CN117218425A - Power generation loss analysis method and system for photovoltaic power station - Google Patents
Power generation loss analysis method and system for photovoltaic power station Download PDFInfo
- Publication number
- CN117218425A CN117218425A CN202311173599.4A CN202311173599A CN117218425A CN 117218425 A CN117218425 A CN 117218425A CN 202311173599 A CN202311173599 A CN 202311173599A CN 117218425 A CN117218425 A CN 117218425A
- Authority
- CN
- China
- Prior art keywords
- dust
- loss
- power station
- analysis
- sample
- 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 143
- 238000004458 analytical method Methods 0.000 title claims abstract description 124
- 239000000428 dust Substances 0.000 claims abstract description 465
- 238000009826 distribution Methods 0.000 claims abstract description 204
- 238000004364 calculation method Methods 0.000 claims abstract description 20
- 238000000034 method Methods 0.000 claims description 51
- 238000012549 training Methods 0.000 claims description 45
- 238000012544 monitoring process Methods 0.000 claims description 25
- 238000004590 computer program Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 13
- 238000003891 environmental analysis Methods 0.000 claims description 12
- 238000010801 machine learning Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 8
- 238000013135 deep learning Methods 0.000 claims description 6
- 238000003066 decision tree Methods 0.000 claims description 4
- 230000008021 deposition Effects 0.000 description 18
- 230000008859 change Effects 0.000 description 16
- 238000010586 diagram Methods 0.000 description 16
- 238000010276 construction Methods 0.000 description 6
- 230000005855 radiation Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 3
- 238000010191 image analysis Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000003042 antagnostic effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Classifications
-
- 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
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a power generation loss analysis method and a system for a photovoltaic power station, and relates to the technical field of photovoltaics. Acquiring a meteorological data sequence within a future preset time range through a meteorological analysis workstation, and carrying out power generation loss identification and dust influence parameter analysis to acquire a meteorological loss coefficient and dust influence parameters; through the dust analysis workstation, a plurality of dust parameters of all photovoltaic panels in the target power station are collected, a dust parameter array is calculated and obtained in combination with dust influence parameters, a dust distribution fitting image is constructed, cascade image generation analysis is carried out, a dust loss coefficient of the target power station is obtained, and a power generation loss coefficient and a power generation loss amount are calculated and obtained in combination with a shielding loss coefficient and a weather loss coefficient. The invention solves the technical problems of incomplete and inaccurate calculation and analysis of the power generation loss of the photovoltaic power station in the prior art.
Description
Technical Field
The invention relates to the technical field of photovoltaics, in particular to a power generation loss analysis system for a photovoltaic power station.
Background
The power generation operation efficiency of the photovoltaic power station is affected by a plurality of factors, such as sunshine hours, radiant quantity, shielding, dust coverage and the like, wherein the dust coverage particularly affects the power generation efficiency, and further affects the power station benefit.
In the prior art, the loss amount of power generation of a photovoltaic power station needs to be analyzed and calculated so as to approximately determine the power generation loss of the photovoltaic power station, and further the loss amount is used as a reference for carrying out relevant work such as maintenance analysis on the power station.
However, in the prior art, the means for analyzing and calculating the power generation loss of the photovoltaic power station generally performs calculation through approximate dust amount sampling, which is not comprehensive enough and not accurate enough, so that the analysis effect of the power generation loss in the construction and use of the photovoltaic power station is poor.
Disclosure of Invention
The application provides a power generation loss analysis method for a photovoltaic power station, which is used for solving the technical problems of incomplete and inaccurate calculation and analysis of the power generation loss of the photovoltaic power station in the prior art.
In a first aspect, the present application provides a method for analyzing power generation loss of a photovoltaic power station, the method being applied to a power generation loss analysis device for construction of the photovoltaic power station, the device including an environmental analysis station, a weather analysis station, a dust analysis station, and a comprehensive loss calculation module, the method comprising:
Acquiring current time information through an environmental analysis workstation, and identifying the time information based on a shielding identifier trained by target power station detection data to obtain a shielding loss coefficient, wherein the target power station is a photovoltaic power station;
collecting a meteorological data sequence within a future preset time range through a meteorological analysis workstation;
carrying out power generation loss identification on the meteorological data sequence, obtaining a meteorological loss coefficient, and carrying out dust influence parameter analysis on the gas phase data sequence to obtain dust influence parameters;
collecting a plurality of dust parameters of partial photovoltaic panels in all photovoltaic panels in a target power station through a dust analysis station, and calculating to obtain a dust parameter array by combining the dust influence parameters;
constructing a dust distribution fitting image according to the dust parameter array, and performing cascading image generation analysis to acquire a dust loss coefficient of the target power station;
and calculating and obtaining a power generation loss coefficient according to the shielding loss coefficient, the meteorological loss coefficient and the dust loss coefficient, and calculating and obtaining the power generation loss.
In a second aspect, the present application provides a power generation loss analysis system for a photovoltaic power plant, the system being connected to a power generation loss analysis apparatus for a photovoltaic power plant, the apparatus comprising an environmental analysis station, a weather analysis station, a dust analysis station, and a comprehensive loss calculation module, the system comprising:
The shielding loss analysis module is used for acquiring current time information through an environmental analysis station, identifying the time information based on a shielding identifier trained by detection data of a target power station, and obtaining a shielding loss coefficient, wherein the target power station is a photovoltaic power station;
the weather data acquisition module is used for acquiring weather data sequences within a future preset time range through a weather analysis workstation;
the weather loss analysis module is used for carrying out power generation loss identification on the weather data sequence, obtaining a weather loss coefficient, and carrying out dust influence parameter analysis on the gas phase data sequence to obtain dust influence parameters;
the dust parameter processing module is used for collecting a plurality of dust parameters of partial photovoltaic panels in all photovoltaic panels in the target power station through the dust analysis station, and calculating and obtaining a dust parameter array by combining the dust influence parameters;
the dust loss analysis module is used for constructing a dust distribution fitting image according to the dust parameter array, generating and analyzing a cascade image, and acquiring a dust loss coefficient of the target power station;
and the power generation loss calculation module is used for calculating and obtaining a power generation loss coefficient according to the shielding loss coefficient, the meteorological loss coefficient and the dust loss coefficient and calculating and obtaining the power generation loss amount.
In a third aspect, a computer device comprises a memory storing a computer program and a processor implementing the steps of the method of the first aspect when the computer program is executed.
In a fourth aspect, a computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the method in the first aspect.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
according to the technical scheme, the current time information is collected in the photovoltaic power station, the power generation loss caused by shielding of surrounding buildings or other objects to the photovoltaic power station is analyzed based on the time, the power generation loss caused by current weather change is collected and analyzed according to weather data, the influence parameters on dust deposition in the photovoltaic power station are analyzed according to the weather data, the dust parameters of part of the collected and obtained photovoltaic panels are corrected and calculated to obtain a dust parameter array, dust with larger influence on the power generation loss is fitted, a dust distribution fitting image is constructed through fitting, cascade image generation and analysis are performed to obtain the power generation loss coefficient of the power station caused by dust, the power generation loss in three aspects is synthesized, and the integral power generation loss in the photovoltaic power station is calculated and obtained. According to the technical scheme, the analysis of the power generation loss is carried out through multiple dimensions, the comprehensiveness of the analysis of the power generation loss of the photovoltaic power can be improved, the influence parameters of the meteorological data analysis on dust deposition can be improved, the accuracy of the dust influence analysis with the largest influence on the power generation loss can be improved, the dust parameters of part of photovoltaic panels are collected to form dust distribution images in a fitting mode, the accuracy and the reliability of the power generation loss through the dust analysis are improved through cascading image analysis, the dust data of all photovoltaic panels are not required to be collected, the cost of the monitoring analysis of the power generation loss is reduced, and the technical effects of improving the comprehensiveness and the accuracy of the power generation loss analysis of a photovoltaic power station are achieved.
Drawings
FIG. 1 is a schematic flow chart of a method for analyzing the power generation loss of a photovoltaic power station;
FIG. 2 is a schematic flow chart of the method for analyzing the power generation loss of a photovoltaic power station to obtain the dust loss coefficient;
fig. 3 is a schematic structural diagram of a power generation loss analysis system for a photovoltaic power station.
Fig. 4 is an internal structural diagram of a computer device in one embodiment.
Reference numerals illustrate: the system comprises a shielding loss analysis module 201, a meteorological data acquisition module 202, a meteorological loss analysis module 203, a dust parameter processing module 204, a dust loss analysis module 205 and a power generation loss calculation module 206.
Detailed Description
The application provides a method and a system for analyzing the power generation loss of a photovoltaic power station, which are used for solving the technical problems of incomplete and inaccurate calculation and analysis of the power generation loss of the photovoltaic power station in the prior art.
Example 1
As shown in fig. 1, an embodiment of the present application provides a power generation loss analysis method for a photovoltaic power station, where the method is applied to a power generation loss analysis device for construction of the photovoltaic power station, and the device includes an environmental analysis station, a weather analysis station, a dust analysis station, and a comprehensive loss calculation module, and the method includes:
S101: acquiring current time information through an environmental analysis workstation, and identifying the time information based on a shielding identifier trained by target power station detection data to obtain a shielding loss coefficient, wherein the target power station is a photovoltaic power station;
in the embodiment of the application, the power generation loss analysis device for the construction of the photovoltaic power station is a device for executing the power generation loss analysis method for the photovoltaic power station, and comprises a memory, a processor and the like, and the device can be used for storing a program required by the execution method and executing the execution process.
The power generation loss analysis device for the construction of the photovoltaic power station comprises an environment analysis station, a weather analysis station, a dust analysis station and a comprehensive loss calculation module which are respectively used for executing different steps in the method.
The target power station is a photovoltaic power station, preferably a power station with larger scale and more photovoltaic panels for absorbing solar radiation to generate power, and the target power station can be blocked, weather and dust deposited and other factors to cause photovoltaic power generation loss.
In the embodiment of the application, the current time information, namely the real-time of the current execution method for power generation loss analysis, is acquired through the environment analysis workstation, wherein the time information can be the time of accurate minutes in one day or the time of accurate days in one year, and the target power station is changed by the solar shielding areas of other buildings, trees, hills and the like in the environment due to the movement of the sun in one day, so that the power generation loss amount is changed. And the change of the area of the shielded area and the change of the sunshine time in the target power station caused by the revolution movement of the sun within one year, thereby causing the change of the power generation loss.
The environment analysis workstation can be equipment with a time acquisition function, and is used for analyzing the power generation loss of the target power station in the current time due to the area of the shielded area in different time by acquiring the current time information.
The method comprises the steps of training a shielding identifier for identifying power generation loss caused by sunlight shielding in a target power station under different time based on detection data of power generation amount detection of the target power station and combining time information acquired by an environmental analysis station, and identifying the current time information to quickly and accurately obtain a power generation loss coefficient caused by sunlight shielding in the target power station under the current time.
Step S101 in the method provided by the embodiment of the present application includes:
acquiring the lost power generation capacity of all the photovoltaic panels in the target power station in a plurality of time periods according to the monitoring data of the target power station, and acquiring a sample time information record and a sample shielding loss coefficient record;
the sample time information record and the sample shielding loss coefficient record are adopted as training data, and a shielding identifier meeting convergence requirements is trained based on machine learning;
and identifying the time information by adopting the shielding identifier to obtain a shielding loss coefficient.
According to the operation monitoring data of the target power station, particularly the operation time monitoring data and the monitoring data of the shielded areas of all photovoltaic panels in the target power station, the monitoring data in a period of time can be collected. Thus, a sample time information record is obtained.
And further calculating and obtaining the lost generated energy of all the photovoltaic panels due to shielding according to the monitoring data of the shielded areas of all the photovoltaic panels, further calculating the ratio of the lost generated energy to the expected generated energy under the expected non-shielding condition, and obtaining a sample shielding loss coefficient record.
The sample time information record and the sample shielding loss coefficient record are used as training data, and a shielding identifier is pointed out based on machine learning, for example, based on a feedforward neural network in the machine learning in the prior art, and the shielding identifier is used for analyzing the photovoltaic power generation loss coefficient of a target power station due to shielding of sunlight according to the time information of the target power station.
The sample time information record and the sample shielding loss coefficient record are used as training data and are respectively used as input and output of the shielding identifier, the shielding identifier is supervised and trained, and network parameters in the shielding identifier are adjusted according to errors of output values and sample values, so that multiple rounds of training are performed until the shielding identifier meets convergence requirements.
Illustratively, the convergence requirement may be an occlusion identifier accuracy of up to 90%.
Based on the shielding identifier meeting the convergence requirement, the current acquired time information can be predicted and identified, and the shielding loss coefficient of the generated energy lost due to the shielding of sunlight in the target power station at the time is obtained.
According to the embodiment of the application, based on machine learning, the implicit relation between the training time and the different power generation loss coefficients caused by the shielding of sunlight in the target power station is used for carrying out recognition prediction on the power generation loss coefficients caused by the shielding of light, so that the accuracy and the efficiency of power generation loss analysis caused by the shielding of sunlight are improved.
S102: collecting a meteorological data sequence within a future preset time range through a meteorological analysis workstation;
in the embodiment of the application, the change of the solar radiation amount can influence the power generation amount of the photovoltaic power station due to the change of weather, so that the meteorological data are further collected, and the proportion of the power generation loss amount of the target power station caused by the meteorological change is analyzed.
And acquiring a meteorological data sequence within a future preset time range through a meteorological analysis workstation, wherein the preset time range is a time range with any time length, namely, the preset time range after the current time can be, for example, one day or a week.
The weather analysis workstation can be software capable of collecting weather data published by a weather publishing website, and collecting weather data within a future preset time range of an area where the target power station is located, for example, the weather analysis workstation can collect the weather data according to a certain frequency, for example, once a day or once an hour, so as to form a weather data sequence.
For example, weather data, temperature, wind direction, wind speed, and the like may be included within the weather data.
S103: carrying out power generation loss identification on the meteorological data sequence, obtaining a meteorological loss coefficient, and carrying out dust influence parameter analysis on the gas phase data sequence to obtain dust influence parameters;
in the embodiment of the application, the meteorological data sequence is subjected to power generation loss identification, namely the proportion of photovoltaic power generation loss caused by the change of the solar radiation amount due to meteorological change.
And, the weather change may cause a change in the dust deposition amount in the power station, for example, in a low humidity weather, the dust deposition amount may be large, and therefore, the influence proportion on the dust deposition amount is analyzed according to the weather data as a dust influence parameter, as part of data of the power generation loss caused by the dust deposition analysis according to the dust later.
And (3) carrying out power generation loss analysis on the meteorological data sequence to obtain the proportion of the power generation loss caused by the meteorological data sequence as a corresponding meteorological loss coefficient. And analyzing dust influence parameters influencing the dust deposition amount of the meteorological data sequence to obtain dust influence parameters under the meteorological data sequence.
Step S103 in the method provided by the embodiment of the present application specifically includes:
acquiring a plurality of sample meteorological data sequences and a plurality of sample meteorological loss coefficients according to the monitoring data of the target power station;
adopting a plurality of sample meteorological data sequences and a plurality of sample meteorological loss coefficients as training data, and training a meteorological loss identifier meeting convergence requirements based on machine learning;
and identifying the meteorological data sequence by adopting a meteorological loss identifier to obtain the meteorological loss coefficient.
According to the embodiment of the application, a plurality of sample meteorological data sequences in different preset time ranges in historical time and the ratio of the lost power generation amount to the expected power generation amount of the target power station due to meteorological change are obtained according to the monitoring data of the target power station and used as a plurality of sample meteorological loss coefficients.
The sample meteorological loss coefficient can be obtained by calculating the variation of solar radiation quantity under different meteorological data sequences and the solar radiation quantity under an ideal photovoltaic power generation state, carrying out simulated photovoltaic power generation analysis based on the variation to obtain simulated loss generated energy, and calculating the ratio of the simulated loss generated energy to the generated energy under the ideal power generation state.
A plurality of sample meteorological data sequences and a plurality of sample meteorological loss coefficients are used as training data, based on machine learning, similarly, a meteorological loss identifier can be constructed based on a feedforward neural network in the prior art, input data of the meteorological loss identifier are meteorological data sequences, output data of the meteorological loss identifier are meteorological loss coefficients, and the meteorological loss identifier is trained until convergence requirements are met.
And identifying the currently acquired meteorological data sequence by adopting the trained meteorological loss identifier, and acquiring a meteorological loss coefficient of the generating capacity of the target power station under the meteorological data sequence.
According to the embodiment of the application, the machine learning technology is adopted, so that the photovoltaic power generation loss caused by the influence of weather on the solar radiation quantity is analyzed according to the weather data sequence, the dimension of the photovoltaic power generation loss analysis can be enriched, and the comprehensiveness and the accuracy of the photovoltaic power generation loss analysis are improved.
Further, step S103 in the method provided by the embodiment of the present application further includes:
according to the dust monitoring data of the target power station, the variation of dust amount under different meteorological data sequences is obtained, and a plurality of sample dust influence parameters are obtained through calculation;
based on a decision tree, adopting meteorological data as decision input, adopting dust influence parameters as decision data, and constructing a dust influence classifier according to a plurality of sample meteorological data sequences and a plurality of sample dust influence parameters;
And adopting a dust influence classifier to carry out decision classification on the meteorological data sequence, and obtaining the dust influence parameters.
According to the embodiment of the application, the change of the dust amount under different meteorological data sequences is obtained according to the dust monitoring data of the target power station, and a plurality of sample dust influence parameters are obtained through calculation.
For example, the dust deposition amount, such as the dust deposition area, on the photovoltaic panel in a predetermined time range, such as 1 day or 1 week, in the ideal power generation state in the target power station may be tested through the test. Further, through tests, dust deposition amounts of the photovoltaic panel in a preset time range under different meteorological data sequences are tested, and then the variation of the dust deposition amounts under the different meteorological data sequences is calculated and obtained, and then the dust deposition rate under the different meteorological data sequences in unit time is calculated and obtained to be used as a plurality of sample dust influence parameters.
Further, based on a decision tree algorithm in the prior art, weather data is adopted as decision input, dust influence parameters are adopted as decision data, wherein different multi-layer decision nodes are constructed according to different types of weather data in the weather data, such as weather data sequences, temperature data sequences and the like, each layer of decision nodes can carry out classification decision according to the input weather data, for example, decision classification is carried out according to first data in the weather data sequences, and a decision output result of a final decision node is constructed according to the mapping relation between different sample dust influence parameters and a plurality of sample weather data sequences, so that a dust influence classifier is constructed.
Alternatively, the dust impact classifier may be constructed using the plurality of sample weather data sequences and the plurality of sample dust impact parameters according to other classification algorithms known in the art.
The dust influence classifier is adopted to carry out decision classification on dust influence parameters on the meteorological data sequence in the current meteorological data sequence, and specifically, multi-layer classification is carried out according to different types of meteorological data in the meteorological data sequence, so that the dust influence parameters in a future preset time range in the target power station are obtained.
According to the embodiment of the application, the influence of the meteorological change on dust deposition in the target power station is analyzed according to the meteorological data sequence to obtain the dust influence parameter, the accuracy of the subsequent analysis of the power generation loss in the target power station according to the dust analysis is improved, the meteorological loss coefficient and the dust influence parameter are obtained through the simultaneous analysis of a group of meteorological data sequences, the data utilization rate is improved, and the comprehensiveness of the photovoltaic power generation loss is improved.
S104: collecting a plurality of dust parameters of partial photovoltaic panels in all photovoltaic panels in a target power station through a dust analysis station, and calculating to obtain a dust parameter array by combining the dust influence parameters;
in the embodiment of the application, the dust analysis workstation is used for collecting a plurality of dust parameters of part of the photovoltaic panels in all the photovoltaic panels in the target power station, and the dust shielding area of part of the photovoltaic panels is collected, and optionally, the dust density or the maximum dust thickness of part of the photovoltaic panels can also be collected.
The dust analysis workstation can comprise image analysis equipment arranged on part of the photovoltaic panels, and can acquire surface images of the photovoltaic panels and perform image recognition so as to acquire a plurality of dust parameters of the part of the photovoltaic panels.
Alternatively, the dust analysis station may also manually collect and analyze dust parameters on portions of the photovoltaic panel.
Preferably, part of the photovoltaic panels are a plurality of photovoltaic panels in the central part of the target power station, and the occupied area of the part of the photovoltaic panels is 1/10 of the occupied area of the target power station.
By collecting dust parameters on part of the photovoltaic panels, the dust parameters are used as a data basis for dust to influence photovoltaic power generation loss analysis, so that the dust parameters on all the photovoltaic panels are prevented from being collected, and the cost of photovoltaic power generation loss analysis is reduced on the basis of ensuring analysis accuracy.
And based on the plurality of dust parameters of the partial photovoltaic panel, respectively carrying out correction calculation processing on the plurality of dust parameters by combining the dust influence parameters obtained by the processing in the above.
Illustratively, the dust influencing parameters include the dust deposition rate under the meteorological data sequence, and in combination with the current dust parameters, the corrected dust parameters on the rear part of the photovoltaic panel, for example after one week or one day, are calculated as a plurality of corrected dust parameters of the part of the photovoltaic panel within a future preset time frame.
Further, a plurality of corrected dust parameters are arranged based on a photovoltaic panel sequence of a portion of the photovoltaic panels that collect the dust parameters to form a dust parameter array.
S105: constructing a dust distribution fitting image according to the dust parameter array, and performing cascading image generation analysis to acquire a dust loss coefficient of the target power station;
according to the embodiment of the application, the dust parameter array is converted into a dust distribution fitting image in the target power station as the dust distribution fitting image, and the dust distribution fitting image is used for generating and analyzing a cascade image of the image so as to analyze the photovoltaic power generation loss proportion caused by dust shielding in the target power station and acquire the dust loss coefficient of the target power station.
As shown in fig. 2, step S105 in the method provided in the embodiment of the present application includes:
acquiring coordinate parameters of a plurality of photovoltaic panels in the target power station, and acquiring a dust distribution fitting image in the target power station by combining the dust parameter array;
training to obtain a dust distribution map generator with K-1 layers of dust distribution map generation channels based on deep learning;
processing the dust distribution fitting image by adopting a dust distribution map generator to obtain a generated fine dust distribution image;
Training to obtain a dust loss identifier;
and identifying the fine dust distribution image by adopting the dust loss identifier to obtain the dust loss coefficient.
In the embodiment of the application, coordinate parameters of a plurality of photovoltaic panels in a target power station are acquired, the position distribution of a part of photovoltaic panels for collecting dust parameters is determined, and the position distribution is used as a plurality of position information of a plurality of pixel points of a fitting image based on the position analysis. Illustratively, according to the position distribution, a mapping relationship between the photovoltaic panel and a plurality of pixel points in the fitting image is constructed.
The coordinate parameters of the photovoltaic panels can be obtained according to construction design data of the photovoltaic power station.
And combining the dust parameter array, and taking the dust parameter of the photovoltaic panel corresponding to each pixel point as the gray value of the pixel point, so as to form a preliminarily fitted dust distribution image. For example, a gray value corresponding to a dust parameter average value in the dust parameter array may be taken as 127, and a rounding may be calculated based on a ratio of other dust parameters to the dust parameter average value, so as to obtain a corresponding gray value as a dust distribution image.
Some photovoltaic panels are preferably a plurality of photovoltaic panels that are integrally formed in a rectangular distribution, and nickable names obtain regular dust distribution images that form a rectangle. Optionally, if the distribution of a part of the photovoltaic panels is irregular, gray values of blank pixel points can be obtained by a linear difference method for part of blank pixel points which do not correspond to the photovoltaic panels in the preliminarily fitted dust distribution image, so that a dust distribution fitting image with a regular shape in a target power station is obtained by fitting, specifically a gray image, wherein the gray value of each pixel point can reflect dust parameters on the photovoltaic panels in the target power station.
Based on the generation countermeasure network in the deep learning, a dust distribution map generator with K-1 layer dust distribution map generation channels is obtained through training, K-1 layer image countermeasure generation is carried out on the dust distribution fitting image through the K-1 layer dust distribution map generation channels, and the generated fine dust distribution image with dust parameters of all areas in the target power station is obtained, wherein K is an integer larger than 1, preferably 10.
The step of the method provided by the embodiment of the application is based on deep learning, training and obtaining a dust distribution map generator with a K-1 layer dust distribution map generation channel, and comprises the following steps:
obtaining K sample dust distribution image sets according to dust monitoring data in the target power station, wherein the sample dust distribution images in the K sample dust distribution image sets are different in size;
constructing K-1 layer dust distribution map generation channels based on a countermeasure generation network, wherein each layer of dust distribution map generation channel comprises a generator and a discriminator, the sizes of an input image and an output image of the K-1 layer dust distribution map generation channel are distributed in a pyramid shape, the size of the input image of the first layer of dust distribution map generation channel is identical to that of the dust distribution fitting image, and the size of the output image of the K-1 layer dust distribution map generation channel is identical to that of the dust distribution fitting image formed by fitting dust parameters of all photovoltaic panels in the target power station;
Performing supervision training on a first layer dust distribution map generation channel by adopting a first sample dust distribution image set and a second sample dust distribution image set in the K sample dust distribution image sets until convergence, wherein the input first sample dust distribution image is added into a random noise input generator, and the generated dust distribution image generated by the generator is judged by a discriminator;
after the first layer dust distribution map generation channel training is completed, training is continued on other K-2 layer dust distribution map generation channels until all the channels are converged.
In the embodiment of the application, K sample dust distribution image sets are acquired according to dust monitoring data in historical time of a target power station, and the sample dust distribution images in the K sample dust distribution image sets are different in size.
The sample dust distribution images in each sample dust distribution image set are fitted based on the method in the previous description, and the fitting is constructed based on dust parameters of different numbers of partial photovoltaic panels in the target power station respectively.
Illustratively, the K sample dust distribution image sets are respectively based on dust parameters collected by photovoltaic panels with distribution areas 1/10, 2/10, 3/10 and … times the area of the target power station, and the distribution areas are respectively rectangular in the target power station. The sample dust distribution images within the K sample dust distribution image sets may be sized to form an upscaled pattern of image pyramid distributions. Optionally, the sample dust distribution image may be obtained by fitting in combination with the interpolation method described above.
Alternatively, dust parameters of all photovoltaic panels in the target power station can be respectively collected at a plurality of time points in the history time, K sample dust distribution images are respectively obtained by fitting, the K sample dust distribution images form image pyramid distribution, and the larger sample dust distribution image comprises the smaller sample dust distribution image. After the dust distribution map generator is trained and acquired, dust parameters of all photovoltaic panels do not need to be continuously acquired.
Based on the countermeasure generation network, a K-1 layer dust distribution diagram generation channel, preferably a 9 layer dust distribution diagram generation channel, is constructed, wherein each layer dust distribution diagram generation channel comprises a generator and a discriminator, the sizes of an input image and an output image of the K-1 layer dust distribution diagram generation channel are distributed in a pyramid shape, the size of the output image of the first layer dust distribution diagram generation channel is identical to that of a dust distribution fitting image, namely, a first sample dust distribution image, the size of the output image is identical to that of a second sample dust distribution image, the size of the input image of the K-1 layer dust distribution diagram generation channel is identical to that of a ninth sample dust distribution image, the size of the output image is identical to that of a dust distribution fitting image formed by fitting dust parameters of all photovoltaic panels in a target power station, namely, a tenth dust distribution fitting image is 10 times the size of the dust distribution fitting image.
The generator in each layer of dust distribution map generation channel comprises a plurality of convolution blocks consisting of convolution kernels, normalization layers and nonlinear variation layers, and the discriminator consists of a plurality of convolution blocks comprising convolution, normalization and nonlinear activation functions, such as sigmoid functions, which can be constructed by those skilled in the art based on the generation of the countermeasure network architecture in the prior art.
Thus, a K-1 layer dust distribution diagram generation channel with a cascade structure is constructed and obtained, and K sample dust distribution image sets are respectively adopted to train the K-1 layer dust distribution diagram generation channel.
The first layer dust distribution diagram generation channel in the K-1 layer dust distribution diagram generation channel is subjected to supervision training by adopting a first sample dust distribution diagram set and a second sample dust distribution diagram set in the K sample dust distribution diagram sets until convergence, wherein the training process is a training process of generating an antagonistic network.
Specifically, the input first sample dust distribution image is added into a generator in a first layer dust distribution map generation channel through random noise input, a generated dust distribution image is obtained through random generation, the generated dust distribution image generated by the generator is input into a discriminator through combining with a second sample dust distribution image comprising the first sample dust distribution image, discrimination is performed through the discriminator, the output of the discriminator is as close to 1 as possible for the generated dust distribution image close to the second sample dust distribution image, and the output of the discriminator is as close to 0 as possible for the generated dust distribution image not close to the second sample dust distribution image.
Thus, training is carried out for a plurality of times until the accuracy of the first layer dust distribution map generation channel meets the requirement, for example, the average value of the discrimination results of the generated dust distribution image generated by the generator in the discriminator reaches 0.9, and then the training is finished.
Based on the same approach, the second and third sample dust distribution image sets may be used to train the second layer dust profile generation channel and train the other K-3 layer dust profile generation channels until all converge.
The dust profile generator is obtained based on the converging K-1 layer dust profile generation channel.
Based on the constructed dust distribution map generator, the dust distribution fitting image can be input into a first layer dust distribution map generation channel for image generation, a generated dust distribution image with the size being 2/10 of the dust parameter fitting image of all photovoltaic panels of the target power station is obtained, the dust distribution map generation image is continuously input into a second layer dust distribution map generation image, a generated dust distribution image with the size being 3/10 of the dust parameter fitting image of all photovoltaic panels of the target power station can be obtained, and finally, based on a k-1 layer dust distribution map generation channel, a final generated dust distribution image, namely a fine dust distribution image, is generated, and the fine dust distribution image is a dust distribution image generated by multistage cascade connection of the dust distribution map generator according to the dust distribution fitting image, and can reflect the dust distribution level on all photovoltaic panels in the current target power station.
Alternatively, the K value may be set to 2, and a dust loss identifier including only one layer of dust distribution map generation channels is constructed, and image countermeasure generation is directly performed on the basis of the inputted dust distribution fitting image, to obtain the fine dust distribution image. However, since more data noise is added to generate a fine dust distribution image from a dust distribution fitted image, the accuracy of the dust loss identifier may be low or a problem of slow convergence may occur.
Based on the fine dust distribution image, photovoltaic power generation loss due to dust shielding in the current target power station is performed.
In the embodiment of the application, the image feature extraction processing identification is carried out on the fine dust distribution image by training the dust loss identifier, the feature reflected by the gray value of each pixel point in the fine dust distribution image is identified, and the power generation loss of the target power station due to the shielding of the dust by the photovoltaic panel under the current dust parameter is obtained.
The method provided by the embodiment of the application further comprises the following steps:
acquiring a fine dust distribution image set of a sample according to dust monitoring data in the target power station;
calculating and acquiring a sample dust loss coefficient set based on power generation data of a target power station under each sample fine dust distribution image;
Constructing a dust loss identifier based on a deep convolution network;
and training the dust loss identifier by adopting the sample fine dust distribution image set and the sample dust loss coefficient set until meeting convergence conditions.
In the embodiment of the application, based on the method disclosed in the foregoing, dust parameters of a plurality of time points of a target power station can be acquired, a dust distribution fitting image is acquired by fitting, and a generated fine dust distribution image is acquired as a sample fine dust distribution image set.
And the simulation of the photovoltaic power generation test can be carried out based on dust parameters at a plurality of time points, the deviation of the generated energy under shielding and the generated energy under an ideal state is obtained, and then the proportion of the photovoltaic power generation loss of the target power station under different dust parameters is calculated and obtained to be used as a sample dust loss coefficient set.
According to the embodiment of the application, the fine dust distribution image is identified through the deep convolution network, and the power generation loss proportion of the target power station under the dust deposition state of the target power station reflected in the fine dust distribution image is output.
Based on the deep convolutional neural network, the dust loss identifier is constructed, wherein the dust loss identifier comprises a plurality of convolutional layers, a pooling layer and a full-connection layer, and convolution characteristic extraction can be carried out on an input fine dust distribution image so as to obtain a corresponding dust loss coefficient through processing.
And training the dust loss identifier by adopting the sample fine dust distribution image set and the sample dust loss coefficient set as training data until meeting convergence conditions. Illustratively, the convergence condition may be that the dust loss coefficient output by the dust loss identifier does not deviate by more than 5% from the sample dust loss coefficient.
And carrying out image processing identification on the fine dust distribution image obtained by the current processing based on the trained dust loss identifier, and obtaining a dust loss coefficient of the generated energy loss caused by dust shielding in the current target power station.
According to the embodiment of the application, the dust parameters are collected on the partial photovoltaic panels in the target power station and used as gray values in the pixel points, the dust distribution fitting image is constructed, the fine dust distribution fitting image comprising the gray values converted by the dust parameters of all the photovoltaic panels is generated through cascading generation of the countermeasure network, the dust loss coefficient is identified, the efficiency and the accuracy of analyzing the photovoltaic power generation loss based on the dust parameters can be improved, only the dust parameters of the partial photovoltaic panels are required to be collected, and the cost of analyzing the power generation loss can be reduced for a large-scale photovoltaic power station.
S106: and calculating and obtaining a power generation loss coefficient according to the shielding loss coefficient, the meteorological loss coefficient and the dust loss coefficient, and calculating and obtaining the power generation loss.
And calculating the total power generation loss coefficient of the target power station based on the shielding loss coefficient, the weather loss coefficient and the dust loss coefficient of the power generation loss of the target power station, which are obtained by analysis in the previous step, due to shielding of external objects, weather change and dust deposition.
For example, the sum of the shielding loss coefficient, the weather loss coefficient, and the dust loss coefficient may be calculated as the power generation loss coefficient. Optionally, the shielding loss coefficient, the meteorological loss coefficient and the dust loss coefficient may overlap partially, and the calculated power generation loss coefficient may be corrected and calculated based on experience of photovoltaic power generation loss analysis in the prior art.
Based on the power generation loss coefficient, the power generation loss quantity predicted and analyzed by the target power station is calculated and obtained by combining the ideal photovoltaic power generation quantity of the target power station, and the calculated power generation loss quantity is used as an analysis result of photovoltaic power generation loss and can be used as reference data for operating and maintaining the photovoltaic power station.
According to the embodiment of the application, through the technical scheme, at least the following technical effects are achieved:
According to the embodiment of the application, the current time information is collected in the photovoltaic power station, the power generation loss caused by shielding of surrounding buildings or other objects to the photovoltaic power station is analyzed based on the time, the power generation loss caused by the current weather change is collected and analyzed according to the weather data, the influence parameters on dust deposition in the photovoltaic power station are analyzed according to the weather data, the dust parameters of part of the collected and obtained photovoltaic panels are corrected and calculated to obtain a dust parameter array, a dust distribution fitting image is constructed, a fine dust distribution fitting image comprising gray values converted by dust parameters of all the photovoltaic panels is generated through cascade generation of an countermeasure network, the dust loss coefficient is identified, the power generation loss coefficient of the power station due to dust is obtained, the power generation loss in three aspects is synthesized, and the overall power generation loss in the photovoltaic power station is calculated and obtained. According to the technical scheme, the analysis of the power generation loss is carried out through multiple dimensions, the comprehensiveness of the analysis of the power generation loss of the photovoltaic power can be improved, the influence parameters of the meteorological data analysis on dust deposition can be improved, the accuracy of the dust influence analysis with the largest influence on the power generation loss can be improved, the dust parameters of part of photovoltaic panels are collected to form dust distribution images in a fitting mode, the accuracy and the reliability of the power generation loss through the dust analysis are improved through cascading image analysis, the dust data of all photovoltaic panels are not required to be collected, the cost of the monitoring analysis of the power generation loss is reduced, and the technical effects of improving the comprehensiveness and the accuracy of the power generation loss analysis of a photovoltaic power station are achieved.
Example two
Based on the same inventive concept as one of the aforementioned embodiments of a power generation loss analysis method for a photovoltaic power plant, as shown in fig. 3, the present application provides a power generation loss analysis system for a photovoltaic power plant, the system being connected to a power generation loss analysis apparatus for a photovoltaic power plant, the apparatus including an environmental analysis station, a weather analysis station, a dust analysis station, a comprehensive loss calculation module, the system comprising:
the shielding loss analysis module 201 is configured to collect current time information through an environmental analysis station, identify the time information based on a shielding identifier trained by detection data of a target power station, and obtain a shielding loss coefficient, where the target power station is a photovoltaic power station;
the weather data acquisition module 202 is configured to acquire a weather data sequence within a future preset time range through a weather analysis workstation;
the weather loss analysis module 203 is configured to perform power generation loss identification on the weather data sequence, obtain a weather loss coefficient, and perform dust influence parameter analysis on the gas phase data sequence to obtain a dust influence parameter;
the dust parameter processing module 204 is configured to collect, by using a dust analysis station, a plurality of dust parameters of a part of photovoltaic panels in all photovoltaic panels in a target power station, and calculate and obtain a dust parameter array in combination with the dust influence parameters;
The dust loss analysis module 205 is configured to construct a dust distribution fitting image according to the dust parameter array, perform cascade image generation analysis, and obtain a dust loss coefficient of the target power station;
the power generation loss calculation module 206 is configured to calculate and obtain a power generation loss coefficient according to the shielding loss coefficient, the weather loss coefficient and the dust loss coefficient, and calculate and obtain a power generation loss amount.
Further, the occlusion loss analysis module 201 is further configured to perform the following steps:
acquiring the lost power generation capacity of all the photovoltaic panels in the target power station in a plurality of time periods according to the monitoring data of the target power station, and acquiring a sample time information record and a sample shielding loss coefficient record;
the sample time information record and the sample shielding loss coefficient record are adopted as training data, and a shielding identifier meeting convergence requirements is trained based on machine learning;
and identifying the time information by adopting the shielding identifier to obtain a shielding loss coefficient.
Further, the weather loss analysis module 203 is further configured to perform the following steps:
acquiring a plurality of sample meteorological data sequences and a plurality of sample meteorological loss coefficients according to the monitoring data of the target power station;
Adopting a plurality of sample meteorological data sequences and a plurality of sample meteorological loss coefficients as training data, and training a meteorological loss identifier meeting convergence requirements based on machine learning;
and identifying the meteorological data sequence by adopting a meteorological loss identifier to obtain the meteorological loss coefficient.
Wherein, still include:
according to the dust monitoring data of the target power station, the variation of dust amount under different meteorological data sequences is obtained, and a plurality of sample dust influence parameters are obtained through calculation;
based on a decision tree, adopting meteorological data as decision input, adopting dust influence parameters as decision data, and constructing a dust influence classifier according to a plurality of sample meteorological data sequences and a plurality of sample dust influence parameters;
and adopting a dust influence classifier to carry out decision classification on the meteorological data sequence, and obtaining the dust influence parameters.
Further, the dust loss analysis module 205 is further configured to perform the following steps:
acquiring coordinate parameters of a plurality of photovoltaic panels in the target power station, and acquiring a dust distribution fitting image in the target power station by combining the dust parameter array;
training to obtain a dust distribution map generator with K-1 layers of dust distribution map generation channels based on deep learning, wherein K is an integer greater than 1;
Processing the dust distribution fitting image by adopting a dust distribution map generator to obtain a generated fine dust distribution image;
training to obtain a dust loss identifier;
and identifying the fine dust distribution image by adopting the dust loss identifier to obtain the dust loss coefficient.
Wherein, based on the deep learning, training acquires the dust distribution map generator with K-1 layer dust distribution map generation channels, comprising:
obtaining K sample dust distribution image sets according to dust monitoring data in the target power station, wherein the sample dust distribution images in the K sample dust distribution image sets are different in size;
constructing K-1 layer dust distribution map generation channels based on a countermeasure generation network, wherein each layer of dust distribution map generation channel comprises a generator and a discriminator, the sizes of an input image and an output image of the K-1 layer dust distribution map generation channel are distributed in a pyramid shape, the size of the input image of the first layer of dust distribution map generation channel is identical to that of the dust distribution fitting image, and the size of the output image of the K-1 layer dust distribution map generation channel is identical to that of the dust distribution fitting image formed by fitting dust parameters of all photovoltaic panels in the target power station;
Performing supervision training on a first layer dust distribution map generation channel by adopting a first sample dust distribution image set and a second sample dust distribution image set in the K sample dust distribution image sets until convergence, wherein the input first sample dust distribution image is added into a random noise input generator, and the generated dust distribution image generated by the generator is judged by a discriminator;
after the first layer dust distribution map generation channel training is completed, training is continued on other K-2 layer dust distribution map generation channels until all the channels are converged.
Wherein, training acquires dust loss recognizer, includes:
acquiring a fine dust distribution image set of a sample according to dust monitoring data in the target power station;
calculating and acquiring a sample dust loss coefficient set based on power generation data of a target power station under each sample fine dust distribution image;
constructing a dust loss identifier based on a deep convolution network;
and training the dust loss identifier by adopting the sample fine dust distribution image set and the sample dust loss coefficient set until meeting convergence conditions.
From the foregoing detailed description of a method for analyzing power generation loss of a photovoltaic power station, those skilled in the art can clearly understand that the method and the system for analyzing power generation loss of a photovoltaic power station in this embodiment, for the device disclosed in the embodiment, since the device corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
For a specific embodiment of a power generation loss analysis system for a photovoltaic power station, reference may be made to the above embodiment of a power generation loss analysis method for a photovoltaic power station, and the description thereof will not be repeated. Each of the above-described modules in a power generation loss analysis device for a photovoltaic power plant may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
Example III
As shown in fig. 4, a computer device is provided, comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the method in embodiment one when the computer program is executed.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 4. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing news data, time attenuation factors and other data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of power loss analysis for a photovoltaic power plant.
It will be appreciated by persons skilled in the art that the architecture shown in fig. 4 is merely a block diagram of some of the architecture relevant to the present inventive arrangements and is not limiting as to the computer device to which the present inventive arrangements are applicable, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
Example IV
The present embodiment provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method of the first embodiment.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (10)
1. A method for analyzing power generation loss of a photovoltaic power station, wherein the method is applied to a power generation loss analysis device constructed in the photovoltaic power station, the device comprises an environment analysis station, a weather analysis station, a dust analysis station and a comprehensive loss calculation module, and the method comprises the following steps:
acquiring current time information through an environmental analysis workstation, and identifying the time information based on a shielding identifier trained by target power station detection data to obtain a shielding loss coefficient, wherein the target power station is a photovoltaic power station;
collecting a meteorological data sequence within a future preset time range through a meteorological analysis workstation;
carrying out power generation loss identification on the meteorological data sequence, obtaining a meteorological loss coefficient, and carrying out dust influence parameter analysis on the gas phase data sequence to obtain dust influence parameters;
collecting a plurality of dust parameters of partial photovoltaic panels in all photovoltaic panels in a target power station through a dust analysis station, and calculating to obtain a dust parameter array by combining the dust influence parameters;
constructing a dust distribution fitting image according to the dust parameter array, and performing cascading image generation analysis to acquire a dust loss coefficient of the target power station;
And calculating and obtaining a power generation loss coefficient according to the shielding loss coefficient, the meteorological loss coefficient and the dust loss coefficient, and calculating and obtaining the power generation loss.
2. The method according to claim 1, characterized in that the method comprises:
acquiring the lost power generation capacity of all the photovoltaic panels in the target power station in a plurality of time periods according to the monitoring data of the target power station, and acquiring a sample time information record and a sample shielding loss coefficient record;
the sample time information record and the sample shielding loss coefficient record are adopted as training data, and a shielding identifier meeting convergence requirements is trained based on machine learning;
and identifying the time information by adopting the shielding identifier to obtain a shielding loss coefficient.
3. The method according to claim 1, characterized in that the method comprises:
acquiring a plurality of sample meteorological data sequences and a plurality of sample meteorological loss coefficients according to the monitoring data of the target power station;
adopting a plurality of sample meteorological data sequences and a plurality of sample meteorological loss coefficients as training data, and training a meteorological loss identifier meeting convergence requirements based on machine learning;
and identifying the meteorological data sequence by adopting a meteorological loss identifier to obtain the meteorological loss coefficient.
4. A method according to claim 3, characterized in that the method comprises:
according to the dust monitoring data of the target power station, the variation of dust amount under different meteorological data sequences is obtained, and a plurality of sample dust influence parameters are obtained through calculation;
based on a decision tree, adopting meteorological data as decision input, adopting dust influence parameters as decision data, and constructing a dust influence classifier according to a plurality of sample meteorological data sequences and a plurality of sample dust influence parameters;
and adopting a dust influence classifier to carry out decision classification on the meteorological data sequence, and obtaining the dust influence parameters.
5. The method according to claim 1, characterized in that the method comprises:
acquiring coordinate parameters of a plurality of photovoltaic panels in the target power station, and acquiring a dust distribution fitting image in the target power station by combining the dust parameter array;
training to obtain a dust distribution map generator with K-1 layers of dust distribution map generation channels based on deep learning, wherein K is an integer greater than 1;
processing the dust distribution fitting image by adopting a dust distribution map generator to obtain a generated fine dust distribution image;
training to obtain a dust loss identifier;
And identifying the fine dust distribution image by adopting the dust loss identifier to obtain the dust loss coefficient.
6. The method according to claim 5, characterized in that the method comprises:
obtaining K sample dust distribution image sets according to dust monitoring data in the target power station, wherein the sample dust distribution images in the K sample dust distribution image sets are different in size;
constructing K-1 layer dust distribution map generation channels based on a countermeasure generation network, wherein each layer of dust distribution map generation channel comprises a generator and a discriminator, the sizes of an input image and an output image of the K-1 layer dust distribution map generation channel are distributed in a pyramid shape, the size of the input image of the first layer of dust distribution map generation channel is identical to that of the dust distribution fitting image, and the size of the output image of the K-1 layer dust distribution map generation channel is identical to that of the dust distribution fitting image formed by fitting dust parameters of all photovoltaic panels in the target power station;
performing supervision training on a first layer dust distribution map generation channel by adopting a first sample dust distribution image set and a second sample dust distribution image set in the K sample dust distribution image sets until convergence, wherein the input first sample dust distribution image is added into a random noise input generator, and the generated dust distribution image generated by the generator is judged by a discriminator;
After the first layer dust distribution map generation channel training is completed, training is continued on other K-2 layer dust distribution map generation channels until all the channels are converged.
7. The method according to claim 6, characterized in that the method comprises:
acquiring a fine dust distribution image set of a sample according to dust monitoring data in the target power station;
calculating and acquiring a sample dust loss coefficient set based on power generation data of a target power station under each sample fine dust distribution image;
constructing a dust loss identifier based on a deep convolution network;
and training the dust loss identifier by adopting the sample fine dust distribution image set and the sample dust loss coefficient set until meeting convergence conditions.
8. A power generation loss analysis system for a photovoltaic power plant, the system being connected to a power generation loss analysis device for the photovoltaic power plant, the device comprising an environmental analysis station, a weather analysis station, a dust analysis station, and a comprehensive loss calculation module, the system comprising:
the shielding loss analysis module is used for acquiring current time information through an environmental analysis station, identifying the time information based on a shielding identifier trained by detection data of a target power station, and obtaining a shielding loss coefficient, wherein the target power station is a photovoltaic power station;
The weather data acquisition module is used for acquiring weather data sequences within a future preset time range through a weather analysis workstation;
the weather loss analysis module is used for carrying out power generation loss identification on the weather data sequence, obtaining a weather loss coefficient, and carrying out dust influence parameter analysis on the gas phase data sequence to obtain dust influence parameters;
the dust parameter processing module is used for collecting a plurality of dust parameters of partial photovoltaic panels in all photovoltaic panels in the target power station through the dust analysis station, and calculating and obtaining a dust parameter array by combining the dust influence parameters;
the dust loss analysis module is used for constructing a dust distribution fitting image according to the dust parameter array, generating and analyzing a cascade image, and acquiring a dust loss coefficient of the target power station;
and the power generation loss calculation module is used for calculating and obtaining a power generation loss coefficient according to the shielding loss coefficient, the meteorological loss coefficient and the dust loss coefficient and calculating and obtaining the power generation loss amount.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311173599.4A CN117218425A (en) | 2023-09-12 | 2023-09-12 | Power generation loss analysis method and system for photovoltaic power station |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311173599.4A CN117218425A (en) | 2023-09-12 | 2023-09-12 | Power generation loss analysis method and system for photovoltaic power station |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117218425A true CN117218425A (en) | 2023-12-12 |
Family
ID=89041940
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311173599.4A Pending CN117218425A (en) | 2023-09-12 | 2023-09-12 | Power generation loss analysis method and system for photovoltaic power station |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117218425A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117574176A (en) * | 2024-01-12 | 2024-02-20 | 江苏无双新能源科技有限公司 | BIPV photovoltaic glass production process optimization method |
| CN118017927A (en) * | 2024-04-10 | 2024-05-10 | 江苏顶晶融合电力工程有限公司 | Automatic cleaning method and system for photovoltaic panel |
| CN118508438A (en) * | 2024-07-17 | 2024-08-16 | 宁波欧达光电有限公司 | Photovoltaic module power loss prediction method, system, equipment and storage medium |
-
2023
- 2023-09-12 CN CN202311173599.4A patent/CN117218425A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117574176A (en) * | 2024-01-12 | 2024-02-20 | 江苏无双新能源科技有限公司 | BIPV photovoltaic glass production process optimization method |
| CN117574176B (en) * | 2024-01-12 | 2024-04-09 | 江苏无双新能源科技有限公司 | BIPV photovoltaic glass production process optimization method |
| CN118017927A (en) * | 2024-04-10 | 2024-05-10 | 江苏顶晶融合电力工程有限公司 | Automatic cleaning method and system for photovoltaic panel |
| CN118017927B (en) * | 2024-04-10 | 2024-06-11 | 江苏顶晶融合电力工程有限公司 | Automatic cleaning method and system for photovoltaic panel |
| CN118508438A (en) * | 2024-07-17 | 2024-08-16 | 宁波欧达光电有限公司 | Photovoltaic module power loss prediction method, system, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6968378B2 (en) | Power prediction method and system of photovoltaic power plant based on operation data of grid interconnection inverter | |
| CN110070226B (en) | Photovoltaic power prediction method and system based on convolutional neural network and meta-learning | |
| CN109842373B (en) | Photovoltaic array fault diagnosis method and device based on space-time distribution characteristics | |
| Raza et al. | An ensemble framework for day-ahead forecast of PV output power in smart grids | |
| CN109873610B (en) | Photovoltaic array fault diagnosis method based on IV characteristic and depth residual error network | |
| CN117218425A (en) | Power generation loss analysis method and system for photovoltaic power station | |
| CN110556820B (en) | Method and device for determining an energy system operating scenario | |
| CN113988477A (en) | Photovoltaic power short-term prediction method and device based on machine learning and storage medium | |
| CN116011608B (en) | Photovoltaic power ultra-short-term prediction method and device, electronic equipment and storage medium | |
| CN111815027A (en) | Photovoltaic station generated power prediction method and system | |
| CN111754026A (en) | Photovoltaic power station group power prediction method and device, computer equipment and storage medium | |
| CN112653198A (en) | Wind power output scene generation method and system based on prediction box | |
| CN113627674A (en) | Distributed photovoltaic power station output prediction method and device and storage medium | |
| CN111242355A (en) | Photovoltaic probability prediction method and system based on Bayesian neural network | |
| CN117540859B (en) | Method and system for estimating generated energy of photovoltaic power station based on neural network | |
| CN115936177A (en) | Photovoltaic output power prediction method and system based on neural network | |
| CN117293809A (en) | Multi-time space scale new energy generation power prediction method based on large model | |
| Fan et al. | Spatial-temporal genetic-based attention networks for short-term photovoltaic power forecasting | |
| CN115618922A (en) | Photovoltaic power prediction method and device, photovoltaic power generation system and storage medium | |
| Buonanno et al. | Comprehensive method for modeling uncertainties of solar irradiance for PV power generation in smart grids | |
| CN114386666A (en) | Wind power plant short-term wind speed prediction method based on space-time correlation | |
| Al-Hilfi et al. | Estimating generated power of photovoltaic systems during cloudy days using gene expression programming | |
| CN111950752A (en) | Photovoltaic power station generating capacity prediction method, device and system and storage medium thereof | |
| CN111126645A (en) | Wind power prediction algorithm based on data mining technology and improved support vector machine | |
| CN117236720A (en) | Photovoltaic power station generated power prediction method utilizing multi-meteorological factor characteristics |
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 |