CN112561058A - Short-term photovoltaic power prediction method based on Stacking-ensemble learning - Google Patents
Short-term photovoltaic power prediction method based on Stacking-ensemble learning Download PDFInfo
- Publication number
- CN112561058A CN112561058A CN202011469711.5A CN202011469711A CN112561058A CN 112561058 A CN112561058 A CN 112561058A CN 202011469711 A CN202011469711 A CN 202011469711A CN 112561058 A CN112561058 A CN 112561058A
- Authority
- CN
- China
- Prior art keywords
- prediction
- input
- gate
- output
- learner
- 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
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"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Development Economics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a short-term photovoltaic power prediction method based on Stacking-ensemble learning. The training principle difference of different depth learning is considered based on the Stacking-ensemble learning short-term photovoltaic power prediction method, advantages of each model are fully exerted, and compared with the traditional single model prediction, the prediction accuracy of the prediction method is obviously improved.
Description
Technical Field
The invention relates to the field of photovoltaic power prediction, in particular to a short-term photovoltaic power prediction method based on stacking-ensemble learning.
Background
At present, photovoltaic power prediction is mainly divided into a physical model, a statistical model and an artificial intelligence model. The physical model researches a meteorological evolution process through mathematical modeling, and predicts according to a photoelectric and wind power conversion physical model without a large amount of data, but the established model is complex, large in calculation amount and poor in anti-interference capability. The statistical model predicts the power by using the statistical relationship among historical measurement data, can effectively solve the problem of prediction delay, has higher requirements on the processing of original data and the stability of time series, and is difficult to reflect the influence of nonlinear factors. Therefore, in recent years, the artificial intelligent model is applied to photovoltaic power prediction with strong nonlinear fitting capability, but most of the traditional machine learning models are shallow neural networks, and deep features of photovoltaic power and other influencing factors such as total solar radiation intensity and scattering level radiation intensity are difficult to mine, so that prediction accuracy and generalization capability are difficult to improve.
Compared with the traditional machine learning, the deep learning technology is less interfered by noise, can fully mine the relevance among data, and provides strong support for the prediction of the power generation power of the renewable energy source. However, only a single mode is adopted for photovoltaic prediction, and due to the large assumed space of the photovoltaic prediction problem, multiple assumptions may reach equal performance on a training set, and if a single model is used, generalization performance may be poor due to randomness.
There are studies on the way of using combined prediction to further improve the model prediction accuracy. Patent document No. CN109767353A (publication date is 2019, 05, and 17) discloses a photovoltaic power generation power prediction method based on a probability distribution function, which adopts a general distribution fitting method to fit to obtain a photovoltaic power generation power probability density function and an accumulated distribution function corresponding to the interval, performs inverse transformation sampling on the photovoltaic power generation power probability distribution density function to obtain a plurality of photovoltaic power generation power values corresponding to the known irradiance, and then calculates an average value of each photovoltaic power generation power value in the photovoltaic output scene set as a photovoltaic power generation power prediction value corresponding to the known irradiance.
However, most of the above combined prediction methods adopt a mean value calculation method to obtain the predicted mean values of different algorithm models or different parameter models of the same type of algorithm, and cannot reflect the differences of data observation of different prediction algorithms, each algorithm cannot train a more excellent model by taking the best advantage and the shortest advantage, and the combined method has no sufficient theoretical support, and the principle is thinner.
Therefore, how to provide a short-term photovoltaic power prediction method for solving the above technical problems becomes a problem to be solved urgently by those skilled in the art.
Disclosure of Invention
The invention provides a short-term photovoltaic power prediction method based on Stacking integrated learning, aiming at overcoming the technical defects of low prediction precision, poor generalization capability and insufficient stability and accuracy of the existing prediction model in the prior art.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a short-term photovoltaic power prediction method based on Stacking-ensemble learning comprises the following steps:
s1, acquiring historical photovoltaic power generation power data and historical weather forecast data and performing data preprocessing, so as to construct a photovoltaic prediction data sample set;
s2, dividing the photovoltaic prediction data sample set into a base model training set and a meta-learner training set;
s3, inputting the base model training set into a plurality of base learners for training, and establishing a plurality of base learner prediction models so as to complete the training of a first layer of prediction models;
s4, inputting the meta-learner training set into a plurality of base learner prediction models of a first layer of prediction models, outputting respective prediction results by each base learner prediction model, and inputting the prediction results of the first layer of prediction models into a meta-learner for training so as to finish training of the meta-learner prediction models in a second layer of prediction models;
and S5, inputting input variables of the day to be predicted into the trained first-layer prediction model and outputting prediction results, and then taking a plurality of prediction results of the first-layer prediction model as input variables of the meta-learner prediction model of the trained second-layer prediction model to obtain the final predicted photovoltaic output power.
Preferably, in step S1, a time series of historical photovoltaic power generation power data and historical weather forecast data is obtained through the data preprocessing, and the time series includes photovoltaic power generation power P, total solar radiation intensity G, scattering level radiation intensity D, wind speed WsTemperature T, relative humidity H, wind direction WdAnd the daily rainfall R, and then taking the time sequence as a photovoltaic prediction data sample set.
Preferably, in step S2, the photovoltaic prediction data sample set S ═ P is formed by using the time series of the historical weather forecast data as input variables of the photovoltaic prediction data sample set and the time series of the historical photovoltaic power generation power as output variables of the photovoltaic prediction data sample setn,Gn,Dn,Wsn,Tn,Hn,Wdn,RnIn which P isnFor the corresponding predicted value, x, of the nth photovoltaic prediction data sample setn＝{Gn,Dn,Wsn,Tn,Hn,Wdn,RnAnd the input feature vector is corresponding to the nth photovoltaic prediction data sample set.
Preferably, in step S3, the plurality of base learners includes a long-short term memory learner, a threshold cyclic unit learner and a cyclic neural network learner.
Preferably, in step S3, the step of building the prediction model of the long-short term memory learner includes:
s311, taking input feature vector xtAs input at time t.
S312, setting the long-term and short-term memory model unit to comprise a forgetting gate, an input gate and an output gate, and setting a forgetting gate unit f at the t-th momenttAnd a t-th time input gate itAnd a t-th time output gate gtCell internal state update ctOutput of prediction model of long-short term memory learning device
Respectively as follows:
wherein sigma is a value converted by sigmoid activating function into a value between 0 and 1, and xtIs the input of the current time, ht-1Is a hidden state at the previous moment, ct-1Is the previous time cell state, Uf、Ui、Uo、UcInput weight matrix W representing respectively forgetting gate, input gate, output gate and cell internal state updatef、Wi、Wo、WcCyclic weight matrices, M, representing respectively the update of the states inside the forgetting gate, the input gate, the output gate and the cellf,Mi,MoCell state weight matrices representing forgetting gate, input gate, output gate, respectively, bf、bi、bo、bcRespectively representing the bias of the forgetting gate, the input gate, the output gate, and the cell internal state update.
S313, taking
And the predicted result is output at the t-th time.
Preferably, in step S3, the step of establishing the threshold cycle unit learner prediction model includes:
s321, taking input feature vector xtAs input at time t.
S322, the threshold circulation unit comprises an input layer, a hidden layer and an output layer; the core of the threshold cycle unit is two gates of a hidden layer, and the influence of information on a final result through control historical data can be selectively caused; wherein the hidden layer comprises an update gate and a reset gate, zt、rtThe refresh gate and the reset gate at time t, respectively, are expressed by equations (6) - (9) and are again based on ztAnd
updating
Wherein sigma is a value converted by the sigmoid activation function into a value between 0 and 1,
and
activation parameters at time t and at time t-1 respectively,
is an activation parameter, Wz，WhRespectively representing cyclic weight matrixes of the update gate and the reset gate, W is a hidden layer-hidden layer connection weight, Uz、UrRepresenting the input weight matrix of the refresh gate and the reset gate respectively, U being the weight matrix of the input-hidden layer connection, bz、brRespectively, the bias of the refresh gate and the reset gate, and b is the bias.
S323, taking
And the predicted result is output at the t-th time.
Preferably, in step S3, the step of building the recurrent neural network learner prediction model includes:
s331, taking input feature vector xtAs input at time t.
S332, the input of the recurrent neural network is xtThe output is
The hidden layer is h, the hidden layer h is called a memory unit and has the capability of storing information, the output of the hidden layer h influences the input of the next moment, and the input of the moment t is assumed to be xtOutput of
Hidden state is
Then
I.e. input x with the current timetIn relation, the hidden state update also at the previous time is as follows:
wherein Q isiRepresents the output of the preceding output unit of the classifier, i represents the class index, C represents the total number of classes, SiRepresenting the ratio of the index of the current element to the sum of the indices of all elements, b and c are offsets, U, V, W are the weight matrices of input-hidden layer, hidden layer-output, hidden layer-hidden layer connections respectively,
represents the t-th hidden layer, and at different times, the values of the weight matrix U, V, W are the same, and f (x) is the tanh function or the ReLU function in the nonlinear activation function.
S333. taking
And the predicted result is output at the t-th time.
Preferably, in step S4, the meta-learner prediction model is a long-short term memory learner prediction model.
Preferably, the day input variable to be predicted is
Compared with the prior art, the technical scheme of the invention has the beneficial effects that: the Stacking-ensemble learning method provided by the invention considers the difference of data observation and training principles of different algorithms, and fully exerts the advantages of each model in the prediction process. And the stronger the learning ability of each base learner is, the lower the correlation degree between the base learners is, and the better the final prediction effect is. Compared with the traditional single model prediction, the short-term photovoltaic power generation prediction method based on the Stacking ensemble learning has higher prediction accuracy.
Drawings
Fig. 1 is a schematic flow diagram of a short-term photovoltaic power prediction method based on Stacking-ensemble learning according to an embodiment of the present invention.
Fig. 2 is a comparison graph of the prediction effect of the short-term photovoltaic power prediction method based on Stacking-ensemble learning and the conventional single model prediction method provided by the embodiment of the invention.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
for the purpose of better illustrating the embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product;
it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
As shown in fig. 1, the short-term photovoltaic power prediction method based on Stacking-ensemble learning provided by the present application includes the following steps:
s1, acquiring historical photovoltaic power generation power data and historical weather forecast data and performing data preprocessing, so as to construct a photovoltaic prediction data sample set;
s2, dividing the data sample set into a base model training set and a meta-learner training set;
s3, inputting the base model training set into a plurality of base learners for training, and establishing a plurality of base learner prediction models so as to complete the training of a first layer of prediction models;
s4, inputting the meta-learner training set into a plurality of base learner prediction models of a first layer of prediction models, outputting respective prediction results by each base learner prediction model, and inputting the prediction results of the first layer of prediction models into a meta-learner for training so as to finish training of a plurality of meta-learner prediction models in a second layer of prediction models;
and S5, inputting the input variables of the days to be predicted into the trained first-layer prediction model and outputting the prediction results, and then taking the prediction results of the first-layer prediction model as the input variables of the meta-learner prediction model of the trained second-layer prediction model to obtain the final predicted photovoltaic output power.
In step S1, a time series of historical photovoltaic power generation power data and historical weather forecast data is obtained through the data preprocessing, where the time series includes photovoltaic power generation power P, total solar radiation intensity G, scattering level radiation intensity D, and wind speed WsTemperature T, relative humidity H, wind direction WdAnd the daily rainfall R, and then taking the time sequence as a photovoltaic prediction data sample set.
In step S1, the photovoltaic power station used for collecting data in this embodiment is an Alice springs photovoltaic power station in australia, the power station is composed of 22 photovoltaic panels with a rated value of 250W, the array rated value is 5.5KW, and grid-connected power generation is performed by an inverter. The temporal resolution was 5min, i.e. 288 data points per day.
In step S2, the time series of the historical weather forecast data is used as the input variable of the photovoltaic prediction data sample set, and the time series of the historical photovoltaic power generation power is used as the output variable of the photovoltaic prediction data sample set. The photovoltaic prediction data samples are aggregated for two years, and 172800 historical photovoltaic power generation power data and weather forecast data are obtained. 86400 samples are taken as a training set of the basic model, and 86400 samples are taken as a training set of the meta-learner. Specifically, a photovoltaic prediction data sample set S ═ { P is formedn,Gn,Dn,Wsn,Tn,Hn,Wdn,RnIn which P isnFor the corresponding predicted value, x, of the nth photovoltaic prediction data sample setn＝{Gn,Dn,Wsn,Tn,Hn,Wdn,RnAnd the input feature vectors are respectively corresponding to the nth photovoltaic prediction data sample set.
In step S3, the plurality of base learners include a Long Short-Term Memory Learner (LSTM), a threshold Recurrent Unit (GRU), and a Recurrent Neural Network learner (RNN).
In step S3, the step of establishing the prediction model of the long-term and short-term memory learner includes:
s311, taking input feature vector xtAs input at time t.
S312, setting the long-term and short-term memory model unit to comprise a forgetting gate, an input gate and an output gate, and setting a forgetting gate unit f at the t-th momenttAnd a t-th time input gate itAnd a t-th time output gate gtCell internal state update ctOutput of prediction model of long-short term memory learning device
Respectively as follows:
wherein sigma is a value converted by sigmoid activating function into a value between 0 and 1, and xtIs the input of the current time, ht-1Is a hidden state at the previous moment, ct-1Is the previous time cell state, Uf、Ui、Uo、UcRespectively showing a forgetting gate, an input gate, an output gate and a sheetInput weight matrix, W, for meta-internal state updatesf、Wi、Wo、WcCyclic weight matrices, M, representing respectively the update of the states inside the forgetting gate, the input gate, the output gate and the cellf,Mi,MoCell state weight matrices representing forgetting gate, input gate, output gate, respectively, bf、bi、bo、bcRespectively representing the bias of the forgetting gate, the input gate, the output gate, and the cell internal state update.
S313, taking
The prediction result 1 is output at time t.
In step S3, the step of establishing the threshold cycle unit learner prediction model includes:
s321, taking input feature vector xtAs input at time t.
S322, the threshold circulation unit comprises an input layer, a hidden layer and an output layer; the core of the threshold cycle unit is two gates of a hidden layer, and the influence of information on a final result through control historical data can be selectively caused; wherein the hidden layer comprises an update gate and a reset gate, zt、rtThe refresh gate and the reset gate at time t, respectively, are expressed by equations (6) - (9) and are again based on ztAnd
updating
Wherein sigma is a value converted by the sigmoid activation function into a value between 0 and 1,
and
activation parameters at time t and at time t-1 respectively,
is an activation parameter, Wz，WhRespectively representing cyclic weight matrixes of the update gate and the reset gate, W is a hidden layer-hidden layer connection weight, Uz、UrRepresenting the input weight matrix of the refresh gate and the reset gate respectively, U being the weight matrix of the input-hidden layer connection, bz、brRespectively, the bias of the refresh gate and the reset gate, and b is the bias.
S323, taking
The prediction result 2 is output at time t.
In step S3, the step of building the recurrent neural network learner prediction model includes:
s331, taking input feature vector xtAs input at time t.
S332, the input of the recurrent neural network is xtThe output is
The hidden layer is h, the hidden layer h is called a memory unit and has the capability of storing information, the output of the hidden layer h influences the input of the next moment, and the input of the moment t is assumed to be xtOutput of
Hidden state is
Then
I.e. input x with the current timetIn relation, the hidden state update also at the previous time is as follows:
wherein Q isiRepresents the output of the preceding output unit of the classifier, i represents the class index, C represents the total number of classes, SiRepresenting the ratio of the index of the current element to the sum of the indices of all elements, b and c are offsets, U, V, W are the weight matrices of input-hidden layer, hidden layer-output, hidden layer-hidden layer connections respectively,
represents the t-th hidden layer, and at different times, the values of the weight matrix U, V, W are the same, and f (x) is the tanh function or the ReLU function in the nonlinear activation function.
S333. taking
The prediction result 3 is output at the t-th time.
In step S4, the meta-learner prediction model is a long-term and short-term memory learner prediction model. In step S5, the input variable of the day to be predicted is
Table 1 shows a comparison of model error indicators provided by the embodiments of the present invention.
From the table 1, it can be seen that the short-term photovoltaic power generation prediction method based on stacking-ensemble learning has higher prediction accuracy.
Referring to fig. 2, it can be seen that the short-term photovoltaic power generation prediction method based on stacking-ensemble learning is closer to the actual value.
The same or similar reference numerals correspond to the same or similar parts;
the terms describing positional relationships in the drawings are for illustrative purposes only and are not to be construed as limiting the patent;
it should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (10)
1. The short-term photovoltaic power prediction method based on Stacking-ensemble learning comprises the following steps:
s1, acquiring historical photovoltaic power generation power data and historical weather forecast data and performing data preprocessing, so as to construct a photovoltaic prediction data sample set;
s2, dividing the photovoltaic prediction data sample set into a base model training set and a meta-learner training set;
s3, inputting the base model training set into a plurality of base learners for training, and establishing a plurality of base learner prediction models so as to complete the training of a first layer of prediction models;
s4, inputting the meta-learner training set into a plurality of base learner prediction models of a first layer of prediction models, outputting respective prediction results by each base learner prediction model, and inputting the prediction results of the first layer of prediction models into a meta-learner for training so as to finish training of the meta-learner prediction models in a second layer of prediction models;
and S5, inputting input variables of the days to be predicted into the trained first-layer prediction model and outputting prediction results, and then taking a plurality of prediction results of the first-layer prediction model as input variables of the meta-learner prediction model of the trained second-layer prediction model to obtain the final predicted photovoltaic output power.
2. The method for short-term photovoltaic power prediction based on Stacking-ensemble learning of claim 1, wherein in step S1, a time series of historical photovoltaic power generation power data and historical weather forecast data is obtained through the data preprocessing, and the time series includes photovoltaic power generation power P, total solar radiation intensity G, scattering level radiation intensity D, wind speed WsTemperature T, relative humidity H, wind direction WdAnd the daily rainfall R, and then taking the time sequence as a photovoltaic prediction data sample set.
3. The Stacking-ensemble learning-based short-term photovoltaic power prediction method according to claim 2, wherein in step S2, the time series of historical weather forecast data is used as input variables of the photovoltaic prediction data sample set, and the time series of historical photovoltaic power generation power is used as output variables of the photovoltaic prediction data sample set.
4. The Stacking-ensemble learning-based short-term photovoltaic power prediction method according to claim 3, wherein a photovoltaic prediction data sample set S ═ { P ═ P is formedn，Gn，Dn，Wsn，Tn，Hn，Wdn，RnIn which P isnFor the predicted value, x, corresponding to the nth sample setn＝{Gn，Dn，Wsn，Tn，Hn，Wdn，RnAnd is the input feature vector corresponding to the nth sample set.
5. The Stacking-ensemble learning based short-term photovoltaic power prediction method according to claim 1, wherein in step S3, the plurality of basis learners includes a long-term short-term memory learner, a threshold cyclic unit learner, and a cyclic neural network learner.
6. The Stacking-ensemble learning-based short-term photovoltaic power prediction method according to claim 5, wherein in step S3, the step of building a long-term short-term memory learner prediction model comprises:
s311, taking input feature vector xtAs input at time t;
s312, setting the long-term and short-term memory model unit to comprise a forgetting gate, an input gate and an output gate, and setting a forgetting gate unit f at the t-th momenttAnd a t-th time input gate itAnd a t-th time output gate gtCell internal state update ctOutput of prediction model of long-short term memory learning device
Respectively as follows:
wherein sigma is a value converted by sigmoid activating function into a value between 0 and 1, and xtIs the input of the current time, ht-1Is a hidden state at the previous moment, ct-1Is the previous time cell state, Uf、Ui、Uo、UcInput weight matrix W representing respectively forgetting gate, input gate, output gate and cell internal state updatef、Wi、Wo、WcCyclic weight matrices, M, representing respectively the update of the states inside the forgetting gate, the input gate, the output gate and the cellf，Mi，MoCell state weight matrices representing forgetting gate, input gate, output gate, respectively, bf、bi、bo、bcBiases representing a forgetting gate, an input gate, an output gate, and a cell internal state update, respectively;
s313, taking
And the predicted result is output at the t-th time.
7. The method for short-term photovoltaic power prediction based on Stacking-ensemble learning of claim 5, wherein in step S3, the step of establishing the threshold cyclic unit learner prediction model comprises:
s321, taking input feature vector xtAs input at time t;
s322, the threshold circulation unit comprises an input layer, a hidden layer and an output layer; the core of the threshold cycle unit is two gates of a hidden layer, and the influence of information on a final result through control historical data can be selectively caused; wherein the hidden layer includes an update gate and a resetDoor, zt、rtThe refresh gate and the reset gate at time t, respectively, are expressed by equations (6) - (9) and are again based on ztAnd
updating
Wherein sigma is a value converted by the sigmoid activation function into a value between 0 and 1,
and
activation parameters at time t and at time t-1 respectively,
is an activation parameter, Wz，WhRespectively representing cyclic weight matrixes of the update gate and the reset gate, W is a hidden layer-hidden layer connection weight, Uz、UrRespectively indicating update gate and update weightInput weight matrix of the input-hidden layer connection, U, bz、brRespectively representing the bias of the update gate and the reset gate, and b is the bias;
s323, taking
And the predicted result is output at the t-th time.
8. The method for short-term photovoltaic power prediction based on Stacking-ensemble learning of claim 5, wherein in step S3, the step of building the recurrent neural network learner prediction model comprises:
s331, taking input feature vector xtAs input at time t;
s332, the input of the recurrent neural network is xtThe output is
The hidden layer is h, the hidden layer h is called a memory unit and has the capability of storing information, the output of the hidden layer h influences the input of the next moment, and the input of the moment t is assumed to be xtOutput of
Hidden state is
Then
I.e. input x with the current timetIn relation, the hidden state update also at the previous time is as follows:
wherein Q isiRepresents the output of the preceding output unit of the classifier, i represents the class index, C represents the total number of classes, SiRepresenting the ratio of the index of the current element to the sum of the indices of all elements, b and c are offsets, U, V, W are the weight matrices of input-hidden layer, hidden layer-output, hidden layer-hidden layer connections respectively,
represents the tth hidden layer, and at different time, the value of the weight matrix U, V, W is the same, f (x) is the tanh function or the ReLU function in the nonlinear activation function;
s333. taking
And the predicted result is output at the t-th time.
9. The Stacking-ensemble learning-based short-term photovoltaic power prediction method according to claim 6, wherein in step S4, the meta-learner prediction model is a long-term short-term memory learner prediction model.
10. The Stacking-ensemble learning-based short-term photovoltaic power prediction method according to claim 2, wherein the daily input variable to be predicted is
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011469711.5A CN112561058B (en) | 2020-12-15 | 2020-12-15 | Short-term photovoltaic power prediction method based on Stacking-integrated learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011469711.5A CN112561058B (en) | 2020-12-15 | 2020-12-15 | Short-term photovoltaic power prediction method based on Stacking-integrated learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112561058A true CN112561058A (en) | 2021-03-26 |
| CN112561058B CN112561058B (en) | 2023-06-06 |
Family
ID=75062981
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011469711.5A Active CN112561058B (en) | 2020-12-15 | 2020-12-15 | Short-term photovoltaic power prediction method based on Stacking-integrated learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112561058B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112990593A (en) * | 2021-03-29 | 2021-06-18 | 广东工业大学 | Transformer fault diagnosis and state prediction method based on CSO-ANN-EL algorithm |
| CN113191091A (en) * | 2021-06-03 | 2021-07-30 | 上海交通大学 | Wind speed prediction method, system and equipment based on hybrid deep learning mechanism |
| CN113820079A (en) * | 2021-07-28 | 2021-12-21 | 中铁工程装备集团有限公司 | Hydraulic cylinder leakage fault diagnosis method based on cyclostationary theory and Stacking model |
| CN113866638A (en) * | 2021-08-24 | 2021-12-31 | 陈九廷 | Battery parameter inference method, device, equipment and medium |
| CN113887798A (en) * | 2021-09-28 | 2022-01-04 | 广东电网有限责任公司 | Power failure grid distribution prediction method, system, equipment and medium under typhoon disaster |
| CN114330935A (en) * | 2022-03-10 | 2022-04-12 | 南方电网数字电网研究院有限公司 | New energy power prediction method and system based on multiple combined strategy integrated learning |
| CN114692999A (en) * | 2022-04-26 | 2022-07-01 | 厦门大学 | Seawater surface temperature prediction method based on ensemble learning |
| CN115577857A (en) * | 2022-11-15 | 2023-01-06 | 南方电网数字电网研究院有限公司 | Energy system output data prediction method and device and computer equipment |
| CN117522626A (en) * | 2023-11-15 | 2024-02-06 | 河北大学 | Photovoltaic output prediction method based on feature selection and abnormal multi-model fusion |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108734331A (en) * | 2018-03-23 | 2018-11-02 | 武汉理工大学 | Short-term photovoltaic power generation power prediction method based on LSTM and system |
| CN110705789A (en) * | 2019-09-30 | 2020-01-17 | 国网青海省电力公司经济技术研究院 | Photovoltaic power station short-term power prediction method |
| CN111008728A (en) * | 2019-11-01 | 2020-04-14 | 深圳供电局有限公司 | Method for predicting short-term output of distributed photovoltaic power generation system |
-
2020
- 2020-12-15 CN CN202011469711.5A patent/CN112561058B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108734331A (en) * | 2018-03-23 | 2018-11-02 | 武汉理工大学 | Short-term photovoltaic power generation power prediction method based on LSTM and system |
| CN110705789A (en) * | 2019-09-30 | 2020-01-17 | 国网青海省电力公司经济技术研究院 | Photovoltaic power station short-term power prediction method |
| CN111008728A (en) * | 2019-11-01 | 2020-04-14 | 深圳供电局有限公司 | Method for predicting short-term output of distributed photovoltaic power generation system |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112990593A (en) * | 2021-03-29 | 2021-06-18 | 广东工业大学 | Transformer fault diagnosis and state prediction method based on CSO-ANN-EL algorithm |
| CN113191091A (en) * | 2021-06-03 | 2021-07-30 | 上海交通大学 | Wind speed prediction method, system and equipment based on hybrid deep learning mechanism |
| CN113191091B (en) * | 2021-06-03 | 2022-12-09 | 上海交通大学 | Wind speed prediction method, system and equipment based on hybrid deep learning mechanism |
| CN113820079A (en) * | 2021-07-28 | 2021-12-21 | 中铁工程装备集团有限公司 | Hydraulic cylinder leakage fault diagnosis method based on cyclostationary theory and Stacking model |
| CN113820079B (en) * | 2021-07-28 | 2024-05-24 | 中铁工程装备集团有限公司 | Hydraulic cylinder leakage fault diagnosis method based on cyclostationary theory and Stacking model |
| CN113866638A (en) * | 2021-08-24 | 2021-12-31 | 陈九廷 | Battery parameter inference method, device, equipment and medium |
| CN113887798A (en) * | 2021-09-28 | 2022-01-04 | 广东电网有限责任公司 | Power failure grid distribution prediction method, system, equipment and medium under typhoon disaster |
| CN114330935A (en) * | 2022-03-10 | 2022-04-12 | 南方电网数字电网研究院有限公司 | New energy power prediction method and system based on multiple combined strategy integrated learning |
| CN114692999A (en) * | 2022-04-26 | 2022-07-01 | 厦门大学 | Seawater surface temperature prediction method based on ensemble learning |
| CN115577857A (en) * | 2022-11-15 | 2023-01-06 | 南方电网数字电网研究院有限公司 | Energy system output data prediction method and device and computer equipment |
| CN117522626A (en) * | 2023-11-15 | 2024-02-06 | 河北大学 | Photovoltaic output prediction method based on feature selection and abnormal multi-model fusion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112561058B (en) | 2023-06-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112561058B (en) | Short-term photovoltaic power prediction method based on Stacking-integrated learning | |
| CN110059878B (en) | Photovoltaic power generation power prediction model based on CNN LSTM and construction method thereof | |
| Wang et al. | Deep learning method based on gated recurrent unit and variational mode decomposition for short-term wind power interval prediction | |
| US20220373984A1 (en) | Hybrid photovoltaic power prediction method and system based on multi-source data fusion | |
| CN109902863B (en) | Wind speed prediction method and device based on multi-factor time-space correlation | |
| CN114626512B (en) | High-temperature disaster forecasting method based on directed graph neural network | |
| CN112488415A (en) | Power load prediction method based on empirical mode decomposition and long-and-short-term memory network | |
| CN115293415A (en) | Multi-wind-farm short-term power prediction method considering time evolution and space correlation | |
| CN112215428B (en) | Photovoltaic power generation power prediction method and system based on error correction and fuzzy logic | |
| CN116894384B (en) | Multi-fan wind speed space-time prediction method and system | |
| CN114462718A (en) | CNN-GRU wind power prediction method based on time sliding window | |
| CN114781723A (en) | Short-term photovoltaic output prediction method based on multi-model fusion | |
| CN116384583A (en) | Photovoltaic power prediction method based on multiple neural networks | |
| CN115115125A (en) | Photovoltaic power interval probability prediction method based on deep learning fusion model | |
| CN115640901A (en) | Small sample load prediction method based on hybrid neural network and generation countermeasure | |
| CN115983494A (en) | Short-term wind power prediction method and system for newly-built small-sample wind power plant | |
| CN116341613A (en) | Ultra-short-term photovoltaic power prediction method based on Informar encoder and LSTM | |
| CN113393119B (en) | Stepped hydropower short-term scheduling decision method based on scene reduction-deep learning | |
| CN112183721B (en) | Construction method of combined hydrological prediction model based on self-adaptive differential evolution | |
| Chen et al. | Short-term Wind Speed Forecasting Based on Singular Spectrum Analysis, Fuzzy C-Means Clustering, and Improved POABP | |
| CN109447843B (en) | Photovoltaic output prediction method and device | |
| CN116187540B (en) | Wind power station ultra-short-term power prediction method based on space-time deviation correction | |
| CN115099497B (en) | CNN-LSTM-based real-time flood forecasting intelligent method | |
| CN116402194A (en) | Multi-time scale load prediction method based on hybrid neural network | |
| CN113902492A (en) | Time-sharing electricity price prediction method and system |
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 |