered-rich electricity price and the like. At the beginning of the electric power market, the price of electricity is generally fixed singly, which results in that the price of electricity cannot reflect the difference of marginal power supply cost at different time intervals. The peak-valley time-of-use electricity price is that one day is divided into three time periods of peak, valley and average, and different electricity prices are implemented by electricity consumption in different time periods. The seasonal electricity price system can improve seasonal load imbalance of the power system, reflect power supply cost in different seasons, and inhibit the load from increasing too fast in summer and winter. The power price of the rich dry is specially designed for water and electricity, one year is divided into a rich water period, a normal water period and a dry water period according to the amount of water, different power prices are carried out in different periods, and the power price of the rich dry is formulated according to the characteristics of low power cost in the rich water period and high power cost in the dry water period. The main effect is to encourage users to reasonably arrange the power utilization time, effectively reduce the peak-valley difference of the power utilization load and improve the stability of the power system. The time-of-use electricity price is priced through the difference in time, the market efficiency is higher than that of a single fixed electricity price, the peak clipping and valley filling can be realized, and the social welfare is increased.
The implementation of the time-of-use electricity price can enable residents to enjoy the benefits on one hand, and is also beneficial to power plants, power grids and low-carbon economy on the other hand. The user changes the electricity utilization mode, so that the electricity is used more at the valley, and less at the peak, and the electricity charge expenditure can be reduced; the peak clipping, valley filling and load balancing effects brought by the implementation of time-of-use electricity prices can be beneficial to power generation enterprises and power grid enterprises; meanwhile, the improvement of the power generation efficiency can reduce carbon emission, protect the environment and benefit the society.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a time-of-use electricity price pricing method suitable for park comprehensive energy supply service.
The technical scheme adopted by the invention for solving the problems is as follows: a time-of-use electricity price pricing method suitable for a park comprehensive energy supply service is characterized by comprising the following steps: carrying out clustering analysis on user groups;
(1) electricity load and demand response characteristics
The demand elasticity matrix is one of typical characterization and analysis methods of user demand response behaviors and consists of self elasticity coefficients and cross elasticity coefficients in different time periods; under the condition of a known demand elasticity matrix, the electric load after the time-of-use electricity price adjustment is represented as follows:
Figure BDA0003385881830000021
Figure BDA0003385881830000022
Figure BDA0003385881830000023
Figure BDA0003385881830000024
Figure BDA0003385881830000025
Figure BDA0003385881830000026
in the formula:
Figure BDA0003385881830000027
-a primary electrical load vector for user k;
Figure BDA0003385881830000028
-time period i user k's original electrical load;
Figure BDA0003385881830000029
-an active electrical load vector for user k after performing a campus time of use price;
Figure BDA00033858818300000210
-active electrical load of user k for period i;
ΔLk-a power load change vector for user k;
Figure BDA00033858818300000211
-diagonal array of original electrical loads of user k, the diagonal elements being the original electrical loads
Figure BDA00033858818300000212
The other elements are all zero;
ηk，i-period and load level of the original electrical load;
Ek-a demand elasticity matrix for user k, the diagonal elements being the self-elasticity coefficients and the other elements being the cross-elasticity coefficients;
εk，iresponsivity e of the period i to itselfk，iiAnd responsiveness e of other periods to period ik，ji；
Δ P-electricity price change vector;
Figure BDA0003385881830000031
-the electricity price for time period i;
Figure BDA0003385881830000032
-current price for time period i;
change of electrical load Δ LkDiagonal matrix dependent on electrical load
Figure BDA0003385881830000033
Requirement elasticity matrix EkAnd a power price change vector delta P which are independent of each other and are power load changes delta Lk3 determinants of (a); wherein, the electrovalence variation vector delta P is an external factor, and the electric load diagonal matrix
Figure BDA0003385881830000034
And the demand elasticity matrix EkIs an internal factor; electric load diagonal array
Figure BDA0003385881830000035
And the demand elasticity matrix EkIs a mathematical representation of the electricity load and demand response characteristics of the user k, and an electricity load matrix of different users
Figure BDA0003385881830000036
And the demand elasticity matrix EkDifferent;
load matrix of electricity
Figure BDA0003385881830000037
Vector η ofk，iAnd the demand elastic matrix EkVector of (e)k，iFully connected to form a feature vector x representing the electricity load and demand response characteristics of user kkBased on the feature vector xkClustering users in the park;
Figure BDA0003385881830000038
feature vector xkHas the following 2 characteristics:
1) high in dimension; electric load matrix
Figure BDA0003385881830000039
And the demand elasticity matrix EkAre all n × n square matrix, eigenvector xkIs 2n2X 1 column vector (if n is 24, then the feature vector xkIs an 1152 × 1 column vector);
2) sparse; electric load matrix
Figure BDA00033858818300000310
The off-diagonal elements of (a) are all zero elements; meanwhile, in general, the demand response is not sensitive in all periods, the periods with larger response degree are often concentrated in the peak-valley or specific period, and the demand elasticity matrix EkIs approximately zero or even zero;
(2) spectral clustering
The spectral clustering is a clustering algorithm based on graph theory, can solve the problems of low clustering speed, poor convergence and the like caused by high dimension, sparseness and even non-convexity of data, and is very suitable for clustering of user groups in a garden; according to graph theory, a graph consists of points and edges; the edge represents the relationship between the points, the size of the edge is a weight and represents the size of the relationship, wherein the radial basis function is one of the most main expression modes of the weight; the graph can be represented by an adjacency matrix, and the adjacency matrix takes the weight between points as an element to reflect the adjacency relation of the points; the subgraph is a graph formed by a subset of points and is a part of the graph, and an adjacent matrix of the subgraph is a sub-matrix of the adjacent matrix of the graph;
Figure BDA0003385881830000041
Figure BDA0003385881830000042
in the formula:
xithe characteristic vector of the point i, namely the characteristic vector representing the electricity utilization load and demand response characteristics of the park user i;
wij-the weights of points i and j, reflecting the degree of similarity of users i and j of the campus;
a-an adjacency matrix;
n-point number, namely the number of park users;
park user feature vector xkDefined as points of the graph, campus user feature vector xkThe similarity between users is defined as a weight, and the problem of clustering of users in a garden is equivalent to the problem of dividing a graph; the divided subgraph is a division of the graph and represents a division mode of a park user group; the clustering target of the community user groups is that the characteristics of the same community user group are as similar as possible, and the characteristics of different community user groups are as different as possible, so that the graph segmentation target is that the sum of weights in sub-graphs is maximized and the sum of weights between the sub-graphs is minimized, and the target function is a canonical cut function (Ncut function);
Figure BDA0003385881830000043
Figure BDA0003385881830000044
Figure BDA0003385881830000045
A1∪A2∪…∪As＝A (13)
Figure BDA0003385881830000046
in the formula:
V(Ar) Sub-diagram ArThe sum of the internal weights;
Figure BDA0003385881830000051
sub-diagram ArAnd the sum of the weights among other subgraphs; g is an Ncut function;
s-number of subgraphs, i.e. the number of user groups in the campus;
the Ncut function is a nonlinear function, the corresponding planning model is a nonlinear planning model, and the solving is difficult; the adjacent matrix is converted into a Laplace matrix, and the algebraic property of the Laplace matrix can be utilized to quickly solve the optimization problem; the Laplace matrix is obtained by subtracting an adjacent matrix from a degree matrix of the adjacent matrix, wherein the degree matrix is a diagonal matrix, and diagonal elements are the sum of all elements in corresponding rows of the adjacent matrix; the specific expression is as follows:
Figure BDA0003385881830000052
Figure BDA0003385881830000053
in the formula (I);
a D-degree matrix;
an L-Laplace matrix;
define subgraph ArIs indicated by a vector frIndicates that it belongs to sub-graph ArThe identification of the point(s), i.e. the identification of the campus users belonging to a certain campus user group; the Ncut function may be expressed as a Laplace matrix and an indicatorQuadratic form of vector; defining unit indication vector h at the same timer＝D1/2frAnd the standard laplacian matrix L' ═ D-1/2LD-1/2The Ncut function can be expressed as:
Figure BDA0003385881830000054
Figure BDA0003385881830000055
Figure BDA0003385881830000056
Figure BDA0003385881830000061
clustering of community user groups based on spectral clustering can be converted into solving the sum of minimum s eigenvalues of a standard Laplace matrix;
Figure BDA0003385881830000062
evaluating the quality of the clustering result of the user groups in the park based on the CH (Calinski-Harabasz) index, and determining the number of the user groups in the optimal park; the CH index value is the ratio of the trace of the covariance matrix in the subgraph and between the subgraphs divided by the corresponding degree of freedom, and the higher the index value is, the better the clustering result of the user group in the garden is indicated;
Figure BDA0003385881830000063
in the formula:
Cs-CH index;
Ms-a covariance matrix within the subgraph;
Nsbetween subgraphsThe covariance matrix of (a);
tr (-) trace of the matrix.
The invention also includes: determining time-sharing time intervals;
clustering the power load data of user groups in different parks, and determining peak-valley time-sharing time periods; the power load data is simple one-dimensional data, the time interval and the clustering quantity are small, and the requirements can be met by utilizing a traditional k-means clustering algorithm; the objective function of the k-means clustering algorithm is the sum of the mean square deviations of the data and the clustering center to which the data belongs; solving a clustering mark of the power load based on a k-means clustering algorithm, and determining a peak-valley time-sharing time interval according to the clustering mark;
Figure BDA0003385881830000064
Figure BDA0003385881830000065
Figure BDA0003385881830000071
Figure BDA0003385881830000072
Figure BDA0003385881830000073
Figure BDA0003385881830000074
TP，h∪TF，h∪TV，h＝{1，2，…，n} (30)
TP，h∩TF，h＝φ
TF，h∩TV，h＝φ (31)
TV，h∩TP，h＝φ
in the formula (I), the compound is shown in the specification,
Figure BDA0003385881830000075
-a cluster label vector of campus user population h commissioned electrical loads;
Figure BDA0003385881830000076
-clustering marks of electricity loads of a community h of users in a time period t;
δP-peak period cluster labels;
δF-a flat period cluster marker;
δh-a valley period cluster marker;
SP，h-a set of peak period cluster labels for a community user population h;
SF，h-a set of flat segment cluster labels for a campus user population h;
SV，h-a set of valley time segment cluster labels for a campus user population h;
TP，h-a set of peak periods for a campus user population h;
TF，h-a flat period set of campus user groups h;
TV，h-a set of valley periods for the campus user population h.
The invention also includes: pricing time-of-use electricity price;
on the basis of community clustering and time-sharing period determination of users in the park, factors such as distributed power supply output, exchange loads inside and outside the park and the like are comprehensively considered, and a park time-sharing electricity price pricing optimization model is constructed;
a. distributed generation output model
1) Distributed wind power generation; the kinetic energy of wind is converted into electric energy by utilizing the fan blades, the output depends on the wind speed, and the output model is as follows;
Figure BDA0003385881830000081
in the formula:
lTW，t-decentralized wind power output for a time period t;
nw-number of decentralized wind power;
vt-wind speed for a time period t;
vI-cut-in wind speed;
vR-rated wind speed;
vO-cut-out wind speed;
lW，S-decentralized wind power rated output;
2) distributed photovoltaics;
the photovoltaic module is utilized to convert solar energy into electric energy, the output depends on the illumination radiation intensity and the temperature, and the output model is as follows:
Figure BDA0003385881830000082
in the formula:
lTP，t-distributed photovoltaic contribution over time period t;
nP-a distributed photovoltaic number;
lP，Sunder standard conditions (intensity of illuminating radiation i)S＝1000W/m2Temperature tSRated output of distributed photovoltaic at 25 ℃;
it-illumination radiation intensity for a time period t;
tt-temperature for a time period t;
εPpower temperature coefficient (generally 0.0039 deg.C)-1)；
b. Load exchange between the inside and outside of the park
Under the background of energy Internet, the resource allocation mode of a park is 'energy utilization nearby + energy coming from a distant place'; the garden distributed power supply is consumed on site firstly, the power utilization requirement of the garden is met, and the redundant part is fed into an external power grid; the power demand of the garden is supplied on the spot by a distributed power supply, and the insufficient part is supplied by an external power grid;
therefore, the load of the grid-connected point of the park is the exchange load inside and outside the park, namely the park electrical load deducts the net load of the output of the distributed power supply;
Figure BDA0003385881830000091
in the formula:
lE，t-exchanging loads inside and outside the campus for a time period t;
Figure BDA0003385881830000092
-the electrical load of user k for a time period t;
when the internal and external exchange load of the park is greater than zero, the power utilization requirement of the park cannot be completely met through the distributed power supply, and the internal and external exchange load of the park is the outsourcing load of the park; when the exchange load inside and outside the park is less than zero, the distributed power supply still remains except for meeting the power utilization requirement of the park, and the absolute value of the exchange load inside and outside the park is the internet access output of the distributed power supply; the difference between the garden outsourcing load and the distributed power supply internet access output is the garden inside and outside exchange load;
Figure BDA0003385881830000093
Figure BDA0003385881830000094
in the formula:
lP，t-the park outsourcing load for time period t;
lG，t-grid-on power of the distributed power supply for a time period t.
The invention also includes: a time-of-use electricity price pricing optimization model;
(1) optimizing variables
The optimization variables of the time-of-use electricity price pricing optimization model are time-of-use electricity price levels of user groups in different parks, and the expression is as follows:
Figure BDA0003385881830000101
in the formula:
Figure BDA0003385881830000102
-electricity prices for the campus user group h for a time period t;
Figure BDA0003385881830000103
-peak period time of day electricity prices for campus user population h;
Figure BDA0003385881830000104
-flat time of day electricity prices for campus user group h;
Figure BDA0003385881830000105
-electricity prices at off-peak hours and time of campus user population h;
(2) objective function
The objective function of the time-of-use electricity price pricing optimization model maximizes the income of a park operator, the income comprises park electricity selling income of a park user, park user capacity income and distributed power supply internet surfing income, and the cost comprises outsourcing electricity quantity and capacity electricity charge;
the electricity selling income of the garden users is the sum of the products of the electric quantity of the garden users in different time periods and the electricity price of the garden users in different time periods; the electricity price of the park at the corresponding time period is expressed as the electricity price of the local consumption electricity of the distributed power supply; the distributed power supply internet income is the sum of the product of the distributed power supply internet electricity quantity and the internet electricity price thereof in different time periods, the internet electricity price refers to the internet electricity price of the coal-fired unit, and the electricity prices in different time periods are the same; the electricity charge of the outsourcing electric quantity is the sum of products of the outsourcing electric quantity of the park and the selling price of the external power grid in different periods, the selling price of the external power grid is the peak-valley price, and the outsourcing electric quantity of the park is respectively settled according to the peak-valley period divided by the selling price of the external power grid;
max RT＝IT-CT＝(IS+IC+IG)-(CP+CC) (38)
Figure BDA0003385881830000106
Figure BDA0003385881830000107
Figure BDA0003385881830000111
Figure BDA0003385881830000112
Figure BDA0003385881830000113
in the formula:
RT-total revenue;
IT-total revenue;
CT-a total cost;
IS-electricity sales revenue for campus users;
IC-campus user capacity revenue;
IG-distributed power grid revenue;
CP-outsourcing electricity charges;
CC-a capacity electricity charge;
iC，k-capacity revenue for user k;
pG-grid-access electricity prices for distributed power sources;
pE，t-external grid sales electricity prices (peak-to-valley electricity prices) for time period t;
pc-capacity electricity prices;
(3) constraint conditions
The constraint conditions of the time-of-use electricity price pricing optimization model comprise time-of-use electricity price, electricity consumption cost, electricity consumption, internal and external exchange load and other constraints;
1) time of use price constraint
In order to promote peak clipping and valley filling of the park, the time-of-use electricity price is decreased in the peak valley leveling period;
Figure BDA0003385881830000121
in the formula:
the descending proportion of the alpha-time-of-use electricity price is 1;
a decreasing proportion 2 of the beta-time-of-use electricity price;
2) cost constraints for electricity
After the park time-of-use electricity price is executed, the electricity utilization cost of the park unit is reduced to a certain extent;
Figure BDA0003385881830000122
in the formula: eta-reduction of the unit electricity cost;
3) constraint of power consumption
In order to guarantee the power consumption requirement of the park, the daily power consumption change range of the park is not large after the time-sharing electricity price of the park is executed;
Figure BDA0003385881830000123
in the formula:
Figure BDA0003385881830000124
-rate of change of daily electricity;
4) internal and external exchange load restraint
The exchange load inside and outside the park is smaller than the capacity of the transformer at the grid-connected point;
-cT≤lE，t≤cT (47)
in the formula:
cT-a grid-tie point transformer capacity;
5) non-negative constraint
Figure BDA0003385881830000125
The invention also includes: solving an algorithm;
the objective function of the time-of-use electricity price pricing optimization model comprises nonlinear functions such as an absolute value and a maximum value, and therefore the optimization model is a nonlinear programming model; because the number of the optimization variables is less, iteration and search can be carried out on a feasible solution space based on a branch-and-bound algorithm.
Compared with the prior art, the invention has the following advantages and effects: the implementation of the time-of-use electricity price can enable residents to enjoy the benefits on one hand, and is also beneficial to power plants, power grids and low-carbon economy on the other hand. The user changes the electricity utilization mode, so that the electricity is used more at the valley, and less at the peak, and the electricity charge expenditure can be reduced; the peak clipping, valley filling and load balancing effects brought by the implementation of the time-of-use electricity price enable power generation enterprises and power grid enterprises to benefit from the peak clipping, valley filling and load balancing effects; meanwhile, the improvement of the power generation efficiency can reduce carbon emission, protect the environment and benefit the society.
Drawings
FIG. 1 is a schematic diagram of clustering of community user groups based on a spectral clustering algorithm;
FIG. 2 is a schematic diagram of determining time-sharing time period based on k-means clustering algorithm;
FIG. 3 is a schematic view of a resource allocation pattern of a campus in the context of an energy Internet;
FIG. 4 is a diagram of campus load before executing campus time-of-use electricity prices;
figure 5 is a schematic diagram of the electrical load of a campus user population.
Detailed Description
The present invention will be described in further detail below by way of examples with reference to the accompanying drawings, which are illustrative of the present invention and are not to be construed as limiting the present invention.
Examples are given.
In this embodiment, a time-of-use electricity price pricing method suitable for a park energy supply service includes:
1. carrying out clustering analysis on user groups;
(1) electricity load and demand response characteristics
The demand elasticity matrix is one of typical characterization and analysis methods of user demand response behaviors and consists of self elasticity coefficients and cross elasticity coefficients in different time periods; under the condition of a known demand elasticity matrix, the electric load after the time-of-use electricity price adjustment is represented as follows:
Figure BDA0003385881830000131
Figure BDA0003385881830000132
Figure BDA0003385881830000133
Figure BDA0003385881830000141
Figure BDA0003385881830000142
Figure BDA0003385881830000143
in the formula:
Figure BDA0003385881830000144
-a primary electrical load vector for user k;
Figure BDA0003385881830000145
-time period i user k's original electrical load;
Figure BDA0003385881830000146
-an active electrical load vector for user k after performing a campus time of use price;
Figure BDA0003385881830000147
-active electrical load of user k for period i;
ΔLk-a power load change vector for user k;
Figure BDA0003385881830000148
-diagonal array of original electrical loads of user k, the diagonal elements being the original electrical loads
Figure BDA0003385881830000149
The other elements are all zero;
ηk，i-period and load level of the original electrical load;
Ek-a demand elasticity matrix for user k, the diagonal elements being the self-elasticity coefficients and the other elements being the cross-elasticity coefficients; epsilonk，iResponsivity e of the period i to itselfk，iiAnd responsiveness e of other periods to period ik，ji；
Δ P-electricity price change vector;
Figure BDA00033858818300001410
-the electricity price for time period i;
Figure BDA00033858818300001411
-current price for time period i;
change of electrical load Δ LkDiagonal matrix dependent on electrical load
Figure BDA0003385881830000159
Requirement elasticity matrix EkAnd a power price change vector delta P which are independent of each other and are power load changes delta Lk3 determinants of (a); wherein, the electrovalence variation vector delta P is an external factor, and the electric load diagonal matrix
Figure BDA0003385881830000151
And the demand elasticity matrix EkIs an internal factor; electric load diagonal array
Figure BDA0003385881830000152
And the demand elasticity matrix EkIs a mathematical representation of the electricity load and demand response characteristics of the user k, and an electricity load matrix of different users
Figure BDA0003385881830000153
And the demand elasticity matrix EkDifferent;
load matrix of electricity
Figure BDA0003385881830000154
Vector η ofk，iAnd the demand elastic matrix EkVector of (e)k，iFully connected to form a feature vector x representing the electricity load and demand response characteristics of user kkBased on the feature vector xkClustering users in the park;
Figure BDA0003385881830000155
feature vector xkHas the following 2 characteristics:
1) high in dimension; electric load matrix
Figure BDA0003385881830000156
And the demand elasticity matrix EkAre all n × n square matrix, eigenvector xkIs 2n2X 1 column vector (if n is 24, then the feature vector xkIs an 1152 × 1 column vector);
2) sparse; electric load matrix
Figure BDA0003385881830000157
The off-diagonal elements of (a) are all zero elements; meanwhile, in general, the demand response is not sensitive in all periods, the periods with larger response degree are often concentrated in the peak-valley or specific period, and the demand elasticity matrix EkIs approximately zero or even zero;
(2) spectral clustering
The spectral clustering is a clustering algorithm based on graph theory, can solve the problems of low clustering speed, poor convergence and the like caused by high dimension, sparseness and even non-convexity of data, and is very suitable for clustering of user groups in a garden; according to graph theory, a graph consists of points and edges; the edge represents the relationship between the points, the size of the edge is a weight and represents the size of the relationship, wherein the radial basis function is one of the most main expression modes of the weight; the graph can be represented by an adjacency matrix, and the adjacency matrix takes the weight between points as an element to reflect the adjacency relation of the points; the subgraph is a graph formed by a subset of points and is a part of the graph, and an adjacent matrix of the subgraph is a sub-matrix of the adjacent matrix of the graph;
Figure BDA0003385881830000158
Figure BDA0003385881830000161
in the formula:
xithe characteristic vector of the point i, namely the characteristic vector representing the electricity utilization load and demand response characteristics of the park user i;
wij-the weights of points i and j, reflecting the degree of similarity of users i and j of the campus;
a-an adjacency matrix;
n-point number, namely the number of park users;
park user feature vector xkDefined as points of the graph, campus user feature vector xkThe similarity between users is defined as a weight, and the problem of clustering of users in a garden is equivalent to the problem of dividing a graph; the divided subgraph is a division of the graph and represents a division mode of a park user group; the clustering target of the community user groups is that the characteristics of the same community user group are as similar as possible, and the characteristics of different community user groups are as different as possible, so that the graph segmentation target is that the sum of weights in sub-graphs is maximized and the sum of weights between the sub-graphs is minimized, and the target function is a canonical cut function (Ncut function);
Figure BDA0003385881830000162
Figure BDA0003385881830000163
Figure BDA0003385881830000164
A1∪A2∪…∪As＝A (13)
Figure BDA0003385881830000165
in the formula:
V(Ar) Sub-diagram ArThe sum of the internal weights;
Figure BDA0003385881830000166
-sum of weights between subgraph Ar and other subgraphs; g is an Ncut function;
s-number of subgraphs, i.e. the number of user groups in the campus;
the Ncut function is a nonlinear function, the corresponding planning model is a nonlinear planning model, and the solving is difficult; the adjacent matrix is converted into a Laplace matrix, and the algebraic property of the Laplace matrix can be utilized to quickly solve the optimization problem; the Laplace matrix is obtained by subtracting an adjacent matrix from a degree matrix of the adjacent matrix, wherein the degree matrix is a diagonal matrix, and diagonal elements are the sum of all elements in corresponding rows of the adjacent matrix; the specific expression is as follows:
Figure BDA0003385881830000171
Figure BDA0003385881830000172
in the formula (I);
a D-degree matrix;
an L-Laplace matrix;
define subgraph ArIs indicated by a vector frIndicates that it belongs to sub-graph ArThe identification of the point(s), i.e. the identification of the campus users belonging to a certain campus user group; the Ncut function may be expressed as a quadratic form of the Laplace matrix and the indicator vector; defining unit indication vector h at the same timer＝D1/2frAnd the standard laplacian matrix L' ═ D-1/2LD-1/2The Ncut function can be expressed as:
Figure BDA0003385881830000173
Figure BDA0003385881830000174
Figure BDA0003385881830000175
Figure BDA0003385881830000176
clustering of community user groups based on spectral clustering can be converted into solving the sum of minimum s eigenvalues of a standard Laplace matrix;
Figure BDA0003385881830000181
evaluating the quality of the clustering result of the user groups in the park based on the CH (Calinski-Harabasz) index, and determining the number of the user groups in the optimal park; the CH index value is the ratio of the trace of the covariance matrix in the subgraph and between the subgraphs divided by the corresponding degree of freedom, and the higher the index value is, the better the clustering result of the user group in the garden is indicated;
Figure BDA0003385881830000182
in the formula:
Cs-CH index;
Ms-a covariance matrix within the subgraph;
Ns-covariance matrix between subgraphs;
tr (·) -traces of the matrix;
in summary, the flow of the spectral clustering algorithm is shown in fig. 1;
2. determining time-sharing time intervals;
clustering the power load data of user groups in different parks, and determining peak-valley time-sharing time periods; the power load data is simple one-dimensional data, the time interval and the clustering quantity are small, and the requirements can be met by utilizing a traditional k-means clustering algorithm; the objective function of the k-means clustering algorithm is the sum of the mean square deviations of the data and the clustering center to which the data belongs; the flow of determining the time-sharing period based on the k-means clustering algorithm is shown in fig. 2;
solving a clustering mark of the power load based on a k-means clustering algorithm, and determining a peak-valley time-sharing time interval according to the clustering mark;
Figure BDA0003385881830000183
Figure BDA0003385881830000184
Figure BDA0003385881830000185
Figure BDA0003385881830000186
Figure BDA0003385881830000187
Figure BDA0003385881830000191
Figure BDA0003385881830000192
TP，h∪TF，h∪TV，h＝{1，2，…，n} (54)
TP，h∩TF，h＝φ
TF，h∩TV，h＝φ (55)
TV，h∩TP，h＝φ
in the formula (I), the compound is shown in the specification,
Figure BDA0003385881830000193
-a cluster label vector of electricity loads for a community user group h;
Figure BDA0003385881830000194
-clustering signature of electricity load for the campus user group h at time period t:
δP-peak period cluster labels;
δF-a flat period cluster marker;
δV-a valley period cluster marker;
SP，h-a set of peak period cluster labels for a community user population h;
SF.h-a set of flat segment cluster labels for a campus user population h;
SV，ha valley time segment clustering mark set of a park user group h;
TP，h-a set of peak periods for a campus user population h;
TF，h-a flat period set of campus user groups h;
TV，h-a set of valley periods for a campus user population h;
3. pricing time-of-use electricity price;
on the basis of community clustering and time-sharing period determination of users in the park, factors such as distributed power supply output, exchange loads inside and outside the park and the like are comprehensively considered, and a park time-sharing electricity price pricing optimization model is constructed;
a. distributed generation output model
1) Distributed wind power generation; the kinetic energy of wind is converted into electric energy by utilizing the fan blades, the output depends on the wind speed, and the output model is as follows;
Figure BDA0003385881830000201
in the formula:
lTW，t-decentralized wind power output for a time period t;
nW-number of decentralized wind power;
vt-wind speed for a time period t;
vI-cut-in wind speed;
vR-rated wind speed;
vO-cut-out wind speed;
lW，Sdecentralized wind power ratingForce;
2) distributed photovoltaics;
the photovoltaic module is utilized to convert solar energy into electric energy, the output depends on the illumination radiation intensity and the temperature, and the output model is as follows:
Figure BDA0003385881830000202
in the formula:
lTP，t-distributed photovoltaic contribution over time period t;
nP-a distributed photovoltaic number;
lP，Sunder standard conditions (intensity of illuminating radiation i)S＝1000W/m2Temperature tSRated output of distributed photovoltaic at 25 ℃;
it-illumination radiation intensity for a time period t;
tt-temperature for a time period t;
εPpower temperature coefficient (generally 0.0039 deg.C)-1)；
b. Load exchange between the inside and outside of the park
As shown in fig. 3, resource allocation patterns of a campus in the context of an energy internet;
under the background of energy Internet, the resource allocation mode of a park is 'energy utilization nearby + energy coming from a distant place'; the garden distributed power supply is consumed on site firstly, the power utilization requirement of the garden is met, and the redundant part is fed into an external power grid; the power demand of the garden is supplied on the spot by a distributed power supply, and the insufficient part is supplied by an external power grid;
therefore, the load of the grid-connected point of the park is the exchange load inside and outside the park, namely the park electrical load deducts the net load of the output of the distributed power supply;
Figure BDA0003385881830000211
in the formula:
lE，t-exchanging loads inside and outside the campus for a time period t;
Figure BDA0003385881830000212
-the electrical load of user k for a time period t;
when the internal and external exchange load of the park is greater than zero, the power utilization requirement of the park cannot be completely met through the distributed power supply, and the internal and external exchange load of the park is the outsourcing load of the park; when the exchange load inside and outside the park is less than zero, the distributed power supply still remains except for meeting the power utilization requirement of the park, and the absolute value of the exchange load inside and outside the park is the internet access output of the distributed power supply; the difference between the garden outsourcing load and the distributed power supply internet access output is the garden inside and outside exchange load;
Figure BDA0003385881830000213
Figure BDA0003385881830000214
in the formula:
lP，t-the park outsourcing load for time period t;
lG，t-grid-access power of the distributed power supply for a time period t;
4. a time-of-use electricity price pricing optimization model;
(1) optimizing variables
The optimization variables of the time-of-use electricity price pricing optimization model are time-of-use electricity price levels of user groups in different parks, and the expression is as follows:
Figure BDA0003385881830000221
in the formula:
Figure BDA0003385881830000222
-electricity prices for the campus user group h for a time period t;
Figure BDA0003385881830000223
-peak period time of day electricity prices for campus user population h;
Figure BDA0003385881830000224
-flat time of day electricity prices for campus user group h;
Figure BDA0003385881830000225
-electricity prices at off-peak hours and time of campus user population h;
(2) objective function
The objective function of the time-of-use electricity price pricing optimization model maximizes the income of a park operator, the income comprises park electricity selling income of a park user, park user capacity income and distributed power supply internet surfing income, and the cost comprises outsourcing electricity quantity and capacity electricity charge;
the electricity selling income of the garden users is the sum of the products of the electric quantity of the garden users in different time periods and the electricity price of the garden users in different time periods; the electricity price of the park at the corresponding time period is expressed as the electricity price of the local consumption electricity of the distributed power supply; the distributed power supply internet income is the sum of the product of the distributed power supply internet electricity quantity and the internet electricity price thereof in different time periods, the internet electricity price refers to the internet electricity price of the coal-fired unit, and the electricity prices in different time periods are the same; the electricity charge of the outsourcing electric quantity is the sum of products of the outsourcing electric quantity of the park and the selling price of the external power grid in different periods, the selling price of the external power grid is the peak-valley price, and the outsourcing electric quantity of the park is respectively settled according to the peak-valley period divided by the selling price of the external power grid;
max RT＝IT-CT＝(IS+IC+IG)-(CP+CC) (62)
Figure BDA0003385881830000226
Figure BDA0003385881830000227
Figure BDA0003385881830000228
Figure BDA0003385881830000229
Figure BDA0003385881830000231
in the formula:
RT-total revenue;
IT-total revenue;
CT-a total cost;
IS-electricity sales revenue for campus users;
IC-campus user capacity revenue;
IG-distributed power grid revenue;
CP-outsourcing electricity charges;
CC-a capacity electricity charge;
ic，k-capacity revenue for user k;
pG-grid-access electricity prices for distributed power sources;
pE，t-external grid sales electricity prices (peak-to-valley electricity prices) for time period t;
pC-capacity electricity prices;
(3) constraint conditions
The constraint conditions of the time-of-use electricity price pricing optimization model comprise time-of-use electricity price, electricity consumption cost, electricity consumption, internal and external exchange load and other constraints;
1) time of use price constraint
In order to promote peak clipping and valley filling of the park, the time-of-use electricity price is decreased in the peak valley leveling period;
Figure BDA0003385881830000232
in the formula:
the descending proportion of the alpha-time-of-use electricity price is 1;
a decreasing proportion 2 of the beta-time-of-use electricity price;
2) cost constraints for electricity
After the park time-of-use electricity price is executed, the electricity utilization cost of the park unit is reduced to a certain extent;
Figure BDA0003385881830000241
in the formula: eta-reduction of the unit electricity cost;
3) constraint of power consumption
In order to guarantee the power consumption requirement of the park, the daily power consumption change range of the park is not large after the time-sharing electricity price of the park is executed;
Figure BDA0003385881830000242
in the formula:
Figure BDA0003385881830000243
-rate of change of daily electricity;
4) internal and external exchange load restraint
The exchange load inside and outside the park is smaller than the capacity of the transformer at the grid-connected point;
-cT≤lE，t≤cT (71)
in the formula:
cT-a grid-tie point transformer capacity;
5) non-negative constraint
Figure BDA0003385881830000244
5. Solving an algorithm;
the objective function of the time-of-use electricity price pricing optimization model comprises nonlinear functions such as an absolute value and a maximum value, and therefore the optimization model is a nonlinear programming model; because the number of the optimization variables is less, iteration and search can be carried out on a feasible solution space based on a branch-and-bound algorithm.
The time period and the electricity price are 2 important parameters of the time-of-use electricity price, and are specific embodiment of differentiated time-of-use electricity prices facing different park user groups. The time-of-use electricity price pricing method for distributed power supply local consumption comprises 3 steps of community user group clustering, time-of-use time interval determination and time-of-use electricity price pricing.
1. Campus user population clustering
The purpose of the community user group clustering is to subdivide the community users to form a community user group, which is the basis for making differential time-of-use electricity prices. And forming a park user group based on a spectral clustering algorithm by taking the power load and demand response data of the park users as input, and determining the power load and demand response characteristics of the park users.
2. Time-sharing period determination
The electricity load characteristics of the user groups in the garden are different, and the peak, the valley and the time-sharing time periods corresponding to the time-sharing electricity price are different. And determining the peak-valley time-sharing time periods of different park user groups based on a k-means clustering algorithm by taking the power load data of the park user groups as input.
3. Park time-of-use electricity price pricing
The method comprises the steps of taking distributed power supply data, peak-to-valley time-sharing time periods, electricity loads and demand response data of a park user group and the like as input, taking park operator income maximization as an objective function, considering constraint conditions such as electricity cost, electricity consumption, internal and external exchange loads and the like, constructing a park time-sharing electricity price pricing optimization model, and solving and determining the time-sharing electricity price of the park user group.
The concrete case is as follows:
1. example arrangement
The voltage grade of a park grid-connected point is 35kV, and the capacity of a transformer is 3150 kVA; the installed capacity of distributed wind power in the garden is 20.0MW, and the installed capacity of distributed photovoltaic is 8.0 MW; 78 users in the garden, with voltage levels of 10kV, among them 52 large industrial users, 26 general industrial and commercial users.
Wind speed, illumination radiation intensity, temperature, electricity load and demand response of park users all have certain randomness, and in order to comprehensively consider various scenes and reduce calculated amount, the scenes are reduced, and data of different scenes are subjected to weighted average processing. After the weighted average processing, the park electrical load, the distributed power supply output and the park internal and external exchange load are shown in fig. 4.
The grid-connection electricity price of the distributed power supply is 0.4153 yuan/kWh, and before park time-sharing electricity price execution, park users settle the electricity price sale according to an external power grid. The decreasing rate α of the time-of-use electricity price is 0.2, the decreasing rate η of the unit electricity cost is 0.01, and the rate of change of the daily electricity consumption
Figure BDA0003385881830000251
2. Analysis of results
The electricity load, peak-to-valley time sharing period of different user groups in the park are shown in fig. 5.
TABLE 1 time-shared periods of campus user groups
Figure BDA0003385881830000252
Figure BDA0003385881830000261
1) Campus user group 1
The user group 1 consists of 19 large industrial users and 16 general industrial and commercial users; the electric load is large in the daytime (09:00-18:00) and small in the rest time periods; the overall response degree to the electricity price is large, the difference degree of the elastic coefficients in different time intervals is large, the elastic coefficients in the early morning time interval (04:00-07:00) and the early evening time interval (18:00-21:00) are small, and the elastic coefficients in the morning time interval (08:00-11:00) and the afternoon time interval (15:00-18:00) are large. The user group 1 is a sensitive user group which uses electricity in the daytime and has a large daytime response degree.
2) Campus user group 2
The user group 2 consists of 21 large industrial users and 9 common industrial and commercial users; the electric load is distributed evenly and has small fluctuation in different time periods; in the overall response degree to the electricity price, the elasticity coefficients are similar in different periods, and the difference degree is small. The user group 2 is a general user group with continuous and stable power utilization and similar response degrees in different periods.
3) Campus user group 3
The user group 3 consists of 12 large industrial users and 1 common industrial and commercial user; the power consumption load is large in the afternoon and evening time period (12:00-22:00), and the power consumption load is small in the rest time periods; the overall response degree to the electricity price is small, the elastic coefficients of the noon time period (10:00-13:00) and the early morning time period (00:00-01:00) are large in the difference degree of the elastic coefficients of different time periods, and the elastic coefficients of the other time periods are small. The user group 3 is an insensitive user group with small response degree in the afternoon and evening power consumption, noon and early morning.
The electricity price package when the park is calculated is shown in the following table.
TABLE 2 park time-of-use electricity price package
Figure BDA0003385881830000262
The comprehensive utilization efficiency of the garden distributed power supply, the friendliness degree with an external power grid and the overall economy are respectively reflected by 3 indexes of the local consumption rate of the distributed power supply, the maximum demand of a transformer at a grid-connected point of the garden and the electricity utilization cost of a unit of the garden, and the change before and after the time-of-use electricity price of the garden is compared and executed, as shown in the following table.
Table 3 change before and after execution of park time electricity prices
Figure BDA0003385881830000271
The park time-of-use electricity price pricing is essentially to subdivide park users and formulate time-of-use electricity prices according to electricity load and demand response characteristics, differently stimulate response potentials of different park user groups and change electricity utilization behaviors, and the following three effects are generated.
1) After the park user responds to the park time-sharing electricity price, the park electricity load is matched with the output of the distributed power supply more, the distributed power supply reduces the power quantity of the internet, local consumption is realized to a greater extent, and the comprehensive utilization efficiency of the park distributed power supply is improved.
2) The local consumption rate of the distributed power supply is improved, power exchange inside and outside the park is reduced, the peak-valley difference of exchange load inside and outside the park is smaller, fluctuation is smaller, the curve is smoother, the maximum demand of a transformer at a grid-connected point of the park is reduced, the load rate is improved, the investment of a power grid is delayed, and the distributed power supply is more friendly to an external power grid.
3) The power production and consumption of the distributed power supply are completed nearby, the power transmission cost is almost zero, and the power supply cost is lower than that of an external power grid. The local consumption rate of the distributed power supply is improved, the unit electricity utilization cost of the park is reduced, and the overall economy of the park is improved.
Those not described in detail in this specification are well within the skill of the art.
Although the present invention has been described with reference to the above embodiments, it should be understood that the scope of the present invention is not limited thereto, and that various changes and modifications can be made by those skilled in the art without departing from the spirit and scope of the present invention.

Claims (5)

1. A time-of-use electricity price pricing method suitable for a park comprehensive energy supply service is characterized by comprising the following steps: carrying out clustering analysis on user groups;
(1) electricity load and demand response characteristics
The demand elasticity matrix is one of typical characterization and analysis methods of user demand response behaviors and consists of self elasticity coefficients and cross elasticity coefficients in different time periods; under the condition of a known demand elasticity matrix, the electric load after the time-of-use electricity price adjustment is represented as follows:
Figure FDA0003385881820000011
Figure FDA0003385881820000012
Figure FDA0003385881820000013
Figure FDA0003385881820000014
Figure FDA0003385881820000015
Figure FDA0003385881820000016
in the formula:
Figure FDA0003385881820000017
-a primary electrical load vector for user k;
Figure FDA0003385881820000018
-time period i user k's original electrical load;
Figure FDA0003385881820000019
-an active electrical load vector for user k after performing a campus time of use price;
Figure FDA00033858818200000110
-active electrical load of user k for period i;
ΔLk-a power load change vector for user k;
Figure FDA0003385881820000021
-diagonal array of original electrical loads of user k, the diagonal elements being the original electrical loads
Figure FDA0003385881820000022
The other elements are all zero;
ηk，i-period and load level of the original electrical load;
Ek-a demand elasticity matrix for user k, the diagonal elements being the self-elasticity coefficients and the other elements being the cross-elasticity coefficients;
εk，iresponsivity e of the period i to itselfk，iiAnd responsiveness e of other periods to period ik，ji；
Δ P-electricity price change vector;
Figure FDA0003385881820000023
-the electricity price for time period i;
Figure FDA0003385881820000024
-current price for time period i;
change of electrical load Δ LkDiagonal matrix dependent on electrical load
Figure FDA0003385881820000025
Requirement elasticity matrix EkAnd a power price change vector delta P which are independent of each other and are power load changes delta Lk3 determinants of (a); wherein, the electrovalence variation vector delta P is an external factor, and the electric load diagonal matrix
Figure FDA0003385881820000026
And the demand elasticity matrix EkIs an internal factor; electric load diagonal array
Figure FDA0003385881820000027
And the demand elasticity matrix EkIs a mathematical representation of the electricity load and demand response characteristics of the user k, and an electricity load matrix of different users
Figure FDA0003385881820000028
And the demand elasticity matrix EkDifferent;
load matrix of electricity
Figure FDA0003385881820000029
Vector η ofk，iAnd the demand elastic matrix EkVector of (e)k，iFully connected to form a feature vector x representing the electricity load and demand response characteristics of user kkBased on the feature vector xkClustering users in the park;
Figure FDA00033858818200000210
feature vector xkHas the following 2 characteristics:
1) high in dimension; electric load matrix
Figure FDA00033858818200000211
And the demand elasticity matrix EkAre all n × n square matrix, eigenvector xkIs 2n2X 1 column vector;
2) sparse; electric load matrix
Figure FDA00033858818200000212
The off-diagonal elements of (a) are all zero elements; meanwhile, in general, the demand response is not sensitive in all periods, the periods with larger response degree are often concentrated in the peak-valley or specific period, and the demand elasticity matrix EkIs approximately zero or even moreTo zero;
(2) spectral clustering
Spectral clustering is a clustering algorithm based on graph theory; according to graph theory, a graph consists of points and edges; the edge represents the relationship between the points, the size of the edge is a weight and represents the size of the relationship, wherein the radial basis function is one of the most main expression modes of the weight; the graph can be represented by an adjacency matrix, and the adjacency matrix takes the weight between points as an element to reflect the adjacency relation of the points; the subgraph is a graph formed by a subset of points and is a part of the graph, and an adjacent matrix of the subgraph is a sub-matrix of the adjacent matrix of the graph;
Figure FDA0003385881820000031
Figure FDA0003385881820000032
in the formula:
xithe characteristic vector of the point i, namely the characteristic vector representing the electricity utilization load and demand response characteristics of the park user i;
wij-the weights of points i and j, reflecting the degree of similarity of users i and j of the campus;
a-an adjacency matrix;
n-point number, namely the number of park users;
park user feature vector xkDefined as points of the graph, campus user feature vector xkThe similarity between users is defined as a weight, and the problem of clustering of users in a garden is equivalent to the problem of dividing a graph; the divided subgraph is a division of the graph and represents a division mode of a park user group; the clustering target of the community user groups is that the characteristics of the same community user group are as similar as possible, and the characteristics of different community user groups are as different as possible, so that the graph segmentation target is that the sum of weights in sub-graphs is maximized and the sum of weights between the sub-graphs is minimized, and the target function is a standard segmentation function;
Figure FDA0003385881820000033
Figure FDA0003385881820000034
Figure FDA0003385881820000035
A1∪A2∪…∪As＝A (13)
Figure FDA0003385881820000036
in the formula:
V(Ar) Sub-diagram ArThe sum of the internal weights;
Figure FDA0003385881820000041
sub-diagram ArAnd the sum of the weights among other subgraphs; g is an Ncut function;
s-number of subgraphs, i.e. the number of user groups in the campus;
the Ncut function is a nonlinear function, the corresponding planning model is a nonlinear planning model, and the solving is difficult; the adjacent matrix is converted into a Laplace matrix, and the algebraic property of the Laplace matrix can be utilized to quickly solve the optimization problem; the Laplace matrix is obtained by subtracting an adjacent matrix from a degree matrix of the adjacent matrix, wherein the degree matrix is a diagonal matrix, and diagonal elements are the sum of all elements in corresponding rows of the adjacent matrix; the specific expression is as follows:
Figure FDA0003385881820000042
Figure FDA0003385881820000043
in the formula (I);
a D-degree matrix;
an L-Laplace matrix;
define subgraph ArIs indicated by a vector frIndicates that it belongs to sub-graph ArThe identification of the point(s), i.e. the identification of the campus users belonging to a certain campus user group; the Ncut function may be expressed as a quadratic form of the Laplace matrix and the indicator vector; defining unit indication vector h at the same timer＝D1/2frAnd the standard laplacian matrix L' ═ D-1/2LD-1/2The Ncut function can be expressed as:
Figure FDA0003385881820000044
Figure FDA0003385881820000045
Figure FDA0003385881820000051
Figure FDA0003385881820000052
clustering of community user groups based on spectral clustering can be converted into solving the sum of minimum s eigenvalues of a standard Laplace matrix;
Figure FDA0003385881820000053
evaluating the quality of the clustering result of the community users based on the CH index, and determining the number of the optimal community users; the CH index value is the ratio of the trace of the covariance matrix in the subgraph and between the subgraphs divided by the corresponding degree of freedom, and the higher the index value is, the better the clustering result of the user group in the garden is indicated;
Figure FDA0003385881820000054
in the formula:
Cs-CH index;
Ms-a covariance matrix within the subgraph;
Ns-covariance matrix between subgraphs;
tr (-) trace of the matrix.
2. The method for pricing electricity-time rates for use in a campus integrated services system of claim 1, further comprising: determining time-sharing time intervals;
clustering the power load data of user groups in different parks, and determining peak-valley time-sharing time periods; the power load data is simple one-dimensional data, the time interval and the clustering quantity are small, and the requirements can be met by utilizing a traditional k-means clustering algorithm; the objective function of the k-means clustering algorithm is the sum of the mean square deviations of the data and the clustering center to which the data belongs; solving a clustering mark of the power load based on a k-means clustering algorithm, and determining a peak-valley time-sharing time interval according to the clustering mark;
Figure FDA0003385881820000055
Figure FDA0003385881820000056
Figure FDA0003385881820000061
TP，h∪TF，h∪TV，h＝{1，2，…，n} (6)
Figure FDA0003385881820000062
in the formula:
Figure FDA0003385881820000063
-a cluster label vector of electricity loads for a community user group h;
Figure FDA0003385881820000064
-clustering marks of electricity loads of a community h of users in a time period t;
δP-peak period cluster labels;
δF-a flat period cluster marker;
δV-a valley period cluster marker;
SP，h-a set of peak period cluster labels for a community user population h;
SF，h-a set of flat segment cluster labels for a campus user population h;
SV，h-a set of valley time segment cluster labels for a campus user population h;
TP，h-a set of peak periods for a campus user population h;
TF，h-a flat period set of campus user groups h;
TV，h-a set of valley periods for the campus user population h.
3. The pricing method for electricity time of use applicable to the park integrated energy supply service according to claim 2, further comprising: pricing time-of-use electricity price;
on the basis of community clustering and time-sharing period determination of users in the park, factors are comprehensively considered, and a park time-sharing electricity price pricing optimization model is constructed;
a. distributed generation output model
1) Distributed wind power generation; the kinetic energy of wind is converted into electric energy by utilizing the fan blades, the output depends on the wind speed, and the output model is as follows;
Figure FDA0003385881820000071
in the formula:
lTW，t-decentralized wind power output for a time period t;
nW-number of decentralized wind power;
vt-wind speed for a time period t;
vI-cut-in wind speed;
vR-rated wind speed;
vO-cut-out wind speed;
lW，s-decentralized wind power rated output;
2) distributed photovoltaics;
the photovoltaic module is utilized to convert solar energy into electric energy, the output depends on the illumination radiation intensity and the temperature, and the output model is as follows:
Figure FDA0003385881820000072
in the formula:
lTP，t-distributed photovoltaic contribution over time period t;
nP-a distributed photovoltaic number;
lP，s-a distributed photovoltaic rated output under standard conditions;
it-illumination radiation intensity for a time period t;
tt-temperature for a time period t;
εP-power temperature coefficient;
b. load exchange between the inside and outside of the park
Under the background of energy Internet, the resource allocation mode of a park is 'energy utilization nearby + energy coming from a distant place'; the garden distributed power supply is consumed on site firstly, the power utilization requirement of the garden is met, and the redundant part is fed into an external power grid; the power demand of the garden is supplied on the spot by a distributed power supply, and the insufficient part is supplied by an external power grid;
therefore, the load of the grid-connected point of the park is the exchange load inside and outside the park, namely the park electrical load deducts the net load of the output of the distributed power supply;
Figure FDA0003385881820000081
in the formula:
lE，t-exchanging loads inside and outside the campus for a time period t;
Figure FDA0003385881820000082
-the electrical load of user k for a time period t;
when the internal and external exchange load of the park is greater than zero, the power utilization requirement of the park cannot be completely met through the distributed power supply, and the internal and external exchange load of the park is the outsourcing load of the park; when the exchange load inside and outside the park is less than zero, the distributed power supply still remains except for meeting the power utilization requirement of the park, and the absolute value of the exchange load inside and outside the park is the internet access output of the distributed power supply; the difference between the garden outsourcing load and the distributed power supply internet access output is the garden inside and outside exchange load;
Figure FDA0003385881820000083
Figure FDA0003385881820000084
in the formula:
lP，t-the park outsourcing load for time period t;
lG，t-grid-on power of the distributed power supply for a time period t.
4. The pricing method for electricity time of use applicable to the park integrated energy supply service according to claim 3, further comprising: a time-of-use electricity price pricing optimization model;
(1) optimizing variables
The optimization variables of the time-of-use electricity price pricing optimization model are time-of-use electricity price levels of user groups in different parks, and the expression is as follows:
Figure FDA0003385881820000091
in the formula:
Figure FDA0003385881820000092
-electricity prices for the campus user group h for a time period t;
Figure FDA0003385881820000093
-peak period time of day electricity prices for campus user population h;
Figure FDA0003385881820000094
-flat time of day electricity prices for campus user group h;
Figure FDA0003385881820000095
-electricity prices at off-peak hours and time of campus user population h;
(2) objective function
The objective function of the time-of-use electricity price pricing optimization model maximizes the income of a park operator, the income comprises park electricity selling income of a park user, park user capacity income and distributed power supply internet surfing income, and the cost comprises outsourcing electricity quantity and capacity electricity charge;
the electricity selling income of the garden users is the sum of the products of the electric quantity of the garden users in different time periods and the electricity price of the garden users in different time periods; the electricity price of the park at the corresponding time period is expressed as the electricity price of the local consumption electricity of the distributed power supply; the distributed power supply internet income is the sum of the product of the distributed power supply internet electricity quantity and the internet electricity price thereof in different time periods, the internet electricity price refers to the internet electricity price of the coal-fired unit, and the electricity prices in different time periods are the same; the electricity charge of the outsourcing electric quantity is the sum of products of the outsourcing electric quantity of the park and the selling price of the external power grid in different periods, the selling price of the external power grid is the peak-valley price, and the outsourcing electric quantity of the park is respectively settled according to the peak-valley period divided by the selling price of the external power grid;
maxRT＝IT-CT＝(IS+IC+IG)-(CP+CC) (14)
Figure FDA0003385881820000096
Figure FDA0003385881820000097
Figure FDA0003385881820000101
Figure FDA0003385881820000102
Figure FDA0003385881820000103
in the formula:
RT-total revenue;
IT-total revenue;
CT-a total cost;
IS-electricity sales revenue for campus users;
IC-campus user capacity revenue;
IG-distributed power grid revenue;
CP-outsourcing electricity charges;
CC-a capacity electricity charge;
iC，k-capacity revenue for user k;
pG-grid-access electricity prices for distributed power sources;
pE，t-external grid sales electricity prices for time period t;
pC-capacity electricity prices;
(3) constraint conditions
The constraint conditions of the time-of-use electricity price pricing optimization model comprise time-of-use electricity price, electricity consumption cost, electricity consumption and internal and external exchange load constraints;
1) time of use price constraint
In order to promote peak clipping and valley filling of the park, the time-of-use electricity price is decreased in the peak valley leveling period;
Figure FDA0003385881820000111
in the formula:
the descending proportion of the alpha-time-of-use electricity price is 1;
a decreasing proportion 2 of the beta-time-of-use electricity price;
2) cost constraints for electricity
After the park time-of-use electricity price is executed, the electricity utilization cost of the park unit is reduced to a certain extent;
Figure FDA0003385881820000112
in the formula: eta-reduction of the unit electricity cost;
3) constraint of power consumption
In order to guarantee the power consumption requirement of the park, the daily power consumption change range of the park is not large after the time-sharing electricity price of the park is executed;
Figure FDA0003385881820000113
in the formula:
Figure FDA0003385881820000114
-rate of change of daily electricity;
4) internal and external exchange load restraint
The exchange load inside and outside the park is smaller than the capacity of the transformer at the grid-connected point;
-cT≤lE，t≤cT (23)
in the formula:
cT-a grid-tie point transformer capacity;
5) non-negative constraint
Figure FDA0003385881820000115
5. The pricing method for electricity time of use applicable to the park integrated energy supply service according to claim 4, further comprising: solving an algorithm;
the objective function of the time-of-use electricity price pricing optimization model comprises nonlinear functions such as an absolute value and a maximum value, and therefore the optimization model is a nonlinear programming model; because the number of the optimization variables is less, iteration and search can be carried out on a feasible solution space based on a branch-and-bound algorithm.
CN202111453603.3A 2021-12-01 2021-12-01 Time-of-use electricity price pricing method suitable for park comprehensive energy supply service Pending CN114399323A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111453603.3A CN114399323A (en) 2021-12-01 2021-12-01 Time-of-use electricity price pricing method suitable for park comprehensive energy supply service

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111453603.3A CN114399323A (en) 2021-12-01 2021-12-01 Time-of-use electricity price pricing method suitable for park comprehensive energy supply service

Publications (1)

Publication Number Publication Date
CN114399323A true CN114399323A (en) 2022-04-26

Family

ID=81225926

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111453603.3A Pending CN114399323A (en) 2021-12-01 2021-12-01 Time-of-use electricity price pricing method suitable for park comprehensive energy supply service

Country Status (1)

Country Link
CN (1) CN114399323A (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106532769A (en) * 2016-11-25 2017-03-22 云南电网有限责任公司电力科学研究院 Method for optimizing demand-side time-of-use power price based on photovoltaic grid-connected uncertainty
AU2020102245A4 (en) * 2019-01-08 2020-10-29 Nanjing Institute Of Technology A grid hybrid rolling dispatching method considering congestion and energy storage tou price
CN112053059A (en) * 2020-09-04 2020-12-08 上海电力交易中心有限公司 Peak-valley load optimization method based on user demand elastic response and time-of-use electricity price

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106532769A (en) * 2016-11-25 2017-03-22 云南电网有限责任公司电力科学研究院 Method for optimizing demand-side time-of-use power price based on photovoltaic grid-connected uncertainty
AU2020102245A4 (en) * 2019-01-08 2020-10-29 Nanjing Institute Of Technology A grid hybrid rolling dispatching method considering congestion and energy storage tou price
CN112053059A (en) * 2020-09-04 2020-12-08 上海电力交易中心有限公司 Peak-valley load optimization method based on user demand elastic response and time-of-use electricity price

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
刘敦楠，徐尔丰，刘明光，周保中，应昱杭，俞秦博: "面向分布式电源就地消纳的园区分时电价定价方法", 电力系统自动化, vol. 44, no. 20, 25 October 2020 (2020-10-25), pages 20 - 25 *
曾鸣, 孙昕, 赵庆波, 赵永亮, 尹佳音, 林海英: "上网侧和销售侧峰谷分时电价联动与整体优化的方法及实证分析", 电力需求侧管理, no. 04, 25 May 2003 (2003-05-25), pages 9 - 13 *

Similar Documents

Publication Publication Date Title
WO2019196375A1 (en) Demand side response-based microgrid optimal unit and time-of-use electricity price optimization method
Hong et al. Optimal sizing of hybrid wind/PV/diesel generation in a stand-alone power system using Markov-based genetic algorithm
Roy Market constrained optimal planning for wind energy conversion systems over multiple installation sites
CN111404193B (en) Data-driven-based microgrid random robust optimization scheduling method
CN113592365B (en) Energy optimization scheduling method and system considering carbon emission and green electricity consumption
CN104598985A (en) Power load forecasting method
CN112950098A (en) Energy planning method and device based on comprehensive energy system and terminal equipment
CN113282646A (en) Daily load curve acquisition method based on power consumption behavior influence factors
CN109818369A (en) A kind of distributed generation resource planing method of meter and fuzzy and stochastic feature of contributing
Bartecka et al. Sizing of prosumer hybrid renewable energy systems in Poland
CN104376373A (en) Distributed power supply planning method based on time sequence characteristic and environmental benefit
Kong et al. Multi-objective optimal allocation and performance evaluation for energy storage in energy systems
CN112994115A (en) New energy capacity configuration method based on WGAN scene simulation and time sequence production simulation
Hong et al. Markov model-based energy storage system planning in power systems
CN112418488A (en) Comprehensive energy system scheduling method and device based on two-stage energy optimization
Steinhilber et al. Indicators assessing the performance of renewable energy support policies in 27 Member States
Guan et al. Chance-constrained optimization of hybrid solar combined cooling, heating and power system considering energetic, economic, environmental, and flexible performances
CN112653195B (en) Method for configuring robust optimization capacity of grid-connected micro-grid
Wang et al. Research on short‐term and mid‐long term optimal dispatch of multi‐energy complementary power generation system
Zhang et al. Research on economic optimal dispatching of microgrid cluster based on improved butterfly optimization algorithm
Anukoolthamchote et al. Net electricity load profiles: Shape and variability considering customer-mix at transformers on the island of Oahu, Hawai'i
CN114399323A (en) Time-of-use electricity price pricing method suitable for park comprehensive energy supply service
Sun et al. Hybrid forecasting model of electric heating load based on CART decision tree regression
Deng et al. Analysis of Renewable Energy Accommodation Capability of Shanxi Power Grid Based on Operation Simulation Method
Zhu et al. Bigdata assisted energy conversion model for innovative city application

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