CN113890027B - Power generation power prediction system and method based on equipment operation state and meteorological parameters - Google Patents
Power generation power prediction system and method based on equipment operation state and meteorological parameters Download PDFInfo
- Publication number
- CN113890027B CN113890027B CN202111214665.9A CN202111214665A CN113890027B CN 113890027 B CN113890027 B CN 113890027B CN 202111214665 A CN202111214665 A CN 202111214665A CN 113890027 B CN113890027 B CN 113890027B
- Authority
- CN
- China
- Prior art keywords
- power
- loss rate
- inverter
- transformer
- photovoltaic module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000010248 power generation Methods 0.000 title claims abstract description 79
- 238000000034 method Methods 0.000 title claims abstract description 11
- 230000007613 environmental effect Effects 0.000 claims abstract description 29
- 238000010191 image analysis Methods 0.000 claims abstract description 20
- 238000001931 thermography Methods 0.000 claims abstract description 16
- 238000003062 neural network model Methods 0.000 claims description 14
- 238000006243 chemical reaction Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000008021 deposition Effects 0.000 claims description 2
- 238000009825 accumulation Methods 0.000 abstract description 13
- 239000000428 dust Substances 0.000 abstract description 11
- 238000012549 training Methods 0.000 description 9
- 239000012080 ambient air Substances 0.000 description 8
- 238000013527 convolutional neural network Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 238000004140 cleaning Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000005022 packaging material Substances 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/381—Dispersed generators
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/22—The renewable source being solar energy
- H02J2300/24—The renewable source being solar energy of photovoltaic origin
- H02J2300/26—The renewable source being solar energy of photovoltaic origin involving maximum power point tracking control for photovoltaic sources
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/50—Photovoltaic [PV] energy
- Y02E10/56—Power conversion systems, e.g. maximum power point trackers
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application relates to the technical field of power generation power prediction of photovoltaic power stations, in particular to a power generation power prediction system and method based on equipment operation states and meteorological parameters. The power generation prediction system based on the equipment running state and the meteorological parameters comprises a plurality of visible light-infrared thermal imaging integrated cameras which are used for shooting a photovoltaic module to obtain an accumulated ash visible light image and a hot spot infrared thermal image; and the image analysis modules are connected with the visible light-infrared thermal imaging integrated cameras in a one-to-one correspondence manner and are used for analyzing the gray-scale visible light image and the hot spot infrared thermal image so as to obtain the power generation loss rate caused by the photovoltaic module. According to the application, the power generation power of the photovoltaic power station is predicted jointly by combining the dust accumulation and hot spot conditions of the photovoltaic module, the maximum power adjustment real-time parameters of the inverter of the photovoltaic module and the efficiency parameters of the transformer on the basis of the environmental meteorological parameters, so that the accuracy and the reliability of the power generation power prediction of the photovoltaic power station are effectively improved.
Description
Technical Field
The invention relates to the technical field of power generation power prediction of photovoltaic power stations, in particular to a power generation power prediction system and method based on equipment operation states and meteorological parameters.
Background
With the large-scale and high-proportion development of wind power and photovoltaic energy sources in China, great challenges are brought to power grid balance and safe, economic and stable operation, and higher technical requirements are provided for the prediction precision and accuracy of the wind-solar power prediction system.
The influencing factors of the generated power of the photovoltaic system comprise meteorological parameters (such as irradiance, ambient air temperature, wind speed and the like), the running state of the photovoltaic module (module dust accumulation, hot spots and the like), the regulating state of the inverter (the inverter regulates the maximum power output according to a control strategy), the efficiency of the transformer and the like. In the running state of the photovoltaic module, the existence of the accumulated ash can greatly influence the absorption of the photovoltaic system to solar irradiation, if the accumulated ash cannot be cleaned in time, the generation efficiency of the photovoltaic module is further reduced, the highest generated power of the module can be attenuated by 40% -60%, and the generated energy is reduced by 20% -30%. It is counted that the direct economic loss of a global solar power station per year due to ineffective cleaning of the photovoltaic module surface or untimely cleaning thereof is up to about $100 billion, and by 2020, this loss is expected to be 1412 billion yuan. In addition, the photovoltaic hot spots can damage packaging materials of photovoltaic cells or components, and normal operation of the photovoltaic array and safety of a photovoltaic power station are affected.
However, the current photovoltaic power station power generation power prediction system mainly considers the influence of meteorological parameters, cloud layers and the like, and less considers the running state of a photovoltaic module and the real-time state of inverter power adjustment, so that the error of the prediction result of the photovoltaic power station power generation power is larger, and the technical requirements of grid connection and scheduling are difficult to meet. For example, patent application No. cn20181036781. X discloses a combined evaluation method and apparatus of a photovoltaic power generation power prediction method, which predicts power generation power according to environmental parameters without considering the operation states of a photovoltaic module, an inverter and a transformer, resulting in inaccurate prediction results of power generation power. For another example, patent application CN201310430694.8 discloses a method and system for predicting power generated by photovoltaic power generation, which predicts power generated according to solar radiation intensity and temperature, and also does not consider the operation states of the photovoltaic module, the inverter and the transformer, resulting in large error in the prediction result of power generated.
Therefore, the prediction precision and accuracy of the generated power of the photovoltaic power station can be ensured only by comprehensively considering the running state parameters of the photovoltaic module, the inverter, the transformer and other equipment and the environmental meteorological parameters.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a system and a method for predicting the generated power based on the equipment running state and the meteorological parameters, which can effectively improve the accuracy and the reliability of the prediction of the generated power of a photovoltaic power station.
The technical scheme adopted for solving the technical problems is as follows: a power generation prediction system based on equipment operation state and meteorological parameters comprises
The plurality of visible light-infrared thermal imaging integrated cameras are used for shooting the photovoltaic module to obtain an ash deposition visible light image and a hot spot infrared thermal image;
The image analysis modules are connected with the visible light-infrared thermal imaging integrated cameras in a one-to-one correspondence manner and are used for analyzing the gray-scale visible light image and the hot spot infrared thermal image so as to obtain the power generation loss rate caused by the photovoltaic module;
The inverter is connected with the photovoltaic module; the inverter parameter collector is connected with the inverter and is used for obtaining inverter power adjustment real-time parameters of the inverter;
The transformer and the transformer parameter collector are connected with the inverter; the transformer parameter collector is connected with the transformer and is used for obtaining transformer efficiency parameters of the transformer;
The weather station is used for acquiring environmental weather parameters;
The data acquisition card is provided with a first data input end connected with the image analysis module, a second data input end connected with the inverter parameter acquisition device, a third data input end connected with the transformer parameter acquisition device and a fourth data input end connected with the weather station, and is used for acquiring the generation power loss rate caused by the photovoltaic module, the inverter power regulation real-time parameter, the transformer efficiency parameter and the environmental weather parameter;
And the industrial control computer is connected with the data output end of the data acquisition card, and predicts the generated power of the photovoltaic power station through the generated power loss rate, the inverter power adjustment real-time parameter, the transformer efficiency parameter and the environmental meteorological parameter caused by the photovoltaic module.
Preferably, the image analysis module analyzes the gray-scale visible light image and the hot spot infrared thermal image through ResNet depth convolution neural network model to obtain the power generation loss rate caused by the photovoltaic module.
Preferably, the ambient weather parameters include solar irradiance, ambient air temperature, and wind speed.
Preferably, the industrial control computer is provided with
The power generation loss rate calculation unit calculates the power generation loss rate caused by inversion and boosting through the inverter power adjustment real-time parameter and the transformer efficiency parameter; specifically, the inverter power is adjusted to obtain the corresponding conversion efficiency a% by real-time parameters, and the transformer efficiency parameter is used to obtain the corresponding conversion efficiency b%, so that the generated power loss rate caused by inversion boosting is 1-a% -b%.
Preferably, the industrial control computer is provided with
The power generation power prediction unit calculates the power generation power of the photovoltaic power station by combining the environmental meteorological parameters, the power generation power loss rate caused by inversion boosting and the power generation power loss rate caused by the photovoltaic module through the deep belief network DBN model.
The power generation power prediction method based on the equipment operation state and the meteorological parameters comprises the following steps of
S1, acquiring a generated power loss rate caused by a photovoltaic module;
s2, acquiring the loss rate of the generated power caused by inversion and boosting of an inverter and a transformer;
s3, acquiring environmental meteorological parameters;
And S4, calculating the power generation of the photovoltaic power station by using the environmental meteorological parameters, the power generation loss rate caused by the photovoltaic module and the power generation loss rate caused by inversion boosting.
Preferably, the S1 specifically comprises
S11, shooting a photovoltaic module through a visible light-infrared thermal imaging integrated camera to obtain an accumulated gray visible light image and a hot spot infrared thermal image;
S12, analyzing the gray visible light image and the hot spot infrared thermal image through a ResNet depth convolution neural network model in an image analysis module to obtain the generation power loss rate caused by the photovoltaic module;
S13, transmitting the generated power loss rate caused by the photovoltaic module to an industrial control computer in real time through a data acquisition card.
Preferably, the S2 specifically comprises
S21, transmitting the inverter power adjustment real-time parameters acquired by the inverter parameter acquisition unit to an industrial control computer through a data acquisition card;
S22, transmitting the transformer efficiency parameters acquired by the transformer parameter acquisition unit to an industrial control computer through a data acquisition card;
S23, calculating the generated power loss rate caused by inversion boosting through a generated power loss rate calculating unit caused by inversion boosting and using an inverter power adjustment real-time parameter and a transformer efficiency parameter.
Preferably, in S23, the inverter power is adjusted to obtain the corresponding conversion efficiency a% and the corresponding conversion efficiency b% according to the transformer efficiency parameter, and the generated power loss rate caused by the inversion boosting is 1-a% ×b%.
Preferably, the step S4 specifically includes calculating the generated power of the photovoltaic power station by combining the deep belief network DBN model in the generated power prediction unit with the environmental meteorological parameters, the generated power loss rate caused by inversion and boosting, and the generated power loss rate caused by the photovoltaic module.
Advantageous effects
According to the application, the generated power of the photovoltaic power station is predicted jointly by combining the dust accumulation and hot spot conditions of the photovoltaic module, the maximum power adjustment real-time parameters of the inverter of the photovoltaic module and the efficiency parameters of the transformer on the basis of the environmental meteorological parameters, so that the accuracy and the reliability of the generated power prediction of the photovoltaic power station are effectively improved;
According to the application, the power generation efficiency loss rate corresponding to different gray deposit and hot spot influence degrees of the photovoltaic module is determined through the ResNet depth convolution neural network model and the photovoltaic module detection image, so that the acquisition of current, voltage, power and other operation parameters in the traditional gray deposit and hot spot analysis is avoided, the subjective error of manual observation and judgment of an operator is eliminated, and the intellectualization, the accuracy and the real-time of the recognition and the analysis of the gray deposit and the hot spot states of the photovoltaic module are realized.
Drawings
FIG. 1 is a schematic diagram of a system for predicting generated power for operating conditions and meteorological parameters of the apparatus of the present invention;
FIG. 2 is a flow chart of a method of generating power prediction for plant operating conditions and meteorological parameters of the present invention.
Detailed Description
The technical scheme of the invention is further described below by the specific embodiments with reference to the accompanying drawings.
Embodiment one: as shown in fig. 1 and 2, the power generation prediction system based on the equipment operation state and the meteorological parameters comprises a plurality of visible light-infrared thermal imaging integrated cameras 1, a plurality of image analysis modules 3, an inverter 4, an inverter parameter collector 4-1, a transformer 5, a transformer parameter collector 5-1, a meteorological station 6, a data acquisition card 7 and an industrial control computer 8.
The visible light-infrared thermal imaging integrated camera 1 is provided with a first lens (visible light lens) and a second lens (infrared thermal lens), wherein the first lens is used for shooting a 'dust accumulation' image of the photovoltaic module 2 (namely, obtaining a dust accumulation visible light image), and the second lens is used for shooting a 'hot spot' image of the photovoltaic module 2 (namely, obtaining a hot spot infrared thermal image). Each photovoltaic panel of the photovoltaic module 2 is correspondingly provided with a visible light-infrared thermal imaging integrated camera 1.
The image analysis modules 3 are connected with the visible light-infrared thermal imaging integrated cameras 1 in a one-to-one correspondence manner and are used for analyzing the gray visible light images and the hot spot infrared thermal images of the corresponding photovoltaic panels so as to obtain the power generation loss rate of each photovoltaic panel in the photovoltaic module 2. And the image analysis module 3 analyzes the gray-deposition visible light image and the hot spot infrared thermal image through ResNet depth convolution neural network model so as to obtain the power generation loss rate caused by the photovoltaic module. The gray visible light image and the hot spot infrared thermal image do not need to be processed before being input into the ResNet depth convolution neural network model. The ResNet deep convolutional neural network model integrates the feature extraction operator and the classifier in an end-to-end multi-layer network, breaks through the limitation of the feature extraction technology in the aspects of image recognition and analysis, and does not need to manually design target image features.
Before ResNet deep convolutional neural network model is really used, a batch of gray visible light images and hot spot infrared thermal images are acquired, the images are divided into a plurality of groups according to the gray accumulation degree and the hot spot degree of a photovoltaic module, and parameters such as current, voltage and power of corresponding image groups are acquired at the same time so as to calculate the average value of the loss rate of the power generation efficiency under each group. Then, the images in each group are randomly split into a training set and a testing set in proportion to complete the deep training of ResNet deep convolutional neural network model.
When the image analysis module 3 is used, the gray-deposition visible light image and the hot spot infrared thermal image can be directly read through the ResNet depth convolution neural network model, the ResNet depth convolution neural network model firstly unifies the image format and the size of the gray-deposition visible light image and the hot spot infrared thermal image, then the image features of the gray-deposition visible light image and the hot spot infrared thermal image are extracted through multi-layer convolution and are unfolded in a full connection mode, finally the probability of each grouping of the images is calculated through a classifier, the power generation efficiency loss rate corresponding to the grouping with the highest possibility is output, and then the correlation between the gray-deposition state image, the hot spot state image and the power generation efficiency loss rate is realized.
The image analysis module 3 can directly detect the influence degree of 'dust accumulation', 'hot spots' of the image analysis from the photovoltaic module through training a mature ResNet depth convolutional neural network model to obtain the corresponding loss rate of the power generation efficiency, so that the acquisition of operation parameters such as current, voltage, power and the like in the traditional dust accumulation analysis is avoided, the subjective error of manually observing and judging the dust accumulation degree by an operator is eliminated, and the intelligent, accurate and real-time identification and analysis of the dust accumulation state of the photovoltaic module are realized.
The inverter 4 is connected with the photovoltaic module 2, and the inverter parameter collector 4-1 is connected with the inverter 4 and is used for obtaining the inverter power adjustment real-time parameter of the inverter 4. The transformer 5 is connected with the inverter 4, and the transformer parameter collector 5-1 is connected with the transformer 5 and is used for obtaining transformer efficiency parameters of the transformer 5.
The weather station 6 is used to obtain environmental weather parameters including solar irradiance, ambient air temperature and wind speed.
The data acquisition card 7 is provided with a first data input end connected with the image analysis module 3, a second data input end connected with the inverter parameter collector 4-1, a third data input end connected with the transformer parameter collector 5-1 and a fourth data input end connected with the weather station 6, and is used for acquiring the generation power loss rate caused by the photovoltaic module, the inverter power regulation real-time parameter, the transformer efficiency parameter and the environmental weather parameter.
The industrial computer 8 is connected with the data output end of the data acquisition card 7, and predicts the generated power of the photovoltaic power station through the generated power loss rate, the inverter power adjustment real-time parameter, the transformer efficiency parameter and the environmental meteorological parameter caused by the photovoltaic module. The industrial control computer 8 is provided with a generated power loss rate calculation unit and a generated power prediction unit which are caused by inversion and boosting.
The power generation power loss rate calculation unit caused by inversion boosting obtains corresponding conversion efficiency a% by adjusting real-time parameters through inverter power, wherein a% is 95-98%; the corresponding conversion efficiency b% and b% are obtained through the efficiency parameters of the transformer, and the generated power loss rate caused by inversion boosting is 1-a% and b%. When a% is 97%, b% is 98%, and the generated power loss rate due to inversion boosting is 5%.
The power generation power prediction unit calculates the power generation power of the photovoltaic power station by combining the environmental meteorological parameters, the power generation power loss rate caused by inversion boosting and the power generation power loss rate caused by the photovoltaic module through the deep belief network DBN model. Before the deep belief network DBN model is actually used, enough data sets are required to be acquired, and each data set comprises solar irradiance, ambient air temperature, wind speed, the generated power loss rate of the photovoltaic panel, the generated power loss rate caused by inversion and boosting of the photovoltaic assembly and the actual photovoltaic generated power of the corresponding photovoltaic panel. These data sets are then split proportionally randomly into training and testing sets to complete the deep training of the deep belief network DBN model.
When the power generation power prediction unit is used, real-time solar irradiance, ambient air temperature, wind speed, power generation power loss rate of the photovoltaic panel and power generation power loss rate caused by inversion and boosting of the photovoltaic module are input through the deep belief network DBN model, the power generation power prediction value of the corresponding photovoltaic panel can be obtained, and then the power generation power of the required photovoltaic power station can be obtained by summing all the power generation power prediction values of the photovoltaic panel. According to the application, the environmental meteorological parameters and the equipment operation state parameters of the photovoltaic module, the inverter and the transformer are combined to jointly predict the power generation power of the photovoltaic power station, so that the accuracy and reliability of the power generation power prediction of the photovoltaic power station are effectively improved.
Embodiment two: as shown in fig. 1 and 2, a power generation power prediction method based on equipment operation state and meteorological parameters performs the following steps through a power generation power prediction system
S1, obtaining the generated power loss rate caused by the photovoltaic module. The step S1 specifically comprises the steps of S11 shooting the photovoltaic module 2 through the visible light-infrared thermal imaging integrated camera 1 to obtain an accumulated gray visible light image and a hot spot infrared thermal image, and S12 analyzing the accumulated gray visible light image and the hot spot infrared thermal image through a ResNet depth convolution neural network model in the image analysis module 3 to obtain the power generation loss rate caused by the photovoltaic module. The visible light-infrared thermal imaging integrated camera 1 is provided with a first lens (visible light lens) and a second lens (infrared thermal lens), wherein the first lens is used for shooting a 'dust accumulation' image of the photovoltaic module 2 (namely, obtaining a dust accumulation visible light image), and the second lens is used for shooting a 'hot spot' image of the photovoltaic module 2 (namely, obtaining a hot spot infrared thermal image). Each photovoltaic panel of the photovoltaic module 2 is correspondingly provided with a visible light-infrared thermal imaging integrated camera 1.
The image analysis modules 3 are connected with the visible light-infrared thermal imaging integrated cameras 1 in a one-to-one correspondence manner and are used for analyzing the gray visible light images and the hot spot infrared thermal images of the corresponding photovoltaic panels so as to obtain the power generation loss rate of each photovoltaic panel in the photovoltaic module 2. And the image analysis module 3 analyzes the gray-deposition visible light image and the hot spot infrared thermal image through ResNet depth convolution neural network model so as to obtain the power generation loss rate caused by the photovoltaic module. The gray visible light image and the hot spot infrared thermal image do not need to be processed before being input into the ResNet depth convolution neural network model. The ResNet deep convolutional neural network model integrates the feature extraction operator and the classifier in an end-to-end multi-layer network, breaks through the limitation of the feature extraction technology in the aspects of image recognition and analysis, and does not need to manually design target image features.
Before ResNet deep convolutional neural network model is really used, a batch of gray visible light images and hot spot infrared thermal images are acquired, the images are divided into a plurality of groups according to the gray accumulation degree and the hot spot degree of a photovoltaic module, and parameters such as current, voltage and power of corresponding image groups are acquired at the same time so as to calculate the average value of the loss rate of the power generation efficiency under each group. Then, the images in each group are randomly split into a training set and a testing set in proportion to complete the deep training of ResNet deep convolutional neural network model.
When the image analysis module 3 is used, the gray-deposition visible light image and the hot spot infrared thermal image can be directly read through the ResNet depth convolution neural network model, the ResNet depth convolution neural network model firstly unifies the image format and the size of the gray-deposition visible light image and the hot spot infrared thermal image, then the image features of the gray-deposition visible light image and the hot spot infrared thermal image are extracted through multi-layer convolution and are unfolded in a full connection mode, finally the probability of each grouping of the images is calculated through a classifier, the power generation efficiency loss rate corresponding to the grouping with the highest possibility is output, and then the correlation between the gray-deposition state image, the hot spot state image and the power generation efficiency loss rate is realized.
S2, obtaining the generated power loss rate caused by inversion and boosting of the inverter and the transformer. The step S2 specifically comprises the step S21 of transmitting the real-time parameters of the inverter power adjustment acquired by the inverter parameter acquisition device 4-1 to the industrial control computer 8 through the data acquisition card 7. S22, the transformer efficiency parameters acquired by the transformer parameter acquisition unit 5-1 are transmitted to the industrial control computer 8 through the data acquisition card 7. S23, calculating the generated power loss rate caused by inversion boosting through a generated power loss rate calculating unit caused by inversion boosting and using an inverter power adjustment real-time parameter and a transformer efficiency parameter. The power generation power loss rate calculation unit caused by inversion boosting obtains corresponding conversion efficiency a% by adjusting real-time parameters through inverter power, wherein a% is 95-98%; the corresponding conversion efficiency b% and b% are obtained through the efficiency parameters of the transformer, and the generated power loss rate caused by inversion boosting is 1-a% and b%. When a% is 97%, b% is 98%, and the generated power loss rate due to inversion boosting is 5%.
S3, acquiring environmental weather parameters. The ambient weather parameters, including solar irradiance, ambient air temperature and wind speed, may be obtained directly from the weather station 6.
And S4, calculating the power generation of the photovoltaic power station by using the environmental meteorological parameters, the power generation loss rate caused by the photovoltaic module and the power generation loss rate caused by inversion boosting. The step S4 specifically comprises the step of calculating the generated power of the photovoltaic power station through a deep belief network DBN model in the generated power prediction unit by combining environmental meteorological parameters, the generated power loss rate caused by inversion boosting and the generated power loss rate caused by the photovoltaic module. Before the deep belief network DBN model is actually used, enough data sets are required to be acquired, and each data set comprises solar irradiance, ambient air temperature, wind speed, the generated power loss rate of the photovoltaic panel, the generated power loss rate caused by inversion and boosting of the photovoltaic assembly and the actual photovoltaic generated power of the corresponding photovoltaic panel. These data sets are then split proportionally randomly into training and testing sets to complete the deep training of the deep belief network DBN model.
When the power generation power prediction unit is used, real-time solar irradiance, ambient air temperature, wind speed, power generation power loss rate of the photovoltaic panel and power generation power loss rate caused by inversion and boosting of the photovoltaic module are input through the deep belief network DBN model, the power generation power prediction value of the corresponding photovoltaic panel can be obtained, and then the power generation power of the required photovoltaic power station can be obtained by summing all the power generation power prediction values of the photovoltaic panel. According to the application, the environmental meteorological parameters and the equipment operation state parameters of the photovoltaic module, the inverter and the transformer are combined to jointly predict the power generation power of the photovoltaic power station, so that the accuracy and reliability of the power generation power prediction of the photovoltaic power station are effectively improved.
The above examples are only illustrative of the preferred embodiments of the present invention and do not limit the spirit and scope of the present invention. Various modifications and improvements of the technical scheme of the present invention will fall within the protection scope of the present invention without departing from the design concept of the present invention, and the technical content of the present invention is fully described in the claims.
Claims (2)
1. The utility model provides a generating power prediction system based on equipment running state and meteorological parameter which characterized in that: comprising
The plurality of visible light-infrared thermal imaging integrated cameras (1) are used for shooting the photovoltaic module (2) to obtain an ash deposition visible light image and a hot spot infrared thermal image;
the image analysis modules (3) are connected with the visible light-infrared thermal imaging integrated cameras (1) in a one-to-one correspondence manner and are used for analyzing the gray-scale visible light image and the hot spot infrared thermal image so as to obtain the generation power loss rate caused by the photovoltaic module;
an inverter (4) and an inverter parameter collector (4-1), wherein the inverter (4) is connected with the photovoltaic module (2); the inverter parameter collector (4-1) is connected with the inverter (4) and is used for acquiring inverter power adjustment real-time parameters of the inverter (4);
The transformer (5) and the transformer parameter collector (5-1), the transformer (5) is connected with the inverter (4); the transformer parameter collector (5-1) is connected with the transformer (5) and is used for obtaining transformer efficiency parameters of the transformer (5);
a weather station (6) for acquiring environmental weather parameters;
The data acquisition card (7) is provided with a first data input end connected with the image analysis module (3), a second data input end connected with the inverter parameter collector (4-1), a third data input end connected with the transformer parameter collector (5-1) and a fourth data input end connected with the weather station (6) and is used for collecting the generation power loss rate, the inverter power regulation real-time parameter, the transformer efficiency parameter and the environmental weather parameter caused by the photovoltaic module;
the industrial control computer (8) is connected with the data output end of the data acquisition card (7) and predicts the generated power of the photovoltaic power station through the generated power loss rate, the inverter power regulation real-time parameter, the transformer efficiency parameter and the environmental meteorological parameter caused by the photovoltaic module;
The image analysis module (3) analyzes the gray visible light image and the hot spot infrared thermal image through ResNet depth convolution neural network model so as to obtain the loss rate of the generated power caused by the photovoltaic module;
the environmental meteorological parameters comprise solar irradiance, environmental air temperature and wind speed;
The industrial control computer (8) is provided with
The power generation loss rate calculation unit calculates the power generation loss rate caused by inversion and boosting through the inverter power adjustment real-time parameter and the transformer efficiency parameter; specifically, the inverter power is used for adjusting real-time parameters to obtain corresponding conversion efficiency a%, and the transformer efficiency parameters are used for obtaining corresponding conversion efficiency b%, so that the generated power loss rate caused by inversion boosting is 1-a%;
The power generation power prediction unit calculates the power generation power of the photovoltaic power station by combining the environmental meteorological parameters, the power generation power loss rate caused by inversion boosting and the power generation power loss rate caused by the photovoltaic module through the deep belief network DBN model.
2. The method for predicting the generated power based on the equipment operation state and the meteorological parameters is characterized by comprising the following steps of: the following steps are performed by the generated power prediction system
S1, acquiring a generated power loss rate caused by a photovoltaic module;
s2, acquiring the loss rate of the generated power caused by inversion and boosting of an inverter and a transformer;
s3, acquiring environmental meteorological parameters;
s4, calculating the power generation of the photovoltaic power station by using the environmental meteorological parameters, the power generation loss rate caused by the photovoltaic module and the power generation loss rate caused by inversion boosting;
the S1 specifically comprises
S11, shooting a photovoltaic module (2) through a visible light-infrared thermal imaging integrated camera (1) to obtain an accumulated gray visible light image and a hot spot infrared thermal image;
S12, analyzing the gray visible light image and the hot spot infrared thermal image through a ResNet depth convolution neural network model in the image analysis module (3) to obtain the generation power loss rate caused by the photovoltaic module;
s13, transmitting the generated power loss rate caused by the photovoltaic module to an industrial control computer (8) in real time through a data acquisition card (7);
The S2 specifically comprises
S21, transmitting the inverter power adjustment real-time parameters acquired by the inverter parameter acquisition device (4-1) to an industrial control computer (8) through a data acquisition card (7);
s22, transmitting the transformer efficiency parameters acquired by the transformer parameter acquisition device (5-1) to an industrial control computer (8) through a data acquisition card (7);
s23, calculating the generated power loss rate caused by inversion boosting through a generated power loss rate calculation unit caused by inversion boosting and using an inverter power adjustment real-time parameter and a transformer efficiency parameter to obtain the generated power loss rate caused by inversion boosting;
The step S23 is specifically that the corresponding conversion efficiency a% is obtained by adjusting the real-time parameters through the inverter power, the corresponding conversion efficiency b% is obtained through the transformer efficiency parameters, and the generated power loss rate caused by inversion boosting is 1-a%;
The step S4 specifically comprises the step of calculating the generated power of the photovoltaic power station through a deep belief network DBN model in the generated power prediction unit by combining environmental meteorological parameters, the generated power loss rate caused by inversion boosting and the generated power loss rate caused by the photovoltaic module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111214665.9A CN113890027B (en) | 2021-10-19 | 2021-10-19 | Power generation power prediction system and method based on equipment operation state and meteorological parameters |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111214665.9A CN113890027B (en) | 2021-10-19 | 2021-10-19 | Power generation power prediction system and method based on equipment operation state and meteorological parameters |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113890027A CN113890027A (en) | 2022-01-04 |
| CN113890027B true CN113890027B (en) | 2024-05-17 |
Family
ID=79003447
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111214665.9A Active CN113890027B (en) | 2021-10-19 | 2021-10-19 | Power generation power prediction system and method based on equipment operation state and meteorological parameters |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113890027B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114123200B (en) * | 2022-01-24 | 2022-04-12 | 国网山西省电力公司晋城供电公司 | Photovoltaic power station dynamic modeling method based on data driving and storage device |
| CN117134352A (en) * | 2023-10-26 | 2023-11-28 | 快电动力(北京)新能源科技有限公司 | Photovoltaic power generation power prediction method and device based on influence of dust accumulation working condition |
| CN118115591B (en) * | 2024-04-28 | 2024-07-05 | 深圳市峰和数智科技有限公司 | Power prediction model training method, photovoltaic power station operation and maintenance method and related devices |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101728984A (en) * | 2010-01-18 | 2010-06-09 | 华北电力大学（保定） | Method for predicting generation power of interconnected photovoltaic power station |
| CN202206330U (en) * | 2011-06-17 | 2012-04-25 | 英利能源（中国）有限公司 | Control device for controlling photovoltaic cell to carry out stepped power generation along with illumination change of environment and solar photovoltaic power generation system |
| CN105811880A (en) * | 2016-05-16 | 2016-07-27 | 安徽思普瑞德新能源科技有限公司 | UAV mounted-based photovoltaic module real-time monitoring system |
| JP2017153300A (en) * | 2016-02-25 | 2017-08-31 | 株式会社ドリーム | Photovoltaic power generation panel inspection device, panel inspection method, and computer program |
| CN109215042A (en) * | 2018-09-28 | 2019-01-15 | 吉林电力股份有限公司科技开发分公司 | A kind of photovoltaic battery panel hot spot effect detection system based on computer vision and its calculation method |
| CN109615629A (en) * | 2018-12-24 | 2019-04-12 | 东北电力大学 | Photovoltaic panel dust stratification status image identifying system and its analysis regulation method based on convolutional neural networks |
| CN112329996A (en) * | 2020-10-26 | 2021-02-05 | 合肥阳光智维科技有限公司 | Inverter mismatch optimization method, device and equipment |
| CN213279182U (en) * | 2020-09-18 | 2021-05-25 | 广州中旭新能源有限公司 | Photovoltaic power generation system with optimized voltage control |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10732319B2 (en) * | 2017-08-30 | 2020-08-04 | International Business Machines Corporation | Forecasting solar power output |
-
2021
- 2021-10-19 CN CN202111214665.9A patent/CN113890027B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101728984A (en) * | 2010-01-18 | 2010-06-09 | 华北电力大学（保定） | Method for predicting generation power of interconnected photovoltaic power station |
| CN202206330U (en) * | 2011-06-17 | 2012-04-25 | 英利能源（中国）有限公司 | Control device for controlling photovoltaic cell to carry out stepped power generation along with illumination change of environment and solar photovoltaic power generation system |
| JP2017153300A (en) * | 2016-02-25 | 2017-08-31 | 株式会社ドリーム | Photovoltaic power generation panel inspection device, panel inspection method, and computer program |
| CN105811880A (en) * | 2016-05-16 | 2016-07-27 | 安徽思普瑞德新能源科技有限公司 | UAV mounted-based photovoltaic module real-time monitoring system |
| CN109215042A (en) * | 2018-09-28 | 2019-01-15 | 吉林电力股份有限公司科技开发分公司 | A kind of photovoltaic battery panel hot spot effect detection system based on computer vision and its calculation method |
| CN109615629A (en) * | 2018-12-24 | 2019-04-12 | 东北电力大学 | Photovoltaic panel dust stratification status image identifying system and its analysis regulation method based on convolutional neural networks |
| CN213279182U (en) * | 2020-09-18 | 2021-05-25 | 广州中旭新能源有限公司 | Photovoltaic power generation system with optimized voltage control |
| CN112329996A (en) * | 2020-10-26 | 2021-02-05 | 合肥阳光智维科技有限公司 | Inverter mismatch optimization method, device and equipment |
Non-Patent Citations (1)
| Title |
|---|
| 基于深度学习的光伏板巡检图像故障检测研究;李露露;《工程科技Ⅱ辑》;20210615;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113890027A (en) | 2022-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113890027B (en) | Power generation power prediction system and method based on equipment operation state and meteorological parameters | |
| CN109615629B (en) | Convolutional neural network-based photovoltaic panel dust accumulation state image recognition system and analysis, regulation and control method thereof | |
| CN105375878B (en) | A kind of method of on-line checking and assessment photovoltaic system efficiency | |
| CN108964606B (en) | Hot spot fault detection method for photovoltaic system | |
| CN107742171B (en) | Photovoltaic power station power generation power prediction method based on mobile shadow image recognition | |
| CN112871938B (en) | Photovoltaic power plant cleans control integration management system | |
| CN114037918A (en) | Photovoltaic hot spot detection method based on unmanned aerial vehicle inspection and image processing | |
| US9773303B2 (en) | Prediction system for short-term solar irradiance and method for operating the same | |
| CN113284124A (en) | Photovoltaic panel defect detection method based on unmanned aerial vehicle vision | |
| CN105262441B (en) | Infrared image-based photovoltaic array fault grading method | |
| CN117220597B (en) | Quick frequency response rate monitoring system of photovoltaic power station | |
| JP3245325U (en) | Solar power plant cluster monitoring system | |
| KR101647345B1 (en) | Apparatus for generating electricity using sunlight with monitoring function | |
| CN115514318A (en) | Photovoltaic power plant monitored control system | |
| CN116503308A (en) | Photovoltaic panel fault identification method based on cascade region convolutional neural network | |
| CN103712685A (en) | Photovoltaic array irradiance measurement identification method | |
| Özer et al. | An approach based on deep learning methods to detect the condition of solar panels in solar power plants | |
| CN116736893B (en) | Intelligent energy management method of optical storage device and optical storage device | |
| CN117200693A (en) | Photovoltaic module fault diagnosis method, device, equipment and storage medium | |
| Hu et al. | A detecting algorithm for occlusion on the surface of photovoltaic modules based on improved YOLOv5 | |
| CN107045282A (en) | Solar automatic tracking system and method based on single chip microcomputer fuzzy control algorithm | |
| CN118040907B (en) | Electric quantity acquisition terminal based on light Fu Bianduan automatic control strategy | |
| Wang et al. | A multi-sensor information fusion monitoring system for photovoltaic power generation | |
| CN109447350A (en) | A kind of prediction technique, system and the platform of the response of solar panel output power | |
| CN116317173B (en) | Energy storage on-line monitoring system applied to photovoltaic project |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |