CN107945046B - New energy power station output data restoration method and device - Google Patents

New energy power station output data restoration method and device Download PDF

Info

Publication number
CN107945046B
CN107945046B CN201610892554.6A CN201610892554A CN107945046B CN 107945046 B CN107945046 B CN 107945046B CN 201610892554 A CN201610892554 A CN 201610892554A CN 107945046 B CN107945046 B CN 107945046B
Authority
CN
China
Prior art keywords
power station
new energy
energy power
output
data
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
CN201610892554.6A
Other languages
Chinese (zh)
Other versions
CN107945046A (en
Inventor
于若英
党东升
王湘艳
董晓晶
赵亮
屈高强
朱凌志
赵利宁
陈宁
田星
施涛
曲立楠
葛路明
韩华玲
彭佩佩
何洁琼
吴福保
丁杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Corp of China SGCC
China Electric Power Research Institute Co Ltd CEPRI
State Grid Ningxia Electric Power Co Ltd
Economic and Technological Research Institute of State Grid Ningxia Electric Power Co Ltd
Original Assignee
State Grid Corp of China SGCC
China Electric Power Research Institute Co Ltd CEPRI
State Grid Ningxia Electric Power Co Ltd
Economic and Technological Research Institute of State Grid Ningxia Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by State Grid Corp of China SGCC, China Electric Power Research Institute Co Ltd CEPRI, State Grid Ningxia Electric Power Co Ltd, Economic and Technological Research Institute of State Grid Ningxia Electric Power Co Ltd filed Critical State Grid Corp of China SGCC
Priority to CN201610892554.6A priority Critical patent/CN107945046B/en
Publication of CN107945046A publication Critical patent/CN107945046A/en
Application granted granted Critical
Publication of CN107945046B publication Critical patent/CN107945046B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Biophysics (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Water Supply & Treatment (AREA)
  • Public Health (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Primary Health Care (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention relates to a method and a device for restoring output data of a new energy power station, wherein the method comprises the following steps: acquiring abnormal output data of the new energy power station; determining similar days and related power stations of the new energy power station; restoring abnormal output data of the new energy power station according to the similar day of the new energy power station and/or the related power station; the method provided by the invention can correct the abnormal data of the new energy power station by using the similar day of the new energy power station and the related power stations, and improves the accuracy and the reliability of the identification and repair of the abnormal data of the new energy power station.

Description

New energy power station output data restoration method and device
Technical Field
The invention relates to the technical field of new energy power generation, in particular to a method and a device for restoring output data of a new energy power station.
Background
At present, the new energy power generation technology in China is rapidly developed, the scale of a new energy power station is larger and larger, but the large-scale new energy power generation grid connection can affect links such as power grid planning, production and operation.
Because reasons such as new energy power station equipment trouble, communication interruption and error code, the new energy electricity generation operating data of monitored control system storage often has unusually, and to photovoltaic power plant, the common condition that leads to the data anomaly includes: (1) the communication interruption between the monitoring host and the data acquisition system is caused by the influence of a complex environment, so that the data is lost; (2) the sensitivity and precision of instruments such as monitoring equipment, a sensor and the like are reduced due to the change of the working environment of the equipment; (3) error codes occur in the data input, conversion and transmission processes or the data acquisition equipment cannot be normally applied; (4) in the data processing process, a unified normalized data processing mode is not adopted, and data expression is improper, and missing items, wrong items and the like of the data are caused artificially.
The abnormal data can cause inaccurate extraction of the new energy power generation operation rule, further influence power grid planning or scheduling operation decision, and possibly threaten the safety and stability of the power grid in serious cases. The abnormal data is usually repaired by adopting the trend of the data in a single day, or the abnormal value is repaired according to the average value of the data of the power station, so that the repairing precision is not high, and the data quality is poor.
Disclosure of Invention
The invention provides a method and a device for restoring output data of a new energy power station, and aims to correct abnormal data of the new energy power station by using a similar day of the new energy power station and a related power station, and improve the accuracy and the reliability of identification and restoration of the abnormal data of the new energy power station.
The purpose of the invention is realized by adopting the following technical scheme:
in a new energy power plant output data restoration method, the improvement comprising:
acquiring abnormal output data of the new energy power station;
determining similar days and related power stations of the new energy power station;
and restoring the abnormal output data of the new energy power station according to the similar day of the new energy power station and/or the related power station.
Preferably, the abnormal output data of the new energy power station includes: lost data, out-of-limit data, and invalid data.
Further, the method can be used for preparing a novel materialThe output time sequence of the new energy power station X is set as (X)1,x2...xi...xn) Wherein x isiThe output of the ith time point of the new energy power station X belongs to n, and n is the total time point number of the output data collected by the new energy power station X;
if it is
Figure BDA0001129543200000027
X is theniIs the lost data;
if xi＞xh|xi＜xlThen xiIs the out-of-limit data, wherein xhIs the upper limit value of the output of the new energy power station X, XlThe output lower limit value of the new energy power station X is obtained;
if xi-x0iI > σ or xiCorresponding normalized value
Figure BDA0001129543200000021
Satisfy the requirement of
Figure BDA0001129543200000022
X is theniIs the invalid data, wherein x0iThe theoretical output of the ith time point of the new energy power station X is sigma which is a first invalid data threshold value,
Figure BDA0001129543200000023
and epsilon is a normalized value corresponding to the output of the ith time point of the adjacent power station of the new energy power station X, and epsilon is a second invalid data threshold value.
Further, the theoretical output X of the ith time point of the new energy power station X is determined according to the following formula (1)0i：
x0i＝Ii·UDi (1)
In the formula (1), IiFor photovoltaic cell output current at moment i, UDiConnecting the terminal voltage of a grid-connected point to the photovoltaic cell at the ith moment;
determining the corresponding output force of the ith time point of the adjacent power station of the new energy power station X according to the following formula (2)Normalized value
Figure BDA0001129543200000024
Figure BDA0001129543200000025
In the formula (2), xAiIs the output of the ith time point of the adjacent power station of the new energy power station X, XAmaxIs the maximum output value, X, of the adjacent power station of the new energy power station XAminThe minimum value of the output of the adjacent power station of the new energy power station X is obtained;
determining the output X of the ith time point of the new energy power station X according to the following formula (3)iCorresponding normalized value
Figure BDA0001129543200000028
Figure BDA0001129543200000026
In the formula (3), xXmaxIs the maximum output value, X, of the new energy power station XXminAnd the minimum value of the output of the new energy power station X is obtained.
Further, the distance d between the new energy power station and the adjacent power station satisfies the following conditions: d is less than or equal to D, wherein D is the maximum geographical distance between the new energy power station and the adjacent power station.
Preferably, the determining the similar day of the new energy power station and the related power stations comprises:
enabling output data of the new energy power station on the ith day to have abnormal data;
acquiring weather data of the new energy power station, and determining weather data W of the new energy power station on the ith day by utilizing a similarity coefficientiAnd the weather data W of the j th day of the new energy power stationjDegree of similarity of
Figure BDA0001129543200000031
If it is
Figure BDA0001129543200000032
Satisfy the requirement of
Figure BDA0001129543200000033
The j day is the similar day of the i day, wherein alpha is a similar day threshold, j is an element [1,2]N is a positive integer;
if the new energy power station has a similar day, acquiring output data X of the new energy power station on the similar day and output data Y of an adjacent new energy power station of the new energy power station on the similar day, and determining similarity r of the output data X of the new energy power station on the similar day and the output data Y of the adjacent new energy power station of the new energy power station on the similar day by using a similarity coefficientXYIf r isXYSatisfy rXYIf the value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station, wherein the value beta is a related power station threshold value;
if the new energy power station has no similar day, acquiring historical output data A of the new energy power station and historical output data B of an adjacent new energy power station of the new energy power station, and determining the similarity r between the historical output data A of the new energy power station and the historical output data B of the adjacent new energy power station of the new energy power station by using a similarity coefficientABIf r isABSatisfy rABAnd if the beta value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station.
Further, the weather data includes: season type, weather type, maximum air temperature, minimum air temperature, and average air temperature information.
Preferably, the repairing the abnormal output data of the new energy power station according to the similar day of the new energy power station and/or the related power station includes:
if the new energy power station has a similar day and related power stations, taking the output data of the new energy power station on the similar day and the output data of the related power stations of the new energy power station on the similar day as BP neural network training samples, taking the output data of the related power stations of the new energy power station on the day to be repaired of the new energy power station as BP neural network input variables, and acquiring BP neural network output variables by using a BP neural network model, namely the repaired output data of the new energy power station on the day to be repaired;
if the new energy power station has a similar day and does not have a related power station, taking the output data and the weather data of the new energy power station on the similar day as BP neural network training samples, taking the output data of the new energy power station on the similar day and the weather data of the new energy power station on a day to be repaired as BP neural network input variables, and acquiring the BP neural network output variables, namely the repair output data of the new energy power station on the day to be repaired by using a BP neural network model;
if the new energy power station has no similar day and has related power stations, historical output data of the related power stations of the new energy power station is used as a BP neural network training sample, output data of the related power stations of the new energy power station on a day to be repaired of the new energy power station is used as a BP neural network input variable, and a BP neural network model is used for obtaining a BP neural network output variable, namely repair output data of the new energy power station on the day to be repaired.
In a new energy power station output data restoration device, the improvement comprising:
the acquisition module is used for acquiring abnormal output data of the new energy power station;
the determining module is used for determining the similar days of the new energy power station and the related power stations;
and the restoration module is used for restoring the abnormal output data of the new energy power station according to the similar day of the new energy power station and/or the related power station.
Preferably, the abnormal output data of the new energy power station includes: lost data, out-of-limit data, and invalid data.
Further, the output time sequence of the new energy power station X is set as (X)1,x2...xi...xn) Wherein x isiFor new energy power station XThe output of i time points, i belongs to n, and n is the total time point number of the output data collected by the new energy power station X;
if it is
Figure BDA0001129543200000046
X is theniIs the lost data;
if xi＞xh|xi＜xlThen xiIs the out-of-limit data, wherein xhIs the upper limit value of the output of the new energy power station X, XlThe output lower limit value of the new energy power station X is obtained;
if xi-x0iI > σ or xiCorresponding normalized value
Figure BDA0001129543200000041
Satisfy the requirement of
Figure BDA0001129543200000042
X is theniIs the invalid data, wherein x0iThe theoretical output of the ith time point of the new energy power station X is sigma which is a first invalid data threshold value,
Figure BDA0001129543200000043
and epsilon is a normalized value corresponding to the output of the ith time point of the adjacent power station of the new energy power station X, and epsilon is a second invalid data threshold value.
Further, the theoretical output X of the ith time point of the new energy power station X is determined according to the following formula (1)0i：
x0i＝Ii·UDi (1)
In the formula (1), IiFor photovoltaic cell output current at moment i, UDiConnecting the terminal voltage of a grid-connected point to the photovoltaic cell at the ith moment;
determining a normalized value corresponding to the output force of the ith time point of the adjacent power station of the new energy power station X according to the following formula (2)
Figure BDA0001129543200000044
Figure BDA0001129543200000045
In the formula (2), xAiIs the output of the ith time point of the adjacent power station of the new energy power station X, XAmaxIs the maximum output value, X, of the adjacent power station of the new energy power station XAminThe minimum value of the output of the adjacent power station of the new energy power station X is obtained;
determining the output X of the ith time point of the new energy power station X according to the following formula (3)iCorresponding normalized value
Figure BDA0001129543200000052
Figure BDA0001129543200000051
In the formula (3), xXmaxIs the maximum output value, X, of the new energy power station XXminAnd the minimum value of the output of the new energy power station X is obtained.
Further, the distance d between the new energy power station and the adjacent power station satisfies the following conditions: d is less than or equal to D, wherein D is the maximum geographical distance between the new energy power station and the adjacent power station.
Preferably, the determining module includes:
the first determining unit is used for enabling the output data of the new energy power station ith day to have abnormal data, acquiring the weather data of the new energy power station, and determining the weather data W of the new energy power station ith day by utilizing the similarity coefficientiAnd the weather data W of the j th day of the new energy power stationjDegree of similarity of
Figure BDA0001129543200000053
If it is
Figure BDA0001129543200000054
Satisfy the requirement of
Figure BDA0001129543200000055
The j day is the similar day of the i day, wherein alpha is a similar day threshold, j is an element [1,2]N is a positive integer;
a second determining unit, configured to, if there is a similar day in the new energy power station, obtain output data X of the new energy power station on the similar day and output data Y of an adjacent new energy power station of the new energy power station on the similar day, and determine, by using a similarity coefficient, a similarity r between the output data X of the new energy power station on the similar day and the output data Y of the adjacent new energy power station of the new energy power station on the similar dayXYIf r isXYSatisfy rXYIf the value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station, wherein the value beta is a related power station threshold value;
a third determining unit, configured to, if there is no similar day in the new energy power station, obtain historical output data a of the new energy power station and historical output data B of an adjacent new energy power station of the new energy power station, and determine a similarity r between the historical output data a of the new energy power station and the historical output data B of the adjacent new energy power station of the new energy power station by using a similarity coefficientABIf r isABSatisfy rABAnd if the beta value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station.
Further, the weather data includes: season type, weather type, maximum air temperature, minimum air temperature, and average air temperature information.
Preferably, the repair module includes:
the first restoration unit is used for taking the output data of the new energy power station on the similar day and the output data of the related power stations of the new energy power station on the similar day as BP neural network training samples if the new energy power station has the similar day and the related power stations, taking the output data of the related power stations of the new energy power station on the day to be restored of the new energy power station as BP neural network input variables, and acquiring the BP neural network output variables by using a BP neural network model, namely the restoration output data of the new energy power station on the day to be restored;
the second restoration unit is used for taking the output data and the weather data of the new energy power station on the similar day as BP neural network training samples, taking the output data of the new energy power station on the similar day and the weather data of the new energy power station on the day to be restored as BP neural network input variables, and acquiring the BP neural network output variables, namely the restoration output data of the new energy power station on the day to be restored by using a BP neural network model if the new energy power station has the similar day and does not have related power stations;
and the third repairing unit is used for taking historical output data of the related power stations of the new energy power station as BP neural network training samples, taking output data of the related power stations of the new energy power station on the day to be repaired of the new energy as BP neural network input variables, and acquiring the BP neural network output variables, namely the repairing output data of the new energy power station on the day to be repaired by using a BP neural network model if the new energy power station does not have similar days and has related power stations.
The invention has the beneficial effects that:
according to the method and the device for restoring the output data of the new energy power station, the output data of the new energy power station on the similar day, the output data of the related power station on the similar day or the historical output data of the related power station are used for restoring the abnormal data of the new energy power station by acquiring the similar day and the related power station of the new energy power station; compared with the prior art, the application scene of new energy power station data restoration is expanded, the abnormal data type can be judged more accurately, and the accuracy and the reliability of the new energy power station abnormal data restoration are improved.
Drawings
FIG. 1 is a flow chart of a new energy power station output data restoration method according to the invention;
FIG. 2 is a graph illustrating a sunrise power curve of a new energy power station including lost data according to an embodiment of the present invention;
FIG. 3 is a graph illustrating a daily output curve of a new energy power station including out-of-limit data according to an embodiment of the present invention;
FIG. 4 is a graph illustrating a sunrise curve of a new energy power station including invalid data according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of the output data of the BK power station 2013 at 1 month and 8 days in accordance with the embodiment of the invention;
FIG. 6 is a graph of the original output and the restored output of the BK power station at 1 month and 8 days in the embodiment of the invention;
fig. 7 is a schematic structural diagram of a new energy power station output data restoration device according to the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The method for restoring the output data of the new energy power station considers and analyzes the correlation of the new energy power station and the data of the adjacent power stations, can judge the type of abnormal data more accurately, and is beneficial to improving the identification accuracy of the abnormal data, as shown in fig. 1, and comprises the following steps:
101. acquiring abnormal output data of the new energy power station;
102. determining similar days and related power stations of the new energy power station;
103. and restoring the abnormal output data of the new energy power station according to the similar day of the new energy power station and/or the related power station.
In step 101, the abnormal output data of the new energy power station includes: lost data, out-of-limit data, and invalid data.
Specifically, due to the reasons of equipment failure, communication interruption, error codes and the like of the new energy power station, the new energy power generation operation data stored in the monitoring system is often abnormal, and in the technical scheme provided by the invention, the abnormal output data of the new energy power station can be obtained in the following mode:
the output time sequence of the new energy power station X is set as (X)1,x2...xi...xn) Wherein x isiThe output of the ith time point of the new energy power station X belongs to n, and n is the total time point number of the output data collected by the new energy power station X;
if it is
Figure BDA0001129543200000074
X is theniIs the lost data;
for example, as shown in fig. 2, a new energy power station is operated at 6: 00 to 22: sunrise curve of 00, wherein 14: 00 to 16: the output data in the 00 time period is lost data;
if xi＞xh|xi＜xlThen xiIs the out-of-limit data, wherein xhIs the upper limit value of the output of the new energy power station X, XlThe output lower limit value of the new energy power station X is obtained;
for example, as shown in fig. 3, a new energy power station is in 6: 00 to 22: sunrise curve of 00, wherein 10: 00 to 12: the output data in the 00 time period is out-of-limit data;
if xi-x0iI > σ or xiCorresponding normalized value
Figure BDA0001129543200000071
Satisfy the requirement of
Figure BDA0001129543200000072
X is theniIs the invalid data, wherein x0iThe theoretical output of the ith time point of the new energy power station X is sigma which is a first invalid data threshold value,
Figure BDA0001129543200000073
a normalized value corresponding to the output of the ith time point of the adjacent power station of the new energy power station X is obtained, and epsilon is second invalid dataAnd (4) a threshold value.
For example, as shown in fig. 4, a new energy power station is in 6: 00 to 22: sunrise curve of 00, wherein 12: 00 to 14: the output data in the 00 time period is invalid data;
wherein, distance d between new forms of energy power station and its adjacent power station satisfies: d is less than or equal to D, and D is the maximum geographical distance between the new energy power station and the adjacent power station.
In specific implementation, because the arrangement of the photovoltaic arrays and the system efficiency are relatively fixed in a certain period of time, and the output is determined under certain conditions of illumination and temperature, the theoretical output value of the power station can be calculated by using data such as illumination intensity and temperature collected by the power station, and the theoretical output value X of the ith time point of the new energy power station X can be determined according to the following formula (1)0i：
x0i＝Ii·UDi (1)
In the formula (1), UDiConnecting the photovoltaic cell to the voltage of the grid-connected point at the ith momentiOutputting current for the photovoltaic cell at the ith moment;
in the prior art, the output current I of the photovoltaic cell can be obtained by the following general formula:
Figure BDA0001129543200000081
in the above formula, I is the output current of the photovoltaic cell, U is the output voltage of the photovoltaic cell, S is the illumination intensity, and the standard condition is 1000W/m2T is the surface temperature of the battery, TrefTaking the temperature as a reference temperature under standard conditions, and taking the temperature of 25 ℃ IscFor short-circuit current of photovoltaic cells, I0Is diode reverse saturation current, Q is charge capacity and is 1.69 multiplied by 1019C, N is diode discharge coefficient, K is Boltzmann constant and is 1.38 multiplied by 1023J/K, RshIs the internal equivalent resistance of the battery.
The output of the adjacent power stations generally has similar trend, so that whether the new energy power station has invalid data or not is determined, and the normalization value corresponding to the output of the new energy power station and the adjacent power stations can be further usedThe comparison value of the normalized value corresponding to the output of (a) is determined, and in specific implementation, the normalized value corresponding to the output of the ith time point of the adjacent power station of the new energy power station X can be determined according to the following formula (2)
Figure BDA0001129543200000082
Figure BDA0001129543200000083
In the formula (2), xAiIs the output of the ith time point of the adjacent power station of the new energy power station X, XAmaxIs the maximum output value, X, of the adjacent power station of the new energy power station XAminThe minimum value of the output of the adjacent power station of the new energy power station X is obtained;
the output X of the new energy power station X at the ith time point can be determined according to the following formula (3)iCorresponding normalized value
Figure BDA0001129543200000085
Figure BDA0001129543200000084
In the formula (3), xXmaxIs the maximum output value, X, of the new energy power station XXminAnd the minimum value of the output of the new energy power station X is obtained.
After acquiring abnormal data of a new energy power station, acquiring a similar day of the new energy power station through correlation of weather data of the new energy power station, acquiring relevant power stations of the new energy power station through output data of adjacent power stations of the new energy power station, and aiming at two random variables { (x)i,yi) The correlation between them is usually described by correlation coefficients, such as: pearson Product Moment Correlation Coefficient (PPMCC), Kendall rank order correlation coefficient (Kendall's tau, KT) and Spearman's rho, SR)The data recovery method can adopt Pearson's product rotation coefficient (PPMCC) to obtain random variable { (x)i,yi) Correlation between rp(X, Y), defined as:
Figure BDA0001129543200000091
in the above formula, X and Y respectively represent two random variables whose correlation is to be determined, n is the number of elements in X and Y, and X isiIs the ith element in X, yiIs the ith element in Y, xaveIs the arithmetic mean of X, yaveThe average value of the weather data of the new energy power station on the day to be repaired and the historical weather data of the new energy power station on the day to be repaired are the arithmetic average value of Y, and X and Y can respectively represent the weather data of the new energy power station on the day to be repaired and the historical weather data of the new energy power station, or the output data of the new energy power station on the day to be repaired and the output data of the adjacent power stations; the correlation coefficient is between-1 and 1, and when r is 0, X and Y are called to be uncorrelated; when r is 1, X and Y are called to be completely related, and in this case, a linear functional relationship exists between X and Y; r is<1, the variation of X causes the partial variation of Y, the larger the absolute value of r is, the larger the variation of X causes the variation of Y is, and r>0.8 is called highly correlated, when r<When the correlation is low at 0.3 and moderate at other times, the similar day and the related power stations of the new energy power station are determined by the following modes when the correlation is particularly implemented:
enabling output data of the new energy power station on the ith day to have abnormal data;
acquiring weather data of the new energy power station, and determining weather data W of the new energy power station on the ith day by utilizing a similarity coefficientiAnd the weather data W of the j th day of the new energy power stationjDegree of similarity of
Figure BDA0001129543200000092
If it is
Figure BDA0001129543200000093
Satisfy the requirement of
Figure BDA0001129543200000094
The j day is the similar day of the i day, wherein alpha is a similar day threshold, j is an element [1,2]N is a positive integer;
if the new energy power station has a similar day, acquiring output data X of the new energy power station on the similar day and output data Y of an adjacent new energy power station of the new energy power station on the similar day, and determining similarity r of the output data X of the new energy power station on the similar day and the output data Y of the adjacent new energy power station of the new energy power station on the similar day by using a similarity coefficientXYIf r isXYSatisfy rXYIf the value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station, wherein the value beta is a related power station threshold value;
if the new energy power station has no similar day, acquiring historical output data A of the new energy power station and historical output data B of an adjacent new energy power station of the new energy power station, and determining the similarity r between the historical output data A of the new energy power station and the historical output data B of the adjacent new energy power station of the new energy power station by using a similarity coefficientABIf r isABSatisfy rABAnd if the beta value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station.
Wherein the weather data comprises: season type, weather type, maximum air temperature, minimum air temperature, and average air temperature information, and the specific description of the weather type is shown in table 1:
TABLE 1 weather types
Figure BDA0001129543200000101
Further, in consideration of the complex nonlinearity of the output of the new energy power station, in the technical scheme provided by the invention, the acquired similar days of the new energy power station and related power stations can be used, a BP neural network algorithm is adopted to realize the restoration of the abnormal data of the new energy power station, and the specific implementation comprises the following restoration processes:
if the new energy power station has a similar day and related power stations, taking the output data of the new energy power station on the similar day and the output data of the related power stations of the new energy power station on the similar day as BP neural network training samples, taking the output data of the related power stations of the new energy power station on the day to be repaired of the new energy power station as BP neural network input variables, and acquiring BP neural network output variables by using a BP neural network model, namely the repaired output data of the new energy power station on the day to be repaired;
if the new energy power station has a similar day and does not have a related power station, taking the output data and the weather data of the new energy power station on the similar day as BP neural network training samples, taking the output data of the new energy power station on the similar day and the weather data of the new energy power station on a day to be repaired as BP neural network input variables, and acquiring the BP neural network output variables, namely the repair output data of the new energy power station on the day to be repaired by using a BP neural network model;
if the new energy power station has no similar day and has related power stations, historical output data of the related power stations of the new energy power station is used as a BP neural network training sample, output data of the related power stations of the new energy power station on a day to be repaired of the new energy power station is used as a BP neural network input variable, and a BP neural network model is used for obtaining a BP neural network output variable, namely repair output data of the new energy power station on the day to be repaired.
For example, the BP neural network training samples, input and output variables for each scenario are determined as follows in table 2:
TABLE 2 BP neural network training samples, input and output variables under various scenarios
Figure BDA0001129543200000102
Figure BDA0001129543200000111
The specific steps of implementing data repair by using the BP neural network algorithm may include:
(1) and initializing a weight w, and setting the current as the t-th iteration.
(2) And sequentially inputting P samples, setting the current input sample as the P-th sample, and calculating the output and the back propagation error of each layer.
(3) If P < P, P ═ P +1, go to step (2), otherwise go to step (4).
(4) And adjusting the connection weight of each layer according to a weight adjustment formula.
(5) And (3) calculating the output, the reverse transmission error and the network total error E (T) of each layer according to the new connection weight, if E (T) < epsilon or T > T, terminating the training process, otherwise, if T ═ T +1, and going to the step (2) to perform a new round of training.
The invention also provides an embodiment of the new energy power station output data restoration method, and by taking the photovoltaic power station as an example, four photovoltaic power station output measured data are selected for analysis. The four power stations are BK, HH, BC and SG, and the installed capacities are respectively 10MW, 250MW, 20MW and 25 MW. The data recording time interval is from 1 month and 6 days to 10 days in 2013, and the sampling interval is 15 minutes. With the BK power station as a research object, for the purpose of comparing data restoration effects, a BK power station 2013 year 1 month 8 daily output data abnormity and missing scene is constructed, as shown in FIG. 5, the BK power station data abnormity and missing scene comprises the following steps: (1) deleting 10-11 th of data; (2) modifying the data between 11 and 12 times to be 0.6 times of the original data; (3) the data at 15-16 are modified to an outlier greater than the installed capacity.
In the embodiment provided by the invention, abnormal data of the photovoltaic power station can be identified by adopting a machine learning mode, and the constructed abnormal data is classified by a support vector machine to identify abnormal points. The time interval is set to be 0-23: 45, the time interval is 15 minutes, the total number of 96 points is obtained, and the output set X is { X ═ X1,x2,…，x96In which the set of output forces X10-11＝{x41,x42,x43,x44,x45},
Figure BDA0001129543200000113
Belonging to a lost data, a contribution set X11-12＝{x46,x47,x48,x49,x50The normalized output difference at the same time as the HH station is shown in table 3:
TABLE 3 normalized output difference between BK and HH power stations
Figure BDA0001129543200000112
Let ε equal to 0.1, x46，x47，x48，x49May be invalid data and need further analysis, and therefore, from the irradiance over this period, the theoretical output value of the BK power station is calculated as shown in table 4:
TABLE 4 comparison table of theoretical value and actual value results of power station
Figure BDA0001129543200000121
Setting the threshold value sigma to be 1MW, knowing that x46, x47, x48 and x49 are abnormal values according to the judgment method of the invalid data, and needing to repair the abnormal data;
set of forces X15-16＝{x61,x62,x63,x64,x65}，x61＝10.50，x62＝12.21，x63＝13.01，x64＝11.71，x6513.42, belonging to out-of-limit data, abnormal data repair is needed.
According to the new energy power station output data restoration method flow, the similar day of the BK power station is determined to be 1 month and 6 days, the related power station is the BC power station, and the output data of the BK power station and the BC power station in 1 month and 6 days are used as input to train a neural network model. The corrected BK power station 1 month and 8 days output curve and the original curve are obtained through the model, and are shown in FIG. 6.
As can be seen from fig. 6, the data restoration accuracy is good, and the restored output and the original output have a small deviation, where the original output data of the BK power station, the restored output data, and the output data restored by the conventional interpolation method are shown in table 5:
TABLE 5 BK original output data of power station, restored output data, and restored output data by conventional interpolation
Figure BDA0001129543200000122
Compared with the traditional interpolation method, the output data obtained by considering the restoration of the output correlation of the power station is closer to the original value, and the neural network restoration method of the output correlation of the power station has good restoration precision.
The invention also provides a new energy power station output data restoration device, as shown in fig. 7, the device includes:
the acquisition module is used for acquiring abnormal output data of the new energy power station;
the determining module is used for determining the similar days of the new energy power station and the related power stations;
and the restoration module is used for restoring the abnormal output data of the new energy power station according to the similar day of the new energy power station and/or the related power station.
Preferably, the abnormal output data of the new energy power station includes: lost data, out-of-limit data, and invalid data.
Wherein, the output time sequence of the new energy power station X is set as (X)1,x2...xi...xn) Wherein x isiThe output of the ith time point of the new energy power station X belongs to n, and n is the total time point number of the output data collected by the new energy power station X;
if it is
Figure BDA0001129543200000137
X is theniIs the lost data;
if xi＞xh|xi＜xlThen xiIs the out-of-limit data, wherein xhIs the upper limit value of the output of the new energy power station X, XlThe output lower limit value of the new energy power station X is obtained;
if xi-x0iI > σ or xiCorresponding normalized value
Figure BDA0001129543200000131
Satisfy the requirement of
Figure BDA0001129543200000132
X is theniIs the invalid data, wherein x0iThe theoretical output of the ith time point of the new energy power station X is sigma which is a first invalid data threshold value,
Figure BDA0001129543200000133
and epsilon is a normalized value corresponding to the output of the ith time point of the adjacent power station of the new energy power station X, and epsilon is a second invalid data threshold value.
Further, the theoretical output X of the ith time point of the new energy power station X is determined according to the following formula (1)0i：
x0i＝Ii·UDi (1)
In the formula (1), IiFor photovoltaic cell output current at moment i, UDiConnecting the terminal voltage of a grid-connected point to the photovoltaic cell at the ith moment;
determining a normalized value corresponding to the output force of the ith time point of the adjacent power station of the new energy power station X according to the following formula (2)
Figure BDA0001129543200000134
Figure BDA0001129543200000135
In the formula (2), xAiIs the output of the ith time point of the adjacent power station of the new energy power station X, XAmaxIs the maximum output value, X, of the adjacent power station of the new energy power station XAminThe minimum value of the output of the adjacent power station of the new energy power station X is obtained;
determining the output X of the ith time point of the new energy power station X according to the following formula (3)iCorresponding normalized value
Figure BDA0001129543200000136
Figure BDA0001129543200000141
In the formula (3), xXmaxIs the maximum output value, X, of the new energy power station XXminAnd the minimum value of the output of the new energy power station X is obtained.
The distance d between the new energy power station and the adjacent power station meets the following requirements: d is less than or equal to D, wherein D is the maximum geographical distance between the new energy power station and the adjacent power station.
The determining module comprises:
the first determining unit is used for enabling the output data of the new energy power station ith day to have abnormal data, acquiring the weather data of the new energy power station, and determining the weather data W of the new energy power station ith day by utilizing the similarity coefficientiAnd the weather data W of the j th day of the new energy power stationjDegree of similarity of
Figure BDA0001129543200000142
If it is
Figure BDA0001129543200000143
Satisfy the requirement of
Figure BDA0001129543200000144
The j day is the similar day of the i day, wherein alpha is a similar day threshold, j is an element [1,2]N is a positive integer;
a second determining unit, configured to, if there is a similar day in the new energy power station, obtain output data X of the new energy power station on the similar day and output data Y of an adjacent new energy power station of the new energy power station on the similar day, and determine, by using a similarity coefficient, a similarity r between the output data X of the new energy power station on the similar day and the output data Y of the adjacent new energy power station of the new energy power station on the similar dayXYIf r isXYSatisfy rXYIf the value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station, wherein the value beta is a related power station threshold value;
a third determining unit, configured to, if there is no similar day in the new energy power station, obtain historical output data a of the new energy power station and historical output data B of an adjacent new energy power station of the new energy power station, and determine a similarity r between the historical output data a of the new energy power station and the historical output data B of the adjacent new energy power station of the new energy power station by using a similarity coefficientABIf r isABSatisfy rABAnd if the beta value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station.
Wherein the weather data comprises: season type, weather type, maximum air temperature, minimum air temperature, and average air temperature information.
The repair module includes:
the first restoration unit is used for taking the output data of the new energy power station on the similar day and the output data of the related power stations of the new energy power station on the similar day as BP neural network training samples if the new energy power station has the similar day and the related power stations, taking the output data of the related power stations of the new energy power station on the day to be restored of the new energy power station as BP neural network input variables, and acquiring the BP neural network output variables by using a BP neural network model, namely the restoration output data of the new energy power station on the day to be restored;
the second restoration unit is used for taking the output data and the weather data of the new energy power station on the similar day as BP neural network training samples, taking the output data of the new energy power station on the similar day and the weather data of the new energy power station on the day to be restored as BP neural network input variables, and acquiring the BP neural network output variables, namely the restoration output data of the new energy power station on the day to be restored by using a BP neural network model if the new energy power station has the similar day and does not have related power stations;
and the third repairing unit is used for taking historical output data of the related power stations of the new energy power station as BP neural network training samples, taking output data of the related power stations of the new energy power station on the day to be repaired of the new energy as BP neural network input variables, and acquiring the BP neural network output variables, namely the repairing output data of the new energy power station on the day to be repaired by using a BP neural network model if the new energy power station does not have similar days and has related power stations.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.

Claims (14)

1. The method for restoring the output data of the new energy power station is characterized by comprising the following steps:
acquiring abnormal output data of the new energy power station;
determining similar days and related power stations of the new energy power station;
restoring abnormal output data of the new energy power station according to the similar day of the new energy power station and/or the related power station;
the repairing of the abnormal output data of the new energy power station according to the similar days of the new energy power station and/or the related power stations comprises the following steps:
if the new energy power station has a similar day and related power stations, taking the output data of the new energy power station on the similar day and the output data of the related power stations of the new energy power station on the similar day as BP neural network training samples, taking the output data of the related power stations of the new energy power station on the day to be repaired of the new energy power station as BP neural network input variables, and acquiring BP neural network output variables by using a BP neural network model, namely the repaired output data of the new energy power station on the day to be repaired;
if the new energy power station has a similar day and does not have a related power station, taking the output data and the weather data of the new energy power station on the similar day as BP neural network training samples, taking the output data of the new energy power station on the similar day and the weather data of the new energy power station on a day to be repaired as BP neural network input variables, and acquiring the BP neural network output variables, namely the repair output data of the new energy power station on the day to be repaired by using a BP neural network model;
if the new energy power station has no similar day and has related power stations, historical output data of the related power stations of the new energy power station is used as a BP neural network training sample, output data of the related power stations of the new energy power station on a day to be repaired of the new energy power station is used as a BP neural network input variable, and a BP neural network model is used for obtaining a BP neural network output variable, namely repair output data of the new energy power station on the day to be repaired.
2. The method of claim 1, wherein the abnormal output data of the new energy power station comprises: lost data, out-of-limit data, and invalid data.
3. The method of claim 2, wherein the energy output time sequence of the new energy power station X is set to (X)1,x2...xi...xn) Wherein x isiThe output of the ith time point of the new energy power station X belongs to n, and n is the total time point number of the output data collected by the new energy power station X;
if it is
Figure FDA0003211274650000011
X is theniIs the lost data;
if xi＞xh|xi＜xlThen xiIs the out-of-limit data, wherein xhIs the upper limit value of the output of the new energy power station X, XlThe output lower limit value of the new energy power station X is obtained;
if xi-x0i|＞Sigma or xiCorresponding normalized value
Figure FDA0003211274650000012
Satisfy the requirement of
Figure FDA0003211274650000013
X is theniIs the invalid data, wherein x0iThe theoretical output of the ith time point of the new energy power station X is sigma which is a first invalid data threshold value,
Figure FDA0003211274650000021
and epsilon is a normalized value corresponding to the output of the ith time point of the adjacent power station of the new energy power station X, and epsilon is a second invalid data threshold value.
4. The method according to claim 3, characterized in that the theoretical output X at the ith time point of the new energy plant X is determined according to the following formula (1)0i：
x0i＝Ii·UDi (1)
In the formula (1), IiFor photovoltaic cell output current at moment i, UDiConnecting the terminal voltage of a grid-connected point to the photovoltaic cell at the ith moment;
determining a normalized value corresponding to the output force of the ith time point of the adjacent power station of the new energy power station X according to the following formula (2)
Figure FDA0003211274650000022
Figure FDA0003211274650000023
In the formula (2), xAiIs the output of the ith time point of the adjacent power station of the new energy power station X, XAmaxIs the maximum output value, X, of the adjacent power station of the new energy power station XAminThe minimum value of the output of the adjacent power station of the new energy power station X is obtained;
determining the output X of the ith time point of the new energy power station X according to the following formula (3)iCorresponding normalized value
Figure FDA0003211274650000024
Figure FDA0003211274650000025
In the formula (3), xXmaxIs the maximum output value, X, of the new energy power station XXminAnd the minimum value of the output of the new energy power station X is obtained.
5. The method according to claim 3, characterized in that the distance d between the new energy plant and its neighbouring plants is such that: d is less than or equal to D, wherein D is the maximum geographical distance between the new energy power station and the adjacent power station.
6. The method of claim 1, wherein the determining similar days and related power stations for the new energy power station comprises:
enabling output data of the new energy power station on the ith day to have abnormal data;
acquiring weather data of the new energy power station, and determining weather data W of the new energy power station on the ith day by utilizing a similarity coefficientiAnd the weather data W of the j th day of the new energy power stationjDegree of similarity of
Figure FDA0003211274650000026
If it is
Figure FDA0003211274650000027
Satisfy the requirement of
Figure FDA0003211274650000028
The j day is the similar day of the i day, wherein alpha is a similar day threshold, j is an element [1,2]N is a positive integer;
if soIf the new energy power station has a similar day, acquiring output data X of the new energy power station on the similar day and output data Y of an adjacent new energy power station of the new energy power station on the similar day, and determining similarity r of the output data X of the new energy power station on the similar day and the output data Y of the adjacent new energy power station of the new energy power station on the similar day by using a similarity coefficientXYIf r isXYSatisfy rXYIf the value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station, wherein the value beta is a related power station threshold value;
if the new energy power station has no similar day, acquiring historical output data A of the new energy power station and historical output data B of an adjacent new energy power station of the new energy power station, and determining the similarity r between the historical output data A of the new energy power station and the historical output data B of the adjacent new energy power station of the new energy power station by using a similarity coefficientABIf r isABSatisfy rABAnd if the beta value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station.
7. The method of claim 6, wherein the weather data comprises: season type, weather type, maximum air temperature, minimum air temperature, and average air temperature information.
8. A new energy power station output data prosthetic devices, its characterized in that, the device includes:
the acquisition module is used for acquiring abnormal output data of the new energy power station;
the determining module is used for determining the similar days of the new energy power station and the related power stations;
the restoration module is used for restoring the abnormal output data of the new energy power station according to the similar day of the new energy power station and/or the related power station;
the repair module includes:
the first restoration unit is used for taking the output data of the new energy power station on the similar day and the output data of the related power stations of the new energy power station on the similar day as BP neural network training samples if the new energy power station has the similar day and the related power stations, taking the output data of the related power stations of the new energy power station on the day to be restored of the new energy power station as BP neural network input variables, and acquiring the BP neural network output variables by using a BP neural network model, namely the restoration output data of the new energy power station on the day to be restored;
the second restoration unit is used for taking the output data and the weather data of the new energy power station on the similar day as BP neural network training samples, taking the output data of the new energy power station on the similar day and the weather data of the new energy power station on the day to be restored as BP neural network input variables, and acquiring the BP neural network output variables, namely the restoration output data of the new energy power station on the day to be restored by using a BP neural network model if the new energy power station has the similar day and does not have related power stations;
and the third repairing unit is used for taking historical output data of the related power stations of the new energy power station as BP neural network training samples, taking output data of the related power stations of the new energy power station on the day to be repaired of the new energy as BP neural network input variables, and acquiring the BP neural network output variables, namely the repairing output data of the new energy power station on the day to be repaired by using a BP neural network model if the new energy power station does not have similar days and has related power stations.
9. The apparatus of claim 8, wherein the abnormal output data of the new energy power station comprises: lost data, out-of-limit data, and invalid data.
10. The apparatus of claim 9, wherein the timing sequence of the new energy plant X is set to (X)1,x2...xi...xn) Wherein x isiThe output of the ith time point of the new energy power station X belongs to n, and n is the total time point number of the output data collected by the new energy power station X;
if it is
Figure FDA0003211274650000041
X is theniIs the lost data;
if xi＞xh|xi＜xlThen xiIs the out-of-limit data, wherein xhIs the upper limit value of the output of the new energy power station X, XlThe output lower limit value of the new energy power station X is obtained;
if xi-x0iI > σ or xiCorresponding normalized value
Figure FDA0003211274650000042
Satisfy the requirement of
Figure FDA0003211274650000043
X is theniIs the invalid data, wherein x0iThe theoretical output of the ith time point of the new energy power station X is sigma which is a first invalid data threshold value,
Figure FDA0003211274650000044
and epsilon is a normalized value corresponding to the output of the ith time point of the adjacent power station of the new energy power station X, and epsilon is a second invalid data threshold value.
11. The apparatus according to claim 10, wherein the theoretical output X at the ith time point of the new energy power station X is determined according to the following formula (1)0i：
x0i＝Ii·UDi (1)
In the formula (1), IiFor photovoltaic cell output current at moment i, UDiConnecting the terminal voltage of a grid-connected point to the photovoltaic cell at the ith moment;
determining a normalized value corresponding to the output force of the ith time point of the adjacent power station of the new energy power station X according to the following formula (2)
Figure FDA0003211274650000045
Figure FDA0003211274650000046
In the formula (2), xAiIs the output of the ith time point of the adjacent power station of the new energy power station X, XAmaxIs the maximum output value, X, of the adjacent power station of the new energy power station XAminThe minimum value of the output of the adjacent power station of the new energy power station X is obtained;
determining the output X of the ith time point of the new energy power station X according to the following formula (3)iCorresponding normalized value
Figure FDA0003211274650000047
Figure FDA0003211274650000048
In the formula (3), xXmaxIs the maximum output value, X, of the new energy power station XXminAnd the minimum value of the output of the new energy power station X is obtained.
12. The arrangement as claimed in claim 10, characterized in that the distance d between the new energy plant and its neighbouring plants is such that: d is less than or equal to D, wherein D is the maximum geographical distance between the new energy power station and the adjacent power station.
13. The apparatus of claim 8, wherein the determining module comprises:
the first determining unit is used for enabling the output data of the new energy power station ith day to have abnormal data, acquiring the weather data of the new energy power station, and determining the weather data W of the new energy power station ith day by utilizing the similarity coefficientiAnd the weather data W of the j th day of the new energy power stationjDegree of similarity of
Figure FDA0003211274650000051
If it is
Figure FDA0003211274650000052
Satisfy the requirement of
Figure FDA0003211274650000053
The j day is the similar day of the i day, wherein alpha is a similar day threshold, j is an element [1,2]N is a positive integer;
a second determining unit, configured to, if there is a similar day in the new energy power station, obtain output data X of the new energy power station on the similar day and output data Y of an adjacent new energy power station of the new energy power station on the similar day, and determine, by using a similarity coefficient, a similarity r between the output data X of the new energy power station on the similar day and the output data Y of the adjacent new energy power station of the new energy power station on the similar dayXYIf r isXYSatisfy rXYIf the value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station, wherein the value beta is a related power station threshold value;
a third determining unit, configured to, if there is no similar day in the new energy power station, obtain historical output data a of the new energy power station and historical output data B of an adjacent new energy power station of the new energy power station, and determine a similarity r between the historical output data a of the new energy power station and the historical output data B of the adjacent new energy power station of the new energy power station by using a similarity coefficientABIf r isABSatisfy rABAnd if the beta value is more than or equal to beta, the adjacent new energy power station of the new energy power station is a related power station of the new energy power station.
14. The apparatus of claim 13, wherein the weather data comprises: season type, weather type, maximum air temperature, minimum air temperature, and average air temperature information.
CN201610892554.6A 2016-10-12 2016-10-12 New energy power station output data restoration method and device Active CN107945046B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610892554.6A CN107945046B (en) 2016-10-12 2016-10-12 New energy power station output data restoration method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610892554.6A CN107945046B (en) 2016-10-12 2016-10-12 New energy power station output data restoration method and device

Publications (2)

Publication Number Publication Date
CN107945046A CN107945046A (en) 2018-04-20
CN107945046B true CN107945046B (en) 2021-12-03

Family

ID=61928913

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610892554.6A Active CN107945046B (en) 2016-10-12 2016-10-12 New energy power station output data restoration method and device

Country Status (1)

Country Link
CN (1) CN107945046B (en)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108776683B (en) * 2018-06-01 2022-01-21 广东电网有限责任公司 Electric power operation and maintenance data cleaning method based on isolated forest algorithm and neural network
CN109583629A (en) * 2018-11-09 2019-04-05 广东电网有限责任公司电力调度控制中心 Improved similar historical day short-term load forecasting method and device based on deviation self-correcting
CN111191864A (en) * 2018-11-15 2020-05-22 中国电力科学研究院有限公司 Photovoltaic output data restoration method and system
CN110704406B (en) * 2019-08-30 2020-12-15 珠海格力电器股份有限公司 Energy data processing method, device and equipment
CN111681408B (en) * 2020-06-08 2021-01-26 广东电网有限责任公司电力科学研究院 Data restoration processing method and device
CN112200464B (en) * 2020-10-14 2023-04-28 国网山东省电力公司聊城供电公司 Correction method and system for photovoltaic power station output data considering spatial correlation
CN116709062B (en) * 2023-08-07 2023-10-20 安徽融兆智能有限公司 Electricity consumption information acquisition equipment with detection function

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105631558A (en) * 2016-03-22 2016-06-01 国家电网公司 BP neural network photovoltaic power generation system power prediction method based on similar day

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105631558A (en) * 2016-03-22 2016-06-01 国家电网公司 BP neural network photovoltaic power generation system power prediction method based on similar day

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于最优交集相似日选取的短期母线负荷综合预测;孙谦等;《中国电机工程学报》;20130205;第33卷(第04期);第126-134页 *
风电场输出功率异常数据识别与重构方法研究;朱倩雯等;《电力系统保护与控制》;20150201;第43卷(第03期);第38-45页 *

Also Published As

Publication number Publication date
CN107945046A (en) 2018-04-20

Similar Documents

Publication Publication Date Title
CN107945046B (en) New energy power station output data restoration method and device
Abbassi et al. Identification of unknown parameters of solar cell models: A comprehensive overview of available approaches
CN109873610B (en) Photovoltaic array fault diagnosis method based on IV characteristic and depth residual error network
Mekki et al. Artificial neural network-based modelling and fault detection of partial shaded photovoltaic modules
US20180046924A1 (en) Whole-life-cycle power output classification prediction system for photovoltaic systems
AU2016202556A1 (en) Model-based power estimation of photovoltaic power generation system
CN111695736B (en) Photovoltaic power generation short-term power prediction method based on multi-model fusion
Lu et al. A hybrid approach for day-ahead forecast of PV power generation
Wang et al. Performance assessment of photovoltaic modules based on daily energy generation estimation
CN111327271A (en) Photovoltaic array fault diagnosis method based on semi-supervised extreme learning machine
Ciabattoni et al. Solar irradiation forecasting using RBF networks for PV systems with storage
CN113037214B (en) Photovoltaic module shadow shielding fault diagnosis method based on S-V curve
Guerriero et al. Innovative algorithm for true maximum detection based on individual PV panel sensor network
CN116565864B (en) Photovoltaic power generation power forecasting method based on PCA-RBF algorithm
Omer et al. Adaptive boosting and bootstrapped aggregation based ensemble machine learning methods for photovoltaic systems output current prediction
Akbar et al. Optimization of accurate estimation of single diode solar photovoltaic parameters and extraction of maximum power point under different conditions
CN114819382B (en) LSTM-based photovoltaic power prediction method
CN116317937A (en) Distributed photovoltaic power station operation fault diagnosis method
Brofferio et al. An in-hand model of photovoltaic modules and/or strings for numerical simulation of renewable-energy electric power systems
KR20230089082A (en) Multilingual speech recognition artificial intelligence model(Wav2Byte) device and method thereof
Lie Power Prediction of Photovoltaic System using Neural Network Models
Muñoz et al. Fault detection and isolation in a photovoltaic system
CN111177973A (en) Photovoltaic array online modeling method based on reinforcement learning
Yagli et al. Using combinational methods for forecast improvement in PV power plants
Hu et al. A fault diagnosis method for photovoltaic modules based on transfer long short-term memory neural network

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