CN114943371A - Principal component analysis-decision tree-based photovoltaic power station power generation prediction method and device - Google Patents
Principal component analysis-decision tree-based photovoltaic power station power generation prediction method and device Download PDFInfo
- Publication number
- CN114943371A CN114943371A CN202210499754.0A CN202210499754A CN114943371A CN 114943371 A CN114943371 A CN 114943371A CN 202210499754 A CN202210499754 A CN 202210499754A CN 114943371 A CN114943371 A CN 114943371A
- Authority
- CN
- China
- Prior art keywords
- data
- power generation
- training
- power station
- decision tree
- 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
- 238000010248 power generation Methods 0.000 title claims abstract description 88
- 238000003066 decision tree Methods 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012549 training Methods 0.000 claims abstract description 89
- 238000009434 installation Methods 0.000 claims abstract description 35
- 238000001514 detection method Methods 0.000 claims abstract description 20
- 238000000513 principal component analysis Methods 0.000 claims abstract description 16
- 238000012360 testing method Methods 0.000 claims abstract description 15
- 238000012795 verification Methods 0.000 claims abstract description 15
- 230000005855 radiation Effects 0.000 claims description 11
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 230000008901 benefit Effects 0.000 description 12
- 210000004027 cell Anatomy 0.000 description 7
- 239000000428 dust Substances 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 210000003850 cellular structure Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 229910021419 crystalline silicon Inorganic materials 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000009423 ventilation Methods 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- 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
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Human Resources & Organizations (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Photovoltaic Devices (AREA)
Abstract
The invention provides a photovoltaic power station power generation prediction method and device based on principal component analysis-decision tree, wherein the method comprises the following steps: acquiring basic installation information of a historical distributed photovoltaic power station, historical power generation data and historical weather detection data, and constructing a photovoltaic prediction data sample set; respectively carrying out principal component analysis on the preset meteorological data and the photovoltaic module technical specification parameters to determine meteorological factors and module technical specification parameters which have the greatest influence on the generated energy; determining complete algorithm training data and a data standard format required by an algorithm; dividing the complete algorithm training data into a training group, a verification group and a test group; constructing a decision tree algorithm training model, importing a training set into the decision tree algorithm training model for training, verifying the training set by using a verification set, and confirming the accuracy and generalization capability of the model by using a test set to obtain a final distributed photovoltaic power station daily power generation amount prediction model; and (4) performing power generation prediction of the photovoltaic power station by using the final daily power generation amount prediction model of the distributed photovoltaic power station.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a photovoltaic power station power generation prediction method and device based on principal component analysis-decision tree (PCA-DTR).
Background
The photovoltaic power generation process is a multivariable coupling nonlinear random process and has the defects of intermittence, uncontrollable property, uncertainty of time and space and the like. In recent years, with the continuous improvement of the grid-connected photovoltaic permeability in China, the photovoltaic power generation has great influence on the safe, economic and stable operation of a power system, so that the accurate prediction of the power generation amount of a grid-connected photovoltaic power station is urgently needed. The influence factors of the existing model for predicting the daily power generation of the distributed photovoltaic power station are not comprehensive enough.
Disclosure of Invention
The present invention aims to provide a method and apparatus for predicting power generation in a photovoltaic power plant based on principal component analysis-decision trees that overcomes, or at least partially addresses, the above-mentioned problems.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
one aspect of the invention provides a photovoltaic power station power generation prediction method based on principal component analysis-decision tree, which comprises the following steps: acquiring basic installation information, generating capacity data and historical weather detection data of a historical distributed photovoltaic power station, and constructing a photovoltaic prediction data sample set; respectively carrying out principal component analysis on the preset meteorological data and the photovoltaic module technical specification parameters to determine meteorological factors and module technical specification parameters which have the greatest influence on the generated energy; determining complete algorithm training data and a data standard format required by an algorithm, wherein the complete algorithm training data comprises: the meteorological factors and the technical specification parameters of the components, the installation angles of the components, the arrangement intervals of the components, the serial-parallel connection modes of the components and the shielding coefficients are analyzed by the principal components; dividing the complete algorithm training data into a training group, a verification group and a test group; building a decision tree algorithm training model, importing the training set into the decision tree algorithm training model for training, verifying the training set by using the verification set, and confirming the accuracy and generalization capability of the model by using a test set to obtain a final distributed photovoltaic power station daily power generation amount prediction model, wherein the input of the decision tree algorithm training model comprises the complete algorithm training data, and the output of the decision tree algorithm training model comprises the daily power generation amount of the photovoltaic power station; and predicting the power generation of the photovoltaic power station by using the final daily power generation amount prediction model of the distributed photovoltaic power station.
After acquiring basic installation information, generating capacity data and historical weather detection data of a historical distributed photovoltaic power station, and before constructing a photovoltaic prediction data sample set, the method further comprises the following steps: and performing data preprocessing on the basic installation information, the generated energy data and the historical weather detection data.
Wherein, the preset meteorological data comprises: solar radiation amount, temperature, humidity, wind speed and wind direction.
Wherein, the photovoltaic module technical specification parameters include: photoelectric conversion efficiency, rated power, size, half-chip or full-chip architecture.
The invention provides a photovoltaic power station power generation prediction device based on principal component analysis-decision tree, which comprises the following components: the building module is used for obtaining basic installation information, generating capacity data and historical weather detection data of a historical distributed photovoltaic power station and building a photovoltaic prediction data sample set; the principal component analysis module is used for respectively carrying out principal component analysis on the preset meteorological data and the technical specification parameters of the photovoltaic module and determining meteorological factors and the technical specification parameters of the module which have the largest influence on the generated energy; a determining module, configured to determine complete algorithm training data and a data standard format required by an algorithm, where the complete algorithm training data includes: the meteorological factors and the technical specification parameters of the components, the installation angles of the components, the arrangement intervals of the components, the serial-parallel connection modes of the components and the shielding coefficients are analyzed by the principal components; the dividing module is used for dividing the complete algorithm training data into a training group, a verification group and a test group; the training module is used for constructing a decision tree algorithm training model, importing the training set into the decision tree algorithm training model for training, verifying the training set by using the verification set, and confirming the accuracy and generalization capability of the model by using the test set to obtain a final distributed photovoltaic power station daily power generation amount prediction model, wherein the input of the decision tree algorithm training model comprises the complete algorithm training data, and the output of the decision tree algorithm training model comprises the daily power generation amount of the photovoltaic power station; and the prediction module is used for predicting the power generation of the photovoltaic power station by using the final daily power generation amount prediction model of the distributed photovoltaic power station.
The building module is further used for performing data preprocessing on the basic installation information, the power generation amount data and the historical weather detection data after the basic installation information, the power generation amount data and the historical weather detection data of the historical distributed photovoltaic power station are obtained and before a photovoltaic prediction data sample set is built.
Wherein, the preset meteorological data comprises: solar radiation amount, temperature, humidity, wind speed and wind direction.
Wherein the photovoltaic module specification parameters include: photoelectric conversion efficiency, power rating, size, half-chip or full-chip architecture.
Therefore, according to the photovoltaic power station power generation prediction method and device based on the principal component analysis-decision tree, provided by the invention, the meteorological factors and the component technical specification parameters which have the greatest influence on the power generation amount are extracted by using the principal component analysis, the daily power generation amount is used as an output amount by combining the component installation angle, the component arrangement distance, the component series-parallel connection mode and the shielding coefficient of a site selection area in the construction of the distributed photovoltaic power station, the model for predicting the power generation amount through the influence factors is finally obtained by introducing the training of the decision tree algorithm model, and the problem that the existing prediction model cannot predict the power generation amount accurately is solved by using the principal component analysis-decision tree algorithm mixed model.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on the drawings without creative efforts.
FIG. 1 is a flow chart of a method for predicting power generation of a photovoltaic power plant based on principal component analysis-decision tree according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for predicting power generation of a photovoltaic power plant, in particular based on a principal component analysis-decision tree, according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a photovoltaic power plant power generation prediction apparatus based on principal component analysis-decision tree according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
Fig. 1 and fig. 2 show a flowchart of a principal component analysis-decision tree-based photovoltaic power station power generation prediction method provided by an embodiment of the present invention, and referring to fig. 1 and fig. 2, the principal component analysis-decision tree-based photovoltaic power station power generation prediction method provided by an embodiment of the present invention includes:
and S1, obtaining basic installation information, generated energy data and historical weather detection data of the historical distributed photovoltaic power station, and constructing a photovoltaic prediction data sample set.
As an optional implementation manner of the embodiment of the present invention, after obtaining basic installation information and power generation amount data of a historical distributed photovoltaic power station and historical weather detection data, before constructing a photovoltaic prediction data sample set, the method for predicting power generation of a photovoltaic power station based on a principal component analysis-decision tree provided in the embodiment of the present invention further includes: and carrying out data preprocessing on the installation basic information, the generating capacity data and the historical weather detection data.
Specifically, the method comprises the steps of obtaining historical distributed photovoltaic power station installation basic information, generating capacity data and historical weather detection data, and conducting data preprocessing, so that a photovoltaic prediction data sample set is constructed.
And S2, performing principal component analysis on the preset meteorological data and the technical specification parameters of the photovoltaic module respectively, and determining meteorological factors and module technical specification parameters which have the greatest influence on the generated energy.
As an optional implementation manner of the embodiment of the present invention, the preset meteorological data includes: solar radiation amount, temperature, humidity, wind speed and wind direction. The technical specification parameters of the photovoltaic module comprise: photoelectric conversion efficiency, power rating, size, half-chip or full-chip architecture.
Specifically, principal component analysis is carried out on several types of meteorological data, a group of variables possibly having correlation are converted into a group of linearly uncorrelated variables through orthogonal transformation, and then meteorological factors which have the largest influence on the generated energy are selected; and performing principal component analysis on the characteristics of the photovoltaic module, and selecting the characteristics of the module which have the greatest influence on the power generation amount.
S3, determining complete algorithm training data and a data standard format required by the algorithm, wherein the complete algorithm training data comprises: meteorological factors and component technical specification parameters analyzed by the main components, component installation angles, component arrangement intervals, a series-parallel connection mode of the components and shielding coefficients.
Specifically, meteorological factors and component technical specification parameters, component installation angles, component arrangement intervals, a series-parallel connection mode of the components and shielding coefficients, which are analyzed by the principal components, are used as inputs, and the power generation amount of the photovoltaic power station is used as an output, so that complete algorithm training data is obtained.
S4, dividing the complete algorithm training data into a training group, a verification group and a test group;
s5, constructing a decision tree algorithm training model, importing a training set into the decision tree algorithm training model for training, verifying the training set by using a verification set, and confirming the accuracy and generalization capability of the model by using a test set to obtain a final distributed photovoltaic power station daily power generation amount prediction model, wherein the input of the decision tree algorithm training model comprises complete algorithm training data, and the output of the decision tree algorithm training model comprises the daily power generation amount of the photovoltaic power station.
Specifically, the factors that affect the power generation of a photovoltaic power plant are numerous and may include:
1. amount of solar radiation
The photovoltaic system has only about 10% utilization efficiency of solar radiation energy (solar cell efficiency, module combination loss, dust loss, control inverter loss, line loss, storage battery efficiency). The power generation capacity of the photovoltaic power station is directly related to the solar radiation quantity, and the radiation intensity and the spectral characteristics of the sun change along with meteorological conditions.
2. Inclination angle of solar cell module
When the photovoltaic module is tiled, the inclination angle is 0 degree; when the photovoltaic module is vertical to the ground (such as a south vertical face of a building), the inclination angle is 0 degree, the photovoltaic module is installed towards the south, and the installation inclination angle is 0-90 degrees: when the installation is carried out in the north direction, the installation inclination angle is between 0 degrees and minus 90 degrees. Different power generation quantities are different in different regions and at different inclination angles.
3. Photoelectric conversion efficiency of solar cell module
The capacity of the photovoltaic module to convert solar energy into electrical energy, i.e., the power generation capacity of the photovoltaic module.
4. Combined loss
Current loss due to current differences of the components occurs whenever the components are connected in series; voltage loss is caused by voltage difference of components when the components are connected in parallel; the combination loss can reach more than 8 percent, and the standard regulation of the Chinese engineering construction standardization association is less than 10 percent.
Wherein:
(1) in order to reduce the combination loss, the components with consistent current should be strictly selected to be connected in series before the installation of the power station;
(2) the attenuation characteristics of the assembly are as uniform as possible. According to the regulation of national standard GB/T-9535, the maximum output power of the solar cell module is detected after testing under the specified condition, and the attenuation of the maximum output power cannot exceed 8%;
(3) an isolation diode is sometimes necessary.
5. Temperature characteristic
The temperature rises by 1 ℃, and the crystalline silicon solar cell: the maximum output power is reduced by 0.04%, the open-circuit voltage is reduced by 0.04% (-2 mv/DEG C), and the short-circuit current is increased by 0.04%. To avoid the influence of temperature on the power production, good ventilation conditions of the assembly should be maintained.
The index belongs to the factory data of the components.
6. Loss of dust
Dust losses from the plant can reach 6% and therefore the components need to be wiped off frequently.
7. MPPT tracking
Maximum Power Point Tracking (MPPT) is an application from the viewpoint of solar battery application, namely, tracking of a maximum power point of a solar battery. And the MPPT function of the grid-connected system is completed in the inverter. It has recently been investigated to place it inside a dc combiner box.
8. Line loss
The line loss of the direct current loop and the alternating current loop of the system is controlled within 5 percent. Therefore, a conductive wire with good conductivity is adopted in design, and the conductive wire needs to have a sufficient diameter. The construction does not allow the material to be reduced by stealing the work. Special care must be taken in maintaining the system to ensure that the connectors and terminals are secure.
9. Controller, inverter efficiency
The voltage drop of the charging and discharging loop of the controller is not more than 5% of the system voltage. The efficiency of grid-connected inverters is currently greater than 95%.
10. Shade
Under certain conditions, some cells in a photovoltaic system are shielded by other objects around them, causing local shadows, which cause some cells to heat, resulting in a phenomenon known as "hot spots". If the shadow effect is not eliminated and exists for a long time, when the hot spot effect reaches a certain degree, the welding point on the component is melted and the grid line is damaged, so that the whole solar cell component is scrapped. The serious shielding not only influences the service life of the junction box and the assembly, but also seriously influences the power generation, and according to measurement and calculation, the power generation of a power station is reduced by about 20-30% due to the fact that very little tree shadow and wire shadow exist in a photovoltaic system.
The method has the advantages that the generated energy of the distributed photovoltaic power station is influenced a lot, the generated energy is greatly different from the expected generated energy according to the data analysis of the established photovoltaic project, and in order to more accurately estimate and predict the generated energy, the method utilizes a decision tree algorithm to establish a model to realize the prediction of the generated energy. Sufficient data volume is needed for realizing prediction, a training set and a verification set are divided, and the input quantity and the output quantity of a training model are determined.
Input quantity:
1. the method comprises the steps of analyzing main components of meteorological data (solar radiation quantity, temperature, humidity, wind direction and wind speed), and then selecting factors with large influence.
2. The photovoltaic module technical specification parameters (photoelectric conversion efficiency, size, power, half-piece or full-piece structure) are analyzed by principal component analysis, and then factors with larger influence are selected.
3. Angle of assembly installation
4. Pitch of arrangement of components
5. Series-parallel connection of components, and layout of components
6. Coefficient of occlusion
Output quantity: daily generated energy.
And S6, performing photovoltaic power station power generation prediction by using the final distributed photovoltaic power station daily power generation prediction model.
Therefore, by using the method for predicting the power generation of the photovoltaic power station based on the principal component analysis-decision tree provided by the embodiment of the invention, the meteorological factors and the technical specification parameters of the components which have the greatest influence on the power generation amount are extracted by using the principal component analysis, the daily power generation amount is used as an output amount by combining the component installation angle, the component arrangement distance, the serial-parallel connection mode of the components and the shielding coefficient of a site selection area in the construction of the distributed photovoltaic power station, the model for predicting the power generation amount through the influence factors is finally obtained by introducing the training of the decision tree algorithm model, and the daily power generation amount of the distributed photovoltaic power station can be predicted by using the mixed model of the principal component analysis-decision tree algorithm, so that the problem that the existing prediction model cannot predict the power generation amount accurately is solved.
The embodiment of the invention provides a photovoltaic power station power generation prediction method based on principal component analysis-decision tree, which has the following advantages:
the method has the advantages that 1, the generated energy influence factors are comprehensive, and the factors with larger influence are screened.
The method has the advantages that 2, the prediction of the daily power generation amount of the distributed photovoltaic power station under different conditions can be realized by inputting the influence factors.
Advantage 3, the method is highly interpretable.
Advantage 4, the model is easy to train, can adapt to the characteristic added fast.
Fig. 3 is a schematic structural diagram of a power generation prediction apparatus of a photovoltaic power station based on a principal component analysis-decision tree according to an embodiment of the present invention, in which the above method is applied, and the following is a brief description of the structure of the power generation prediction apparatus of the photovoltaic power station based on the principal component analysis-decision tree, and other things are not the least, please refer to the related description in the power generation prediction method of the photovoltaic power station based on the principal component analysis-decision tree, see fig. 3, and the power generation prediction apparatus of the photovoltaic power station based on the principal component analysis-decision tree according to the embodiment of the present invention includes:
the building module is used for obtaining basic installation information, generating capacity data and historical weather detection data of a historical distributed photovoltaic power station and building a photovoltaic prediction data sample set;
the principal component analysis module is used for respectively carrying out principal component analysis on the preset meteorological data and the technical specification parameters of the photovoltaic module and determining meteorological factors and the technical specification parameters of the module which have the greatest influence on the generated energy;
a determining module, configured to determine complete algorithm training data and a data standard format required by an algorithm, where the complete algorithm training data includes: the meteorological factors and the technical specification parameters of the components, the installation angles of the components, the arrangement intervals of the components, the serial-parallel connection modes of the components and the shielding coefficients are analyzed by the principal components;
the dividing module is used for dividing the complete algorithm training data into a training group, a verification group and a test group;
the training module is used for constructing a decision tree algorithm training model, importing the training set into the decision tree algorithm training model for training and verifying the training set by using the verification set, and then confirming the accuracy and the generalization capability of the model by using the test set to obtain a final distributed photovoltaic power station daily power generation amount prediction model, wherein the input of the decision tree algorithm training model comprises the complete algorithm training data, and the output of the decision tree algorithm training model comprises the daily power generation amount of the photovoltaic power station;
and the prediction module is used for predicting the power generation of the photovoltaic power station by using the final daily power generation amount prediction model of the distributed photovoltaic power station.
As an optional implementation manner of the embodiment of the present invention, the construction module is further configured to perform data preprocessing on the basic installation information, the power generation amount data, and the historical weather detection data after the basic installation information, the power generation amount data, and the historical weather detection data of the historical distributed photovoltaic power station are acquired and before the photovoltaic prediction data sample set is constructed.
As an optional implementation manner of the embodiment of the present invention, the preset meteorological data includes: solar radiation amount, temperature, humidity, wind speed and wind direction.
As an optional implementation manner of the embodiment of the present invention, the technical specification parameters of the photovoltaic module include: photoelectric conversion efficiency, power rating, size, half-chip or full-chip architecture.
Therefore, by using the photovoltaic power station power generation prediction device based on the principal component analysis-decision tree provided by the embodiment of the invention, the meteorological factors and the component technical specification parameters which have the greatest influence on the power generation amount are extracted by using the principal component analysis, the module installation angle, the module arrangement distance, the module series-parallel connection mode and the shielding coefficient of a site selection area of the distributed photovoltaic power station in the construction are combined to be used as predicted input quantity, the daily power generation amount is used as output quantity, the decision tree algorithm model is introduced for training, the model for predicting the power generation amount through the influence factors is finally obtained, and the daily power generation amount of the distributed photovoltaic power station can be predicted by using the principal component analysis-decision tree algorithm mixed model, so that the problem that the existing prediction model cannot predict the power generation amount accurately is solved.
The embodiment of the invention provides a photovoltaic power station power generation prediction device based on principal component analysis-decision tree, which has the following advantages:
the method has the advantages that 1, the generated energy influence factors are comprehensive, and the factors with larger influence are screened.
The method has the advantages that 2, the prediction of the daily power generation amount of the distributed photovoltaic power station under different conditions can be realized by inputting the influence factors.
Advantage 3, the method is highly interpretable.
Advantage 4, the model is easy to train and can adapt to the added features quickly.
Advantage 5, the device universality is stronger.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (8)
1. A photovoltaic power station power generation prediction method based on principal component analysis-decision tree is characterized by comprising the following steps:
acquiring basic installation information, generating capacity data and historical weather detection data of a historical distributed photovoltaic power station, and constructing a photovoltaic prediction data sample set;
respectively carrying out principal component analysis on the preset meteorological data and the photovoltaic module technical specification parameters to determine meteorological factors and module technical specification parameters which have the greatest influence on the generated energy;
determining complete algorithm training data and a data standard format required by the algorithm, wherein the complete algorithm training data comprises: the meteorological factors and the technical specification parameters of the components, the installation angles of the components, the arrangement intervals of the components, the serial-parallel connection modes of the components and the shielding coefficients are analyzed by the principal components;
dividing the complete algorithm training data into a training group, a verification group and a test group;
building a decision tree algorithm training model, importing the training set into the decision tree algorithm training model for training, verifying the training set by using the verification set, and confirming the accuracy and generalization capability of the model by using a test set to obtain a final distributed photovoltaic power station daily power generation amount prediction model, wherein the input of the decision tree algorithm training model comprises the complete algorithm training data, and the output of the decision tree algorithm training model comprises the daily power generation amount of the photovoltaic power station;
and predicting the power generation of the photovoltaic power station by using the final daily power generation amount prediction model of the distributed photovoltaic power station.
2. The method of claim 1, wherein after obtaining the historical distributed photovoltaic power plant basic installation information and power generation data and the historical weather detection data, before constructing the photovoltaic prediction data sample set, the method further comprises:
and performing data preprocessing on the basic installation information, the generated energy data and the historical weather detection data.
3. The method of claim 1, wherein the pre-set weather-like data comprises: solar radiation amount, temperature, humidity, wind speed and wind direction.
4. The method of claim 1, wherein the photovoltaic module specification parameters comprise: photoelectric conversion efficiency, power rating, size, half-chip or full-chip architecture.
5. A photovoltaic power station power generation prediction device based on principal component analysis-decision tree is characterized by comprising:
the building module is used for obtaining basic installation information, generating capacity data and historical weather detection data of a historical distributed photovoltaic power station and building a photovoltaic prediction data sample set;
the principal component analysis module is used for respectively carrying out principal component analysis on the preset meteorological data and the technical specification parameters of the photovoltaic module and determining meteorological factors and the technical specification parameters of the module which have the greatest influence on the generated energy;
a determining module, configured to determine complete algorithm training data and a data standard format required by an algorithm, where the complete algorithm training data includes: the meteorological factors and the technical specification parameters of the components, the installation angles of the components, the arrangement intervals of the components, the series-parallel connection modes of the components and the shielding coefficients are analyzed from the principal components;
the dividing module is used for dividing the complete algorithm training data into a training group, a verification group and a test group;
the training module is used for constructing a decision tree algorithm training model, importing the training set into the decision tree algorithm training model for training and verifying the training set by using the verification set, and then confirming the accuracy and the generalization capability of the model by using the test set to obtain a final distributed photovoltaic power station daily power generation amount prediction model, wherein the input of the decision tree algorithm training model comprises the complete algorithm training data, and the output of the decision tree algorithm training model comprises the daily power generation amount of the photovoltaic power station;
and the prediction module is used for predicting the power generation of the photovoltaic power station by using the final daily power generation amount prediction model of the distributed photovoltaic power station.
6. The device of claim 5, wherein the construction module is further configured to perform data preprocessing on the basic installation information and power generation amount data and the historical weather detection data after obtaining the basic installation information and power generation amount data and the historical weather detection data of the historical distributed photovoltaic power station and before constructing the photovoltaic prediction data sample set.
7. The apparatus of claim 5, wherein the pre-set weather-like data comprises: solar radiation amount, temperature, humidity, wind speed and wind direction.
8. The apparatus of claim 5, wherein the photovoltaic module specification parameters comprise: photoelectric conversion efficiency, power rating, size, half-chip or full-chip architecture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210499754.0A CN114943371A (en) | 2022-05-09 | 2022-05-09 | Principal component analysis-decision tree-based photovoltaic power station power generation prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210499754.0A CN114943371A (en) | 2022-05-09 | 2022-05-09 | Principal component analysis-decision tree-based photovoltaic power station power generation prediction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114943371A true CN114943371A (en) | 2022-08-26 |
Family
ID=82907739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210499754.0A Pending CN114943371A (en) | 2022-05-09 | 2022-05-09 | Principal component analysis-decision tree-based photovoltaic power station power generation prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114943371A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115587644A (en) * | 2022-08-30 | 2023-01-10 | 浙江正泰智维能源服务有限公司 | Photovoltaic power station performance parameter prediction method, device, equipment and medium |
| CN117196180A (en) * | 2023-10-17 | 2023-12-08 | 无锡市广盈电力设计有限公司 | Distribution line photovoltaic collection point site selection method containing high-proportion distributed photovoltaic |
| CN117522156A (en) * | 2023-10-17 | 2024-02-06 | 江苏尚诚能源科技有限公司 | Distributed photovoltaic prediction evaluation method and system based on big data analysis |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103440526A (en) * | 2013-07-05 | 2013-12-11 | 北京建筑大学 | Power generation prediction method and device based on building information model |
| CN107909218A (en) * | 2017-12-06 | 2018-04-13 | 南通大学 | A kind of short-term method for forecasting photovoltaic power generation quantity |
| CN110147912A (en) * | 2019-05-30 | 2019-08-20 | 江苏工程职业技术学院 | A kind of power generation prediction method and device based on Building Information Model |
| CN111738522A (en) * | 2020-06-28 | 2020-10-02 | 电子科技大学中山学院 | Photovoltaic power generation power prediction method, storage medium and terminal equipment |
| CN111815027A (en) * | 2020-06-09 | 2020-10-23 | 山东大学 | Photovoltaic station generated power prediction method and system |
| US20210194424A1 (en) * | 2019-04-25 | 2021-06-24 | Shandong University | Method and system for power prediction of photovoltaic power station based on operating data of grid-connected inverters |
| CN114329985A (en) * | 2021-12-30 | 2022-04-12 | 山东浪潮科学研究院有限公司 | Photovoltaic power generation prediction method, device, equipment and storage medium |
| CN114330100A (en) * | 2021-11-30 | 2022-04-12 | 国网辽宁省电力有限公司锦州供电公司 | Short-term photovoltaic power probability interval prediction method |
-
2022
- 2022-05-09 CN CN202210499754.0A patent/CN114943371A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103440526A (en) * | 2013-07-05 | 2013-12-11 | 北京建筑大学 | Power generation prediction method and device based on building information model |
| CN107909218A (en) * | 2017-12-06 | 2018-04-13 | 南通大学 | A kind of short-term method for forecasting photovoltaic power generation quantity |
| US20210194424A1 (en) * | 2019-04-25 | 2021-06-24 | Shandong University | Method and system for power prediction of photovoltaic power station based on operating data of grid-connected inverters |
| CN110147912A (en) * | 2019-05-30 | 2019-08-20 | 江苏工程职业技术学院 | A kind of power generation prediction method and device based on Building Information Model |
| CN111815027A (en) * | 2020-06-09 | 2020-10-23 | 山东大学 | Photovoltaic station generated power prediction method and system |
| CN111738522A (en) * | 2020-06-28 | 2020-10-02 | 电子科技大学中山学院 | Photovoltaic power generation power prediction method, storage medium and terminal equipment |
| CN114330100A (en) * | 2021-11-30 | 2022-04-12 | 国网辽宁省电力有限公司锦州供电公司 | Short-term photovoltaic power probability interval prediction method |
| CN114329985A (en) * | 2021-12-30 | 2022-04-12 | 山东浪潮科学研究院有限公司 | Photovoltaic power generation prediction method, device, equipment and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| 蒋浩等: "主成分分析结合神经网络的光伏发电量预测", 《电力系统及其自动化学报》 * |
| 阿迪娜·卡尼: ""基于DS-WMC综合预测模型的光伏电站短期功率研究"", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
| 龚莺飞 等: ""光伏功率预测技术"", 《电力系统自动化》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115587644A (en) * | 2022-08-30 | 2023-01-10 | 浙江正泰智维能源服务有限公司 | Photovoltaic power station performance parameter prediction method, device, equipment and medium |
| CN115587644B (en) * | 2022-08-30 | 2024-06-11 | 浙江正泰智维能源服务有限公司 | Photovoltaic power station performance parameter prediction method, device, equipment and medium |
| CN117196180A (en) * | 2023-10-17 | 2023-12-08 | 无锡市广盈电力设计有限公司 | Distribution line photovoltaic collection point site selection method containing high-proportion distributed photovoltaic |
| CN117522156A (en) * | 2023-10-17 | 2024-02-06 | 江苏尚诚能源科技有限公司 | Distributed photovoltaic prediction evaluation method and system based on big data analysis |
| CN117196180B (en) * | 2023-10-17 | 2024-04-26 | 无锡市广盈电力设计有限公司 | Distribution line photovoltaic collection point site selection method containing high-proportion distributed photovoltaic |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114943371A (en) | Principal component analysis-decision tree-based photovoltaic power station power generation prediction method and device | |
| Chen et al. | Forecasting based power ramp-rate control for PV systems without energy storage | |
| Jamil et al. | Performance ratio analysis based on energy production for large-scale solar plant | |
| Arévalo et al. | Smart monitoring method for photovoltaic systems and failure control based on power smoothing techniques | |
| Ramu et al. | An IoT‐based smart monitoring scheme for solar PV applications | |
| CN115409209A (en) | Distributed photovoltaic panel abnormity detection method and device and storage medium | |
| Taghezouit et al. | Model-based fault detection in photovoltaic systems: A comprehensive review and avenues for enhancement | |
| CN107482681B (en) | Solar cell set based on Internet and monitoring system and monitoring method thereof | |
| Lodhi et al. | Modelling and experimental characteristics of photovoltaic modules in typical days at an actual photovoltaic power station | |
| Zou et al. | Performance of a grid-connected BIPV system at Jiangsu Province, China | |
| Besheer et al. | Status Monitoring and Performance Investigation of a 5.1 kW Rooftop Grid-Tie Photovoltaic Energy System in Egypt | |
| Chen et al. | The design and implementation of dust monitoring system for photovoltaic power generation | |
| CN110059974B (en) | Photovoltaic socket type selection method based on photovoltaic optimized access | |
| Zhang et al. | Lifetime evaluation method of PV inverter considering the influence of fundamental frequency and low frequency junction temperature | |
| Muhammad et al. | Fault detection using acceptance ratio analysis on polycrystalline grid-connected photovoltaics system | |
| Elgamal et al. | Performance Metrics for Photovoltaic 4.0 Systems: 27.5 kW Solar Station as an Example | |
| CN107657346B (en) | Photovoltaic power station sub-transient potential parameter prediction method | |
| Sathiracheewin et al. | Short term analysis of PV system performance in different areas of Thailand | |
| Otani et al. | Field experience with large‐scale implementation of domestic PV systems and with large PV systems on buildings in Japan | |
| Mihov et al. | Performance Analysis of a Grid-Connected PV Park in Ruse, Bulgaria | |
| Belik et al. | Prediction of the degradation process of mono-Si photovoltaic panels in Ukrainian and Czech conditions | |
| Li et al. | Research on Grid-connected PV Power Station Condition Evaluation Technology Based on SCADA and Hierarchical Progression | |
| Xiaoli et al. | Condition monitoring and diagnosis for grid-connected photovoltaic power system | |
| Wang et al. | A multi-sensor information fusion monitoring system for photovoltaic power generation | |
| Yao et al. | Application of Photovoltaic Systems in Field Observation and Research Stations: Research on the Relationship between Power Generation Scale and Electricity Consumption to Improve Photovoltaic Application in Field Observation Stations |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220826 |