CN110942196B - Predicted irradiation correction method and device - Google Patents
Predicted irradiation correction method and device Download PDFInfo
- Publication number
- CN110942196B CN110942196B CN201911183469.2A CN201911183469A CN110942196B CN 110942196 B CN110942196 B CN 110942196B CN 201911183469 A CN201911183469 A CN 201911183469A CN 110942196 B CN110942196 B CN 110942196B
- Authority
- CN
- China
- Prior art keywords
- prediction
- error
- irradiation
- predicted
- time
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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 application provides a method and a device for correcting predicted irradiation, wherein an error state transition probability matrix is established according to actual measurement errors between predicted fitting irradiation and actual measurement irradiation corresponding to each time node, a predicted error corresponding to the next predicted time is predicted according to the error state transition probability matrix, and the predicted fitting irradiation at the next predicted time is corrected by using the predicted error. And after the actual measurement irradiation of the prediction day is obtained, the error state transition matrix of the corresponding time node is updated by using the actual measurement error of the prediction day, so that the phenomenon that the error state transition probability matrix has larger errors due to less historical data is avoided. In addition, the influence of the cloud cover on the error between the prediction fitting irradiation and the prediction irradiation is discharged, so that the error state transition probability matrix only contains the transition situation of random errors, the prediction of the random errors is more accurate, and the finally obtained prediction correction irradiation is more accurate.
Description
Technical Field
The invention belongs to the technical field of photovoltaics, and particularly relates to a method and a device for correcting predicted irradiation.
Background
With the rapid development of photovoltaic power generation technology, more and more photovoltaic power stations begin to access the power grid. However, due to randomness and fluctuation of photovoltaic power generation, a series of problems can be brought to safe operation of a power grid, and therefore light abandoning phenomena occur in many regions. In order to improve the photovoltaic power generation capability of a power grid, the power generation capability of a photovoltaic power station needs to be predicted in advance through an accurate photovoltaic power generation power prediction model. Therefore, the phenomenon of light abandoning can be reduced, and the photovoltaic power station scheduling plan can be conveniently made by the power grid.
The photovoltaic power generation power prediction model is generally a model established by utilizing actual measurement effective irradiation of a photovoltaic power station, and predicted irradiation obtained by prediction such as numerical weather forecast is substituted into the photovoltaic power generation power prediction model to predict the photovoltaic power generation power of a future date. Under the condition that the model algorithm and the model data quality are not changed, the more the predicted irradiation is close to the actually measured effective irradiation, the higher the prediction precision of the model is. Therefore, the accuracy of the predicted irradiation input by the correction model can improve the prediction accuracy of the photovoltaic power generation power. At present, a high-precision predicted irradiation correction method is needed to obtain high-precision predicted irradiation.
Disclosure of Invention
In view of the above, an object of the present application is to provide a method and an apparatus for correcting predicted irradiation, so as to solve the technical problem of low prediction accuracy of predicted irradiation for photovoltaic power generation, and a specific technical solution thereof is as follows:
in a first aspect, the present application provides a method for correcting a predicted irradiation, which includes:
acquiring predicted irradiation and predicted meteorological data corresponding to each predicted time of a predicted day;
inputting the predicted irradiation and the predicted meteorological data corresponding to the same prediction moment into a pre-obtained irradiation fitting model to obtain the predicted fitting irradiation corresponding to the prediction moment;
taking the error state of the historical moment nearest to the first prediction moment of the prediction day as an initial state, and sequentially obtaining the prediction error corresponding to each prediction moment according to the error state transition probability matrix corresponding to the corresponding time node; the error state transition probability matrix represents the transition probability of the error between the prediction fitting irradiation and the actually measured irradiation from the current error state to other error states;
correcting the predicted fitting irradiation at the predicted time by using the predicted error corresponding to each predicted time to obtain the predicted corrected irradiation corresponding to the predicted time;
after actual measurement irradiation corresponding to each prediction time of the prediction day is obtained, calculating an error between the actual measurement irradiation corresponding to the same prediction time and prediction fitting irradiation to obtain an actual measurement error sequence;
actual measurement errors corresponding to the moment when the actual measurement cloud amount data is larger than the cloud amount threshold value are removed from the actual measurement error sequence to obtain an effective actual measurement error sequence, and the error state of each actual measurement error in the effective actual measurement error sequence is determined;
and updating the error state transition probability matrix corresponding to the corresponding time node by using the error state of the actual measurement error corresponding to at least one prediction day.
In a possible implementation manner of the first aspect, updating the error state transition probability matrix by using an error state of an actual measurement error corresponding to at least one prediction day includes:
for any prediction time, counting the error state of an actual measurement error corresponding to the prediction time on at least one prediction day and the error state corresponding to the next prediction time of the prediction time;
updating the times of the error state corresponding to the prediction time in the historical data, and updating the transition times of the error state of the prediction time to other error states at the next prediction time;
and calculating to obtain a new error state transition probability corresponding to the prediction time according to the updated error state times corresponding to the prediction time and the updated error state transition times.
In another possible implementation manner of the first aspect, the process of establishing an error state transition probability matrix corresponding to each time node includes:
acquiring actual measurement errors between historical actual measurement irradiation and historical fitting irradiation of a same time node in historical data to obtain a historical actual measurement error sequence;
acquiring actually measured cloud amount data of each time node in the historical data, removing historical actually measured errors corresponding to the moment when the actually measured cloud amount data is larger than a cloud amount threshold value from the historical actually measured error sequence to obtain an effective historical actually measured error sequence, and dividing error data in the effective historical actually measured error sequence into different error states;
extracting actual measurement errors of the same time node corresponding to each day from the effective historical actual measurement error sequence to obtain each error state corresponding to each time node;
and establishing an error state transition probability matrix corresponding to the previous time node in the two adjacent time nodes according to each error state corresponding to any two adjacent time nodes.
In another possible implementation manner of the first aspect, taking an error state of a history time closest to a first predicted time of the predicted day as an initial state, and obtaining a predicted error corresponding to each predicted time in sequence according to an error state transition probability matrix corresponding to a corresponding time node, includes:
taking an error state of a history moment closest to a first prediction moment of the prediction day as an initial state, and obtaining a prediction error corresponding to the first prediction moment according to an error state transition probability matrix corresponding to the history moment;
and for any other prediction time except the first prediction time in the prediction day, obtaining the prediction error of the current prediction time according to the prediction error of the previous prediction time and the error state transition probability matrix corresponding to the previous prediction time.
In another possible implementation manner of the first aspect, the obtaining a prediction error at a current prediction time according to a prediction error at a previous prediction time and an error state transition probability matrix corresponding to a time node that is the same as the previous prediction time includes:
searching a target error state with the maximum transition probability corresponding to the prediction error state at the previous prediction time from an error state transition probability matrix corresponding to the time node with the same previous prediction time;
and calculating the average value of all error values in the target error state as a prediction error corresponding to the current prediction time.
In yet another possible implementation manner of the first aspect, the process of establishing an irradiation fitting model includes:
acquiring historical measured irradiation, historical predicted irradiation and historical predicted meteorological data in historical data of the same season;
and learning fitting parameters between the historical actual measurement irradiation and the historical prediction irradiation and between the historical prediction meteorological data by using a machine learning algorithm to obtain an irradiation fitting model corresponding to the season.
In another possible implementation manner of the first aspect, the correcting the predicted fitting irradiation at the predicted time by using the prediction error corresponding to each predicted time to obtain the predicted corrected irradiation corresponding to the predicted time includes:
and calculating the sum of the prediction error corresponding to the same prediction moment and the prediction fitting irradiation to obtain the prediction correction irradiation corresponding to the prediction moment.
In a second aspect, the present application further provides a predicted irradiation modification apparatus, including:
the data acquisition module is used for acquiring predicted irradiation and predicted meteorological data corresponding to each predicted time of a predicted day;
the irradiation fitting module is used for inputting the predicted irradiation and the predicted meteorological data corresponding to the same prediction moment into a pre-obtained irradiation fitting model to obtain the predicted fitting irradiation corresponding to the prediction moment;
the error prediction module is used for taking the error state of the historical moment nearest to the first prediction moment of the prediction day as an initial state and sequentially obtaining the prediction error corresponding to each prediction moment according to the error state transition probability matrix corresponding to the corresponding time node; the error state transition probability matrix represents the transition probability of the error between the prediction fitting irradiation and the actually measured irradiation from the current error state to other error states;
and the irradiation correction module is used for correcting the predicted fitting irradiation at each predicted time by using the prediction error corresponding to each predicted time to obtain the predicted corrected irradiation corresponding to the predicted time.
The actual measurement error acquisition module is used for calculating the error between the actual measurement irradiation and the predicted fitting irradiation corresponding to the same prediction moment after the actual measurement irradiation corresponding to each prediction moment of the prediction day is obtained, so as to obtain an actual measurement error sequence;
the first cloud cover interference data removing module is used for removing actual measurement errors corresponding to the moment when the actual measurement cloud cover data is larger than the cloud cover threshold value from the actual measurement error sequence to obtain an effective actual measurement error sequence, and determining the error state of each actual measurement error in the effective actual measurement error sequence;
and the matrix updating module is used for updating the error state transition probability matrix by using the error state of the actual measurement error corresponding to at least one prediction day.
In a possible implementation manner of the second aspect, the matrix updating module includes:
the statistic submodule is used for counting the error state of an actual measurement error corresponding to the prediction time on at least one prediction day and the error state corresponding to the next prediction time of the prediction time at any prediction time;
the error transfer frequency updating submodule is used for updating the frequency of the error state corresponding to the prediction time in the historical data and updating the transfer frequency of the error state of the prediction time to be transferred to other error states at the next prediction time;
and the transition probability calculation submodule is used for calculating and obtaining a new error state transition probability corresponding to the prediction time according to the updated error state transition times and the updated error state transition times corresponding to the prediction time.
In a third aspect, the present application further provides an apparatus, comprising: at least one processor, and at least one memory coupled to the processor;
wherein the memory has stored therein program instructions;
the processor is configured to call program instructions in the memory to perform the method of predicting irradiation modification according to any one of the first aspect.
The predicted irradiation correction method provided by the invention obtains the predicted irradiation and predicted meteorological data corresponding to each predicted time of the predicted day, and then obtains the predicted fitted irradiation corresponding to the time according to the predicted irradiation and predicted meteorological data at the same time by using the irradiation fitting model. And according to the error state transition probability matrix corresponding to the historical moment, obtaining a prediction error corresponding to each prediction moment, and then according to the prediction error, correcting the prediction fitting irradiation at the moment to obtain the prediction correction irradiation. The method comprises the steps of establishing an error state transition probability matrix through actual measurement errors between prediction fitting irradiation and actual measurement irradiation corresponding to each time node, predicting a prediction error corresponding to the next prediction moment according to the error state transition probability matrix, and correcting the prediction fitting irradiation at the next prediction moment by using the prediction error. Moreover, after the actual measurement irradiation of the prediction day is obtained, the error state transition matrix of the corresponding time node is updated by using the actual measurement error of the prediction day, so that the phenomenon that the created error state transition probability matrix has larger error due to less historical data is avoided, the accuracy of the error state transition probability matrix is improved, and the accuracy of the prediction correction irradiation is finally improved. In addition, the error data of cloud cover interference is removed when the error state transition probability matrix is established and updated, and the influence of the cloud cover on the error between the prediction fitting irradiation and the prediction irradiation is eliminated, so that the error state transition probability matrix only contains the transition situation of random errors, and the prediction of the random errors is more accurate.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flow chart of a method for correcting predicted irradiance according to an embodiment of the present invention;
fig. 2 is a flowchart of a process of establishing an error state transition probability matrix corresponding to each time node according to an embodiment of the present invention;
FIG. 3 is a schematic overall flow chart of an irradiation correction method provided by an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an irradiation modification apparatus according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a matrix update module according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of another irradiation modification apparatus provided in an embodiment of the present invention.
Detailed Description
The difference between the predicted irradiation obtained by prediction such as numerical weather forecast and the actually measured effective irradiation of the power station is mainly caused by the following three reasons: (1) The measured effective irradiation is the total inclined plane irradiation of the photovoltaic module, and the predicted irradiation is the total horizontal plane irradiation generally; (2) The actual measurement effective irradiation and the prediction irradiation come from two different data sources, and the distribution conditions of the data are possibly different; (3) Actually measured effective irradiation contains the influence of various influencing factors at the current moment on irradiation, and the influence of the environment and meteorological parameters at the current moment is difficult to accurately reflect by predicting irradiation.
For the predicted irradiation correction, correction is performed in different manners mainly based on the above three differences. For the difference caused by the first reason, the predicted total irradiation of the horizontal plane is converted into the total irradiation of the inclined plane consistent with the inclination angle of the photovoltaic module through a physical model formula at present. For the difference caused by the second and the third methods, the difference of the historical data of the second and the third methods can be combined together by a certain method to correct the predicted irradiation by combining the influence of factors such as the current time environment, weather and the like.
In short-term power prediction, a power prediction result of 0-24 hours in the future needs to be reported within a certain limited time period in the previous day, so that the predicted irradiation in the future needs to be acquired in the previous day, and the real-time change situation of environmental and meteorological influence factors on the predicted date cannot be acquired at the moment. Therefore, for the predicted irradiation correction of the short-term power prediction, the emphasis is on correcting the first difference and the second difference, and the first difference is corrected by using a physical model formula, which is not described herein again.
In the existing method for correcting the predicted irradiation, one mode is to use the irradiation mean value of similar dates and same time with the same weather type in history to perform weighted calculation with the predicted value of the irradiation at the predicted time, so as to achieve the purpose of correcting the predicted irradiation, and ensure that the predicted irradiation and the actually measured effective irradiation have similar distribution. However, the method does not consider the influence of meteorological factors such as cloud cover, and the irradiation quantity at the historical similar time may have a great difference from the irradiation at the current time. Therefore, using historical similar time exposures for correction may result in some error in the corrected predicted exposure. Moreover, the irradiation data only at the historical similar time has the problem of insufficient data amount.
And the other mode is to establish a self-irradiation fitting model for the errors of actually measured effective irradiation and predicted irradiation, calculate the predicted error of the predicted day by using the error of the predicted day before and correct the predicted irradiation. However, in this method, the influence of factors such as environment and weather on the predicted day is not considered, the error change is random, and the scheme assumes the change of the error as linear change, so the obtained predicted error may have a certain error with the actual error, and further the irradiation predicted value may have a certain error.
In order to solve the problems, the invention provides a predicted irradiation correction method, which establishes an error state transition probability matrix by counting actual measurement errors between predicted fitting irradiation and actual measurement irradiation corresponding to each time node, wherein the error state transition probability matrix is used for representing the transition probability of random errors between the predicted fitting irradiation and the actual measurement irradiation from the current error state to other error states. The error data of cloud cover interference is removed when the error state transition probability matrix is established, the influence of the cloud cover on the error between the prediction fitting irradiation and the prediction irradiation is eliminated, the error state transition probability matrix only contains the transition situation of random errors, the prediction of the random errors is more accurate, and the finally obtained prediction correction irradiation is more accurate. And then, predicting a prediction error corresponding to the next prediction moment according to the error state transition probability matrix, and correcting the prediction fitting irradiation of the next prediction moment by using the prediction error, wherein the finally obtained prediction correction irradiation is more accurate.
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.
Fig. 1 shows a flowchart of a predicted irradiation correction method provided by the present invention, which is used for correcting predicted irradiation of a predicted day provided by weather forecast and the like to obtain more accurate predicted irradiation, so that irradiation data input by a photovoltaic power prediction model is more accurate, and finally, photovoltaic power generation power predicted by the photovoltaic power generation power prediction model is more accurate. As shown in fig. 1, the method includes:
and S110, acquiring predicted irradiation and predicted meteorological data corresponding to each predicted time of the predicted day.
The predicted irradiation corresponding to each predicted time of the predicted day can be provided by numerical weather forecast.
The predicted meteorological data can comprise wind speed, temperature, humidity, air pressure and other data; the predictive weather data may also be provided by numerical weather forecasts.
And S120, inputting the predicted irradiation and the predicted meteorological data corresponding to the same prediction moment into a pre-obtained irradiation fitting model to obtain the predicted fitting irradiation corresponding to the prediction moment.
The solar irradiance has larger difference in different seasons, so that corresponding irradiation fitting models are respectively established for data in different seasons, and when the irradiation obtained by weather forecast is corrected, the predicted fitting irradiation corresponding to the predicted irradiation is obtained according to the irradiation fitting model in the season where the predicted day is located.
And the irradiation fitting model is obtained by irradiation prediction according to history, weather prediction data and actual measurement irradiation training. And inputting the three types of data into a machine learning model, so that the machine learning model learns to obtain a fitting relation between the historical measured irradiation and the historical predicted meteorological data, and further obtain an irradiation fitting model.
When the irradiation fitting model is used, the predicted irradiation and the predicted meteorological data are used as the input of the irradiation fitting model, and the output of the model is the predicted fitting irradiation.
In one embodiment, the process of establishing an irradiance fit model may include:
1) Acquiring historical measured irradiation, historical predicted irradiation and historical predicted meteorological data in historical data of the same season;
the historical forecast irradiation refers to irradiation at the historical moment provided by a numerical weather forecast before the historical moment; the historical actual measurement irradiation refers to the actual measured effective irradiation at the historical moment; the historical predicted weather data is predicted weather data of weather data at a historical time such as weather forecast before the historical time.
2) And learning fitting parameters between the historical actual measurement irradiation and the historical prediction irradiation and between the historical prediction meteorological data by using a machine learning algorithm to obtain an irradiation fitting model corresponding to the season.
Preprocessing various data obtained in the step 1), taking historical actual measurement irradiation as a dependent variable, taking historical predicted irradiation and historical predicted meteorological data as independent variables, and learning a fitting relation between the historical actual measurement irradiation and the historical predicted meteorological data by a machine learning model to obtain an irradiation fitting model.
And inputting the predicted irradiation and the predicted meteorological data into an irradiation fitting model, and outputting corresponding predicted fitting irradiation.
And S130, taking the error state of the historical moment nearest to the first prediction moment of the prediction day as an initial state, and sequentially obtaining the prediction error corresponding to each prediction moment according to the error state transition probability matrix corresponding to the corresponding time node.
The error state transition probability matrix represents the transition probability of the random error between the predicted fitting irradiation and the actually measured irradiation from the error state at the current moment to other error states at the next moment, and the error data of cloud interference is removed when the error state transition probability matrix is obtained.
The error state transition probability matrix is established based on the Markov state transition matrix, the state of the next moment can be predicted according to the current state of the event, and the state of the next moment is only related to the state of the current moment and is unrelated to the state before the current moment, so that the 'non-aftereffect' is met. Because the irradiation data is easily influenced by random factors such as environment, weather and the like, the change has uncertainty, and the characteristic of 'no after-effect' of Markov only needs to consider the transition probability between states and is suitable for a random process, an error state transition probability matrix is introduced in the process of predicting irradiation correction.
And regarding the first predicted time of the predicted day, taking the error state of the historical time closest to the predicted time as an initial error state, and obtaining the predicted error of the predicted time by using the error state transition probability matrix corresponding to the historical time.
In a possible implementation manner, the error state with the maximum transition probability corresponding to the initial error state is searched from the error state transition probability matrix, and the error state is determined as the error state corresponding to the prediction time.
For example, assume that data is collected every 15min interval, i.e. 15min interval between two adjacent time nodes. If the current time is 12: irradiation of 00, then the predicted time is 13 1 And searching E from the error state transition probability matrix corresponding to 11 1 Transition probabilities to other error states, and determines the error state with the largest transition probability as 13.
And for the non-first prediction time of the prediction day, obtaining the prediction error of the current prediction time according to the prediction error of the previous prediction time and the error state transition probability matrix corresponding to the prediction time.
For example, the prediction time is 13. Then, according to the error states corresponding to 13. By analogy, the error states corresponding to 13.
In an application scenario, the error state with the maximum probability value in the error state transition probability matrix corresponding to the previous prediction time may include two or more error states, and then one of the two or more error states with the maximum probability value is randomly selected as the error state corresponding to the current prediction time.
The error state is a set including a plurality of measured errors. After the error state corresponding to the prediction time is obtained, the average value of all error values contained in the error state is calculated to be used as the prediction error of the prediction time.
And S140, correcting the predicted fitting irradiation at the predicted time by using the prediction error corresponding to each predicted time to obtain the predicted corrected irradiation corresponding to the predicted time.
And calculating to obtain the predicted corrected irradiation by using the predicted fitting irradiation obtained in the step S120 and the predicted error obtained in the step S130. Specifically, the sum of the predicted fitting irradiation and the prediction error corresponding to the same prediction time is calculated as the predicted corrected irradiation corresponding to the prediction time.
And after the actual measurement irradiation at the prediction moment is obtained, updating the error state transition probability matrix corresponding to the moment according to the actual measurement error between the actual measurement irradiation at the prediction moment and the prediction fitting irradiation.
S150, after the actual measurement irradiation corresponding to each prediction time of the prediction day is obtained, calculating the error between the actual measurement irradiation corresponding to the same prediction time and the prediction fitting irradiation to obtain an actual measurement error sequence.
And S160, actual measurement errors corresponding to the moment when the actual measurement cloud amount data is larger than the cloud amount threshold value are removed from the actual measurement error sequence to obtain an effective actual measurement error sequence, and the error state of each actual measurement error in the effective actual measurement error sequence is determined.
And S170, updating the error state transition probability matrix corresponding to the corresponding time node by using the error state to which the actual measurement error corresponding to at least one prediction day belongs.
In the predicted irradiation correction method provided by this embodiment, an error state transition probability matrix is established according to the actual measurement error between the predicted fitting irradiation and the actual measurement irradiation corresponding to each time node, a predicted error corresponding to the next predicted time is predicted according to the error state transition probability matrix, and the predicted fitting irradiation at the next predicted time is corrected by using the predicted error. And after the actual measurement irradiation of the prediction day is obtained, the error state transition matrix of the corresponding time node is updated by using the actual measurement error of the prediction day, so that the phenomenon that the created error state transition probability matrix has larger error due to less historical data is avoided. In addition, error data of cloud amount interference is eliminated when the error state transition probability matrix is established, and the influence of the cloud amount on the error between the prediction fitting irradiation and the prediction irradiation is eliminated, so that the error state transition probability matrix only contains the transition condition of random errors, the random errors are predicted more accurately, and the finally obtained prediction correction irradiation is more accurate.
In one embodiment of the present invention, as shown in fig. 2, the process of establishing the error state transition probability matrix corresponding to each time node is as follows:
s210, actual measurement errors between historical actual measurement irradiation and historical fitting irradiation of the same time node in historical data are obtained, and a historical actual measurement error sequence is obtained.
And inputting the historical predicted irradiation and the historical meteorological data into an irradiation fitting model to obtain historical fitting irradiation.
And calculating a relative error between the historical measured irradiation and the historical fitting irradiation corresponding to the same time node in the historical data, namely a historical measured error. According to a large amount of historical data, historical actual measurement errors, namely historical actual measurement error sequences, corresponding to each time node (namely data acquisition time node) in one day can be obtained.
S220, actual measurement cloud amount data of each time node in the historical data are obtained, historical actual measurement errors corresponding to the time when the actual measurement cloud amount data are larger than a cloud amount threshold value are removed from the historical actual measurement error sequence, and an effective historical actual measurement error sequence is obtained.
As described above, in a weather with a large cloud amount, an error between an actually measured irradiation value and a predicted irradiation value is mainly caused by the cloud amount, and if the error at that time is added as a random error to an error state transition probability matrix, the error state transition probability matrix is affected. Therefore, it is necessary to determine whether to record the error at the time into the error state transition probability matrix according to the cloud amount data at the corresponding time.
In an embodiment of the present invention, a cloud cover threshold may be set, and if the actually measured cloud cover data at a certain time is greater than the cloud cover threshold, the historical actually measured error data corresponding to the certain time is removed, so as to obtain an effective historical actually measured error sequence.
And S230, dividing the error data in the effective historical measured error sequence into different error states.
Clustering or other partitioning methods may be employed to partition these errors into different error states.
S240, extracting the actual measurement error of the same time node corresponding to each day from the effective historical actual measurement error sequence to obtain each error state corresponding to each time node.
Since the historical data includes data acquired for many days, the actual measurement error of the same time node corresponding to each day needs to be extracted from the error sequence, the actual measurement error state corresponding to the time node is obtained, and finally the error state corresponding to the historical data of each time node is obtained.
And S250, establishing an error state transition probability matrix corresponding to the previous time node in the two adjacent time nodes according to each error state corresponding to any two adjacent time nodes.
And after the error states corresponding to the i moment and the i +1 moment are processed, establishing an error state transition probability matrix of the i moment by using the error states corresponding to the two time nodes.
The following describes the process of establishing the error state transition probability matrix with reference to a specific example:
assuming that the time interval of the time node for collecting data is 15min (96 time nodes in total in one day), an error state transition probability matrix is established for each time node.
Taking the establishment process of the error state transition probability matrix at 12 am.
Note that, in a day 12State E 1 And 12, the error state is E 2 Then, the error state is called as 12 1 Transfer to E 2 。
1) And collecting actual measurement errors (namely relative errors between actual measurement irradiation and prediction fitting irradiation) corresponding to each time node in the historical data, and dividing the errors into different error states by adopting a clustering or other dividing method, wherein each error state is equivalent to an error set.
2) And 12, error state E of every day in statistical historical data i Number of occurrences N i Wherein i =1,2, …, m, m is the total number of error states.
3) And calculating a transition probability matrix of 12.
First, the state E of 12 i At 12 1 ,E 2 ,…,E m Number of times N i1 ,N i2 ,…,N im The last step knowing the state E i Number of occurrences N i Then, the state transition probability of the error is calculated:
p in formula 1 ij Indicating error by state E i Transfer to E j I, j =1,2, …, m; n is a radical of hydrogen ij : error from state E i Transfer to E j The number of times of (c); n is a radical of i For errors in state E i The number of times of (c); n is a radical of ij Is the error is formed by i Transfer to E j The number of times.
Obtaining a state transition probability matrix corresponding to 12:
in addition, the process of updating the error state transition probability matrix in the foregoing embodiment is similar to the establishing process shown in fig. 2, and in a possible implementation manner of the present invention, the implementation process of step S170 may be as follows:
for any prediction time, counting the error state of an actual measurement error corresponding to the prediction time on at least one prediction day and the error state corresponding to the next prediction time of the prediction time; then, updating the number of times of the error state corresponding to the prediction time in the historical data, and updating the number of times of transition of the error state of the prediction time to other error states at the next prediction time; and finally, calculating to obtain a new error state transition probability corresponding to the prediction time according to the updated error state times corresponding to the prediction time and the updated error state transition times.
The following describes the updating process of the error state transition probability matrix with reference to a specific example:
assume that 12 1 Transition to State E 2 The probability of (c) is:
N 12 for 12 in the history data 1 Transition to State E 2 The number of times of (c); n is a radical of hydrogen 1 For state E in the history data 1 Total number of occurrences.
When data of one day is newly added, the error state of 12 to 12 1 To E 2 Then E is 1 To E 2 The probability of (d) is updated as:
in other embodiments of the present invention, other updating methods can be used besides the calculation method shown in the above formula 2.
The above-mentioned updating process is updated every day, and in other embodiments, the error state transition probability matrix may be updated by accumulating data for a certain period of time and then using the data for the period of time.
When the data volume is small, a large error exists in the established error state transition probability matrix, on one hand, the probability calculated by using the small data volume is not universal, and on the other hand, the situation that an error state appearing at a certain time does not exist in the state transition probability matrix may exist, namely, the error state does not appear in the historical data.
When the error state transition probability is established, the influence of the error between the actual measurement irradiation and the prediction irradiation on the accuracy of the data in the error state transition probability matrix caused by the cloud cover is eliminated, so that the data in the error state transition probability matrix is more accurate, the error obtained by prediction is more accurate, and the correction accuracy of the prediction irradiation is finally improved. And the existing error state transition probability matrix at the corresponding moment is updated by utilizing the newly obtained actual measurement error between the prediction fitting irradiation and the actual measurement irradiation, so that the error state transition probability matrix is prevented from having larger errors, and the accuracy of the error state transition probability matrix is further improved.
Referring to fig. 3, an overall flowchart of the irradiation correction method provided by the present invention is shown, as shown in fig. 3, the whole process of the irradiation correction method is divided into three parts, the first part is a training model, the second part is a correction process, and the third part is an update process (i.e., an error state transition probability matrix update process).
The correction process and the update process, i.e., the process model training process shown in fig. 1, i.e., the process of training the irradiation fitting model, and the process of establishing the error state transition probability matrix shown in fig. 2, are not described herein again.
It should be noted that the model training process is performed off-line, and the modification process and the update process are completed on-line.
Corresponding to the embodiment of the predicted irradiation correction method, the application also provides an embodiment of a predicted irradiation correction device.
Referring to fig. 4, a schematic structural diagram of a predicted irradiation modification apparatus provided in an embodiment of the present invention is shown, where the apparatus includes:
and the data acquisition module 110 is configured to acquire predicted irradiation and predicted meteorological data corresponding to each predicted time of the predicted day.
And the irradiation fitting module 120 is configured to input the predicted irradiation and the predicted meteorological data corresponding to the same prediction time into a pre-obtained irradiation fitting model to obtain the predicted fitting irradiation corresponding to the prediction time.
And the irradiation fitting model is obtained by training according to historical predicted irradiation, historical predicted meteorological data and historical measured irradiation in the same season as the predicted day.
In one possible implementation, the process of establishing an irradiation fitting model includes:
and obtaining historical actual measurement irradiation, historical prediction irradiation and historical prediction meteorological data in historical data of the same season. And then, learning fitting parameters between the historical measured irradiation and the historical predicted irradiation and between the historical predicted meteorological data by using a machine learning algorithm to obtain an irradiation fitting model corresponding to the season.
And the error prediction module 130 is configured to use an error state of a historical time closest to a first prediction time of the prediction day as an initial state, and obtain a prediction error corresponding to each prediction time in sequence according to the error state transition probability matrix corresponding to the corresponding time node.
The error state transition probability matrix represents the transition probability of the random error between the predicted fitting irradiation and the actually measured irradiation from the current error state to other error states, and error data of cloud interference is removed when the error state transition probability matrix is obtained; each error state contains a plurality of errors.
In an embodiment of the present invention, the error prediction module 130 is specifically configured to:
taking an error state of a history time closest to a first prediction time of a prediction day as an initial state, and obtaining a prediction error corresponding to the first prediction time according to an error state transition probability matrix corresponding to the history time;
and for any other prediction time except the first prediction time in the prediction day, obtaining the prediction error of the current prediction time according to the prediction error of the previous prediction time and the error state transition probability matrix corresponding to the previous prediction time.
In a possible implementation manner, a target error state with the maximum transition probability corresponding to the prediction error state at the previous prediction time is searched from an error state transition probability matrix corresponding to the time node with the same previous prediction time; then, the average value of the error values in the target error state is calculated as the prediction error corresponding to the current prediction time.
And the irradiation correction module 140 is configured to correct the predicted fitting irradiation at the predicted time by using the prediction error corresponding to each predicted time to obtain a predicted corrected irradiation corresponding to the predicted time.
Specifically, the sum of the prediction error and the prediction fitting irradiation corresponding to the same prediction time is calculated to obtain the prediction correction irradiation corresponding to the prediction time.
The actual measurement error acquisition module 150 is configured to calculate an error between actual measurement irradiation and predicted fitting irradiation corresponding to the same prediction time after actual measurement irradiation corresponding to each prediction time of the prediction day is obtained, so as to obtain an actual measurement error sequence;
a first cloud cover interference data eliminating module 160, configured to eliminate, from the actual measurement error sequence, an actual measurement error corresponding to a time when the actual measurement cloud cover data is greater than the cloud cover threshold, to obtain an effective actual measurement error sequence, and determine an error state to which each actual measurement error in the effective actual measurement error sequence belongs;
and the matrix updating module 170 is configured to update the error state transition probability matrix by using an error state of the actually measured error corresponding to at least one prediction day.
In a possible implementation manner, as shown in fig. 5, the matrix updating module 170 specifically includes:
the statistic submodule 171 is configured to, for any one of the prediction times, count an error state to which an actual measurement error corresponding to the prediction time on at least one prediction day belongs, and an error state corresponding to a prediction time next to the prediction time.
And an error transition number updating submodule 172, configured to update the number of times of the error state corresponding to the prediction time in the history data, and update the number of transitions that the error state of the prediction time transitions to another error state at the next prediction time.
The transition probability calculation sub-module 173 is configured to calculate a new error state transition probability corresponding to the prediction time according to the updated number of error states corresponding to the prediction time and the updated number of error state transitions.
The predicted irradiation correction device provided in this embodiment establishes an error state transition probability matrix according to the actual measurement error between the predicted fitting irradiation and the actual measurement irradiation corresponding to each time node, predicts the predicted error corresponding to the next predicted time according to the error state transition probability matrix, and corrects the predicted fitting irradiation at the next predicted time by using the predicted error. And after the actual measurement irradiation of the prediction day is obtained, the error state transition matrix of the corresponding time node is updated by using the actual measurement error of the prediction day, so that the phenomenon that the created error state transition probability matrix has larger error due to less historical data is avoided, the accuracy of the error state transition probability matrix is improved, and the accuracy of the prediction correction irradiation is finally improved. In addition, error data of cloud cover interference is removed when an error state transition probability matrix is established, and the influence of the cloud cover on the error between the prediction fitting irradiation and the prediction irradiation is eliminated, so that the error state transition probability matrix only contains the transition situation of random errors, the prediction of the random errors is more accurate, and the finally obtained prediction correction irradiation is more accurate.
In an embodiment of the present invention, as shown in fig. 6, the apparatus shown in fig. 4 further includes:
a historical error obtaining module 210, configured to obtain, by the 150, an actual measurement error between historical actual measurement irradiation and historical fitting irradiation of a same time node in the historical data, to obtain a historical actual measurement error sequence;
and a second cloud cover interference data removing module 220, configured to obtain measured cloud cover data of each time node in the historical data, remove a historical measured error corresponding to a time when the measured cloud cover data is greater than a cloud cover threshold from the historical measured error sequence, obtain an effective historical measured error sequence, and divide error data in the effective historical measured error sequence into different error states.
An error state dividing module 230, configured to extract actual measurement errors of the same time node corresponding to each day from the effective historical actual measurement error sequence, to obtain each error state corresponding to each time node;
a state matrix establishing module 240, configured to establish an error state transition probability matrix corresponding to a previous time node in any two adjacent time nodes according to each error state corresponding to the two adjacent time nodes.
The error state transition probability matrix can be updated according to real-time data, when the error state transition probability matrix is updated by using the real-time data, cloud interference data also needs to be provided, effective actual measurement error data is obtained, finally, the error state transition probability matrix is updated by using the error state of the actual measurement error corresponding to at least one prediction day, and the specific updating process is described earlier and is not repeated here.
The irradiation correction device provided by the embodiment eliminates the influence of the error between the actual measurement irradiation and the prediction irradiation on the accuracy of the data in the error state transition probability matrix caused by the cloud cover when the error state transition probability is established, so that the data in the error state transition probability matrix is more accurate, the error obtained by prediction is more accurate, and the correction accuracy of the prediction irradiation is finally improved.
On the other hand, the embodiment of the invention also provides a device, which can be a PC or a server or other devices with computing capability. The apparatus includes at least one processor, and at least one memory coupled to the processor. The memory stores program instructions, and the processor is configured to call the program instructions in the memory to execute any one of the above embodiments of the predicted irradiation modification method.
In a typical configuration, the device may also include input/output interfaces, network interfaces, and the like.
The memory may include volatile memory in a computer readable medium, random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), including at least one memory chip. The memory is an example of a computer-readable medium.
While, for purposes of simplicity of explanation, the methodologies are shown and described as a series of acts, it will be appreciated by those skilled in the art that the claimed subject matter is not limited by the order of acts, as some steps may, in accordance with the claimed subject matter, occur in other orders and/or concurrently. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.
It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other. For the device-like embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The steps in the method of the embodiments of the present application may be sequentially adjusted, combined, and deleted according to actual needs.
The device and the modules and sub-modules in the terminal in the embodiments of the present application can be combined, divided and deleted according to actual needs.
In the several embodiments provided in the present application, it should be understood that the disclosed terminal, apparatus and method may be implemented in other manners. For example, the above-described terminal embodiments are merely illustrative, and for example, the division of a module or a sub-module is only one logical division, and there may be other divisions when the terminal is actually implemented, for example, a plurality of sub-modules or modules may be combined or integrated into another module, or some features may be omitted or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or modules, and may be in an electrical, mechanical or other form.
The modules or sub-modules described as separate parts may or may not be physically separate, and parts that are modules or sub-modules may or may not be physical modules or sub-modules, may be located in one place, or may be distributed over a plurality of network modules or sub-modules. Some or all of the modules or sub-modules can be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, each functional module or sub-module in the embodiments of the present application may be integrated into one processing module, or each module or sub-module may exist alone physically, or two or more modules or sub-modules may be integrated into one module. The integrated modules or sub-modules may be implemented in the form of hardware, or may be implemented in the form of software functional modules or sub-modules.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (10)
1. A method of predictive irradiance correction, comprising:
acquiring predicted irradiation and predicted meteorological data corresponding to each predicted time of a predicted day;
inputting the predicted irradiation and the predicted meteorological data corresponding to the same prediction moment into a pre-obtained irradiation fitting model to obtain the predicted fitting irradiation corresponding to the prediction moment; the irradiation fitting model is obtained according to historical predicted irradiation, historical predicted meteorological data and historical measured irradiation training;
taking the error state of the historical moment nearest to the first prediction moment of the prediction day as an initial state, and sequentially obtaining the prediction error corresponding to each prediction moment according to the error state transition probability matrix corresponding to the corresponding time node; the error state transition probability matrix represents the transition probability of the error between the predicted fitting irradiation and the measured irradiation from the current error state to other error states;
correcting the predicted fitting irradiation at each predicted time by using the prediction error corresponding to each predicted time to obtain the predicted corrected irradiation corresponding to the predicted time;
after actual measurement irradiation corresponding to each prediction time of the prediction day is obtained, calculating the error between the actual measurement irradiation corresponding to the same prediction time and the prediction fitting irradiation to obtain an actual measurement error sequence;
actual measurement errors corresponding to the moment when the actual measurement cloud amount data is larger than the cloud amount threshold value are removed from the actual measurement error sequence to obtain an effective actual measurement error sequence, and the error state of each actual measurement error in the effective actual measurement error sequence is determined;
and updating the error state transition probability matrix corresponding to the corresponding time node by using the error state of the actual measurement error corresponding to at least one prediction day.
2. The method of claim 1, wherein updating the error state transition probability matrix with an error state of the measured error for at least one prediction day comprises:
for any prediction time, counting the error state of an actual measurement error corresponding to the prediction time on at least one prediction day and the error state corresponding to the next prediction time of the prediction time;
updating the number of times of the error state corresponding to the prediction time in the historical data, and updating the number of times of transition of the error state of the prediction time to other error states at the next prediction time;
and calculating to obtain a new error state transition probability corresponding to the prediction time according to the updated error state times corresponding to the prediction time and the updated error state transition times.
3. The method according to claim 1 or 2, wherein the process of establishing the error state transition probability matrix corresponding to each time node comprises:
acquiring actual measurement errors between historical actual measurement irradiation and historical fitting irradiation of a same time node in historical data to obtain a historical actual measurement error sequence;
acquiring actually measured cloud amount data of each time node in the historical data, removing historical actually measured errors corresponding to the moment when the actually measured cloud amount data is larger than a cloud amount threshold value from the historical actually measured error sequence to obtain an effective historical actually measured error sequence, and dividing error data in the effective historical actually measured error sequence into different error states;
extracting actual measurement errors of the same time node corresponding to each day from the effective historical actual measurement error sequence to obtain each error state corresponding to each time node;
and establishing an error state transition probability matrix corresponding to the previous time node in the two adjacent time nodes according to each error state corresponding to any two adjacent time nodes.
4. The method according to claim 1 or 2, wherein the step of obtaining the prediction error corresponding to each prediction time by taking the error state of the history time nearest to the first prediction time of the prediction day as an initial state and sequentially according to the error state transition probability matrix corresponding to the corresponding time node comprises the steps of:
taking an error state of a history moment closest to a first prediction moment of the prediction day as an initial state, and obtaining a prediction error corresponding to the first prediction moment according to an error state transition probability matrix corresponding to the history moment;
and for any other prediction time except the first prediction time in the prediction day, obtaining the prediction error of the current prediction time according to the prediction error of the previous prediction time and the error state transition probability matrix corresponding to the previous prediction time.
5. The method according to claim 4, wherein obtaining the prediction error at the current prediction time according to the prediction error at the previous prediction time and the error state transition probability matrix corresponding to the time node that is the same as the previous prediction time comprises:
searching a target error state with the maximum transition probability corresponding to the prediction error state at the previous prediction time from an error state transition probability matrix corresponding to the time node with the same previous prediction time;
and calculating the average value of all error values in the target error state as a prediction error corresponding to the current prediction time.
6. The method of claim 1, wherein the process of establishing an irradiance fit model comprises:
acquiring historical measured irradiation, historical predicted irradiation and historical predicted meteorological data in historical data of the same season;
and learning fitting parameters between the historical measured irradiation and the historical predicted irradiation and between the historical predicted meteorological data by using a machine learning algorithm to obtain an irradiation fitting model corresponding to the season.
7. The method of claim 1, wherein the correcting the predicted fitting irradiance at the predicted time with the prediction error corresponding to each predicted time to obtain the predicted corrected irradiance at the predicted time comprises:
and calculating the sum of the prediction error corresponding to the same prediction moment and the prediction fitting irradiation to obtain the prediction correction irradiation corresponding to the prediction moment.
8. A predicted irradiation correction apparatus, comprising:
the data acquisition module is used for acquiring predicted irradiation and predicted meteorological data corresponding to each predicted time of a predicted day;
the irradiation fitting module is used for inputting the predicted irradiation and the predicted meteorological data corresponding to the same prediction moment into a pre-obtained irradiation fitting model to obtain the predicted fitting irradiation corresponding to the prediction moment; the irradiation fitting model is obtained according to historical predicted irradiation, historical predicted meteorological data and historical actual measurement irradiation training;
the error prediction module is used for taking an error state of a historical moment nearest to the first prediction moment of the prediction day as an initial state and sequentially obtaining a prediction error corresponding to each prediction moment according to an error state transition probability matrix corresponding to the corresponding time node; the error state transition probability matrix represents the transition probability of the error between the prediction fitting irradiation and the actually measured irradiation from the current error state to other error states;
the irradiation correction module is used for correcting the predicted fitting irradiation at the predicted time by using the predicted error corresponding to each predicted time to obtain the predicted corrected irradiation corresponding to the predicted time;
the actual measurement error acquisition module is used for calculating the error between the actual measurement irradiation and the predicted fitting irradiation corresponding to the same prediction moment after the actual measurement irradiation corresponding to each prediction moment of the prediction day is obtained, so as to obtain an actual measurement error sequence;
the first cloud cover interference data removing module is used for removing actual measurement errors corresponding to the moment when the actual measurement cloud cover data is larger than the cloud cover threshold value from the actual measurement error sequence to obtain an effective actual measurement error sequence, and determining the error state of each actual measurement error in the effective actual measurement error sequence;
and the matrix updating module is used for updating the error state transition probability matrix by using the error state of the actually measured error corresponding to at least one prediction day.
9. The apparatus of claim 8, wherein the matrix update module comprises:
the statistic submodule is used for counting the error state of an actual measurement error corresponding to the prediction time on at least one prediction day and the error state corresponding to the next prediction time of the prediction time at any prediction time;
the error transfer frequency updating submodule is used for updating the frequency of the error state corresponding to the prediction time in the historical data and updating the transfer frequency of the error state of the prediction time to be transferred to other error states at the next prediction time;
and the transition probability calculation submodule is used for calculating and obtaining a new error state transition probability corresponding to the prediction time according to the updated error state transition times and the updated error state transition times corresponding to the prediction time.
10. An electronic device, comprising: at least one processor, and at least one memory coupled to the processor;
wherein the memory has stored therein program instructions;
the processor is configured to call program instructions in the memory to perform the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911183469.2A CN110942196B (en) | 2019-11-27 | 2019-11-27 | Predicted irradiation correction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911183469.2A CN110942196B (en) | 2019-11-27 | 2019-11-27 | Predicted irradiation correction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110942196A CN110942196A (en) | 2020-03-31 |
| CN110942196B true CN110942196B (en) | 2023-04-07 |
Family
ID=69908306
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911183469.2A Active CN110942196B (en) | 2019-11-27 | 2019-11-27 | Predicted irradiation correction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110942196B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105787594A (en) * | 2016-02-29 | 2016-07-20 | 南京航空航天大学 | Irradiation prediction method based on multivariate time series and regression analysis |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150009375A (en) * | 2013-07-16 | 2015-01-26 | 한국전자통신연구원 | Method and system for predicting power consumption |
-
2019
- 2019-11-27 CN CN201911183469.2A patent/CN110942196B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105787594A (en) * | 2016-02-29 | 2016-07-20 | 南京航空航天大学 | Irradiation prediction method based on multivariate time series and regression analysis |
Non-Patent Citations (2)
| Title |
|---|
| 基于马尔科夫链误差修正的光伏发电预测;段俊东等;《河南理工大学学报(自然科学版)》(第01期);全文 * |
| 微电网光伏发电的Adaboost天气聚类超短期预测方法;谭津;《电力系统自动化》;20171110;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110942196A (en) | 2020-03-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106779223B (en) | Photovoltaic system power generation real-time prediction method and device | |
| CN106685674B (en) | Method and device for predicting network event and establishing network event prediction model | |
| US20220294218A1 (en) | Method and system for predicting regional short-term energy power by taking weather into consideration | |
| JP2014021555A (en) | Natural energy amount prediction device | |
| CN111861023A (en) | Statistical-based hybrid wind power prediction method and device | |
| CN111680398B (en) | Single machine performance degradation prediction method based on Holt-windows model | |
| CN116227637A (en) | Active power distribution network oriented refined load prediction method and system | |
| CN116069095A (en) | Machine room environment adjusting method, equipment and medium | |
| CN111709569A (en) | Method and device for predicting and correcting output power of wind power plant | |
| CN114819374A (en) | Regional new energy ultra-short term power prediction method and system | |
| CN112612781A (en) | Data correction method, device, equipment and medium | |
| CN112580844A (en) | Meteorological data processing method, device, equipment and computer readable storage medium | |
| CN110942196B (en) | Predicted irradiation correction method and device | |
| CN111159900B (en) | Method and device for predicting wind speed of fan | |
| CN112540159A (en) | Nuclear power plant atmospheric diffusion prediction correction method, system, medium and electronic equipment | |
| CN112036634A (en) | Photovoltaic power generation power determination method, prediction system and readable storage medium | |
| CN116845868A (en) | Distributed photovoltaic power prediction method and device, electronic equipment and storage medium | |
| CN115907131A (en) | Method and system for building electric heating load prediction model in northern area | |
| CN115908051A (en) | Method for determining energy storage capacity of power system | |
| CN114971062A (en) | Photovoltaic power prediction method and device | |
| CN114169590A (en) | Reservoir warehousing runoff forecasting and correcting method and system based on multi-scenario division | |
| CN114417653A (en) | Uncertainty-based satellite assembly circuit board thermal reliability analysis method | |
| CN113568067A (en) | Numerical weather forecasting method and device, computer storage medium and electronic equipment | |
| CN112787882A (en) | Internet of things edge traffic prediction method, device and equipment | |
| KR20210132503A (en) | National power demand forecasting using cyclic neural network model |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20221216 Address after: Room 423, Sunshine Power R&D Center Building, No. 1699, Xiyou Road, High-tech Zone, Hefei City, Anhui Province, 230088 Applicant after: Sunshine Hui Carbon Technology Co.,Ltd. Address before: 230088 No. 1699 Xiyou Road, Hefei High-tech Zone, Anhui Province Applicant before: SUNGROW POWER SUPPLY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |