CN110543988A - Photovoltaic short-term output prediction system and method based on XGboost algorithm - Google Patents
Photovoltaic short-term output prediction system and method based on XGboost algorithm Download PDFInfo
- Publication number
- CN110543988A CN110543988A CN201910804739.0A CN201910804739A CN110543988A CN 110543988 A CN110543988 A CN 110543988A CN 201910804739 A CN201910804739 A CN 201910804739A CN 110543988 A CN110543988 A CN 110543988A
- Authority
- CN
- China
- Prior art keywords
- data
- prediction
- training
- data set
- module
- 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.)
- Pending
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 65
- 238000007781 pre-processing Methods 0.000 claims abstract description 21
- 238000007418 data mining Methods 0.000 claims abstract description 11
- 238000005065 mining Methods 0.000 claims abstract description 8
- 238000009412 basement excavation Methods 0.000 claims abstract description 3
- 238000012545 processing Methods 0.000 claims description 32
- 230000006870 function Effects 0.000 claims description 23
- 238000012360 testing method Methods 0.000 claims description 22
- 238000012795 verification Methods 0.000 claims description 21
- 230000002159 abnormal effect Effects 0.000 claims description 17
- 238000011156 evaluation Methods 0.000 claims description 17
- 238000010248 power generation Methods 0.000 claims description 16
- 238000003066 decision tree Methods 0.000 claims description 14
- 238000013139 quantization Methods 0.000 claims description 7
- 238000013079 data visualisation Methods 0.000 claims description 6
- 238000005192 partition Methods 0.000 claims description 6
- 238000000638 solvent extraction Methods 0.000 claims description 6
- 238000012804 iterative process Methods 0.000 claims description 4
- 230000005855 radiation Effects 0.000 claims description 4
- 238000002790 cross-validation Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000004146 energy storage Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention relates to a photovoltaic short-term output prediction system and method based on an XGboost algorithm, wherein the system comprises a data mining unit and a secondary depth mining unit, and the data mining unit comprises: the data preprocessing module is used for preprocessing input characteristic data; the data set dividing module is used for dividing the input characteristic data which is preprocessed in the data preprocessing module into a plurality of data sets; the secondary depth excavation unit includes: the prediction module is used for carrying out model training on the prediction model by utilizing the characteristic data to obtain a trained prediction model; the training module is used for predicting the photovoltaic short-term output by utilizing the trained prediction model and outputting a prediction result comprising the photovoltaic output power; the prediction model adopts an XGboost algorithm and adopts a CART tree as a base learner. Compared with the prior art, the method has the advantages of high efficiency of a prediction algorithm, accurate prediction result and the like.
Description
Technical Field
the invention relates to the field of new energy power generation of a power system, in particular to a photovoltaic short-term output prediction system and method based on an XGboost algorithm.
background
The new energy power generation has high volatility and randomness due to the inherent properties, power difference can be caused by insufficient output and energy storage or overload operation, so that a plurality of negative effects and even breakdown are brought to a power system, the generated power at the future moment can be obtained through prediction of the new energy power generation, so that the power difference is obtained, the intelligent scheduling of the microgrid system is realized, and the electric energy quality and the operation stability of the system are improved.
currently, short-term power generation prediction refers to prediction of power generation conditions from hours to days in the future. Short-term photovoltaic output prediction plays an important role in micro-grids and power systems. Under the current environmental protection pressure and the national support condition for new energy power generation, accurate power generation prediction can efficiently use new energy power generation equipment, the method is an important basis for realizing micro-grid power balance and large power grid cooperative power generation, and scientific guidance is provided for micro-grid scheduling arrangement.
at present, methods for predicting short-term photovoltaic output mainly comprise a time sequence method, an SVM (support vector machine), a BP (back propagation) neural network, an optimization algorithm based on an improved neural network and the like. The time series method is based on a linear model, and has defects in the field of processing multidimensional nonlinearity; the SVM has better generalization capability, but the model training time is long, and the requirements on short-term prediction are difficult to meet; BP has good learning ability and nonlinear expression ability, but is easy to fall into local optimum, and the convergence rate can not meet the requirement.
at present, prediction models are built more and more by using deep learning methods such as a deep neural network, and the performance of the prediction models in processing high-dimensional and nonlinear problems is particularly prominent. However, the advantages of the short-term output prediction model of the correlation algorithm based on the deep learning theory are generally based on the data set basis with high dimensionality and high data volume, and the correlation algorithm is long in calculation time, slow in convergence and easy to overfit.
the XGboost algorithm integrates a plurality of works of a gradient lifting algorithm, and a large amount of optimization is performed on engineering application, so that the XGboost algorithm is one of the most successful machine learning methods at present. The novel photovoltaic output prediction model based on the XGboost (extreme Gradient boosting) algorithm has great breakthrough in the aspects of convergence, calculation speed and data set dependency, the prediction result is accurate and efficient, the XGboost algorithm based on the integrated learner theory has many advantages in photovoltaic short-term output prediction, but the XGboost algorithm is not specifically applied to the field of new energy power generation at present.
disclosure of Invention
the invention aims to overcome the defects in the prior art and provide a photovoltaic short-term output prediction system and method based on an XGboost algorithm.
The purpose of the invention can be realized by the following technical scheme:
A photovoltaic short-term output prediction system based on an XGboost algorithm comprises a data mining unit and a secondary depth mining unit.
the data mining unit comprises:
the data preprocessing module is used for preprocessing input characteristic data including historical power generation data, weather and solar radiation; the preprocessing comprises the steps of carrying out abnormal value processing and missing value processing on input characteristic data.
The data preprocessing module comprises a data visualization sub-module, a data quantization sub-module, an abnormal value processing sub-module and a missing value processing sub-module, wherein the data visualization sub-module converts feature data of text types into chart data, the data quantization sub-module reasonably performs fuzzification and normalization processing on a plurality of feature data, the abnormal value processing sub-module effectively corrects the abnormal values in the feature data, and the missing value processing sub-module performs mean interpolation processing on the missing values in the feature data.
The data set dividing module is used for dividing the input characteristic data which is preprocessed in the data preprocessing module into a test data set, a verification data set and a training data set; the characteristic data in the training data set are used for training the prediction model, the training model is obtained when the objective function is minimum, the characteristic data in the verification data set are used for perfecting the training model, the trained prediction model is obtained when the evaluation index meets the required value, the characteristic data in the test data set pass through the trained prediction model, and finally the prediction result of the photovoltaic short-term output is output.
furthermore, the data set partitioning module partitions the data set through a cross validation method, evenly partitions the data set into multiple parts, takes most of the data as training data, and takes the rest of the data as test data to perform experiments.
further, the objective function includes a loss function and a regularization term, the loss function is a root mean square error with a first order term and a second order term, the regularization term relates complexity of the model, and the evaluation index includes a root mean square error RMSE and a mean absolute error percentage MAPE.
The secondary depth excavation unit includes:
The prediction module is used for carrying out model training on the prediction model by utilizing the characteristic data in the training data set and the verification data set to obtain a trained prediction model;
The training module is used for inputting the trained prediction model by using the characteristic data in the test data set, performing photovoltaic short-term output prediction and outputting a prediction result including photovoltaic output power;
the prediction model adopts an XGboost algorithm.
further, the XGboost algorithm adopts a CART tree as a base learner.
A photovoltaic short-term output prediction method based on an XGboost algorithm comprises the following steps:
1) A data preprocessing step: taking all input characteristic data including historical power generation data, weather and solar radiation as effective characteristic data, removing abnormal values of the effective characteristic data, and performing vacancy value processing by using a mean interpolation method;
2) A data set dividing step: dividing the characteristic data set into a training data set, a verification data set and a test data set by using a cross verification method;
3) training: training a prediction model by using the feature data in the verification data set and the training data set to obtain a trained prediction model;
4) a prediction step: and predicting the photovoltaic short-term output by using the characteristic data in the test data set, and outputting the prediction of the photovoltaic output power.
Further, the training step specifically includes:
301) Adjusting parameters of the XGboost prediction model, including tree depth, learning rate and iteration times;
302) building a prediction model by using the characteristic data in the training data set to obtain a training model;
303) Judging whether the target function reaches the minimum, if so, executing step 304), and if not, executing step 302);
304) Inputting the characteristic data in the verification data set into a training model, calculating an evaluation index, judging whether the evaluation index meets a required value, if so, saving the prediction model, and if not, executing the step 301).
further, the prediction model is built as an iterative process, a new decision tree is generated for each iteration, and the specific steps of one iterative process include:
3021) calculating the first and second derivatives of the loss function at each training sample point before the start of each iteration;
3022) Generating a new decision tree through a greedy strategy, and calculating a predicted value corresponding to each leaf node through a parameter value corresponding to the leaf node;
3023) adding the newly generated decision tree to the training model.
Further, the predicting step specifically includes:
401) inputting the characteristic data in the test data set into the prediction model stored in the step 304);
402) Outputting a predicted output value comprising photovoltaic output power;
403) And calculating an evaluation index according to the processing predicted value and the real output value.
Compared with the prior art, the invention has the following advantages:
1) the prediction algorithm is efficient: the deep learning method is applied to the field of new energy power generation, the XGboost algorithm is adopted, the CART tree is used as a base learner, and compared with other deep learning methods, the deep learning method has the advantages of good self-learning effect, high model solving speed, reduced interdependence among characteristics and the like;
2) the prediction result is accurate: compared with other prediction models, the prediction model based on the XGboost algorithm has lower RMSE and MAPE index values, provides more convenient, accurate and efficient prediction effect for the design of a photovoltaic system, and has higher practical value in practical application;
3) the data adaptability is high: by arranging the data mining unit, cleaning and basic processing are carried out on the original data, including dividing a data set, screening characteristic values, normalizing the characteristic values and the like, so that the test data are more suitable for subsequent photovoltaic prediction.
Drawings
FIG. 1 is a schematic diagram of the framework and flow of the present invention;
FIG. 2 is a diagram of an algorithm model according to the present invention.
The system comprises a data mining unit 1, a data mining unit 2, a secondary deep mining unit 11, a data preprocessing module 12, a data set partitioning module 21, a prediction module 22 and a training module.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments.
Examples
as shown in fig. 1, the invention provides a photovoltaic short-term output prediction system based on an XGBoost algorithm, which includes a data mining unit 1 and a secondary deep mining unit 2, where the data mining unit 1 includes a data preprocessing module 11 and a data set partitioning module 12, and the secondary deep mining unit 2 is machine learning algorithm software designed for photovoltaic output prediction, and includes a training module 21 and a prediction module 22.
The data preprocessing module 11 completes abnormal value processing and vacancy value processing on input feature data, the data set dividing module 12 divides the preprocessed feature data into a test data set, a verification data set and a training data set through a cross verification method, the training module 21 performs model training and model excellent performance evaluation on a prediction model based on an XGboost algorithm by using the feature data in the training data set and the verification data set to obtain a trained prediction model, the prediction module 22 performs photovoltaic short-term output prediction on the feature data in the test data set by using the prediction model trained in the training module 21, and outputs a prediction result, namely photovoltaic output power at an integral point moment.
In this embodiment, the data set partitioning module partitions the data set by a ten-fold cross-validation method, equally partitions the data set into ten parts according to the size, and performs an experiment by taking nine parts of the data set as training data and one part as test data in turn.
the data preprocessing module 11 is an integration of many functional sub-modules, including a data visualization sub-module, a data quantization sub-module, an abnormal value processing sub-module, and a missing value processing sub-module. The data visualization submodule converts the data of the text type into chart data, so that visualization experience is enhanced, and abnormal values and missing values can be conveniently and accurately positioned; the data quantization submodule reasonably fuzzifies and normalizes a plurality of characteristic data including weather types, season types and time periods of one day; the abnormal value processing sub-module effectively corrects the abnormal values in the data set, generally eliminates the whole sample corresponding to the abnormal values in each characteristic, and prevents the model from solving distortion and operation errors; the missing value processing sub-module generally performs mean interpolation processing on the missing data in the data set.
taking the weather type in the feature data as an example, the data quantization submodule performs data processing by using a fuzzy algorithm and a normalization algorithm, performs fuzzy operation based on an algorithm principle and an empirical value, and sets the weather type as a weather type in a clear day: 1, cloudy: 0.7, cloudy day: 0.4, rainy day: 0] into the interval [0,1 ]. Other characteristic data such as season type, time of day and the like are similarly processed.
the secondary deep mining unit 2 realizes secondary deep mining on the feature data after preprocessing and dividing the data set, and solves the prediction model through the training module 21 and the prediction module 22 to obtain the photovoltaic short-term output condition. The prediction model adopts an XGboost algorithm to solve the problem of supervised learning, the XGboost is an integrated learner in the field of machine learning, a CART tree is adopted as a base learner in the prediction model, and an objective function is defined.
The target function comprises a loss function and a regular term, the loss function adopts Mean Squared Error (MSE) with a first-order term and a second-order term to avoid model under-fitting, the regular term is associated with the complexity of the model, and the minimum regular term is optimized to reduce the complexity of the model and avoid model over-fitting.
the objective function is derived in a multi-step mode by combining the algorithm with the application scene, and the parameters of the objective function are derived results of the formula or the formula. The derivation of the objective function is:
wherein t is the number of iteration steps, n is the number of decision trees, l is the training loss, R is the regular term, f is each decision tree, and y is the addition model composed of decision trees f.
the smaller the value of the objective function is, the better the prediction effect of the prediction model is. Meanwhile, the Root Mean Square Error (RMSE) and mean absolute error percentage (MAPE) are selected as evaluation indexes of the effect by the prediction model. The smaller the root mean square error RMSE and the average absolute error percentage MAPE is, the more accurate the prediction is, and the better the adopted prediction model is.
in the training module 21, the feature data in the training data set is used for training the prediction model, the training is completed when the objective function is minimum, so as to obtain the training model, the feature data in the verification data set is used for improving the training model, the trained prediction model is obtained when the evaluation index meets the required value, and finally the prediction model is stored. In the prediction module 22, the feature data in the test data set is used for finally outputting the prediction result of the photovoltaic short-term output through the finally saved prediction model.
as shown in fig. 2, the XGBoost algorithm of this embodiment is an ensemble learner composed of weak learners, and is equivalent to an addition model composed of K decision trees.
the following table is an illustrative table of data input and output of the photovoltaic output prediction system of this embodiment. The input data comprises characteristics of historical integral point photovoltaic power generation values, average, highest and lowest temperatures, month, day and time, illumination intensity, humidity, seasonal type, day weather type and photovoltaic panel area in the day, and a historical data set comprising all the input characteristics is processed by a prediction model to obtain an output parameter, namely integral point photovoltaic output power.
As shown in fig. 1, the invention further provides a method for working the photovoltaic short-term output prediction system based on the XGBoost algorithm, and the specific flow is as follows:
Data preprocessing: and by using the advantage that the XGboost algorithm does not depend on the correlation among the features, all the features are defaulted as effective features, and only abnormal value elimination and vacancy value processing based on a mean interpolation method are carried out on the sample data.
data set partitioning: dividing a data set into a training data set, a verification data set and a test data set according to a cross verification method;
Model training: inputting feature data in the preprocessed training data set into an XGboost prediction model with initial parameters adjusted, and calculating parameter distribution of a decision tree layer by using a greedy algorithm to minimize a target function to obtain a training model; inputting the characteristic data in the verification data set into a training model, calculating evaluation indexes RMSE and MAPE according to the output predicted value and the real output value to judge the excellence of the model, if the required value cannot be reached, continuing optimizing parameters until a better effect is reached, and storing the prediction model;
and (3) output prediction: inputting the characteristic data in the test data set into a stored prediction model for output prediction, and calculating an evaluation index according to the output prediction value and the real output value: RMSE and MAPE.
the system adopts a prediction model meeting the evaluation index requirements or the optimal evaluation indexes to predict the photovoltaic output.
In the process of building a prediction model based on the XGboost algorithm, the algorithm generates a new decision tree in each iteration, and the specific process is as follows:
3021) Calculating the first and second derivatives of the loss function at each training sample point before the start of each iteration;
3022) Generating a new decision tree through a greedy strategy, and calculating a predicted value corresponding to each leaf node through a parameter value corresponding to the leaf node;
3023) Adding the newly generated decision tree to the training model.
while the invention has been described with reference to specific embodiments, the invention is not limited thereto, and those skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. the photovoltaic short-term output prediction system based on the XGboost algorithm is characterized by comprising a data mining unit and a secondary depth mining unit, wherein the data mining unit comprises:
The data preprocessing module is used for preprocessing input characteristic data including historical power generation data, weather and solar radiation; the preprocessing comprises the steps of carrying out abnormal value processing and missing value processing on input characteristic data;
The data set dividing module is used for dividing the input characteristic data which is preprocessed in the data preprocessing module into a test data set, a verification data set and a training data set;
the secondary depth excavation unit includes:
The prediction module is used for carrying out model training on the prediction model by utilizing the characteristic data in the training data set and the verification data set to obtain a trained prediction model;
The training module is used for inputting the trained prediction model by using the characteristic data in the test data set, performing photovoltaic short-term output prediction and outputting a prediction result including photovoltaic output power;
The prediction model adopts an XGboost algorithm.
2. the XGboost algorithm-based photovoltaic short-term output prediction system as claimed in claim 1, wherein the feature data in the training data set is used for training a prediction model, the training model is obtained when an objective function is minimum, the feature data in the verification data set is used for perfecting the training model, the trained prediction model is obtained when an evaluation index meets a required value, and the feature data in the test data set passes through the trained prediction model to finally output the prediction result of the photovoltaic short-term output.
3. The XGboost algorithm-based photovoltaic short-term output prediction system as claimed in claim 2, wherein the objective function comprises a loss function and a regularization term, the loss function is a root mean square error with a first order term and a second order term, the regularization term relates complexity of a model, and the evaluation index comprises a Root Mean Square Error (RMSE) and a mean absolute error percentage (MAPE).
4. The XGboost algorithm-based photovoltaic short-term output prediction system as claimed in claim 1, wherein the data pre-processing module comprises a data visualization sub-module, a data quantization sub-module, an abnormal value processing sub-module and a missing value processing sub-module, the data visualization sub-module converts text type feature data into chart data, the data quantization sub-module reasonably performs fuzzification and normalization processing on a plurality of feature data, the abnormal value processing sub-module effectively corrects abnormal values in the feature data, and the missing value processing sub-module performs mean interpolation processing on missing values in the feature data.
5. the XGboost algorithm-based photovoltaic short-term output prediction system as claimed in claim 1, wherein the data set partitioning module partitions the data set by a cross-validation method, equally partitions the data set into a plurality of parts, uses a part of the data as training data and the remaining part of the data as test data, and the number of the parts of the data as training data is greater than the number of the parts of the data as test data.
6. The XGboost algorithm-based photovoltaic short-term output prediction system as claimed in claim 1, wherein the XGboost algorithm uses CART tree as a base learner.
7. A prediction method of a photovoltaic short-term output prediction system based on an XGboost algorithm and applied to any one of claims 1-6, characterized by comprising the following steps:
1) a data preprocessing step: taking all input characteristic data including historical power generation data, weather and solar radiation as effective characteristic data, removing abnormal values of the effective characteristic data, and performing vacancy value processing by using a mean interpolation method;
2) A data set dividing step: dividing the characteristic data set into a training data set, a verification data set and a test data set by using a cross verification method;
3) Training: training a prediction model by using the feature data in the verification data set and the training data set to obtain a trained prediction model;
4) a prediction step: and predicting the photovoltaic short-term output by using the characteristic data in the test data set, and outputting the prediction of the photovoltaic output power.
8. The XGboost algorithm-based photovoltaic short-term output prediction method according to claim 7, wherein the training step specifically comprises:
301) Adjusting parameters of the XGboost prediction model, including tree depth, learning rate and iteration times;
302) building a prediction model by using the characteristic data in the training data set to obtain a training model;
303) judging whether the target function reaches the minimum, if so, executing step 304), and if not, executing step 302);
304) Inputting the characteristic data in the verification data set into a training model, calculating an evaluation index, judging whether the evaluation index meets a required value, if so, saving the prediction model, and if not, executing the step 301).
9. The XGboost algorithm-based photovoltaic short-term output prediction method according to claim 8, wherein the prediction model is built as an iterative process, a new decision tree is generated for each iteration, and the specific steps of one iterative process include:
3021) calculating the first and second derivatives of the loss function at each training sample point before the start of each iteration;
3022) Generating a new decision tree through a greedy strategy, and calculating a predicted value corresponding to each leaf node through a parameter value corresponding to the leaf node;
3023) adding the newly generated decision tree to the training model.
10. The XGboost algorithm-based photovoltaic short-term output prediction method according to claim 7, wherein the prediction step specifically comprises:
401) Inputting the characteristic data in the test data set into the prediction model stored in the step 304);
402) Outputting a predicted output value comprising photovoltaic output power;
403) And calculating an evaluation index according to the processing predicted value and the real output value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910804739.0A CN110543988A (en) | 2019-08-28 | 2019-08-28 | Photovoltaic short-term output prediction system and method based on XGboost algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910804739.0A CN110543988A (en) | 2019-08-28 | 2019-08-28 | Photovoltaic short-term output prediction system and method based on XGboost algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110543988A true CN110543988A (en) | 2019-12-06 |
Family
ID=68712254
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910804739.0A Pending CN110543988A (en) | 2019-08-28 | 2019-08-28 | Photovoltaic short-term output prediction system and method based on XGboost algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110543988A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111369070A (en) * | 2020-03-13 | 2020-07-03 | 西安理工大学 | Envelope clustering-based multimode fusion photovoltaic power prediction method |
CN111626508A (en) * | 2020-05-27 | 2020-09-04 | 华东师范大学 | Rail transit vehicle-mounted data prediction method based on xgboost model |
CN112364477A (en) * | 2020-09-29 | 2021-02-12 | 中国电器科学研究院股份有限公司 | Outdoor empirical prediction model library generation method and system |
CN113222208A (en) * | 2021-03-26 | 2021-08-06 | 辽宁工程技术大学 | Ada-XGboost-based traffic accident prediction system |
CN113658002A (en) * | 2021-08-17 | 2021-11-16 | 中国平安财产保险股份有限公司 | Decision tree-based transaction result generation method and device, electronic equipment and medium |
CN113947201A (en) * | 2021-08-02 | 2022-01-18 | 国家电投集团电站运营技术(北京)有限公司 | Training method and device for power decomposition curve prediction model and storage medium |
CN114240003A (en) * | 2022-02-23 | 2022-03-25 | 泰豪软件股份有限公司 | New energy output prediction method, system, storage medium and equipment |
CN114819248A (en) * | 2022-01-20 | 2022-07-29 | 山东浪潮科学研究院有限公司 | Wind power generation capacity prediction method and device based on random forest |
CN116061742A (en) * | 2022-10-25 | 2023-05-05 | 广州汇锦能效科技有限公司 | Charging control method and system for electric automobile in time-of-use electricity price photovoltaic park |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105929216A (en) * | 2016-04-08 | 2016-09-07 | 南京航空航天大学 | Satellite power source main bus-bar current interval prediction method |
CN109727066A (en) * | 2018-12-27 | 2019-05-07 | 浙江华云信息科技有限公司 | A kind of big industrial electricity consumers load forecasting method based on XGBoost algorithm |
CN109858974A (en) * | 2019-02-18 | 2019-06-07 | 重庆邮电大学 | Automobile-used family identification model construction method and recognition methods are purchased |
CN109902874A (en) * | 2019-02-28 | 2019-06-18 | 武汉大学 | A kind of micro-capacitance sensor photovoltaic power generation short term prediction method based on deep learning |
-
2019
- 2019-08-28 CN CN201910804739.0A patent/CN110543988A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105929216A (en) * | 2016-04-08 | 2016-09-07 | 南京航空航天大学 | Satellite power source main bus-bar current interval prediction method |
CN109727066A (en) * | 2018-12-27 | 2019-05-07 | 浙江华云信息科技有限公司 | A kind of big industrial electricity consumers load forecasting method based on XGBoost algorithm |
CN109858974A (en) * | 2019-02-18 | 2019-06-07 | 重庆邮电大学 | Automobile-used family identification model construction method and recognition methods are purchased |
CN109902874A (en) * | 2019-02-28 | 2019-06-18 | 武汉大学 | A kind of micro-capacitance sensor photovoltaic power generation short term prediction method based on deep learning |
Non-Patent Citations (3)
Title |
---|
SJPLJR: ""训练样本和测试样本"", 《HTTPS://BLOG.CSDN.NET/SJPLJR/ARTICLE/DETAILS/70169157》 * |
视界IT: ""XGBoost"", 《HTTPS://BLOG.CSDN.NET/WEIXIN_41725746/ARTICLE/DETAILS/89195333》 * |
钱仲文,陈浩: ""一种基于XGBoost算法的月度负荷预测方法"", 《浙江电力》 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111369070A (en) * | 2020-03-13 | 2020-07-03 | 西安理工大学 | Envelope clustering-based multimode fusion photovoltaic power prediction method |
CN111369070B (en) * | 2020-03-13 | 2023-06-27 | 西安理工大学 | Multimode fusion photovoltaic power prediction method based on envelope clustering |
CN111626508A (en) * | 2020-05-27 | 2020-09-04 | 华东师范大学 | Rail transit vehicle-mounted data prediction method based on xgboost model |
CN111626508B (en) * | 2020-05-27 | 2023-12-22 | 华东师范大学 | Track traffic vehicle-mounted data prediction method based on xgboost model |
CN112364477A (en) * | 2020-09-29 | 2021-02-12 | 中国电器科学研究院股份有限公司 | Outdoor empirical prediction model library generation method and system |
CN112364477B (en) * | 2020-09-29 | 2022-12-06 | 中国电器科学研究院股份有限公司 | Outdoor empirical prediction model library generation method and system |
CN113222208A (en) * | 2021-03-26 | 2021-08-06 | 辽宁工程技术大学 | Ada-XGboost-based traffic accident prediction system |
CN113947201A (en) * | 2021-08-02 | 2022-01-18 | 国家电投集团电站运营技术(北京)有限公司 | Training method and device for power decomposition curve prediction model and storage medium |
CN113658002B (en) * | 2021-08-17 | 2023-07-28 | 中国平安财产保险股份有限公司 | Transaction result generation method and device based on decision tree, electronic equipment and medium |
CN113658002A (en) * | 2021-08-17 | 2021-11-16 | 中国平安财产保险股份有限公司 | Decision tree-based transaction result generation method and device, electronic equipment and medium |
CN114819248A (en) * | 2022-01-20 | 2022-07-29 | 山东浪潮科学研究院有限公司 | Wind power generation capacity prediction method and device based on random forest |
CN114240003A (en) * | 2022-02-23 | 2022-03-25 | 泰豪软件股份有限公司 | New energy output prediction method, system, storage medium and equipment |
CN116061742A (en) * | 2022-10-25 | 2023-05-05 | 广州汇锦能效科技有限公司 | Charging control method and system for electric automobile in time-of-use electricity price photovoltaic park |
CN116061742B (en) * | 2022-10-25 | 2024-05-03 | 广州汇锦能效科技有限公司 | Charging control method and system for electric automobile in time-of-use electricity price photovoltaic park |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110543988A (en) | Photovoltaic short-term output prediction system and method based on XGboost algorithm | |
Saez-de-Ibarra et al. | Co-optimization of storage system sizing and control strategy for intelligent photovoltaic power plants market integration | |
Reinoso et al. | Simulation of photovoltaic centrals with dynamic shading | |
CN102930358B (en) | A kind of neural net prediction method of photovoltaic power station power generation power | |
CN105631558A (en) | BP neural network photovoltaic power generation system power prediction method based on similar day | |
CN104601104B (en) | Ultra-short term photovoltaic prediction method with ARMA correction based on LSSVM | |
CN114048896B (en) | Method, system, equipment and medium for predicting photovoltaic power generation data | |
CN111191854A (en) | Photovoltaic power generation prediction model and method based on linear regression and neural network | |
CN104050517A (en) | Photovoltaic power generation forecasting method based on GRNN | |
Huang et al. | A novel power output model for photovoltaic systems | |
CN113268931B (en) | Method and system for reconstructing photovoltaic array based on multi-target slime optimization algorithm | |
Nematirad et al. | Solar radiation forecasting using artificial neural networks considering feature selection | |
CN112149264A (en) | Active power distribution network planning method based on improved Harris eagle optimization algorithm | |
CN112149905A (en) | Photovoltaic power station short-term power prediction method based on wavelet transformation and wavelet neural network | |
CN115115127A (en) | Low-voltage transformer area intelligent regulation and control method and system based on deep learning | |
CN112927097A (en) | Photovoltaic power generation short-term prediction method based on GRA-ABC-Elman model | |
CN108846529A (en) | A kind of generated energy forecasting system | |
CN113837434A (en) | Solar photovoltaic power generation prediction method and device, electronic equipment and storage medium | |
CN116485139A (en) | Short-term photovoltaic power generation amount prediction method based on multi-feature fusion | |
Alanazi et al. | Day-ahead solar forecasting based on multi-level solar measurements | |
Habib et al. | Quasi-dynamic load and battery sizing and scheduling for stand-alone solar system using mixed-integer linear programming | |
CN116054129A (en) | Multi-electrolytic-tank combined operation power distribution method based on photovoltaic hydrogen production | |
CN116307111A (en) | Reactive load prediction method based on K-means clustering and random forest algorithm | |
Yuan et al. | A novel hybrid short-term wind power prediction framework based on singular spectrum analysis and deep belief network utilized improved adaptive genetic algorithm | |
CN114372615A (en) | Short-term photovoltaic power prediction method and system considering spatial correlation |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191206 |