CN116187559A - Centralized wind power ultra-short-term power prediction method, system and cloud platform - Google Patents
Centralized wind power ultra-short-term power prediction method, system and cloud platform Download PDFInfo
- Publication number
- CN116187559A CN116187559A CN202310148695.7A CN202310148695A CN116187559A CN 116187559 A CN116187559 A CN 116187559A CN 202310148695 A CN202310148695 A CN 202310148695A CN 116187559 A CN116187559 A CN 116187559A
- Authority
- CN
- China
- Prior art keywords
- station
- ultra
- short
- power
- measured
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000011144 upstream manufacturing Methods 0.000 claims abstract description 63
- 238000005259 measurement Methods 0.000 claims abstract description 36
- 238000005457 optimization Methods 0.000 claims abstract description 7
- 230000005540 biological transmission Effects 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 14
- 238000002955 isolation Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 125000004122 cyclic group Chemical group 0.000 claims description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000009977 dual effect Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000002441 reversible effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/14—Multichannel or multilink protocols
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/02—Network architectures or network communication protocols for network security for separating internal from external traffic, e.g. firewalls
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04Q—SELECTING
- H04Q9/00—Arrangements in telecontrol or telemetry systems for selectively calling a substation from a main station, in which substation desired apparatus is selected for applying a control signal thereto or for obtaining measured values therefrom
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention discloses a centralized wind power ultra-short-term power prediction method, a system and a cloud platform, which comprise the following steps: obtaining an actual measurement wind direction of a station to be measured, calculating to obtain a dominant wind direction of the station to be measured, and calculating to obtain an upstream reference station according to the dominant wind direction; obtaining model input data and inputting a plurality of preset ultra-short-term prediction models to predict, so as to obtain a plurality of wind power ultra-short-term prediction powers of the station to be detected; and carrying out optimal prediction power optimization on the plurality of wind power ultra-short-term prediction powers to obtain the optimal ultra-short-term prediction power, and transmitting the optimal ultra-short-term prediction power to the station to be detected through the double links. According to the method, all wind farm data are concentrated to the cloud, the problems that an existing ultra-short-term prediction method lacks a historical data set and data among stations are not shared are solved, weather actual measurement data of other stations can be added according to the spatial correlation of wind speeds during ultra-short-term modeling, and the prediction accuracy is effectively improved.
Description
Technical Field
The invention relates to the technical field of wind power ultra-short-term power prediction, in particular to a centralized wind power ultra-short-term power prediction method, a system and a cloud platform.
Background
Wind power generation has obvious randomness, volatility and uncertainty, and greatly influences the stability and reliability of the operation of a power grid. In order to reduce the influence of such random fluctuations as much as possible, it is necessary to be able to accurately predict the dynamic change of the wind power in the future, and power prediction is an important technical means to solve this problem. In 2015, "two rules" issued by northwest energy monitoring bureau clearly predicts new energy power as a thin item for examination; along with the large-scale grid connection of new energy, in the national standard technical provision for wind farm access to electric power system (GB/T19963.1-2021) published in 2021, higher requirements are put forward on power prediction performance, and the 4h prediction accuracy of a power prediction result is improved from 85% to 87% before only in terms of wind power ultra-short-term prediction, so that the prediction difficulty is greatly improved.
At present, the power grid generally requires that the wind power plant report ultra-short-term predicted power every 15min, including predicted power for a time length of 4 hours in the future from the current moment, and the wind power plant is checked by the predicted power of the last point, namely the 4 th hour. The main stream ultra-short term power prediction generally adopts a one-station one-strategy and locally deployed mode, models are built for a single station, then the single model is deployed on the station to perform prediction locally, and prediction data are generated at 15min intervals to report to a power grid.
Although the ultra-short term power prediction mode is beneficial to stably acquiring real-time data such as wind power plant operation data, meteorological monitoring data and the like, the method is often limited in the aspects of improving accuracy by the following points: firstly, due to the limitation of station resources, only a single ultra-short-term prediction model can be deployed, and the storage time limit of historical data is not too long; secondly, in the model updating period, because the geographical positions of all stations are scattered, the model of each station cannot be optimized and iterated in time; thirdly, the data source is that the mode of a single station cannot share the related data of other wind farms, and the prediction modeling considering the space-time correlation cannot be performed.
Disclosure of Invention
In order to solve the technical problems, the invention provides a centralized wind power ultra-short-term power prediction method, a centralized wind power ultra-short-term power prediction system and a cloud platform, all wind power plant data are gathered to the cloud platform, prediction is performed in a plurality of modeling modes, and then a prediction result is preferentially reported, so that the accuracy of wind power ultra-short-term power prediction can be effectively improved.
In a first aspect, the invention provides a centralized wind power ultra-short term power prediction method, which comprises the following steps:
acquiring respective wind turbine generator set operation data and weather actual measurement data from each station through dual-link transmission;
acquiring the measured wind direction of a station to be measured from the meteorological measured data, calculating the dominant wind direction of the station to be measured according to the measured wind direction, and calculating the upstream reference station corresponding to the station to be measured according to the dominant wind direction;
obtaining model input data and inputting a plurality of preset ultra-short-term prediction models to predict, so as to obtain a plurality of wind power ultra-short-term prediction powers of the station to be detected, wherein the model input data comprises the wind turbine running data, the weather actual measurement data and weather prediction data of the station to be detected and the weather actual measurement data of the upstream reference station;
and according to the ultra-short-term prediction accuracy, carrying out optimal prediction power optimization on a plurality of wind power ultra-short-term prediction powers to obtain optimal ultra-short-term prediction powers, and transmitting the optimal ultra-short-term prediction powers to the station to be detected through the double links.
Further, the step of calculating the dominant wind direction of the station to be measured according to the actually measured wind direction and calculating the upstream reference station corresponding to the station to be measured according to the dominant wind direction includes:
counting the measured wind direction of the station to be measured in a first historical period according to the direction sector to obtain wind direction frequency, and taking the measured wind direction with the highest wind direction frequency as the dominant wind direction of the station to be measured;
selecting a plurality of upstream stations in the upstream wind direction range of the station to be measured from each station according to the dominant wind direction;
and calculating a pearson correlation coefficient between the measured wind speed of the station to be measured and the measured wind speed of the upstream station, and selecting an upstream reference station corresponding to the station to be measured from the upstream station according to the pearson correlation coefficient.
Further, the step of calculating a pearson correlation coefficient between the measured wind speed of the station to be measured and the measured wind speed of the upstream station, and selecting an upstream reference station corresponding to the station to be measured from the upstream station according to the pearson correlation coefficient includes:
acquiring a first actually measured wind speed sequence of the station to be measured in a second historical period and a second actually measured wind speed sequence of the upstream station in a third historical period, wherein the third historical period is a period corresponding to a first preset time length for moving the second historical period forwards;
respectively calculating pearson correlation coefficients between the first measured wind speed sequence and the second measured wind speed sequence according to different first preset time lengths, taking the maximum value of a plurality of pearson correlation coefficients as the correlation coefficient of the upstream station, and taking the first preset time length corresponding to the correlation coefficient as the delay time length of the wind speed between the station to be measured and the upstream station;
respectively calculating a difference value of a second preset time length minus the delay time length corresponding to each upstream station, and taking the upstream station with the difference value within a preset range and the maximum correlation coefficient as an upstream reference station corresponding to the station to be detected;
wherein the pearson correlation coefficient is calculated using the following formula:
wherein X is i For the i-th element in the first measured wind speed sequence,
for the elemental average value of the first measured wind speed sequence, Y i For the ith element in the second measured wind speed sequence,/-Can->
And n is the number of elements in the first measured wind speed sequence and the second measured wind speed sequence.
Further, the step of optimizing the optimal predicted power for the plurality of wind power ultra-short-term predicted powers according to the ultra-short-term predicted accuracy, and obtaining the optimal ultra-short-term predicted power comprises the following steps:
setting a time period according to a date to be detected, and acquiring the actual power of the historical wind power of the station to be detected in the time period and the ultra-short period predicted power of the historical wind power predicted by each ultra-short period predicted model, wherein the time period comprises a same-ratio period and a ring-ratio period;
calculating to obtain a daily ultra-short-term prediction accuracy according to the historical wind power actual power and the historical wind power ultra-short-term prediction power, and calculating to obtain an average accuracy loss value according to the daily ultra-short-term prediction accuracy and the days of the time period;
and taking the weighted sum of the average accuracy loss value corresponding to the same ratio period and the average accuracy loss value corresponding to the ring ratio period as an objective function, respectively calculating objective function values of the ultra-short-term prediction models, taking the ultra-short-term prediction model with the largest objective function value as an optimal prediction model, and taking the wind power ultra-short-term prediction power obtained by prediction of the optimal prediction model as optimal ultra-short-term prediction power.
Further, the daily ultra-short term prediction accuracy is calculated by adopting the following formula:
in which Q m,d For model M m Ultra short term prediction accuracy on day d, p' m,i For model M m Historical wind power ultra-short-term predicted power at ith time on day d, p i Cap is total capacity of the assembly machine of the station, and n is total time number in a day;
calculating the average accuracy loss value by adopting the following formula:
L avg ＝L/t
wherein L is an accuracy loss value, Q ref The accuracy rate reaches the standard value for ultra-short-term prediction, t is the number of days in a time period, L avg The average accuracy loss value;
the objective function is calculated using the following formula:
y＝min(aL 1,avg +bL 2,avg )
wherein L is 1,avg For model M m Average accuracy loss value, L, over the period of the same ratio 2,avg For model M m The average accuracy loss value in the cyclic ratio period, and a and b are weights of the average accuracy loss value in the same ratio period and the average accuracy loss value in the cyclic ratio period respectively.
Further, after the optimal ultrashort-term predicted power is issued to the station to be tested through the double link, the method further comprises:
counting the times of taking each ultra-short-term prediction model as the optimal prediction model in a preset period, taking the ultra-short-term prediction model with the largest times as a local prediction model, and deploying the local prediction model to the station to be detected;
judging whether the time difference between the current time and the next station reporting time is smaller than a time threshold value, if so, judging whether the issued optimal ultra-short-term predicted power is received;
and if the optimal ultra-short-term predicted power is not received, ultra-short-term power prediction is carried out through the local prediction model, and a prediction result is reported to the power grid.
Further, the dual-link transmission includes:
the wind turbine generator system operation data and weather actual measurement data of each station are respectively transmitted to a third station area from a second station area through forward isolation and to a second centralized control area through a special power line as station data;
the station data transmitted to the station three areas are transmitted to a cloud platform through a firewall;
the station data transmitted to the centralized control second area are transmitted to the centralized control third area through forward isolation, and are transmitted to the cloud platform from the centralized control third area through a firewall.
In a second aspect, the present invention provides a centralized wind power ultra-short term power prediction system, the system comprising:
the data transmission module is used for acquiring the running data and the weather actual measurement data of the respective wind turbine generator from each station through double-link transmission;
the reference station selection module is used for acquiring the measured wind direction of the station to be measured from the meteorological measured data, calculating the dominant wind direction of the station to be measured according to the measured wind direction, and calculating the upstream reference station corresponding to the station to be measured according to the dominant wind direction;
the ultra-short-term prediction module is used for acquiring model input data and inputting a plurality of preset ultra-short-term prediction models to predict, so as to obtain a plurality of wind power ultra-short-term prediction powers of the station to be detected, wherein the model input data comprises the wind turbine running data, the weather actual measurement data and weather prediction data of the station to be detected and the weather actual measurement data of the upstream reference station;
and the prediction optimizing module is used for optimizing the optimal prediction power of the wind power short-term prediction power according to the ultra-short-term prediction accuracy, obtaining the optimal ultra-short-term prediction power, and transmitting the optimal ultra-short-term prediction power to the station to be detected through the double link.
In a third aspect, an embodiment of the present invention further provides a cloud platform, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the steps of the method when executing the computer program.
The invention provides a centralized wind power ultra-short-term power prediction method, a system and a cloud platform. According to the method, data of all wind power plants are converged to the cloud platform through the dual links, a plurality of prediction models are deployed at the cloud for prediction, and optimizing and reporting are carried out according to the prediction results.
Drawings
FIG. 1 is a flow chart of a centralized wind power ultra-short term power prediction method in an embodiment of the invention;
FIG. 2 is a schematic diagram of a structure for uploading data through a dual link in an embodiment of the present invention;
FIG. 3 is a schematic diagram of a directional sector for counting dominant wind directions in an embodiment of the invention;
FIG. 4 is a schematic diagram of a structure of data transmission through dual links according to an embodiment of the present invention;
FIG. 5 is a schematic structural diagram of a centralized wind power ultra-short term power prediction system in an embodiment of the invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a method for predicting ultra-short term power of concentrated wind power according to a first embodiment of the present invention includes steps S10 to S40:
and step S10, acquiring respective wind turbine generator operation data and weather actual measurement data from each station through dual-link transmission.
In order to solve the problem that the existing ultra-short-term power prediction mode cannot share other wind power plant related data and cannot perform prediction modeling considering space-time correlation, the method collects related data of all stations through a cloud platform, namely, wind turbine running data, weather actual measurement data and real-time data in the past period of each station are uploaded to the cloud, the period is determined according to actual requirements, for example, in order to predict accuracy, the recent historical data for a few months need to be uploaded, wherein the wind turbine running data comprises cabin anemometer wind speed, wind direction, actual power and the like, and the weather actual measurement data comprises temperature, relative humidity, pressure, high-level wind speed, wind direction and the like.
Interruption of the data transmission link is the biggest problem faced by the centralized ultra-short term power prediction mode. If the uploading link fails, the cloud end cannot receive the actual measurement data of the station, and the prediction accuracy of the cloud end ultra-short-term model is affected; if the downlink fails, the cloud optimal ultra-short-term prediction result cannot be issued to the station, and the station does not have reasonable data to report, so that the accuracy of the ultra-short-term prediction data reported to the power grid is reduced. In order to ensure that the cloud can timely receive the latest actual measurement data, the invention adopts a double-link mode to transmit the data to a cloud platform, namely, two data transmission links of a station-cloud and a station-centralized control-cloud are established, the possibility of data interruption is reduced as much as possible, and the stability of uploading and downloading is ensured, as shown in fig. 2, the first link is that a station two area transmits the data to a station three area through forward isolation, and the station three area uploads the data to the cloud platform through a firewall; the second link is a station two-zone and transmits data to a centralized control two-zone through a special electric line, the centralized control two-zone is transmitted to a centralized control three-zone through forward isolation, and then the centralized control three-zone uploads the data to a cloud platform through a firewall, and the subsequent data processing and prediction are performed on the cloud platform, so that the limitation of a local station on hardware resources and data resources is overcome.
Step S20, obtaining the measured wind direction of the station to be measured from the meteorological measured data, calculating the dominant wind direction of the station to be measured according to the measured wind direction, and calculating the upstream reference station corresponding to the station to be measured according to the dominant wind direction.
Further, in order to improve accuracy of ultra-short term power prediction, the invention considers space-time correlation, and includes the steps of:
step S201, according to the direction sector, counting the measured wind direction of the station to be tested in the first history period to obtain wind direction frequency, and taking the measured wind direction with the highest wind direction frequency as the dominant wind direction of the station to be tested.
In order to facilitate explanation, the station a is used to represent the station to be tested, the measured wind direction of the station a in the weather measured data in the past period, such as several days, is obtained from the data stored in the cloud platform, please refer to fig. 3, the wind directions of all time points in the period are counted according to the direction sector, so as to obtain the wind direction frequency of the period, and the wind direction with the highest frequency is used as the dominant wind direction of the station a in the counted period, wherein the time point can be determined according to the reporting time required by the power grid, and the target power grid generally requires that the wind power plant report the ultra-short term predicted power every 15 minutes, so that the time point in the invention can be counted by 15 minutes.
Step S202, selecting a plurality of upstream stations in the upstream wind direction range of the station to be tested from all stations according to the dominant wind direction.
Step S203, calculating a pearson correlation coefficient between the measured wind speed of the station to be measured and the measured wind speed of the upstream station, and selecting an upstream reference station corresponding to the station to be measured from the upstream station according to the pearson correlation coefficient.
According to the obtained dominant wind direction, stations in the upstream wind direction range of the station A can be screened, wind moves to the upstream stations of the station A according to the spatial correlation of wind speed, and then reaches the station A, so that the measured wind speed of the upstream station in the past time period can influence the wind speed of the station A in the future time period, and the station with the largest influence on the station A is selected from the upstream stations in the next step, and the method comprises the following specific steps of:
step S2031, obtaining a first actually measured wind speed sequence of the station to be measured in a second history period and a second actually measured wind speed sequence of the upstream station in a third history period, where the third history period is a period corresponding to a first preset time period when the second history period is moved forward.
If the distance between the station A and the station B is S, the measured wind speed of the station A in the past period T is obtained, the measured wind speed of the station B in the corresponding period T is obtained, and the distance S is referred to, for example, if the period T represents 3-point to 6-point period and the period DeltaT is 3 hours, the period T-DeltaT represents 1-point to 3-point period, and due to different wind speeds, the DeltaT is variable within a certain range, so that different periods corresponding to DeltaT can be selected according to actual conditions to obtain the wind speed, and generally the DeltaT can be variable in hours, such as a range of [ 0-12 h ].
Step S2032, respectively calculating pearson correlation coefficients between the first actually measured wind speed sequence and the second actually measured wind speed sequence according to different first preset durations, taking the maximum value of a plurality of pearson correlation coefficients as the correlation coefficient of the upstream station, and taking the first preset duration corresponding to the correlation coefficient as the delay duration of the wind speed between the station to be measured and the upstream station; .
Since wind speeds are usually counted at time points such as 15min, the wind speed of the T period is actually a wind speed including a plurality of time points corresponding to each other, and these wind speeds may form a wind speed sequence, that is, there is a measured wind speed sequence X for the past period T for the station a, and there is a plurality of measured wind speed sequences Y for the past period T- Δt for the station B according to the difference of Δt, and then the pearson correlation coefficients between the sequences X and Y are calculated respectively, as shown in the following formula:
wherein X is i For the i-th element in the first measured wind speed sequence,
for the elemental average value of the first measured wind speed sequence, Y i For the ith element in the second measured wind speed sequence,/-Can->
And n is the number of elements in the first measured wind speed sequence and the second measured wind speed sequence.
And selecting DeltaT corresponding to the sequence with the maximum wind speed of the phase relation number from a plurality of pearson correlation coefficients between the sequence X and a plurality of sequences Y as the delay time of the wind speed between the station A and the station B, wherein the maximum pearson correlation coefficient can be used as the correlation coefficient of the station B.
Step S2033, respectively calculating a difference value of the second preset duration minus the delay duration corresponding to each upstream station, and taking the upstream station with the difference value within the preset range and the maximum correlation coefficient as the upstream reference station corresponding to the station to be measured.
According to the steps, the pearson correlation coefficient and the delay time length between the station A and the plurality of upstream stations B are calculated respectively, then the station B with the delay time length close to the second preset time length and the highest correlation coefficient is selected as the upstream reference station of the station A, wherein the second preset time length is usually set to be the ultra-short-term power prediction time length, for example, the ultra-short-term prediction power required by the current power grid to be reported by the wind power plant is the prediction power with the future 4 hours from the current moment, and therefore the second preset time length can be set to be 4 hours. Through the steps, the reference station with the greatest influence on the station to be detected can be selected, and the prediction power can be optimally predicted from the spatial correlation.
Step S30, obtaining model input data and inputting a plurality of preset ultra-short-term prediction models to predict, so as to obtain a plurality of wind power ultra-short-term prediction powers of the station to be detected, wherein the model input data comprises the wind turbine generator set operation data, the weather actual measurement data and weather prediction data of the station to be detected and the weather actual measurement data of the upstream reference station.
The method comprises the steps of inquiring predicted power of a station A in a future T1 period, namely, the time length required to be reported, such as weather predicted data in a future 4-hour period, acquiring fan operation data and weather actual measurement data in a past period from a cloud platform, wherein the past period is selected according to the requirement of an ultra-short-term prediction model, and acquiring weather actual measurement data of an upstream reference station B in a past T1-delta T period, wherein the past T1-delta T period is a period after the previous reported period is advanced by delta T. And then taking the obtained data as the input of a model, respectively inputting the obtained data into a plurality of pre-deployed ultra-short-term prediction models to predict, thereby obtaining the ultra-short-term prediction power in the future period T1, wherein the input data can be flexibly adapted according to different prediction models, and the ultra-short-term prediction models can adopt a plurality of existing ultra-short-term prediction models and are not described too much.
And S40, carrying out optimal prediction power optimization on a plurality of wind power ultra-short-term prediction powers according to the ultra-short-term prediction accuracy, obtaining optimal ultra-short-term prediction powers, and transmitting the optimal ultra-short-term prediction powers to the station to be detected through the double link.
After obtaining a plurality of ultra-short-term predicted powers, performing competitive optimization on a plurality of models, namely performing prediction performance evaluation on a plurality of prediction models based on time periods, wherein the specific steps are as follows:
and S401, setting a time period according to a date to be detected, and acquiring the actual power of the historical wind power of the station to be detected in the time period and the ultra-short-period predicted power of the historical wind power predicted by each ultra-short-period prediction model, wherein the time period comprises a same-ratio period and a ring-ratio period.
The time period is set according to the date to be measured, assuming that the current date is D, D1 days before and after the same day of the last year can be selected as the same period, D2 days before the current date can be selected as the ring period, wherein D1 and D2 can be the same days, such as one week or half month, different days can be selected according to actual conditions, and the performance of each prediction model can be evaluated in the set same period and the ring period.
And step S402, calculating to obtain a daily ultra-short-term prediction accuracy according to the historical wind power actual power and the historical wind power ultra-short-term prediction power, and calculating to obtain an average accuracy loss value according to the daily ultra-short-term prediction accuracy and the days of the time period.
Inquiring actual power of a wind power plant in the same cycle and the ring cycle and ultra-short-period prediction power of each ultra-short-period prediction model, and then according to the following formula, the daily ultra-short-period prediction accuracy in the time cycle is:
in which Q m,d For model M m Ultra short term prediction accuracy on day d, p' m,i For model M m Historical wind power ultra-short-term predicted power at ith time on day d, p i For the historical wind power actual power at the ith moment, cap is the final assembly machine of the stationCapacity, n, is the total number of times of day.
The time in the above formula is a time point, and still taking 15min as an example, the time point n of one day is 96, which is 24×60/15=96.
Then calculating an accuracy loss value according to the ultra-short-term prediction accuracy of each day, and further calculating an average accuracy loss value according to the days of the time period:
L avg ＝L/t
wherein L is an accuracy loss value, Q ref The accuracy rate reaches the standard value for ultra-short-term prediction, t is the number of days in a time period, L avg Is the average accuracy loss value. Wherein Q is ref Can be set to 85% according to the current requirements of the power grid.
And S403, respectively calculating objective function values of the ultra-short-term prediction models by taking the weighted sum of the average accuracy loss value corresponding to the same ratio period and the average accuracy loss value corresponding to the ring ratio period as an objective function, taking the ultra-short-term prediction model with the largest objective function value as an optimal prediction model, and taking the wind power ultra-short-term prediction power obtained by prediction of the optimal prediction model as optimal ultra-short-term prediction power.
Calculating an average accuracy loss value in the same ratio period and an average accuracy loss value in the ring ratio period through the formula, carrying out weighted summation on the two average accuracy loss values, and taking the two average accuracy loss values as an objective function, namely:
y＝min(aL 1,avg +bL 2,avg )
wherein L is 1,avg For model M m Average accuracy loss value, L, over the period of the same ratio 2,avg For model M m The average accuracy loss value in the cyclic ratio period, and a and b are weights of the average accuracy loss value in the same ratio period and the average accuracy loss value in the cyclic ratio period respectively.
Wherein, a and b are self-defined weight values which can be set to 0.5 and 0.5 or other proportional weights, and the weight values are self-defined according to the importance of the same-ratio period and the ring-ratio period. According to the objective function, selecting a prediction model with the minimum objective function value as an optimal prediction model, and issuing a prediction result of the optimal prediction model as optimal ultra-short-term prediction power.
The method for carrying out ultra-short-term power prediction on the cloud platform overcomes the problem that the station end is limited by hardware resources, predicts through a plurality of ultra-short-term prediction models, then selects and reports the optimal result, fully utilizes the advantages of a plurality of prediction models, can select a more used prediction model according to different meteorological conditions, and solves the defect that a single model cannot adapt to all meteorological conditions, so that the result of each prediction can keep higher accuracy.
Referring to fig. 4, for the optimal ultra-short-term predicted power, the optimal ultra-short-term predicted power is also issued from the cloud platform to the station in a dual-link manner, and the same as the uploading link is adopted, the first issuing link is that the cloud platform issues data to the station three region through the firewall, the station three region generates a file and transmits the file to the station two region through reverse isolation of the station, and the station two region analyzes the file to acquire the data and reports the data; the second downlink is that the cloud platform transmits data to the centralized control three areas through a firewall, the centralized control three areas generate files and then transmit the files to the centralized control two areas through centralized control reverse isolation, then the centralized control two areas transmit the data to the station two areas through the power special line for reporting, and the station can report ultra-short-term predicted power to the power grid at regular time according to time requirements.
Further, in consideration of burstiness of network and equipment faults and uncertainty of recovery, a set of ultra-short-term prediction models are deployed locally at a station, wherein the models are one of a plurality of ultra-short-term prediction models of a cloud platform, which is stable in statistics in a certain period, for example, the number of times that each ultra-short-term prediction model is taken as an optimal prediction model in a period of time is counted, the ultra-short-term prediction model with the largest number of times is deployed locally and is taken as a final redundancy reporting model when data delivery failure of the cloud occurs, namely, a traditional ultra-short-term prediction model is taken as a bottom protection measure, so that accuracy of ultra-short-term reporting results is guaranteed.
The station reports the ultra-short-term prediction power once every 15min according to the current power grid requirement, at this moment, the station can judge whether the time difference between the current time and the next reporting time is smaller than a time threshold value through one cycle, and because a certain time is required to be reserved for data acquisition and reporting, the time threshold value can be set to be a few minutes, when the time difference is smaller than the threshold value, the station can prepare to report the prediction result, firstly, whether the optimal ultra-short-term prediction power issued by the cloud platform is received or not is inquired, if yes, the optimal ultra-short-term prediction power issued by the cloud platform is directly uploaded to the power grid, if not received, the locally deployed ultra-short-term prediction model is immediately started for prediction, and the prediction result is reported to the power grid.
According to the centralized wind power ultra-short-term power prediction method provided by the embodiment, all wind power plant data are concentrated to the cloud, so that the problems that an existing ultra-short-term prediction method lacks a historical data set and data among stations are not shared are solved, weather actual measurement data of other stations can be added according to the spatial correlation of wind speeds when an ultra-short-term prediction model is built, and especially the method is greatly helpful for weather conditions which are difficult to predict in numerical weather prediction such as sudden change of wind speeds, and the prediction accuracy is effectively improved.
Referring to fig. 5, based on the same inventive concept, a centralized wind power ultra-short term power prediction system according to a second embodiment of the present invention includes:
the data transmission module 10 is used for acquiring respective wind turbine running data and weather actual measurement data from each station through dual-link transmission;
the reference station selection module 20 is configured to obtain an actual measurement wind direction of a station to be measured from the meteorological actual measurement data, calculate a dominant wind direction of the station to be measured according to the actual measurement wind direction, and calculate an upstream reference station corresponding to the station to be measured according to the dominant wind direction;
the ultra-short-term prediction module 30 is configured to obtain model input data and input a plurality of preset ultra-short-term prediction models to perform prediction, so as to obtain a plurality of wind power ultra-short-term prediction powers of the station to be tested, where the model input data includes the wind turbine running data, the weather actual measurement data, the weather prediction data, and the weather actual measurement data of the upstream reference station;
and the prediction optimizing module 40 is configured to perform optimal prediction power optimization on a plurality of wind power ultra-short term prediction powers according to the ultra-short term prediction accuracy, obtain an optimal ultra-short term prediction power, and send the optimal ultra-short term prediction power to the station to be tested through the double link.
The technical features and technical effects of the centralized wind power ultra-short-term power prediction system provided by the embodiment of the invention are the same as those of the method provided by the embodiment of the invention, and are not repeated here. All or part of each module in the centralized wind power ultra-short-term power prediction system can be realized by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In addition, the embodiment of the invention also provides a cloud platform which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the steps of the method when executing the computer program.
In summary, the embodiment of the invention provides a centralized wind power ultra-short-term power prediction method, a system and a cloud platform, wherein the method obtains respective wind turbine running data and weather actual measurement data from each station through dual-link transmission; acquiring the measured wind direction of a station to be measured from the meteorological measured data, calculating the dominant wind direction of the station to be measured according to the measured wind direction, and calculating the upstream reference station corresponding to the station to be measured according to the dominant wind direction; obtaining model input data and inputting a plurality of preset ultra-short-term prediction models to predict, so as to obtain a plurality of wind power ultra-short-term prediction powers of the station to be detected, wherein the model input data comprises the wind turbine running data, the weather actual measurement data and weather prediction data of the station to be detected and the weather actual measurement data of the upstream reference station; and according to the ultra-short-term prediction accuracy, carrying out optimal prediction power optimization on a plurality of wind power ultra-short-term prediction powers to obtain optimal ultra-short-term prediction powers, and transmitting the optimal ultra-short-term prediction powers to the station to be detected through the double links. According to the method, all wind farm data are concentrated to the cloud, the problems that an existing ultra-short-term prediction method lacks a historical data set and data among stations are not shared are solved, weather actual measurement data of other stations can be added according to the spatial correlation of wind speeds during ultra-short-term modeling, and the method is particularly helpful for weather conditions which are difficult to predict in numerical weather prediction such as sudden change of wind speeds, and the prediction accuracy is effectively improved.
In this specification, each embodiment is described in a progressive manner, and all the embodiments are directly the same or similar parts referring to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments. It should be noted that, any combination of the technical features of the foregoing embodiments may be used, and for brevity, all of the possible combinations of the technical features of the foregoing embodiments are not described, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples represent only a few preferred embodiments of the present application, which are described in more detail and are not thereby to be construed as limiting the scope of the invention. It should be noted that modifications and substitutions can be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and substitutions should also be considered to be within the scope of the present application. Therefore, the protection scope of the patent application is subject to the protection scope of the claims.
Claims (9)
1. The centralized wind power ultra-short-term power prediction method is characterized by comprising the following steps of:
acquiring respective wind turbine generator set operation data and weather actual measurement data from each station through dual-link transmission;
acquiring the measured wind direction of a station to be measured from the meteorological measured data, calculating the dominant wind direction of the station to be measured according to the measured wind direction, and calculating the upstream reference station corresponding to the station to be measured according to the dominant wind direction;
obtaining model input data and inputting a plurality of preset ultra-short-term prediction models to predict, so as to obtain a plurality of wind power ultra-short-term prediction powers of the station to be detected, wherein the model input data comprises the wind turbine running data, the weather actual measurement data and weather prediction data of the station to be detected and the weather actual measurement data of the upstream reference station;
and according to the ultra-short-term prediction accuracy, carrying out optimal prediction power optimization on a plurality of wind power ultra-short-term prediction powers to obtain optimal ultra-short-term prediction powers, and transmitting the optimal ultra-short-term prediction powers to the station to be detected through the double links.
2. The method for predicting ultra-short term power of concentrated wind power according to claim 1, wherein the steps of calculating a dominant wind direction of the station to be measured according to the measured wind direction and calculating an upstream reference station corresponding to the station to be measured according to the dominant wind direction include:
counting the measured wind direction of the station to be measured in a first historical period according to the direction sector to obtain wind direction frequency, and taking the measured wind direction with the highest wind direction frequency as the dominant wind direction of the station to be measured;
selecting a plurality of upstream stations in the upstream wind direction range of the station to be measured from each station according to the dominant wind direction;
and calculating a pearson correlation coefficient between the measured wind speed of the station to be measured and the measured wind speed of the upstream station, and selecting an upstream reference station corresponding to the station to be measured from the upstream station according to the pearson correlation coefficient.
3. The method for predicting ultra-short term power of concentrated wind power according to claim 2, wherein the step of calculating a pearson correlation coefficient between the measured wind speed of the station to be measured and the measured wind speed of the upstream station, and selecting an upstream reference station corresponding to the station to be measured from the upstream station according to the pearson correlation coefficient comprises:
acquiring a first actually measured wind speed sequence of the station to be measured in a second historical period and a second actually measured wind speed sequence of the upstream station in a third historical period, wherein the third historical period is a period corresponding to a first preset time length for moving the second historical period forwards;
respectively calculating pearson correlation coefficients between the first measured wind speed sequence and the second measured wind speed sequence according to different first preset time lengths, taking the maximum value of a plurality of pearson correlation coefficients as the correlation coefficient of the upstream station, and taking the first preset time length corresponding to the correlation coefficient as the delay time length of the wind speed between the station to be measured and the upstream station;
respectively calculating a difference value of a second preset time length minus the delay time length corresponding to each upstream station, and taking the upstream station with the difference value within a preset range and the maximum correlation coefficient as an upstream reference station corresponding to the station to be detected;
wherein the pearson correlation coefficient is calculated using the following formula:
wherein X is i For the i-th element in the first measured wind speed sequence,
for the elemental average value of the first measured wind speed sequence, Y i For the ith element in the second measured wind speed sequence,/-Can->
And n is the number of elements in the first measured wind speed sequence and the second measured wind speed sequence.
4. The method for predicting ultra-short term power of concentrated wind power according to claim 1, wherein the step of optimizing the optimal predicted power for the plurality of ultra-short term predicted powers according to the ultra-short term prediction accuracy, to obtain the optimal ultra-short term predicted power comprises:
setting a time period according to a date to be detected, and acquiring the actual power of the historical wind power of the station to be detected in the time period and the ultra-short period predicted power of the historical wind power predicted by each ultra-short period predicted model, wherein the time period comprises a same-ratio period and a ring-ratio period;
calculating to obtain a daily ultra-short-term prediction accuracy according to the historical wind power actual power and the historical wind power ultra-short-term prediction power, and calculating to obtain an average accuracy loss value according to the daily ultra-short-term prediction accuracy and the days of the time period;
and taking the weighted sum of the average accuracy loss value corresponding to the same ratio period and the average accuracy loss value corresponding to the ring ratio period as an objective function, respectively calculating objective function values of the ultra-short-term prediction models, taking the ultra-short-term prediction model with the largest objective function value as an optimal prediction model, and taking the wind power ultra-short-term prediction power obtained by prediction of the optimal prediction model as optimal ultra-short-term prediction power.
5. The method for predicting ultra-short term power of concentrated wind power according to claim 4, wherein the daily ultra-short term prediction accuracy is calculated by adopting the following formula:
in which Q m,d For model M m Ultra short term prediction accuracy on day d, p' m,i For model M m Historical wind power ultra-short-term predicted power at ith time on day d, p i Cap is total capacity of the assembly machine of the station, and n is total time number in a day;
calculating the average accuracy loss value by adopting the following formula:
L avg ＝L/t
wherein L is an accuracy loss value, Q ref The accuracy rate reaches the standard value for ultra-short-term prediction, t is the number of days in a time period, L avg The average accuracy loss value;
the objective function is calculated using the following formula:
y＝min(aL 1，avg +bL 2，avg )
wherein L is 1,avg For model M m Average accuracy loss value, L, over the period of the same ratio 2,avg For model M m The average accuracy loss value in the cyclic ratio period, and a and b are weights of the average accuracy loss value in the same ratio period and the average accuracy loss value in the cyclic ratio period respectively.
6. The method for predicting ultra-short term power of concentrated wind power according to claim 4, further comprising, after said issuing said optimal ultra-short term predicted power to said station under test via said duplex link:
counting the times of taking each ultra-short-term prediction model as the optimal prediction model in a preset period, taking the ultra-short-term prediction model with the largest times as a local prediction model, and deploying the local prediction model to the station to be detected;
judging whether the time difference between the current time and the next station reporting time is smaller than a time threshold value, if so, judging whether the issued optimal ultra-short-term predicted power is received;
and if the optimal ultra-short-term predicted power is not received, ultra-short-term power prediction is carried out through the local prediction model, and a prediction result is reported to the power grid.
7. The method for predicting ultra-short term power of concentrated wind power according to claim 1, wherein the double-link transmission comprises:
the wind turbine generator system operation data and weather actual measurement data of each station are respectively transmitted to a third station area from a second station area through forward isolation and to a second centralized control area through a special power line as station data;
the station data transmitted to the station three areas are transmitted to a cloud platform through a firewall;
the station data transmitted to the centralized control second area are transmitted to the centralized control third area through forward isolation, and are transmitted to the cloud platform from the centralized control third area through a firewall.
8. A centralized wind power ultra-short term power prediction system, comprising:
the data transmission module is used for acquiring the running data and the weather actual measurement data of the respective wind turbine generator from each station through double-link transmission;
the reference station selection module is used for acquiring the measured wind direction of the station to be measured from the meteorological measured data, calculating the dominant wind direction of the station to be measured according to the measured wind direction, and calculating the upstream reference station corresponding to the station to be measured according to the dominant wind direction;
the ultra-short-term prediction module is used for acquiring model input data and inputting a plurality of preset ultra-short-term prediction models to predict, so as to obtain a plurality of wind power ultra-short-term prediction powers of the station to be detected, wherein the model input data comprises the wind turbine running data, the weather actual measurement data and weather prediction data of the station to be detected and the weather actual measurement data of the upstream reference station;
and the prediction optimizing module is used for optimizing the optimal prediction power of the wind power ultra-short-term prediction power according to the ultra-short-term prediction accuracy, obtaining the optimal ultra-short-term prediction power, and transmitting the optimal ultra-short-term prediction power to the station to be detected through the double links.
9. A cloud platform comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310148695.7A CN116187559B (en) | 2023-02-21 | 2023-02-21 | Centralized wind power ultra-short-term power prediction method, system and cloud platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310148695.7A CN116187559B (en) | 2023-02-21 | 2023-02-21 | Centralized wind power ultra-short-term power prediction method, system and cloud platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116187559A true CN116187559A (en) | 2023-05-30 |
| CN116187559B CN116187559B (en) | 2024-03-15 |
Family
ID=86450258
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310148695.7A Active CN116187559B (en) | 2023-02-21 | 2023-02-21 | Centralized wind power ultra-short-term power prediction method, system and cloud platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116187559B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116661831A (en) * | 2023-07-21 | 2023-08-29 | 国能日新科技股份有限公司 | Automatic upgrading system and method for power generation station algorithm |
Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090195275A1 (en) * | 2008-02-01 | 2009-08-06 | Daniel Joseph Friedman | Technique for efficiently managing both short-term and long-term frequency adjustments of an electronic circuit clock signal |
| US20100180694A1 (en) * | 2009-01-20 | 2010-07-22 | Richard Duncan Ely, III | Bernoulli wind prediction system |
| CN102269124A (en) * | 2011-06-30 | 2011-12-07 | 内蒙古电力勘测设计院 | Ultra-short term wind power station generated power forecasting system |
| CN103745274A (en) * | 2014-01-08 | 2014-04-23 | 贵州电力试验研究院 | Short-term power prediction method applied to dispersed wind power |
| US20150039228A1 (en) * | 2012-02-13 | 2015-02-05 | State Grid Corporation Of China | Ultra-short-term forecasting method including real-time monitoring of the effect of upper and lower courses |
| US20150160373A1 (en) * | 2013-12-07 | 2015-06-11 | Cardinal Wind, Inc. | Computer-implemented data analysis methods and systems for wind energy assessments |
| CN106529741A (en) * | 2016-12-12 | 2017-03-22 | 国网电力科学研究院 | Space relevant characteristic-based ultra-short-period wind power prediction method |
| CN109784563A (en) * | 2019-01-18 | 2019-05-21 | 南方电网科学研究院有限责任公司 | A kind of ultra-short term power forecasting method based on virtual anemometer tower technology |
| US20190258235A1 (en) * | 2016-09-14 | 2019-08-22 | flexciton Limited | Automated supervisory control system |
| CN111144663A (en) * | 2019-12-31 | 2020-05-12 | 上海电力大学 | Ultra-short-term wind power prediction method for offshore wind farm considering output fluctuation process |
| CN111478312A (en) * | 2019-11-20 | 2020-07-31 | 国网河北省电力有限公司电力科学研究院 | Comprehensive energy cluster coordination control method for improving power grid stability |
| CN111582544A (en) * | 2020-04-08 | 2020-08-25 | 湖北省气象服务中心（湖北省专业气象服务台） | Power grid load prediction method, device and system under new energy grid-connected condition |
| CN112417028A (en) * | 2020-11-26 | 2021-02-26 | 国电南瑞科技股份有限公司 | Wind speed time sequence characteristic mining method and short-term wind power prediction method |
| CN112668775A (en) * | 2020-12-25 | 2021-04-16 | 西安翔迅科技有限责任公司 | Air quality prediction method based on time sequence convolution network algorithm |
| CN112700050A (en) * | 2021-01-04 | 2021-04-23 | 南京南瑞继保电气有限公司 | Method and system for predicting ultra-short-term 1 st point power of photovoltaic power station |
| CN113722903A (en) * | 2021-08-24 | 2021-11-30 | 华北电力大学 | Photo-thermal power generation capacity configuration method for full-renewable energy source sending-end system |
| CN114707734A (en) * | 2022-04-12 | 2022-07-05 | 华润电力技术研究院有限公司 | Method, device and medium for predicting ultra-short-term wind power |
| CN114844032A (en) * | 2022-03-28 | 2022-08-02 | 华能宁南风力发电有限公司 | Wind power plant wind power prediction method and system |
| CN115374995A (en) * | 2022-06-22 | 2022-11-22 | 国网浙江省电力有限公司丽水供电公司 | Distributed photovoltaic and small wind power station power prediction method |
| US20220373984A1 (en) * | 2021-05-19 | 2022-11-24 | Shandong University | Hybrid photovoltaic power prediction method and system based on multi-source data fusion |
| US20230037193A1 (en) * | 2021-07-26 | 2023-02-02 | Dalian University Of Technology | Wind power output interval prediction method |
-
2023
- 2023-02-21 CN CN202310148695.7A patent/CN116187559B/en active Active
Patent Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090195275A1 (en) * | 2008-02-01 | 2009-08-06 | Daniel Joseph Friedman | Technique for efficiently managing both short-term and long-term frequency adjustments of an electronic circuit clock signal |
| US20100180694A1 (en) * | 2009-01-20 | 2010-07-22 | Richard Duncan Ely, III | Bernoulli wind prediction system |
| CN102269124A (en) * | 2011-06-30 | 2011-12-07 | 内蒙古电力勘测设计院 | Ultra-short term wind power station generated power forecasting system |
| US20150039228A1 (en) * | 2012-02-13 | 2015-02-05 | State Grid Corporation Of China | Ultra-short-term forecasting method including real-time monitoring of the effect of upper and lower courses |
| US20150160373A1 (en) * | 2013-12-07 | 2015-06-11 | Cardinal Wind, Inc. | Computer-implemented data analysis methods and systems for wind energy assessments |
| CN103745274A (en) * | 2014-01-08 | 2014-04-23 | 贵州电力试验研究院 | Short-term power prediction method applied to dispersed wind power |
| US20190258235A1 (en) * | 2016-09-14 | 2019-08-22 | flexciton Limited | Automated supervisory control system |
| CN106529741A (en) * | 2016-12-12 | 2017-03-22 | 国网电力科学研究院 | Space relevant characteristic-based ultra-short-period wind power prediction method |
| CN109784563A (en) * | 2019-01-18 | 2019-05-21 | 南方电网科学研究院有限责任公司 | A kind of ultra-short term power forecasting method based on virtual anemometer tower technology |
| CN111478312A (en) * | 2019-11-20 | 2020-07-31 | 国网河北省电力有限公司电力科学研究院 | Comprehensive energy cluster coordination control method for improving power grid stability |
| CN111144663A (en) * | 2019-12-31 | 2020-05-12 | 上海电力大学 | Ultra-short-term wind power prediction method for offshore wind farm considering output fluctuation process |
| CN111582544A (en) * | 2020-04-08 | 2020-08-25 | 湖北省气象服务中心（湖北省专业气象服务台） | Power grid load prediction method, device and system under new energy grid-connected condition |
| CN112417028A (en) * | 2020-11-26 | 2021-02-26 | 国电南瑞科技股份有限公司 | Wind speed time sequence characteristic mining method and short-term wind power prediction method |
| CN112668775A (en) * | 2020-12-25 | 2021-04-16 | 西安翔迅科技有限责任公司 | Air quality prediction method based on time sequence convolution network algorithm |
| CN112700050A (en) * | 2021-01-04 | 2021-04-23 | 南京南瑞继保电气有限公司 | Method and system for predicting ultra-short-term 1 st point power of photovoltaic power station |
| US20220373984A1 (en) * | 2021-05-19 | 2022-11-24 | Shandong University | Hybrid photovoltaic power prediction method and system based on multi-source data fusion |
| US20230037193A1 (en) * | 2021-07-26 | 2023-02-02 | Dalian University Of Technology | Wind power output interval prediction method |
| CN113722903A (en) * | 2021-08-24 | 2021-11-30 | 华北电力大学 | Photo-thermal power generation capacity configuration method for full-renewable energy source sending-end system |
| CN114844032A (en) * | 2022-03-28 | 2022-08-02 | 华能宁南风力发电有限公司 | Wind power plant wind power prediction method and system |
| CN114707734A (en) * | 2022-04-12 | 2022-07-05 | 华润电力技术研究院有限公司 | Method, device and medium for predicting ultra-short-term wind power |
| CN115374995A (en) * | 2022-06-22 | 2022-11-22 | 国网浙江省电力有限公司丽水供电公司 | Distributed photovoltaic and small wind power station power prediction method |
Non-Patent Citations (10)
| Title |
|---|
| 史坤鹏;赵伟;李婷;刘梦华;王泽一;: "计及熵权指标及关联度排序的风电历史数据挖掘", 电测与仪表, no. 04, 25 February 2017 (2017-02-25) * |
| 叶林;张慈杭;汤涌;孙舶皓;仲悟之;蓝海波;刘辉;李湃;黄越辉;: "多时空尺度协调的风电集群有功分层预测控制方法", 中国电机工程学报, no. 13 * |
| 叶林;杨丹萍;赵永宁;: "风电场风能资源评估的测量―关联―预测方法综述", 电力系统自动化, no. 03, 10 February 2016 (2016-02-10) * |
| 张雨薇;王庆杰;李江涛;李文军;: "基于多维特征分析的链式电网短期用电量预测研究", 电网与清洁能源, no. 11 * |
| 易跃春;马月;王霁雪;李桂敏;秦潇;陈文凯;: "风电场超短期风功率预测问题研究", 水力发电, no. 07 * |
| 林涛;董栅;秦冬阳;马同宽;: "基于支持向量回归的风电场短期功率预测", 中南民族大学学报(自然科学版), no. 04 * |
| 王红刚;李彬;: "基于融合长短时记忆网络的风电场超短期风速预测研究", 可再生能源, no. 01, 20 January 2020 (2020-01-20) * |
| 许沛华;陈正洪;谷春;詹天成;: "风电功率预测预报系统的设计与开发", 水电能源科学, no. 03, 25 March 2013 (2013-03-25) * |
| 赏益;高志球;韩威;: "基于RF和GRU组合算法的超短期风电功率预测", 中国科技论文, no. 09 * |
| 郎伟明;麻向津;周博文;杨东升;罗艳红;刘林奇;: "基于LSTM和非参数核密度估计的风电功率概率区间预测", 智慧电力, no. 02 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116661831A (en) * | 2023-07-21 | 2023-08-29 | 国能日新科技股份有限公司 | Automatic upgrading system and method for power generation station algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116187559B (en) | 2024-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Farag et al. | A two ways communication-based distributed control for voltage regulation in smart distribution feeders | |
| CN109426892A (en) | System and method for the distribution load forecast in power grid | |
| CN106707166B (en) | A kind of output of wind electric field method of evaluating characteristic | |
| CN116187559B (en) | Centralized wind power ultra-short-term power prediction method, system and cloud platform | |
| US20110202191A1 (en) | Energy generating system and control thereof | |
| CN103527406A (en) | Wind park control system | |
| CN112855408B (en) | Early warning method and early warning device for hydroelectric generating set | |
| EP4058855B1 (en) | Turbine monitoring and maintenance | |
| Amayri et al. | Condition based maintenance of wind turbine systems considering different turbine types | |
| Ernst et al. | Online-monitoring and prediction of wind power in german transmission system operation centres | |
| EP2381094A1 (en) | Energy network and control thereof | |
| CN102953936A (en) | Unit commitment for wind power generation | |
| Yang et al. | Sub-minute probabilistic solar forecasting for real-time stochastic simulations | |
| EP3741991B1 (en) | Method for dynamic real-time optimization of the performance of a wind park and wind park | |
| CN114738204A (en) | Intelligent early warning method and system for wind generating set | |
| CN117039852A (en) | Power prediction method, device, equipment and storage medium of wind-solar-water complementary power station | |
| EP3966907B1 (en) | Apparatus and method for dynamic forecasting, aggregation, and validation | |
| CN113610285A (en) | Power prediction method for distributed wind power | |
| Hosseini Dolatabadi et al. | Enhancing photovoltaic farm capacity estimation: A comprehensive analysis with a novel approach | |
| Fonte et al. | ANN approach to WECS power forecast | |
| Agabus et al. | ESTIMATION OF WIND POWER PRODUCTION THROUGH SHORT-TERM FORECAST. | |
| Fang et al. | The Shutdown Risk Assessment of Wind Farms under Typhoon Disaster | |
| Madsen et al. | Scheduling data collection for remote control of wind turbines | |
| CN117094848A (en) | Wind power plant wind measurement data interpolation method, device, equipment and storage medium | |
| Nguyen et al. | Fast and Reliable Modelling of Offshore Wind Generation for Adequacy Studies |
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 |