CN114004399A - Power generation loss prediction method and device and electronic equipment - Google Patents
Power generation loss prediction method and device and electronic equipment Download PDFInfo
- Publication number
- CN114004399A CN114004399A CN202111266020.XA CN202111266020A CN114004399A CN 114004399 A CN114004399 A CN 114004399A CN 202111266020 A CN202111266020 A CN 202111266020A CN 114004399 A CN114004399 A CN 114004399A
- Authority
- CN
- China
- Prior art keywords
- power generation
- component
- solar
- predicted
- picture
- 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 133
- 238000000034 method Methods 0.000 title claims abstract description 75
- 238000004140 cleaning Methods 0.000 claims abstract description 25
- 238000012549 training Methods 0.000 claims description 35
- 230000008021 deposition Effects 0.000 claims description 33
- 239000000428 dust Substances 0.000 claims description 33
- 238000012360 testing method Methods 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 19
- 238000007781 pre-processing Methods 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 8
- 238000012805 post-processing Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 238000009434 installation Methods 0.000 claims description 5
- 238000009825 accumulation Methods 0.000 claims description 4
- 238000003672 processing method Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 10
- 239000004071 soot Substances 0.000 description 7
- 238000013135 deep learning Methods 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 6
- 238000003860 storage Methods 0.000 description 5
- 238000012423 maintenance Methods 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 238000009472 formulation Methods 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000003344 environmental pollutant Substances 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 231100000719 pollutant Toxicity 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 241001270131 Agaricus moelleri Species 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 210000003608 fece Anatomy 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Primary Health Care (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Photovoltaic Devices (AREA)
Abstract
The invention discloses a method and a device for predicting power generation loss and electronic equipment. The power generation amount loss prediction method includes: acquiring a shot photovoltaic module picture, and acquiring irradiance and a solar incident angle during shooting; identifying a component area to be predicted in the photovoltaic component picture; calculating the average gray scale of the component area to be predicted; and predicting the power generation capacity loss according to the average gray scale of the area of the component to be predicted, the irradiance and the solar incident angle. The embodiment of the invention can improve the prediction precision of the power generation loss and is beneficial to scientifically making a component cleaning plan.
Description
Technical Field
The embodiment of the invention relates to the technical field of distributed energy, in particular to a method and a device for predicting power generation loss and electronic equipment.
Background
The subassembly of photovoltaic power plant adheres to pollutants such as dust, bird's droppings, snow easily, and the pollutant not only can influence photovoltaic module generating efficiency, causes the loss of subassembly generated energy, also can cause the influence to the quality safety of subassembly, regularly carries out the washing of subassembly and is the necessary work of photovoltaic fortune dimension. How to scientifically make a cleaning plan by measuring the power generation loss and the cleaning cost is an intelligent cleaning problem focused by the current photovoltaic operation and maintenance manufacturer, and the evaluation and prediction of the power generation loss of the dust deposition component is the core content of the problem.
At present, in the operation and maintenance process of a photovoltaic power station, the prediction mode of the power generation loss of a component comprises the following steps: and prediction according to meteorological data and historical data recorded by a station end and prediction according to the current dust deposition degree. The method has the advantages that the number of characteristic points needing to be collected is predicted according to meteorological data and station-side historical data, the meteorological data are prediction data, and the data accuracy is limited by the accuracy of collection equipment, so that the accuracy of power generation loss prediction is difficult to guarantee by the scheme. However, for the scheme for predicting the current soot formation degree, the prior art generally focuses on how to more accurately judge the current soot formation degree, and the prediction is performed through the relationship between the soot formation degree and the power generation loss without considering other influence factors on the power generation loss, so that the prediction accuracy of the scheme on the power generation loss is difficult to guarantee. In summary, the conventional prediction method of the power generation loss of the module has the problem of low prediction accuracy of the power generation loss, and influences the formulation of the module cleaning plan.
Disclosure of Invention
The embodiment of the invention provides a method and a device for predicting power generation loss and electronic equipment, which are used for improving the prediction precision of the power generation loss and are beneficial to scientifically making a component cleaning plan.
In a first aspect, an embodiment of the present invention provides a method for predicting a power generation loss, including:
acquiring a shot photovoltaic module picture, and acquiring irradiance and a solar incident angle during shooting;
identifying a component area to be predicted in the photovoltaic component picture;
calculating the average gray scale of the component area to be predicted;
and predicting the power generation capacity loss according to the average gray scale of the area of the component to be predicted, the irradiance and the solar incident angle.
Optionally, the component area to be predicted is identified by a component identification model;
the acquisition process of the component identification model comprises the following steps:
acquiring a photovoltaic module picture set; wherein, a part of the pictures in the photovoltaic module picture set are used as a training set, and the rest of the pictures are used as a test set;
training the training set to obtain a preliminary identification model;
judging whether the preliminary identification model meets the index requirements or not according to the test set; if so, taking the preliminary identification model as the component identification model; otherwise, continuing to train the training set.
Optionally, before training the training set, the method further includes: preprocessing the photovoltaic module picture set;
correspondingly, before identifying the component region to be predicted in the photovoltaic component picture, the method further comprises the following steps: and preprocessing the photovoltaic module picture by adopting the same processing method.
Optionally, the pre-processing comprises: at least one of image enhancement processing, normalization processing, and scale transformation.
Optionally, predicting the power generation loss by using a power generation loss prediction model;
the process of obtaining the power generation loss prediction model comprises the following steps:
acquiring a generating capacity loss data set; the power generation capacity loss data set comprises photovoltaic module pictures at different dust accumulation degrees, corresponding power generation capacity loss data, corresponding irradiance and corresponding solar incident angles;
extracting a component region in the photovoltaic component picture, and calculating the average gray scale of the component region;
and obtaining the power generation loss prediction model according to the average gray scale of the assembly area, the irradiance, the solar incident angle and the power generation loss data.
Optionally, the obtaining of the power generation amount loss data set includes:
selecting a cleaning component, acquiring a shot cleaning component picture, and acquiring generated energy loss data, irradiance and a solar incident angle of the cleaning component during shooting;
selecting a control assembly to perform a dust deposition test; and acquiring the shot contrast assembly pictures at preset time intervals, and acquiring the generated energy loss data, the irradiance and the solar incident angle of the contrast assembly during shooting.
Optionally, the acquiring the sun incident angle at the time of shooting includes:
calculating a time angle according to the real solar time during shooting;
calculating the solar declination angle according to the number of days from one month to one day of the year when shooting is carried out;
calculating a solar altitude angle according to the solar declination angle, the geographic latitude of the assembly and the time angle;
calculating a solar azimuth angle according to the solar altitude angle, the geographic latitude of the assembly and the solar declination angle;
and calculating the solar incident angle according to the solar altitude angle, the included angle between the assembly installation and the ground, the solar azimuth angle and the assembly azimuth angle.
Optionally, before calculating the average gray scale of the component region to be predicted, the method further includes:
and carrying out post-processing on the component area to be predicted.
Optionally, the post-processing comprises: at least one of component contour extraction, noise point removal, misrecognized region removal, and component region refinement extraction.
In a second aspect, an embodiment of the present invention further provides an electric power generation amount loss prediction apparatus, including:
the acquisition module is used for acquiring a shot photovoltaic assembly picture and acquiring irradiance and a solar incident angle during shooting;
the identification module is used for identifying the area of the component to be predicted in the photovoltaic component picture;
the calculation module is used for calculating the average gray scale of the component area to be predicted;
and the prediction module is used for predicting the power generation loss according to the average gray scale of the area of the component to be predicted, the irradiance and the solar incident angle.
In a third aspect, an embodiment of the present invention further provides an electronic device, including:
the shooting device is used for shooting a picture of the photovoltaic module;
a memory, a processor, and a computer program stored on the memory and executable on the processor; the processor implements the power generation amount loss prediction method as provided in any of the embodiments of the present invention when executing the program.
According to the method for predicting the power generation capacity loss, on the basis that the average gray level of the area of the component to be predicted is used for representing the gray level of the component, the irradiance and the solar incident angle at the moment of picture shooting of the photovoltaic component are introduced as parameters for predicting the power generation capacity loss of the component, so that the influence of the solar incident angle on the irradiation attenuation rate of the gray component on the photovoltaic surface in different time periods is considered, and the accuracy of predicting the power generation capacity loss can be improved. Meanwhile, compared with the prior art, the embodiment of the invention introduces the solar incident angle parameter, avoids the problem of poor scheme applicability caused by time difference in different longitude and latitude areas due to the fact that a timestamp is directly used as a parameter, and enables the prediction method to have high universality, universality and practicability. Therefore, compared with the prior art, the embodiment of the invention can improve the prediction precision of the power generation loss and is beneficial to scientifically making a component cleaning plan.
Drawings
Fig. 1 is a schematic flow chart of a method for predicting power generation loss according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a process of obtaining a component recognition model according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating another component recognition model obtaining process according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a process for obtaining a model for predicting power generation loss according to an embodiment of the present invention;
FIG. 5 is a schematic flow chart of another method for predicting power generation loss according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electric power generation amount loss prediction apparatus according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
As described in the background art, the conventional method of predicting the power generation amount loss of the module has a problem of low accuracy of predicting the power generation amount loss, and the cause of the problem will be specifically described below.
For photovoltaic power station components, the way of generating capacity loss prediction and cleaning scheme formulation in the prior art comprises:
1. the method comprises the steps of predicting the loss of the power generation capacity of the assembly caused by future dust deposition by acquiring weather data such as weather, irradiation, humidity and wind direction, and station-side acquired data such as current and voltage of the assembly and historical power generation capacity, and formulating a cleaning scheme based on the loss. However, the scheme has more characteristic points to be acquired, and the accuracy of meteorological data is limited by the accuracy of acquisition equipment.
2. And predicting by acquiring the current dust deposition degree of the component and utilizing the relation between the dust deposition degree and the power generation loss. As described in the background art, the prior art focuses on how to more accurately determine the current soot deposition level, and the existing soot deposition level determination scheme includes:
1) and the dust deposition degree of the assembly is artificially judged through inspection. However, the artificial judgment has subjectivity, and the judgment result is rough; and if the component needs to be cleaned directly through manual judgment, the loss of the generated energy cannot be quantified at all.
2) And installing a dust detection device. But the equipment cost is higher and the subsequent operation and maintenance cost is high.
3) The image sampling is carried out on the assembly, the identification and judgment of the dust accumulation degree are carried out, namely, the image information of the dust deposition assembly is introduced, and the dust deposition degree of the assembly is output through the image identification technology and is used as a subsequent judgment basis or a predicted characteristic factor. However, the scheme usually requires a collecting camera to be installed above the component, is greatly restricted by shooting equipment, and has certain hardware cost and installation and maintenance cost.
Therefore, in summary, the scheme 2 does not consider other influence factors on the power generation loss, and the prediction accuracy is difficult to ensure.
Based on the research, the embodiment of the invention provides a power generation loss prediction method which can be executed by a power generation loss prediction device, is suitable for power generation loss prediction of dust deposition components of a photovoltaic power station in various environments, and facilitates formulation of a component cleaning scheme. Fig. 1 is a schematic flow chart of a method for predicting power generation loss according to an embodiment of the present invention. Referring to fig. 1, the power generation amount loss prediction method includes the steps of:
and S110, acquiring a shot photovoltaic assembly picture, and acquiring irradiance and a solar incident angle during shooting.
The shooting tool of the photovoltaic module picture can be specifically adapted according to on-site shooting equipment, for example, a mobile phone, a camera, an unmanned aerial vehicle or a camera is used, and the photovoltaic module picture only contains components needing to be detected. Therefore, the scheme does not depend on fixed shooting equipment, and has high flexibility and high practicability. Illustratively, irradiance may be measured by a horizontal irradiator.
Irradiance is directly related to the power generation of the assembly; and the solar incident angle has certain influence on the irradiation attenuation rate of the dust deposition assembly, so that the generated energy of the assembly is influenced. Specifically, the attenuation rate of the module radiation caused by dust deposition tends to decrease and then increase along with the change of the solar incident angle of the light receiving surface of the module, and the attenuation rate rapidly increases after the incident angle increases to 60 degrees. That is, the attenuation rate of the radiation acquired by the assembly is related to the incident angle of the sun, and the attenuation rate of the soot-deposited assembly is higher than that of the clean assembly. The embodiment introduces the parameter of the solar incident angle, so that the power generation loss can be more accurately predicted.
And S120, identifying the area of the component to be predicted in the photovoltaic component picture.
The area of the component to be predicted can be an image part occupied by the component to be predicted in a plurality of components covered by the photovoltaic component picture; it may further be an effective part of the image portion occupied by the component to be predicted, i.e. an image portion within the outer contour of the component to be predicted.
And S130, calculating the average gray scale of the component area to be predicted.
The average gray scale of the area of the component to be predicted can reflect the gray deposition degree of the component to be predicted and is used as a parameter for predicting the power generation loss.
And S140, predicting the power generation loss according to the average gray scale, the irradiance and the solar incident angle of the area of the component to be predicted.
The relationship between the average gray scale, the irradiance and the solar incident angle and the power generation loss can be obtained according to experimental data or station-side historical data, and the specific obtaining mode is not limited here.
According to the method for predicting the power generation capacity loss, on the basis that the average gray level of the area of the component to be predicted is used for representing the gray level of the component, the irradiance and the solar incident angle at the moment of picture shooting of the photovoltaic component are introduced as parameters for predicting the power generation capacity loss of the component, so that the influence of the solar incident angle on the irradiation attenuation rate of the gray component on the photovoltaic surface in different time periods is considered, and the accuracy of predicting the power generation capacity loss can be improved. Meanwhile, compared with the prior art, the embodiment of the invention introduces the solar incident angle parameter, avoids the problem of poor scheme applicability caused by time difference in different longitude and latitude areas due to the fact that a timestamp is directly used as a parameter, and enables the prediction method to have high universality, universality and practicability. Therefore, compared with the prior art, the embodiment of the invention can improve the prediction precision of the power generation loss and is beneficial to scientifically making a component cleaning plan.
In addition to the above embodiments, the method for acquiring the solar incident angle at the time of photographing may be: calculating a time angle according to the real solar time during shooting; calculating the solar declination angle according to the number of days from one month to one day of the year when shooting is carried out; calculating a solar altitude angle according to the solar declination angle, the geographic latitude and the time angle of the assembly; calculating a solar azimuth angle according to the solar altitude angle, the geographic latitude of the assembly and the solar declination angle; and calculating the solar incident angle according to the solar altitude, the included angle between the assembly installation and the ground, the solar azimuth angle and the assembly azimuth angle.
The embodiment calculates the solar incident angle by integrating the shooting time, various directions and angles of the sun and various directions and angles of the assembly, so that the calculation result of the solar incident angle is accurate and reliable.
Specifically, the solar incident angle (denoted as θ) is calculated according to the following formula:
t＝(t_hour-12)*15° (1)
sinHs＝sinφ*sind+cosφ*cosδ*cost (3)
cosθ＝cos(90°-Hs)*cosβ+sin(90°-Hs)sinβcos(As-A) (5)
in the above formulas, t represents a time angle; t _ hour represents the real solar time during shooting, namely the real solar time corresponding to the hour of the picture shooting moment of the photovoltaic module; δ represents the solar declination angle; t _ day represents the number of days from the month of the year at the time of shooting; hsRepresenting the solar altitude; phi represents the geographic latitude of the component; a. thesRepresenting the sun azimuth; beta denotes the assembly mountingThe ground included angle; a denotes the component azimuth.
On the basis of the above embodiments, optionally, the component region to be predicted is identified by a component identification model, so as to realize high-precision extraction of the component region to be predicted. The following describes the acquisition process of the component recognition model.
Fig. 2 is a schematic diagram of an acquisition process of a component identification model according to an embodiment of the present invention. Referring to fig. 2, in one embodiment, optionally, the obtaining of the component recognition model comprises the following steps:
s210, acquiring a photovoltaic module picture set; and the partial pictures in the photovoltaic module picture set are used as a training set, and the rest pictures are used as a test set.
The photovoltaic module picture set can be obtained by selecting part of sample pictures from a large number of photovoltaic module pictures shot under dimensions of various times, environments, shooting angles and the like; the method has the advantages that the photovoltaic module pictures covering different scenes in the range as much as possible are selected, and the component identification and segmentation accuracy of the obtained component identification model in the complex scene of the photovoltaic power station can be ensured. Illustratively, the selection rules include, but are not limited to, the following:
1) selecting photovoltaic module pictures with various resolutions;
2) selecting photovoltaic module pictures shot under different brightness and contrast;
3) selecting photovoltaic module pictures shot at different shooting angles, wherein the pictures comprise a close view and a long view and comprise modules with different sizes and shapes;
4) selecting a photovoltaic module picture shot in an open environment, for example, selecting a photovoltaic module picture containing interference features except for a module, for example, features of people, buildings, vegetation, mountains, lakes, sky and the like; or, selecting a photovoltaic module picture containing characteristics which are easy to be confused with the module, such as characteristics including a window, a roof plastic plate, an air conditioner outdoor unit and the like.
5) Selecting photovoltaic module pictures in different states, for example, selecting normal (clean) module pictures and dust-deposition module pictures in different degrees to form a photovoltaic module picture set; and selecting the pictures of the photovoltaic modules with different colors, corner points and side lines.
And S220, training a training set to obtain a primary recognition model.
Wherein, the training set can be trained by adopting a deep learning network.
S230, judging whether the preliminary identification model meets the index requirements or not according to the test set; if yes, go to S240; otherwise, execution continues with S220.
The generalization performance can be used as an index for determining whether the model is qualified, such as pixel accuracy and cross-over ratio.
And S240, taking the preliminary identification model as a component identification model.
The embodiment of the invention realizes the acquisition of the component identification model through S210-S240. Due to different application scenes, the photovoltaic power station has great difference of power station environments, such as roofs, water surfaces, mountainous regions, fields and other environments, and therefore the image recognition precision of the photovoltaic modules under the general scene is influenced by environmental factors. Self-adaption and high-precision component segmentation cannot be achieved based on the traditional image recognition segmentation technology, noise is inevitably introduced when the ash deposition state of the component is predicted on the basis, and the final generated energy loss prediction precision is affected. In the embodiment, self-adaption and high-precision segmentation of the to-be-predicted assembly can be achieved through the assembly recognition model, error recognition caused by an environment background is reduced, and the assembly recognition precision and the generating capacity loss prediction accuracy are improved. Moreover, the photovoltaic component picture set can contain different components of different photovoltaic power stations as much as possible, so that the scheme can better realize the adaptation of the scene of the universal power station.
Fig. 3 is a schematic diagram of another component recognition model obtaining process according to an embodiment of the present invention. Referring to fig. 3, in one embodiment, optionally, the obtaining of the component recognition model comprises the following steps:
and S310, acquiring a photovoltaic module picture set.
And the partial pictures in the photovoltaic module picture set are used as a training set, and the rest pictures are used as a test set. After the photovoltaic module picture set is obtained, all pictures in the photovoltaic module picture set can be labeled one by one according to preset calibration rules, and labeled label contents can be, for example, the outline shape, the component type and the like of components in the pictures, so that preparation is made for model training and optimization.
And S320, preprocessing the photovoltaic assembly picture set.
In the step, the same preprocessing method is adopted for the pictures in the training set and the test set, and the picture set of the photovoltaic module, the training set and the test set in each subsequent step are preprocessed picture sets. Illustratively, the preprocessing includes: at least one of image enhancement processing, normalization processing, and scale transformation.
And S330, training the training set by adopting a deep learning network to obtain a primary recognition model.
Illustratively, a specific semantic recognition model Unet deep learning model can be adopted for training, the amount of the labeled data set required by the model is much less than that of other semantic recognition tasks, so that a heavy labeling process is avoided, and the model training process is facilitated to be simplified. Moreover, because the low-level features of the components, such as color, contour, texture and the like, are obvious, the reliability of the training result is not influenced by the model.
Specifically, a deep learning Unet network may be used as a base network for component identification, a resnet50 as a skeleton network for component identification network, and imagenet pre-training weights for initialization work of model parameters.
S340, judging whether the generalization performance of the primary recognition model meets the requirement; if yes, go to S350; otherwise, S330 is performed.
Wherein, when the generalization performance exceeds a preset threshold, the generalization performance can be considered to meet the requirement. Illustratively, the generalization performance may include pixel accuracy (denoted PA) and a cross-over ratio (denoted IoU), calculated according to the following formula:
in the above formulas, TP, TN, FP, and FN all represent statistics of data prediction conditions, and are positive number for positive sample prediction, negative number for negative sample prediction, positive number for negative sample prediction, and negative number for positive sample prediction, respectively.
And S350, taking the preliminary identification model as a component identification model.
The embodiment of the invention realizes the acquisition of the component identification model based on the deep learning method through S310-S350, and has high universality and adaptability.
On the basis of the above embodiments, alternatively, the power generation amount loss is predicted by using the power generation amount loss prediction model, and the influence factor of the solar incident angle is introduced when the power generation amount loss prediction model is formed, thereby improving the prediction accuracy of the power generation amount loss prediction model. Next, a process of obtaining the power generation amount loss prediction model will be described.
S410, acquiring a power generation loss data set; the power generation capacity loss data set comprises photovoltaic module pictures at different dust deposition degrees, corresponding power generation capacity loss data, corresponding irradiance and corresponding solar incident angles.
The photovoltaic module pictures in the power generation loss data set can only comprise target module images for constructing the power generation loss model, and a large number of irrelevant features are not required to be included when the module identification model is constructed, so that the reliability of the finally formed power generation loss prediction model is ensured. The power generation capacity loss data corresponding to the photovoltaic module picture is power generation capacity loss data in a module dust deposition state when the photovoltaic module picture is shot; the irradiance corresponding to the photovoltaic assembly picture is the irradiance when the photovoltaic assembly picture is shot; the solar incident angle corresponding to the photovoltaic module picture is the solar incident angle when the photovoltaic module picture is shot.
Alternatively, the process of acquiring the power generation amount loss data set includes:
and selecting a cleaning assembly, acquiring a shot cleaning assembly picture, and acquiring the generated energy loss data, irradiance and solar incident angle of the cleaning assembly during shooting.
Selecting a control assembly to perform a dust deposition test; and acquiring the shot contrast assembly pictures at preset time intervals, and acquiring the generated energy loss data, the irradiance and the solar incident angle of the contrast assembly during shooting.
The method has the advantages that the dust deposition degree of the assembly can be accurately controlled by carrying out the dust deposition test, and compared with the method of directly collecting the dust deposition assembly running in the power station, the reliability of the generating capacity loss prediction model can be improved through the dust deposition test. On the other hand, the reliability of representing the accumulated dust degree by using the average gray level of the image can be verified by using the accumulated dust test, and the average gray level calculation process is adaptively adjusted. Illustratively, during the soot deposition test, a time stamp may be recorded to assist in recording the degree of soot deposition; for example, the longer the deposition test time, the more the deposition, and the greater the deposition degree.
And S420, extracting the component area in the photovoltaic component picture.
This step can be understood as the elimination of invalid areas, the identification and segmentation of the image area occupied by the component itself in the photovoltaic component picture. Only the gray value of the assembly area is calculated, so that unnecessary environmental factors can be effectively avoided, and the accuracy of the power generation loss prediction model is improved. Illustratively, this step may adopt the existing identification method in the prior art to extract the component region; alternatively, the component region extraction may be performed using a component recognition model.
And S430, calculating the average gray scale of the component area.
The average gray scale of the building region can be obtained by dividing the sum of the gray scales of all pixel points of the component region by the number of the pixel points of the component region.
Specifically, the average gray level (denoted as avg _ gary) is calculated as follows:
in the above formulas, Grayi,jRepresenting the gray value, R, corresponding to the pixel point (i, j) in the component regioni,j、Gi,j、Bi,jRespectively representing RGB color space components of pixel points (i, j) in the component region; m and n respectively represent the number of pixel points in the length direction of the component region and the number of pixel points in the width direction of the component region.
And S440, obtaining a power generation loss prediction model according to the average gray scale, the irradiance, the solar incident angle and the power generation loss data of the assembly area.
The relationship among the average gray scale, the irradiance, the solar incident angle and the power generation loss data can be established by adopting a depth regression model, and a power generation loss prediction model is obtained through training and optimization and is used for subsequent prediction.
According to the embodiment of the invention, the power generation loss prediction model is constructed based on the depth regression model through S410-S440, and the solar incident angle factor is included in the model construction, so that the model has higher applicability to scenes with different longitudes, latitudes and time ranges.
Fig. 5 is a schematic flow chart of another method for predicting power generation loss according to an embodiment of the present invention. Referring to fig. 5, on the basis of the above embodiments, optionally, the power generation amount loss prediction method includes:
and S510, acquiring the shot photovoltaic module picture.
The step can be specifically that the staff uses the shooting equipment to shoot the photovoltaic module picture, and transmits the picture to the cloud server, and the generated energy loss prediction device downloads the picture from the cloud server.
S520, preprocessing the photovoltaic module picture.
In the step, the photovoltaic module picture is preprocessed by adopting the same processing method as the method for preprocessing the photovoltaic module picture set in the process of training the module recognition model, so that the accuracy of the recognition result of the module area to be predicted is improved.
And S530, identifying the area of the component to be predicted in the photovoltaic component picture by adopting the component identification model.
And S540, performing post-processing on the component area to be predicted.
The step is a process of optimizing the recognition result of the component recognition model so as to further improve the accuracy of the final prediction result. Illustratively, the post-processing includes: and the method comprises at least one of component contour extraction, noise point removal (namely, removing noise points with the contour area smaller than a certain threshold), false identification region removal and component region fine extraction (namely, extracting component regions in the graph according to the positions of the identification labels).
And S550, calculating the average gray scale of the component area to be predicted.
And S560, obtaining irradiance and solar incident angle during shooting.
Illustratively, when a picture of the photovoltaic module is taken, a shooting timestamp may be recorded.
And S570, adopting a power generation loss prediction model to predict the power generation loss of the component according to the average gray scale, the irradiance and the solar incident angle of the component region to be predicted.
According to the embodiment of the invention, the power generation loss prediction of the photovoltaic module is realized through S510-S570, and the prediction result can be used as the basis for the subsequent cleaning judgment. The embodiment of the invention provides a component dust accumulation identification and power generation loss evaluation method based on deep learning in a general scene. Firstly, in the embodiment, the component identification model constructed based on deep learning is adopted to extract the to-be-predicted component region of the obtained photovoltaic component picture, and then the power generation loss prediction model constructed based on the deep regression model is adopted to predict the power generation loss. The embodiment of the invention can realize the adaptation of the general power station scene, and compared with the prior art, the invention realizes the self-adaption and high-precision component region segmentation and the accumulated dust state identification, reduces the false identification caused by the environmental background, and improves the component identification precision and the accuracy of the power generation loss prediction. The method does not need to install a fixed sampling camera to obtain the photovoltaic module picture, and can be specifically adapted according to on-site shooting equipment, such as a mobile phone, a camera and the like. When the power generation loss is predicted, the method incorporates the solar incident angle factor, so that the prediction model has higher applicability to scenes with different longitudes, latitudes and time ranges. Therefore, the prediction method has the advantages of high universality, high flexibility, expandability, low cost, high practicability and the like.
The embodiment of the invention also provides a device for predicting the loss of the generated energy, which is used for executing the method for predicting the loss of the generated energy provided by any embodiment of the invention and has corresponding beneficial effects. Fig. 6 is a schematic structural diagram of an electric power generation amount loss prediction apparatus according to an embodiment of the present invention. Referring to fig. 6, the power generation amount loss prediction apparatus includes: an acquisition module 610, an identification module 620, a calculation module 630, and a prediction module 640.
The obtaining module 610 is configured to obtain a photographed photovoltaic module picture, and obtain irradiance and a solar incident angle during photographing. The identification module 620 is configured to identify a component region to be predicted in the photovoltaic component picture. The calculating module 630 is used for calculating the average gray scale of the component region to be predicted. The prediction module 640 is used for predicting the power generation loss according to the average gray scale of the component area to be predicted, the irradiance and the solar incident angle.
The power generation amount loss prediction device provided by the embodiment of the invention comprises: the device comprises an acquisition module, an identification module, a calculation module and a prediction module, wherein on the basis of representing the component dust deposition degree by using the average gray level of a component region to be predicted, the irradiance and the solar incident angle at the moment of taking a picture of a photovoltaic component are introduced as parameters for predicting the power generation loss of the component, so that the influence of the solar incident angle on the dust deposition component irradiation attenuation rate in the photovoltaic surface at different time periods is considered, and the accuracy of the power generation loss prediction can be improved. Meanwhile, compared with the prior art, the embodiment of the invention introduces the solar incident angle parameter, avoids the problem of poor scheme applicability caused by time difference in different longitude and latitude areas due to the fact that a timestamp is directly used as a parameter, and improves the universality, universality and practicability of the device. Therefore, the embodiment of the invention can improve the prediction precision of the power generation loss and is beneficial to scientifically making a component cleaning plan.
On the basis of the foregoing embodiments, optionally, the identification module is specifically configured to identify the component region to be predicted by using a component identification model.
In addition to the above embodiments, the power generation amount loss prediction apparatus may further include: and the component identification model building module is used for building the component identification model. The component recognition model building module is specifically configured to: acquiring a photovoltaic module picture set; wherein, a part of pictures in the photovoltaic module picture set are used as a training set, and the rest of pictures are used as a test set; training a training set to obtain a primary recognition model; judging whether the preliminary identification model meets the index requirements or not according to the test set; if so, taking the preliminary identification model as a component identification model; otherwise, continuing training the training set.
On the basis of the foregoing embodiments, optionally, the component recognition model building module is further configured to pre-process the photovoltaic component picture set before training the training set.
In addition to the above embodiments, the power generation amount loss prediction apparatus may further include: and the preprocessing module is used for preprocessing the photovoltaic component picture by adopting the same processing method as the preprocessing method for the photovoltaic component picture set before identifying the component area to be predicted in the photovoltaic component picture.
On the basis of the above embodiments, optionally, the prediction module is specifically configured to predict the power generation amount loss by using a power generation amount loss prediction model.
In addition to the above embodiments, the power generation amount loss prediction apparatus may further include: and the prediction model building module is used for building a power generation loss prediction model. The prediction model building module is specifically used for acquiring a generating capacity loss data set; the power generation capacity loss data set comprises photovoltaic module pictures at different dust deposition degrees, corresponding power generation capacity loss data, corresponding irradiance and corresponding solar incident angles; extracting a component area in the photovoltaic component picture, and calculating the average gray scale of the component area; and obtaining a power generation loss prediction model according to the average gray scale, irradiance, solar incident angle and power generation loss data of the assembly area.
On the basis of the foregoing embodiments, optionally, the prediction model building module includes: the generating capacity loss data set acquisition unit is used for acquiring a generating capacity loss data set; the method comprises the steps that a cleaning assembly is selected, a shot cleaning assembly picture is obtained, and the generated energy loss data, the irradiance and the solar incident angle of the cleaning assembly during shooting are obtained; selecting a control assembly to perform a dust deposition test; and acquiring the shot contrast assembly pictures at preset time intervals, and acquiring the generated energy loss data, the irradiance and the solar incident angle of the contrast assembly during shooting.
On the basis of the foregoing embodiments, optionally, the obtaining module includes: an incident angle acquisition unit for acquiring a solar incident angle at the time of photographing; the method is specifically used for: calculating a time angle according to the real solar time during shooting; calculating the solar declination angle according to the number of days from one month to one day of the year when shooting is carried out; calculating a solar altitude angle according to the solar declination angle, the geographic latitude and the time angle of the assembly; calculating a solar azimuth angle according to the solar altitude angle, the geographic latitude of the assembly and the solar declination angle; and calculating the solar incident angle according to the solar altitude, the included angle between the assembly installation and the ground, the solar azimuth angle and the assembly azimuth angle.
In addition to the above embodiments, the power generation amount loss prediction apparatus may further include: and the post-processing module is used for performing post-processing on the component area to be predicted before calculating the average gray scale of the component area to be predicted.
The embodiment of the invention also provides the electronic equipment. Fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. Fig. 7 shows a block diagram of an exemplary electronic device suitable for implementing the power generation amount loss prediction method provided by any embodiment of the present invention, and does not set any limit to the function and the range of use of the embodiment of the present invention.
Referring to fig. 7, the electronic device includes a photographing device 72, a memory 71, a processor 70, and a computer program stored on the memory 71 and executable on the processor 70. The shooting device 72 is used for shooting a photovoltaic module picture; the processor 70, when executing a program, implements the power generation amount loss prediction method as provided by any of the embodiments of the present invention.
The electronic device further comprises, exemplarily, an output means 73. The processor 70, the memory 71, the photographing apparatus 72, and the output device 73 in the electronic apparatus may be connected by a bus or other means, and the bus connection is exemplified in fig. 7. Illustratively, the capture device 72 may be a cell phone, camera, or camcorder, among other devices.
The memory 71, as a computer-readable storage medium, may be used to store software programs, computer-executable programs, and modules, such as program instructions/modules (for example, an acquisition module, an identification module, a calculation module, and a prediction module of the electric power generation amount loss prediction apparatus) corresponding to the electric power generation amount loss prediction method in the embodiment of the present invention. The processor 70 executes various functional applications of the device/terminal/server and data processing by executing software programs, instructions, and modules stored in the memory 71, that is, implements the above-described power generation amount loss prediction method.
The memory 71 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 71 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 71 may further include memory located remotely from the processor 70, which may be connected to the electronic device through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The output device 73 may include a display device such as a display screen.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (11)
1. A power generation amount loss prediction method characterized by comprising:
acquiring a shot photovoltaic module picture, and acquiring irradiance and a solar incident angle during shooting;
identifying a component area to be predicted in the photovoltaic component picture;
calculating the average gray scale of the component area to be predicted;
and predicting the power generation capacity loss according to the average gray scale of the area of the component to be predicted, the irradiance and the solar incident angle.
2. The electric power generation amount loss prediction method according to claim 1, characterized in that the component region to be predicted is identified by a component identification model;
the acquisition process of the component identification model comprises the following steps:
acquiring a photovoltaic module picture set; wherein, a part of the pictures in the photovoltaic module picture set are used as a training set, and the rest of the pictures are used as a test set;
training the training set to obtain a preliminary identification model;
judging whether the preliminary identification model meets the index requirements or not according to the test set; if so, taking the preliminary identification model as the component identification model; otherwise, continuing to train the training set.
3. The power generation loss prediction method according to claim 2, further comprising, before training the training set: preprocessing the photovoltaic module picture set;
correspondingly, before identifying the component region to be predicted in the photovoltaic component picture, the method further comprises the following steps: and preprocessing the photovoltaic module picture by adopting the same processing method.
4. The electric power generation amount loss prediction method according to claim 3, characterized in that the preprocessing includes: at least one of image enhancement processing, normalization processing, and scale transformation.
5. The electric power generation amount loss prediction method according to claim 1, characterized in that the electric power generation amount loss is predicted using an electric power generation amount loss prediction model;
the process of obtaining the power generation loss prediction model comprises the following steps:
acquiring a generating capacity loss data set; the power generation capacity loss data set comprises photovoltaic module pictures at different dust accumulation degrees, corresponding power generation capacity loss data, corresponding irradiance and corresponding solar incident angles;
extracting a component region in the photovoltaic component picture, and calculating the average gray scale of the component region;
and obtaining the power generation loss prediction model according to the average gray scale of the assembly area, the irradiance, the solar incident angle and the power generation loss data.
6. The electric power generation amount loss prediction method according to claim 5, wherein the acquiring the electric power generation amount loss data set includes:
selecting a cleaning component, acquiring a shot cleaning component picture, and acquiring generated energy loss data, irradiance and a solar incident angle of the cleaning component during shooting;
selecting a control assembly to perform a dust deposition test; and acquiring the shot contrast assembly pictures at preset time intervals, and acquiring the generated energy loss data, the irradiance and the solar incident angle of the contrast assembly during shooting.
7. The power generation amount loss prediction method according to claim 1, 5, or 6, wherein acquiring the solar incident angle at the time of shooting includes:
calculating a time angle according to the real solar time during shooting;
calculating the solar declination angle according to the number of days from one month to one day of the year when shooting is carried out;
calculating a solar altitude angle according to the solar declination angle, the geographic latitude of the assembly and the time angle;
calculating a solar azimuth angle according to the solar altitude angle, the geographic latitude of the assembly and the solar declination angle;
and calculating the solar incident angle according to the solar altitude angle, the included angle between the assembly installation and the ground, the solar azimuth angle and the assembly azimuth angle.
8. The power generation amount loss prediction method according to claim 1, further comprising, before calculating the average gradation of the component region to be predicted:
and carrying out post-processing on the component area to be predicted.
9. The electric power generation amount loss prediction method according to claim 8, characterized in that the post-processing includes: at least one of component contour extraction, noise point removal, misrecognized region removal, and component region refinement extraction.
10. An electric power generation amount loss prediction device characterized by comprising:
the acquisition module is used for acquiring a shot photovoltaic assembly picture and acquiring irradiance and a solar incident angle during shooting;
the identification module is used for identifying the area of the component to be predicted in the photovoltaic component picture;
the calculation module is used for calculating the average gray scale of the component area to be predicted;
and the prediction module is used for predicting the power generation loss according to the average gray scale of the area of the component to be predicted, the irradiance and the solar incident angle.
11. An electronic device, comprising:
the shooting device is used for shooting a picture of the photovoltaic module;
a memory, a processor, and a computer program stored on the memory and executable on the processor; the processor, when executing the program, implements the power generation amount loss prediction method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111266020.XA CN114004399A (en) | 2021-10-28 | 2021-10-28 | Power generation loss prediction method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111266020.XA CN114004399A (en) | 2021-10-28 | 2021-10-28 | Power generation loss prediction method and device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114004399A true CN114004399A (en) | 2022-02-01 |
Family
ID=79924790
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111266020.XA Pending CN114004399A (en) | 2021-10-28 | 2021-10-28 | Power generation loss prediction method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114004399A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115532753A (en) * | 2022-09-02 | 2022-12-30 | 阳光新能源开发股份有限公司 | Photovoltaic power station dust loss measuring and calculating method, device, equipment and storage medium |
| CN117634771A (en) * | 2023-10-30 | 2024-03-01 | 中国电力工程顾问集团中南电力设计院有限公司 | Photovoltaic cleaning decision method based on dust monitoring and weather prediction |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108537267A (en) * | 2018-04-02 | 2018-09-14 | 江南大学 | A kind of corncob automatic identifying method based on significance test |
| CN110210060A (en) * | 2019-04-28 | 2019-09-06 | 安徽建筑大学 | The prediction technique of solar energy photovoltaic panel superficial dust degree |
| CN112926798A (en) * | 2021-03-26 | 2021-06-08 | 苏州朋友保科技有限公司 | Method, device, equipment and medium for predicting photovoltaic power generation loss caused by dust |
-
2021
- 2021-10-28 CN CN202111266020.XA patent/CN114004399A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108537267A (en) * | 2018-04-02 | 2018-09-14 | 江南大学 | A kind of corncob automatic identifying method based on significance test |
| CN110210060A (en) * | 2019-04-28 | 2019-09-06 | 安徽建筑大学 | The prediction technique of solar energy photovoltaic panel superficial dust degree |
| CN112926798A (en) * | 2021-03-26 | 2021-06-08 | 苏州朋友保科技有限公司 | Method, device, equipment and medium for predicting photovoltaic power generation loss caused by dust |
Non-Patent Citations (2)
| Title |
|---|
| 张姝伟: "基于视觉的光伏板积灰状态监测与性能评估", 中国优秀硕士学位论文全文数据库（电子期刊）工程科技Ⅱ辑, no. 7, 31 July 2019 (2019-07-31), pages 042 - 335 * |
| 王胜捷等: "光伏组件积灰特性及其透射衰减规律研究", 农业工程学报, no. 22, 30 November 2019 (2019-11-30), pages 242 - 250 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115532753A (en) * | 2022-09-02 | 2022-12-30 | 阳光新能源开发股份有限公司 | Photovoltaic power station dust loss measuring and calculating method, device, equipment and storage medium |
| CN117634771A (en) * | 2023-10-30 | 2024-03-01 | 中国电力工程顾问集团中南电力设计院有限公司 | Photovoltaic cleaning decision method based on dust monitoring and weather prediction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Caldas et al. | Very short-term solar irradiance forecast using all-sky imaging and real-time irradiance measurements | |
| CN109416413B (en) | Solar energy forecast | |
| KR101926544B1 (en) | Method, apparatus and computer program for meteorological observation using weather research forecasting model | |
| CN112422783A (en) | Unmanned aerial vehicle intelligent patrol system based on parking apron cluster | |
| KR101980741B1 (en) | A method, system and program for simulaiton of solar energy generator | |
| CN114004399A (en) | Power generation loss prediction method and device and electronic equipment | |
| Boz et al. | An automated model for rooftop PV systems assessment in ArcGIS using LIDAR. | |
| CN114240868A (en) | Unmanned aerial vehicle-based inspection analysis system and method | |
| US20170140077A1 (en) | Method for assessing efficiency of power generation systems | |
| CN114663827B (en) | Method and system for judging waterlogging and ponding depth of scene | |
| CN107238874A (en) | Based on when scape photo real-time weather detection method and system | |
| US11983813B2 (en) | Method and system for determining solar access of a structure | |
| Dissawa et al. | Sky Image‐Based Localized, Short‐Term Solar Irradiance Forecasting for Multiple PV Sites via Cloud Motion Tracking | |
| CN110852164A (en) | YOLOv 3-based method and system for automatically detecting illegal building | |
| CN117949046B (en) | Cluster networking type marine environment tracking and monitoring method and system for abnormal water clusters | |
| Vega-Garita et al. | A practical method for considering shading on photovoltaics systems energy yield | |
| CN116192005A (en) | Photovoltaic tracking method, system and medium based on micro space-time scale irradiation prediction | |
| CN114926424A (en) | Detection method and electronic equipment | |
| Abouelaziz et al. | Photogrammetry and deep learning for energy production prediction and building-integrated photovoltaics decarbonization | |
| Murakami et al. | Spatiotemporal heatwave risk modeling combining multiple observations | |
| CN114612420A (en) | YOLO algorithm-based visible light defect identification method for power transmission line | |
| Cañadillas et al. | First Results of a Low Cost All-Sky Imager for Cloud Tracking and Intra-Hour Irradiance Forecasting serving a PV-based Smart Grid in La Graciosa Island | |
| CN114612864B (en) | Intelligent black smoke detection method and system based on deep learning segmentation strategy | |
| Dissawa et al. | Sky Image-Based Short-Term Solar Power Forecasting Model | |
| CN118274805B (en) | Surveying and mapping method based on remote sensing big data analysis |
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 | ||
| CB02 | Change of applicant information |
Address after: Room 201 and 301, office building of Sunshine Power Supply Co., Ltd., No. 2, Tianhu Road, High-tech Zone, Hefei City, Anhui Province, 230088 Applicant after: Sunshine Zhiwei Technology Co.,Ltd. Address before: 230088 101, 201, office building of Hefei Haoyang Power Technology Co., Ltd., No.23 Tianzhi Road, high tech Zone, Hefei City, Anhui Province Applicant before: Sunshine Zhiwei Technology Co.,Ltd. Address after: 230088 101, 201, office building of Hefei Haoyang Power Technology Co., Ltd., No.23 Tianzhi Road, high tech Zone, Hefei City, Anhui Province Applicant after: Sunshine Zhiwei Technology Co.,Ltd. Address before: 230088 101, 201, office building of Hefei Haoyang Power Technology Co., Ltd., No.23 Tianzhi Road, high tech Zone, Hefei City, Anhui Province Applicant before: Hefei sunshine Zhiwei Technology Co.,Ltd. |
|
| CB02 | Change of applicant information |