CN112052430B - Photovoltaic station correlation comprehensive evaluation system based on improved TOPSIS - Google Patents

Photovoltaic station correlation comprehensive evaluation system based on improved TOPSIS Download PDF

Info

Publication number
CN112052430B
CN112052430B CN202010824036.7A CN202010824036A CN112052430B CN 112052430 B CN112052430 B CN 112052430B CN 202010824036 A CN202010824036 A CN 202010824036A CN 112052430 B CN112052430 B CN 112052430B
Authority
CN
China
Prior art keywords
photovoltaic
correlation
station
output
index
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010824036.7A
Other languages
Chinese (zh)
Other versions
CN112052430A (en
Inventor
任洲洋
夏威夷
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.)
Chongqing University
Original Assignee
Chongqing University
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 Chongqing University filed Critical Chongqing University
Priority to CN202010824036.7A priority Critical patent/CN112052430B/en
Publication of CN112052430A publication Critical patent/CN112052430A/en
Application granted granted Critical
Publication of CN112052430B publication Critical patent/CN112052430B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • 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
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression 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
    • 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

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Analysis (AREA)
  • Development Economics (AREA)
  • Operations Research (AREA)
  • Health & Medical Sciences (AREA)
  • Pure & Applied Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Mathematical Physics (AREA)
  • Tourism & Hospitality (AREA)
  • Mathematical Optimization (AREA)
  • Marketing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Computational Mathematics (AREA)
  • Educational Administration (AREA)
  • Software Systems (AREA)
  • Quality & Reliability (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Game Theory and Decision Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Databases & Information Systems (AREA)
  • Algebra (AREA)
  • Probability & Statistics with Applications (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Photovoltaic Devices (AREA)

Abstract

The invention discloses a photovoltaic station correlation comprehensive evaluation system based on an improved TOPSIS, which comprises a data input module, a data validity evaluation module, a correlation quantization index system establishment module, an index system evaluation module, a weight calculation module, a photovoltaic station correlation evaluation module and a database; the invention performs rationality test on the index system and the comprehensive method, and has satisfactory distinguishing capability.

Description

Photovoltaic station correlation comprehensive evaluation system based on improved TOPSIS
Technical Field
The invention relates to the field of photovoltaic station correlation evaluation, in particular to a photovoltaic station correlation comprehensive evaluation system based on an improved TOPSIS.
Background
Renewable energy is considered as a solution to alleviate energy crisis and environmental pollution. In recent years, renewable energy sources typified by wind power and photovoltaic are highly valued, and the installed capacity is increasing year by year. However, the intermittent and random nature of the photovoltaic output presents difficulties for system operation and scheduling, thereby limiting the acceptance of photovoltaic sites. Within a certain distance range, the plurality of photovoltaic stations show a certain spatial correlation, and the fluctuation of the station group is amplified after the superposition of the output of the plurality of stations. But at the same time, the spatial correlation law can provide more information for the prediction and simulation of the photovoltaic resource, which is important for improving the accuracy of the prediction and simulation of the photovoltaic resource and further promoting the digestion of the photovoltaic resource. Therefore, the quantitative analysis and evaluation of the correlation of the photovoltaic power station are a significant problem in the field of new energy. In a certain region, the correlation between individual photovoltaic field stations and clusters of photovoltaic field stations helps to reflect the individual contribution to the overall photovoltaic resource of the region. The relevance evaluation and sequencing of the individual photovoltaic power stations can help operators of the power system and traders of the power system to determine which photovoltaic station is system friendly, making operational and investment decisions more scientific and advantageous. Therefore, the research on the photovoltaic station related evaluation can promote the knowledge of the photovoltaic correlation, so that the improvement of the digestion of the photovoltaic resources is of great significance.
However, the existing quantitative evaluation index and comprehensive evaluation method for the photovoltaic correlation have the following disadvantages: 1) Part of the literature uses Spearman and Kendall correlation coefficients for analyzing the accuracy and rationality of photovoltaic output predictions and simulation results. However, the Spearman correlation coefficient and the Kendall correlation coefficient describe rank correlations, and cannot reflect the numerical correlations of the magnitudes of the forces. In addition, in the two correlation coefficients, the importance of each sampling point in the output sequence is equal, and the correlation of the concerned characteristic value (such as the daily maximum output) in engineering practice cannot be represented. 2) At present, only a few studies discuss correlation evaluation indexes of photovoltaic power generation. Patent CN 106126934A discloses a method for obtaining a photovoltaic power generation correlation index, and the correlation between the theoretical output and the relative output of the light Fu Changzhan headroom is described by using a correlation coefficient. The photovoltaic station operation characteristic evaluation index system published in the patent CN 104362621A considers the generated energy. However, the characteristics of the photovoltaic output are not combined, and the actual requirements of the correlation in engineering are not considered. 3) The indexes are not intuitive and concise enough, and the indexes cannot be conveniently and directly used for guiding the scheduling and transaction of the photovoltaic resources.
Disclosure of Invention
The invention aims to provide a photovoltaic station correlation comprehensive evaluation system based on an improved TOPSIS, which comprises a data input module, a data validity evaluation module, a correlation quantization index system establishment module, an index system evaluation module, a weight calculation module, a photovoltaic station correlation evaluation module and a database.
The data input module acquires historical output data of the photovoltaic station and inputs the historical output data to the data validity evaluation module.
The requirements for acquiring the historical output data of the photovoltaic field station by the data input module are as follows: the sampling period of the photovoltaic output data is less than 1h, and the data cover at least one year. The photovoltaic daily power generation amount data sampling period is 1 day, and the data cover at least one year.
And the data validity evaluation module is used for verifying the validity of the historical output data of the photovoltaic station to obtain an effective output data set, and inputting the effective output data set into the correlation quantization index system establishment module.
The data validity evaluation module is used for checking the validity of the historical output data of the photovoltaic station, and comprises the following steps of:
1) judging whether the a-th sunrise force data has the start moment and the end moment of the force output, if so, entering the step 2), otherwise, deleting the a-th sunrise force data, and enabling a=a+1 to enter the step 3). The initial value of a is 1.
2) Judging whether the starting moment and the ending moment of the a-th sunrise force meet the constraint condition (1), if so, storing the a-th sunrise force data in an effective force data set, and entering the step 3). Otherwise, delete the a-th sunrise data and let a=a+1, go to step 3). The elements in the effective force data set are arranged in ascending order according to time sequence.
Wherein t is initiation Indicating the starting moment of the output, t ending To the moment of ending the output, t sunrise And t sunset The sunrise time and the sunset time of the area where the photovoltaic station is located are respectively indicated.
3) Judging whether a > N is true, if so, entering the step 4), otherwise, returning to the step 1). N is the total number of dates of the historical output data of the photovoltaic station.
4) Judging whether the date intervals of two adjacent elements of the effective output data set are smaller than 30, if yes, entering the step 5), otherwise, returning to the data input module to acquire the historical output data of the photovoltaic station again;
5) Judging whether the number of elements in the effective output data set is more than or equal to 70% N, if so, outputting the effective output data set. Otherwise, the data input module is returned to acquire the historical output data of the photovoltaic station again;
the correlation quantization index system establishment module establishes a single-layer tree-shaped photovoltaic station correlation quantization index system according to the effective output data set and inputs the single-layer tree-shaped photovoltaic station correlation quantization index system to the index system assessment module.
The single-layer tree-shaped photovoltaic station correlation quantification index system comprises trend correlation indexes of a daily output sequence and a daily power generation sequence, numerical correlation indexes of the daily output sequence and the daily power generation sequence and time correlation indexes of the daily output sequence and the daily power generation sequence.
The trend correlation index includes a daily output Gini correlation coefficient and a daily power generation amount Gini correlation coefficient.
The numerical correlation index comprises an intra-day output similarity, an intra-day power generation amount similarity, a maximum output duty ratio and a maximum intra-day power generation amount duty ratio.
The time correlation index comprises a large probability peak time difference, a large probability effective time length difference, a maximum peak time deviation and a maximum effective time length difference.
Solar energy output Gini related coefficient gamma of photovoltaic station p,i The following is shown:
wherein, gamma p,i Is the correlation coefficient of the solar output Gini of the ith solar photovoltaic station; r is (r) n The rank of the nth power sampling point in the ith daily output sequence of the photovoltaic station;representing the rank of the nth power sampling point of the average output sequence of all stations in the area; n is the total number of sampling points of the sunrise force sequence; int (·) represents a rounding function;
solar power generation amount Gini related coefficient gamma of photovoltaic station g The following is shown:
wherein R is i The rank of the ith daily power generation amount in the photovoltaic station daily power generation amount annual sequence;a rank indicating an average daily power generation amount of the station in the i-th day area; d represents the total number of days covered by data for correlation evaluation;
the within-day output similarity d p,i The following is shown:
wherein P is the rated power of the photovoltaic field station;representing the average rated power of all stations in the area; p is p i,n The power of the nth sampling point in the day output sequence of the ith day volt-field station; />Power of an nth sampling point of the regional station average output sequence;
solar power generation amount similarity d g The following is shown:
in the formula g i The solar energy generation amount of the ith day in the solar energy generation amount annual sequence of the photovoltaic station;an average daily power generation amount of the station in the ith day area is shown;
the maximum output ratio R p,i The following is shown:
maximum daily power generation rate R g The following is shown:
peak time difference d m,i The following is shown:
wherein m is i Peak time of the ith sunrise force sequence.The peak of the ith output of the group sequence of area lights Fu Changzhan.
The moment of maximum output in the sequence of output in the day, namely the peak time m of the sequence of output in the i day i The following is shown:
m i ＝{m,p m ≥max{p n ,n＝1,2...N}} (9)
effective duration difference d t,i The following is shown:
wherein t is i Is the effective duration of the ith sunrise power sequence of light Fu Changzhan.Is the effective duration of the ith day of the regional light Fu Changzhan group.
Maximum peak time difference d m,max The following is shown:
maximum effective time difference d t,max The following is shown:
where i=1, 2, …, D.
The calculation steps for calculating the large probability peak time difference M (p) are as follows:
1) Time difference d of peak m,i And (5) ascending order arrangement is performed, and a peak time difference cumulative probability curve is drawn. The abscissa of the peak time difference cumulative probability curve is the peak time difference cumulative probability p.
2) The value of the peak time difference integrated probability p is set.
3) The corresponding relation of the time difference accumulated probability curve is taken as the time difference M (p) of the large probability peak.
The calculation step of calculating the large probability effective duration difference D (p') is as follows:
1) Difference of effective duration d t,i And (5) arranging in ascending order, and drawing an effective duration difference cumulative probability curve. The abscissa of the effective duration difference cumulative probability curve is the effective duration difference cumulative probability p'.
2) The value of the effective duration difference cumulative probability p' is set.
3) And taking the corresponding relation of the effective time difference accumulated probability curve as the large probability effective time length difference D (p').
And the index system evaluation module evaluates the received single-layer tree-shaped photovoltaic field station correlation quantization index system.
The index system evaluation module evaluates the received single-layer tree-shaped photovoltaic field station correlation quantitative index system as follows:
1) Respectively calculating the distinguishing degree of each index in the single-layer tree-shaped photovoltaic station correlation quantization index system; wherein the degree of distinction D of the jth index j The following is shown:
wherein x is ij A j index representing an i-th photovoltaic station; w (w) j A weight representing the j-th index; i=1, 2, …, n; n is the number of photovoltaic stations; j=1, 2, …, m; m is the total index number;
2) Sequentially judging whether the distinguishing degree of each index is smaller than a threshold value D min The method comprises the steps of carrying out a first treatment on the surface of the Delete zoneThe graduation being less than threshold D min Is an indicator of (2).
The weight calculation module calculates the combination weight of each index in the photovoltaic station correlation quantization index system and inputs the combination weight to the photovoltaic station correlation evaluation module.
Obtaining the combination weight w of each index j The steps of (a) are as follows:
1) Subjective weight a of j-th index is obtained by expert scoring j 。
2) Calculating entropy e of the j-th index j The method comprises the following steps:
wherein x is ij lnx ij ＝0。
3) Calculating the difference coefficient g of the j-th index j The method comprises the following steps:
g j ＝1-e j (15)
4) By means of the coefficient of difference g j For weight a j Updating to obtain:
wherein w is j The combination weight of the j-th index.
The photovoltaic field station correlation evaluation module evaluates the correlation of the photovoltaic field stations.
The comprehensive evaluation and sequencing method for the correlation of the photovoltaic field station comprises the following steps:
1) Normalizing all indexes in a single-layer tree-shaped photovoltaic station correlation quantization index system to obtain a standardized index matrix X '' n×m ；
Wherein, the index matrix X 'is standardized' n×m The j-th column element of row i is as follows:
2) Establishing a weighted decision matrix V n×m The method comprises the steps of carrying out a first treatment on the surface of the Wherein the weighted decision matrix V n×m The ith row and jth column element v in (b) ij The following is shown:
v ij ＝x′ ij w j (20)
3) Setting an ideal solution Y + And negative ideal solution Y - The method comprises the following steps:
4) Ideal alignment of Y + And negative ideal solution Y - And (3) simplifying to obtain:
5) Calculation of the ith photovoltaic station v ij Euclidean distance d from ideal i+ The method comprises the following steps:
in the formula, v ij Is a matrix V n×m Represents the value of the ith light Fu Changzhan weighted by the jth index;
calculation of the ith photovoltaic station v ij Euclidean distance d from negative ideal solution i- The method comprises the following steps:
6) Calculating the proximity coefficient C of a photovoltaic station i The method comprises the following steps:
approximation coefficient C i And the correlation of the photovoltaic field station and the photovoltaic field station cluster is positively correlated.
The database stores data of the data input module, the correlation quantization index system building module, the index system evaluation module, the data validity evaluation module, the weight calculation module and the photovoltaic station correlation evaluation module.
The invention creates comprehensive, scientific and visual photovoltaic correlation quantization indexes, avoids distortion caused by normalization and homodromous to Euclidean distance, simplifies calculation, and adopts a system of subjective and objective combination of entropy to determine index weights so that the setting of the index weights can be suitable for different application scenes; based on a quantitative analysis system, the index system and the comprehensive method are reasonably tested, and the method has satisfactory distinguishing capability.
Drawings
FIG. 1 is a diagram of an index architecture as disclosed herein.
Fig. 2 is a graph showing sunrise force curves of a photovoltaic station.
FIG. 3 is a peak time bias cumulative probability curve and an effective duration difference cumulative probability curve.
Fig. 4 is a flow chart of an improved TOPSIS method.
Fig. 5 is a schematic diagram of the results of 18 station 10 indicators.
Fig. 6 is a comparison of the ranking results of the modified TOPSIS method with the conventional TOPSIS method.
Detailed Description
The present invention is further described below with reference to examples, but it should not be construed that the scope of the above subject matter of the present invention is limited to the following examples. Various substitutions and alterations are made according to the ordinary skill and familiar means of the art without departing from the technical spirit of the invention, and all such substitutions and alterations are intended to be included in the scope of the invention.
Example 1:
referring to fig. 1 to 4, a photovoltaic station correlation comprehensive evaluation system based on an improved TOPSIS comprises a data input module, a data validity evaluation module, a correlation quantization index system establishment module, an index system evaluation module, a weight calculation module, a photovoltaic station correlation evaluation module and a database.
The data input module acquires historical output data of the photovoltaic station and inputs the historical output data to the data validity evaluation module.
The requirements of the historical output data of the photovoltaic field station are as follows: the sampling period of the photovoltaic output data is less than 1h, and the data cover at least one year. The photovoltaic daily power generation amount data sampling period is 1 day, and the data cover at least one year.
And the data validity evaluation module is used for verifying the validity of the historical output data of the photovoltaic station to obtain an effective output data set, and inputting the effective output data set into the correlation quantization index system establishment module.
The data validity evaluation module is used for checking the validity of the historical output data of the photovoltaic station, and comprises the following steps of:
1) The solar curve of the photovoltaic power is in a single-peak mode, which is similar to the positive half period of the cosine curve shape, under the influence of the solar irradiance period, as shown in fig. 2. The photovoltaic station output generally increases from zero before and after sunrise, reaches a maximum around noon, and finally drops to zero at sunset. And defining the moment when the output of the photovoltaic station changes from zero to positive for the first time in the day as the output starting moment, and the moment when the output of the photovoltaic station changes from zero to zero for the last time in the day as the output ending moment. The duration between the start time of the output and the end time of the output is defined as the effective duration. According to the daily law of photovoltaic output, the output data of each day must have output starting time and output ending time.
Judging whether the a-th sunrise force data has the starting moment and the ending moment of the force, if so, entering the step 2), otherwise, deleting the a-th sunrise force data, and enabling a=a+1 to enter the step 3). The initial value of a is 1.
2) Judging whether the starting moment and the ending moment of the a-th sunrise force meet the constraint condition (1), if so, storing the a-th sunrise force data in an effective force data set, and entering the step 3). Otherwise, delete the a-th sunrise data and let a=a+1, go to step 3). The elements in the effective force data set are arranged in ascending order according to time sequence.
Wherein t is initiation Indicating the starting moment of the output, t ending To the moment of ending the output, t sunrise And t sunset The sunrise time and the sunset time of the area where the photovoltaic station is located are respectively indicated.
3) Judging whether a > N is true, if so, entering the step 4), otherwise, returning to the step 1). N is the total number of dates of the historical output data of the photovoltaic station.
4) Judging whether the date intervals of two adjacent elements of the effective output data set are smaller than 30, if yes, entering step 5), otherwise,
5) In order for the data set to contain enough information to warrant the rationality of the relevant evaluation, the number of invalid days should be less than 30% of the total number of days in a year.
Judging whether the number of elements in the effective output data set is more than or equal to 70% A, if so, outputting the effective output data set. Otherwise, the method is used for controlling the power supply.
The correlation quantization index system establishment module establishes a single-layer tree-shaped photovoltaic station correlation quantization index system according to the effective output data set and inputs the single-layer tree-shaped photovoltaic station correlation quantization index system to the index system assessment module.
The single-layer tree-shaped photovoltaic station correlation quantification index system comprises trend correlation indexes of a daily output sequence and a daily power generation sequence, numerical correlation indexes of the daily output sequence and the daily power generation sequence and time correlation indexes of the daily output sequence and the daily power generation sequence.
The trend correlation index includes a daily output Gini correlation coefficient and a daily power generation amount Gini correlation coefficient.
The numerical correlation index comprises an intra-day output similarity, an intra-day power generation amount similarity, a maximum output duty ratio and a maximum intra-day power generation amount duty ratio.
The time correlation index comprises a large probability peak time difference, a large probability effective time length difference, a maximum peak time deviation and a maximum effective time length difference.
Solar energy output Gini related coefficient gamma of photovoltaic station p,i The following is shown:
wherein, gamma p,i Is the correlation coefficient of the solar output Gini of the ith solar photovoltaic station. P is p i,n Is the rank of the nth power sampling point in the ith day volt-field station day output sequence.A rank of an nth power sample point representing an average output sequence of the regional station. And N is the total number of sampling points of the sunrise force sequence. int (·) represents a rounding function.
Solar power generation amount Gini related coefficient gamma of photovoltaic station g The following is shown:
in the formula g i Is the rank of the ith daily power generation in the annual sequence of the solar power generation of the photovoltaic station.The rank of the average daily power generation amount of the station in the ith day area is represented. D represents the total number of days covered by data for correlation evaluation.
The within-day output similarity d p,i The following is shown:
solar power generation amount similarity d g The following is shown:
the maximum output ratio R p,i The following is shown:
maximum daily power generation rate R g The following is shown:
peak time difference d m,i The following is shown:
wherein m is i Peak time of the ith sunrise force sequence.The peak of the ith output of the group sequence of area lights Fu Changzhan.
The moment of maximum output in the sequence of output in the day, namely the peak time m of the sequence of output in the i day i The following is shown:
m i ＝{m,p m ≥max{p n ,n＝1,2...N}} (9)
effective duration difference d t,i The following is shown:
wherein t is i Is the effective duration of the ith sunrise power sequence of light Fu Changzhan.Is the effective duration of the ith day of the regional light Fu Changzhan group.
Maximum peak time difference d m,max The following is shown:
maximum effective time difference d t,max The following is shown:
where i=1, 2, …, D.
The calculation steps for calculating the large probability peak time difference M (p) are as follows:
1) Time difference d of peak m,i And (5) ascending order arrangement is performed, and a peak time difference cumulative probability curve is drawn. The abscissa of the peak time difference cumulative probability curve is the peak time difference cumulative probability p.
2) The value of the peak time difference integrated probability p is set.
3) The corresponding relation of the time difference accumulated probability curve is taken as the time difference M (p) of the large probability peak.
M (p) represents the correspondence of the time difference cumulative probability curve, where p is the cumulative probability. Then the meaning of M (p) is that the probability that the peak moveout is lower than M (p) is p, in other words that the probability that the peak moveout is (1-p) is higher than M (p). When p is chosen to be a significant probability, such as 98% or 95%, M (p) is referred to as the large probability peak moveout. Similarly, the large probability effective duration difference D (p) may be defined by D t,i And (i=1, 2 … D) drawing the obtained effective duration difference cumulative probability curve.
The calculation step of calculating the large probability effective duration difference D (p') is as follows:
1) Difference of effective duration d t,i And (5) arranging in ascending order, and drawing an effective duration difference cumulative probability curve. The abscissa of the effective duration difference cumulative probability curve is the effective duration difference cumulative probability p'.
2) The value of the effective duration difference cumulative probability p' is set.
3) And taking the corresponding relation of the effective time difference accumulated probability curve as the large probability effective time length difference D (p').
And the index system evaluation module evaluates the received single-layer tree-shaped photovoltaic field station correlation quantization index system.
The index system evaluation module evaluates the received single-layer tree-shaped photovoltaic field station correlation quantitative index system as follows:
and respectively calculating the distinguishing degree of each index in the single-layer tree-shaped photovoltaic station correlation quantization index system. Wherein the degree of distinction D of the jth index j The following is shown:
wherein x is ij The j index representing the i-th photovoltaic station. w (w) j And the weight of the j-th index is represented. i=1, 2, …, n. n is the number of photovoltaic sites. j=1, 2, …, m. m is the total number of indexes.
If D j And if the index is more than or equal to 0.002, the indexes of the index have important functions and cannot be deleted, otherwise, the index is removed from the index system.
The weight calculation module calculates the combination weight of each index in the photovoltaic station correlation quantization index system and inputs the combination weight to the photovoltaic station correlation evaluation module.
Obtaining the combination weight w of each index j The steps of (a) are as follows:
1) Subjective weight a of j-th index is obtained by expert scoring j 。
2) Calculating entropy e of the j-th index j The method comprises the following steps:
wherein x is ij lnx ij ＝0。
3) Calculating the difference coefficient g of the j-th index j The method comprises the following steps:
g j ＝1-e j (15)
4) By means of the coefficient of difference g j For weight a j Updating to obtain:
wherein w is j The combination weight of the j-th index.
The photovoltaic field station correlation evaluation module evaluates the correlation of the photovoltaic field stations.
The comprehensive evaluation and sequencing method for the correlation of the photovoltaic field station comprises the following steps:
1) All indexes in the single-layer tree-shaped photovoltaic station correlation quantization index system are cost indexes, and the improved TOPSIS method is based on the cost indexes, does not need to convert the indexes into benefit indexes in the same direction, and only needs to normalize the indexes.
Normalizing all indexes in a single-layer tree-shaped photovoltaic station correlation quantization index system to obtain a standardized index matrix X '' n×m ；
Wherein, the index matrix X 'is standardized' n×m The j-th column element of row i is as follows:
2) Establishing a weighted decision matrix V n×m The method comprises the steps of carrying out a first treatment on the surface of the Wherein the weighted decision matrix V n×m The ith row and jth column element v in (b) ij The following is shown:
V m×n ＝X' m×n W 1×m (20)
in which W is 1×m And combining the weight matrix for the index.
3) Setting an ideal solution Y + And negative ideal solution Y - The method comprises the following steps:
4) Ideal alignment of Y + And negative ideal solution Y - And (3) simplifying to obtain:
Y + ＝{0,0,…,0} (23)
Y - ＝{1,1,…,1} (24)
5) Calculation of the ith photovoltaic station v ij Euclidean distance d from ideal i+ The method comprises the following steps:
in the formula, v ij Representing the ith photovoltaic field station.
Calculation of the ith photovoltaic station v ij Euclidean distance d from negative ideal solution i- The method comprises the following steps:
6) Calculated light Fu Changzhan v ij Proximity coefficient C to photovoltaic field station clusters i The method comprises the following steps:
approximation coefficient C i And the correlation of the photovoltaic field station and the photovoltaic field station cluster is positively correlated.
The database stores data of the data input module, the correlation quantization index system building module, the index system evaluation module, the data validity evaluation module, the weight calculation module and the photovoltaic station correlation evaluation module. The database is stored in a computer readable storage medium.
Example 2:
referring to fig. 1 to 4, a photovoltaic field station correlation comprehensive evaluation method based on an improved TOPSIS mainly comprises the following steps:
1) And carrying out validity check on the output data of the photovoltaic field station based on the characteristics of the photovoltaic output and the solar radiation law.
The raw data of the correlation evaluation is a sequence of sampled powers covering at least one year. The collected data is valid for generating relevant evaluations only when the following 3 conditions are met:
1.1 Condition 1: daily regularity
According to the daily law of photovoltaic output, the daily output data must have output starting time and output ending time, and at the same time, the daily output starting time and output ending time must also meet the condition of the formula (13).
1.2 Condition 2: time continuity
Temporal continuity means that the effective days of the dataset satisfying condition 1 should have continuity. More specifically, avoiding losing one month of information and ensuring that the correlation contains the complete seasonal signature of photovoltaic power, consecutive inactive daily intervals in the annual dataset should be less than 30 days.
1.3 Condition 3: information adequacy
In order for the data set to contain enough information to warrant the rationality of the relevant evaluation, the number of invalid days should be less than 30% of the total number of days in a year.
2) And establishing a single-layer tree-shaped photovoltaic field station correlation quantization index system according to the photovoltaic output characteristics and actual engineering requirements.
The single-layer tree-shaped photovoltaic station correlation quantization index system comprises 10 indexes for respectively describing trend correlation, numerical correlation and time correlation of a daily output sequence and a daily power generation sequence. The trend correlation index includes a daily output Gini correlation coefficient and a daily power generation amount Gini correlation coefficient. The numerical correlation index includes an intra-day output similarity, an intra-day power generation amount similarity, a maximum output ratio, and a maximum intra-day power generation amount ratio. The time correlation index comprises maximum probability peak time difference, maximum probability effective duration difference, maximum peak time deviation and maximum effective duration difference.
2.1 A daily output Gini correlation coefficient and a daily power generation amount Gini correlation coefficient are calculated.
2.2 Calculating the similarity of the output in the day and the similarity of the power generation in the day.
2.3 A calculation method for calculating a maximum output ratio and a maximum daily power generation amount ratio.
2.4 A large probability peak time difference and a large probability effective time length difference are calculated.
2.5 Calculating the maximum peak time difference and the maximum effective time length difference.
3) The subjective weight of each index is determined by consulting engineering field expert, and the subjective weight is adjusted based on entropy weight method to obtain the subjective and objective combination weight of each index.
4) And quantitatively analyzing the rationality and scientificity of the index system based on four basic requirements and distinction degree.
The distinguishing degree of each index is calculated, so that the distinguishing capability of the index system is quantitatively checked.
5) And scoring and sorting the evaluation objects based on the improved TOPSIS method, and determining the comprehensive evaluation sorting result of the correlation of the photovoltaic field station.
5.1 Establishing a standardized index matrix X' n×m 。
5.2 Calculating a weighted decision matrix V n×m ：
5.3 Setting the ideal solution Y) + And negative ideal solution Y - ：
5.4 The euclidean distance between each evaluation object and the positive/negative ideal solution is calculated.
5.5 Calculating the proximity coefficient of each evaluation object.
Example 3:
referring to fig. 5 and 6, an experiment for verifying a photovoltaic field station correlation comprehensive evaluation system based on an improved TOPSIS mainly comprises the following steps:
1) Acquiring data
15min scale 2017-2018 output data of 18 photovoltaic stations in a certain area in north China.
2) Validity check of data
Validity check is performed according to conditions 1 to 3, and the proportion of invalid days in each station to all days in the original data set is obtained as shown in table 1. For most sites, the availability days occupy a satisfactory proportion in the original dataset. However, in 2017, the percentage of invalid days exceeds 30% for station 5, which means that the data set of station 5 in 2017 does not meet the information adequacy condition, cannot be used to evaluate its relevance, but the data set of station 5 in 2018 can be used for the relevance evaluation. Further, in 2017, the continuous ineffective gap of the station 5, the station 13, and the station 15 exceeds 30 days, making the time continuity unsatisfactory. Thus, comparing the validity data sets of 2017 and 2018, the data set of 2018 was selected to establish a relevant evaluation.
3) Calculation results of various indexes
Fig. 3 shows the performance of 18 stations on each index, and shows that the performance of all photovoltaic power stations is different in index, and the index system is verified to describe different correlation characteristics. In view of the completely different performance of photovoltaic power stations on these indexes, it is necessary to introduce a comprehensive evaluation method, and to integrate all the results to obtain a correlation comprehensive evaluation result. From an visual perspective, station 1, station 4, station 14, and station 16 perform better overall, while station 7 and station 10 perform poorly in a ranked list of most individual metrics. And the comprehensive evaluation results of other stations are difficult to intuitively see by the ordered list of the single indexes.
4) The subjective weight of each index is determined by consulting engineering field expert, and the subjective weight is adjusted based on entropy weight method to obtain the subjective and objective combination weight of each index.
Table 2 shows the weights of the various indices. The second column in the table shows the descending order of the difference coefficients, revealing the magnitude of the adjustment. The subjective weight and the combined weight are compared, and the weights of indexes such as the maximum power ratio and the like are found to be amplified due to the obvious distinguishing capability, and other indexes are found to be reduced due to the poor performance of distinguishing target objects.
5) And quantitatively analyzing the rationality and scientificity of the index system based on four basic requirements and distinction degree.
The discrimination degree of each index is calculated according to the formula (12), thereby quantitatively checking the discrimination capability of the index system. Table 2 shows the discrimination of the respective indices after weighting, and although the weighting assignment of the entropy weighting method amplifies the influence of the respective indices, all the indices have satisfactory discrimination ability. The discrimination coefficients of the indexes all meet D j <0.002 (j=1, 2, …, m). Therefore, all the indexes have a relatively important effect and cannot be deleted.
6) And scoring and sorting the evaluation objects based on the improved TOPSIS method, and determining the comprehensive evaluation sorting result of the correlation of the photovoltaic field station.
6.1 Establishing a standardized index matrix X' n×m 。
6.2 Calculating a weighted decision matrix V n×m 。
6.3 Setting the ideal solution Y) + And negative ideal solution Y - 。
6.4 The euclidean distance between each evaluation object and the positive/negative ideal solution is calculated.
6.5 Calculating the proximity coefficient of each evaluation object.
Table 1 proportion of invalid data in each station 2017 and 2018
Table 2 weights and discrimination of the respective indexes
Table 3 calculation results of various indexes of station 4, station 10, station 12 and station 14
The evaluation result of the improved TOPSIS model is verified by the traditional TOPSIS method. First, the correlation levels of most stations show slight differences or remain unchanged, such as station 1, station 4 and station 7, which indicates that the ordering results of the two methods have some consistency. However, for some photovoltaic stations, the relevant evaluation results show significant differences, such as station 10, station 12, and station 14. In the ranking table generated by the modified TOPSIS, the relevant evaluation was: station 14> station 4> station 12> station 10, the ranking result varies in the ranking table generated by the conventional TOPSIS: station 4> station 14> station 10> station 12. According to the results given in table 3, for most of the indexes, the performances of the stations 10 and 12 are worse than those of the stations 4 and 14. Therefore, their locations in the ranked list are worse than for both station 4 and station 14. Overall, station 4 and station 14 perform differently in most metrics. The advantages of the station 14 are amplified due to the large probability of poor effective duration and large maximum power duty cycle weight in the improved TOPSIS method, which may also explain why the station 12 is in a better ranking than the station 10 in the improved TOPSIS method. Sta 1, sta 4, sta 14, and sta 16 are top in rank, which is consistent with the analysis of single index ranking results.

Claims (4)

1. The photovoltaic station correlation comprehensive evaluation system based on the improved TOPSIS is characterized by comprising a data input module, a data validity evaluation module, a correlation quantization index system establishment module, an index system evaluation module, a weight calculation module, a photovoltaic station correlation evaluation module and a database;
the data input module acquires historical output data of the photovoltaic station and inputs the historical output data to the data validity evaluation module;
the data validity evaluation module is used for verifying the validity of the historical output data of the photovoltaic field station to obtain a valid output data set, and the valid output data set is input to the correlation quantization index system establishment module;
the correlation quantization index system establishment module establishes a single-layer tree-shaped photovoltaic station correlation quantization index system according to the effective output data set and inputs the single-layer tree-shaped photovoltaic station correlation quantization index system to the index system evaluation module;
the index system evaluation module evaluates the received single-layer tree-shaped photovoltaic station correlation quantization index system;
the weight calculation module calculates the combined weight of each index in the photovoltaic station correlation quantization index system and inputs the combined weight to the photovoltaic station correlation evaluation module;
the photovoltaic station correlation evaluation module evaluates the correlation of the photovoltaic stations;
the database stores data of the data input module, the correlation quantization index system building module, the index system evaluation module, the data validity evaluation module, the weight calculation module and the photovoltaic station correlation evaluation module;
the single-layer tree-shaped photovoltaic station correlation quantification index system comprises trend correlation indexes of a daily output sequence and a daily power generation sequence, numerical correlation indexes of the daily output sequence and the daily power generation sequence and time correlation indexes of the daily output sequence and the daily power generation sequence;
the trend correlation index of the daily output sequence and the daily power generation sequence, the numerical correlation index of the daily output sequence and the daily power generation sequence and the time correlation index of the daily output sequence and the daily power generation sequence are cost indexes;
the trend correlation index comprises a daily output Gini correlation coefficient and a daily power generation amount Gini correlation coefficient;
the numerical correlation index comprises an intra-day output similarity, an intra-day power generation amount similarity, a maximum output duty ratio and a maximum intra-day power generation amount duty ratio;
the time correlation index comprises a large probability peak time difference, a large probability effective time length difference, a maximum peak time deviation and a maximum effective time length difference;
solar energy output Gini related coefficient gamma of photovoltaic station p,i The following is shown:
wherein, gamma p,i Is the relation of the output Gini in the day of the ith solar photovoltaic stationCoefficients; r is (r) n The rank of the nth power sampling point in the ith daily output sequence of the photovoltaic station;representing the rank of the nth power sampling point of the average output sequence of all stations in the area; n is the total number of sampling points of the sunrise force sequence; int (·) represents a rounding function;
solar power generation amount Gini related coefficient gamma of photovoltaic station g The following is shown:
wherein R is i The rank of the ith daily power generation amount in the photovoltaic station daily power generation amount annual sequence;a rank indicating an average daily power generation amount of the station in the i-th day area; d represents the total number of days covered by data for correlation evaluation;
the within-day output similarity d p,i The following is shown:
wherein P is the rated power of the photovoltaic field station;representing the average rated power of all stations in the area; p is p i,n The power of the nth sampling point in the day output sequence of the ith day volt-field station; />Power of an nth sampling point of the regional station average output sequence;
solar power generation amount similarity d g The following is shown:
in the formula g i The solar energy generation amount of the ith day in the solar energy generation amount annual sequence of the photovoltaic station;an average daily power generation amount of the station in the ith day area is shown;
the maximum output ratio R p,i The following is shown:
maximum daily power generation rate R g The following is shown:
peak time difference d m,i The following is shown:
wherein m is i Peak time of the ith sunrise force sequence;peak time of ith output of region light Fu Changzhan group sequence;
the moment of maximum output in the sequence of output in the day, namely the peak time m of the sequence of output in the i day i The following is shown:
m i ＝{m,p m ≥max{p n ,n＝1,2...N}} (8)
effective duration difference d t,i The following is shown:
wherein t is i An effective duration of the ith sunrise power sequence of light Fu Changzhan;the effective duration of the ith sunrise sequence for the area light Fu Changzhan group;
maximum peak time difference d m,max The following is shown:
maximum effective time difference d t,max The following is shown:
wherein i=1, 2, …, D;
the index system evaluation module evaluates the received single-layer tree-shaped photovoltaic field station correlation quantitative index system as follows:
a1 Respectively calculating the distinguishing degree of each index in the single-layer tree-shaped photovoltaic station correlation quantization index system; wherein the degree of distinction D of the jth index j The following is shown:
wherein x is ij A j index representing an i-th photovoltaic station; w (w) j A weight representing the j-th index; i=1, 2, …, n; n is the number of photovoltaic stations; j=1, 2, …, m; m is the total index number;
a2 Whether the distinguishing degree of each index is smaller than the threshold value D min The method comprises the steps of carrying out a first treatment on the surface of the The deletion differentiation is less than the threshold D min Is an indicator of (2);
the data validity evaluation module is used for checking the validity of the historical output data of the photovoltaic station, and comprises the following steps of:
b1 Judging whether the a-th sunrise force data has the start moment and the end moment of the force output, if so, entering a step b 2), otherwise, deleting the a-th sunrise force data, and enabling a=a+1 to enter a step b 3); a is 1 as initial value;
b2 Judging whether the starting moment and the ending moment of the a-th sunrise force meet the constraint condition (13), if so, storing the a-th sunrise force data in an effective force data set, and entering a step b 3); otherwise, deleting the a-th sunrise force data, and enabling a=a+1 to enter a step b 3); the elements in the effective output data set are arranged in ascending order according to time sequence;
wherein t is initiation Indicating the starting moment of the output, t ending To the moment of ending the output, t sunrise And t sunset Respectively representing sunrise time and sunset time of the area where the photovoltaic station is located;
b3 Judging whether a > N is true, if so, entering the step b 4), otherwise, returning to the step b 1); n is the total date of the historical output data of the photovoltaic station;
b4 Judging whether the date intervals of two adjacent elements of the effective output data set are smaller than 30, if yes, entering the step b 5), otherwise, returning to the data input module to acquire the historical output data of the photovoltaic station again;
b5 Judging whether the number of elements in the effective output data set is more than or equal to 70% N, if so, outputting the effective output data set; otherwise, the data input module is returned to acquire the historical output data of the photovoltaic station again;
obtaining the combination weight w of each index j The steps of (a) are as follows:
c1 Subjective weight a) of j index obtained by expert scoring j ；
c2 Calculating entropy e of the j-th index j I.e.：
Wherein x is ij lnx ij ＝0；
c3 Calculating the difference coefficient g of the jth index j The method comprises the following steps:
g j ＝1-e j (15)
c4 Using the coefficient of difference g j For weight a j Updating to obtain:
wherein w is j Combining weights for the j-th index;
the comprehensive evaluation and sequencing method for the correlation of the photovoltaic field station comprises the following steps:
d1 Normalizing all indexes in the single-layer tree-shaped photovoltaic station correlation quantization index system to obtain a standardized index matrix X '' n×m ；
Wherein, the index matrix X 'is standardized' n×m The j-th column element of row i is as follows:
d2 A) establishing a weighted decision matrix V n×m The method comprises the steps of carrying out a first treatment on the surface of the Wherein the weighted decision matrix V n×m The ith row and jth column element v in (b) ij The following is shown:
v ij ＝x' ij w j (20)
d3 Setting the ideal solution Y) + And negative ideal solution Y - The method comprises the following steps:
d4 Alignment ideal solution Y + And negative ideal solution Y - And (3) simplifying to obtain:
d5 (ii) calculating the ith photovoltaic field station v ij Euclidean distance d from ideal i+ The method comprises the following steps:
in the formula, v ij Is a matrix V n×m Represents the value of the ith light Fu Changzhan weighted by the jth index;
calculation of the ith photovoltaic station v ij Euclidean distance d from negative ideal solution i- The method comprises the following steps:
d6 Calculating the proximity coefficient C of the photovoltaic station i The method comprises the following steps:
approximation coefficient C i And the correlation of the photovoltaic field station and the photovoltaic field station cluster is positively correlated.
2. The improved TOPSIS-based comprehensive photovoltaic terminal relevance assessment system according to claim 1, wherein the photovoltaic terminal historical output data comprises at least photovoltaic output data, photovoltaic daily power generation data of the last 1 year.
3. The improved TOPSIS-based photovoltaic field station correlation comprehensive assessment system according to claim 1, wherein the calculation step of calculating the large probability peak time difference M (p) is as follows:
1) Time difference d of peak m,i Ascending order arrangement is carried out, and a peak time difference cumulative probability curve is drawn; the abscissa of the peak time difference cumulative probability curve is the peak time difference cumulative probability p;
2) Setting the value of the peak time difference accumulation probability p;
3) The corresponding relation of the time difference accumulated probability curve is taken as the time difference M (p) of the large probability peak.
4. The improved TOPSIS-based photovoltaic field station correlation comprehensive assessment system according to claim 1, wherein the calculation step of calculating the large probability effective duration difference D (p') is as follows:
1) Difference of effective duration d t,i Ascending order arrangement is carried out, and an effective duration difference cumulative probability curve is drawn; the abscissa of the effective duration difference cumulative probability curve is the effective duration difference cumulative probability p';
2) Setting the value of the effective duration difference accumulation probability p';
3) And taking the corresponding relation of the effective time difference accumulated probability curve as the large probability effective time length difference D (p').
CN202010824036.7A 2020-08-17 2020-08-17 Photovoltaic station correlation comprehensive evaluation system based on improved TOPSIS Active CN112052430B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010824036.7A CN112052430B (en) 2020-08-17 2020-08-17 Photovoltaic station correlation comprehensive evaluation system based on improved TOPSIS

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010824036.7A CN112052430B (en) 2020-08-17 2020-08-17 Photovoltaic station correlation comprehensive evaluation system based on improved TOPSIS

Publications (2)

Publication Number Publication Date
CN112052430A CN112052430A (en) 2020-12-08
CN112052430B true CN112052430B (en) 2023-09-29

Family

ID=73600338

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010824036.7A Active CN112052430B (en) 2020-08-17 2020-08-17 Photovoltaic station correlation comprehensive evaluation system based on improved TOPSIS

Country Status (1)

Country Link
CN (1) CN112052430B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116260141B (en) * 2023-05-16 2023-10-20 国网信息通信产业集团有限公司 Reconstruction method and system for power of photovoltaic power station and reconstruction terminal

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104361250A (en) * 2014-11-25 2015-02-18 武汉大学 Photovoltaic grid connected safety evaluation method
CN107480856A (en) * 2017-07-06 2017-12-15 浙江大学 Based on the sale of electricity company power customer appraisal procedure for improving similarity to ideal solution ranking method
CN108717608A (en) * 2018-06-11 2018-10-30 国网山东省电力公司经济技术研究院 Million kilowatt beach photovoltaic plant accesses electric network synthetic decision-making technique and system
CN109193748A (en) * 2018-07-23 2019-01-11 国电南瑞科技股份有限公司 A kind of evaluation method and calculating equipment of photovoltaic digestion capability
CN110110904A (en) * 2019-04-17 2019-08-09 华北电力大学 Consider the integrated energy system optimization method of economy, independence and carbon emission
CN111293681A (en) * 2020-01-22 2020-06-16 重庆大学 Photovoltaic station output fluctuation quantitative evaluation method based on RankBoost

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190114564A1 (en) * 2017-10-18 2019-04-18 United Parcel Service Of America, Inc. Enriched Logistics System for Unmanned Vehicle Delivery of Parcels

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104361250A (en) * 2014-11-25 2015-02-18 武汉大学 Photovoltaic grid connected safety evaluation method
CN107480856A (en) * 2017-07-06 2017-12-15 浙江大学 Based on the sale of electricity company power customer appraisal procedure for improving similarity to ideal solution ranking method
CN108717608A (en) * 2018-06-11 2018-10-30 国网山东省电力公司经济技术研究院 Million kilowatt beach photovoltaic plant accesses electric network synthetic decision-making technique and system
CN109193748A (en) * 2018-07-23 2019-01-11 国电南瑞科技股份有限公司 A kind of evaluation method and calculating equipment of photovoltaic digestion capability
CN110110904A (en) * 2019-04-17 2019-08-09 华北电力大学 Consider the integrated energy system optimization method of economy, independence and carbon emission
CN111293681A (en) * 2020-01-22 2020-06-16 重庆大学 Photovoltaic station output fluctuation quantitative evaluation method based on RankBoost

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
"A Correlation Evaluation Method of PV Power Output Based on ITOPSIS";Weiyi Xia;《Journal of Electrical Engineering & Technology》;2083-2094 *
"基于层析分析和改进逼近理想解法的分层储能系统综合评估";雷琪 等;《电力系统保护与控制》;第45卷(第3期);13-19 *
"户用光伏系统实时监测与运行评价";孙伟健;《中国优秀硕士学位论文全文数据库 工程科技II辑》;C042-956 *

Also Published As

Publication number Publication date
CN112052430A (en) 2020-12-08

Similar Documents

Publication Publication Date Title
CN110610280A (en) Short-term prediction method, model, device and system for power load
Yao et al. Short-term load forecasting method based on feature preference strategy and LightGBM-XGboost
CN112258251B (en) Grey correlation-based integrated learning prediction method and system for electric vehicle battery replacement demand
WO2020098728A1 (en) Cluster load prediction method and apparatus, and storage medium
CN110895773A (en) DBN power grid load prediction method and device based on generalized demand side resources
CN117272850B (en) Elastic space analysis method for safe operation scheduling of power distribution network
CN112052430B (en) Photovoltaic station correlation comprehensive evaluation system based on improved TOPSIS
CN111222700A (en) Day-ahead electricity price probability prediction method based on dynamic network quantile model
CN116826710A (en) Peak clipping strategy recommendation method and device based on load prediction and storage medium
CN115545333A (en) Method for predicting load curve of multi-load daily-type power distribution network
CN112799897A (en) Information management method, management system and storage medium based on big data
CN111798333A (en) Energy utilization evaluation and electricity utilization safety analysis method and system
CN113902181A (en) Short-term prediction method and equipment for common variable heavy overload
CN112288130B (en) New energy consumption calculation method based on two-stage multi-objective optimization
CN117277312A (en) Gray correlation analysis-based power load influence factor method and equipment
CN116629624A (en) Economic evaluation method and terminal
CN112365164A (en) Medium-large energy user energy characteristic portrait method based on improved density peak value fast search clustering algorithm
CN116796892A (en) Short-term electric vehicle charging load probability prediction method based on composite quantile regression
CN108335042B (en) Method for calculating cleaning index of dynamic photovoltaic panel
CN116502939A (en) Load adjustment evaluation method, device, equipment and medium
CN115511230A (en) Electric energy substitution potential analysis and prediction method
CN113837486B (en) RNN-RBM-based distribution network feeder long-term load prediction method
CN116432510A (en) Flexible load prediction method based on ensemble learning-longhorn beetle whisker optimization algorithm
CN112036713A (en) Coal-to-electricity comprehensive benefit evaluation method and system based on principal component analysis
CN112949948A (en) Integrated learning method and system for electric vehicle power conversion demand interval prediction in time-sharing mode

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