Method, system and apparatus for determining photovoltaic panel orientation, and storage medium

Abstract

Provided in the embodiments of the present description are a method, system and apparatus for determining a photovoltaic panel orientation, and a storage medium. The method comprises: acquiring historical data, and training a photovoltaic panel orientation determination model on the basis of the historical data (S310), wherein the historical data comprises historical horizontal total irradiation data; acquiring photovoltaic panel data of a photovoltaic panel in a preset scenario (S320), wherein the photovoltaic panel data in the preset scenario comprises total irradiation data and environmental data, the total irradiation data represents a total irradiation amount, the total irradiation amount is the sum of direct irradiation, diffuse irradiation and reflected irradiation, and the environmental data comprises the ground material of the reflected irradiation and the distance of the reflected irradiation that correspond to the photovoltaic panel; on the basis of the photovoltaic panel data in the preset scenario and the trained photovoltaic panel orientation determination model, determining a photovoltaic panel orientation in which the total irradiation amount satisfies a preset condition (S330); and performing orientation control on the photovoltaic panel in the preset scenario on the basis of the determined photovoltaic panel orientation (S340).

Description

Method, system, device and storage medium for determining orientation of photovoltaic panel TECHNICAL FIELD

[0001] The present specification relates to the field of new energy, and in particular, to a method, system, device and storage medium for determining orientation of photovoltaic panel. BACKGROUND

[0002] Photovoltaic power generation is a very important power generation method in the field of new energy. The core of photovoltaic power generation facilities is photovoltaic panels, i.e. solar panels. In photovoltaic power generation, in order to obtain more total irradiance and provide greater output power, the orientation of the photovoltaic panel needs to be adjusted constantly according to a tracking algorithm to obtain the maximum irradiance. The commonly used tracking algorithm is a sun tracking algorithm, which can obtain the maximum direct irradiance when the photovoltaic panel is perpendicular to the sun. Since the total irradiance is also affected by diffuse irradiance and reflected irradiance, the maximum direct irradiance does not mean the maximum total irradiance, and this algorithm cannot obtain the maximum total irradiance, making it difficult to improve the power generation.

[0003] Therefore, it is desirable to provide a method for determining the orientation of a photovoltaic panel to maximize the obtained total irradiance and improve power generation.

[0004] SUMMARY

[0005] One of the embodiments of the present specification provides a method for determining the orientation of a photovoltaic panel. The method comprises: obtaining historical data and training a photovoltaic panel orientation determination model based on the historical data, the historical data comprising historical horizontal total irradiance data; obtaining photovoltaic panel data of a photovoltaic panel under a preset scene, wherein the photovoltaic panel data under the preset scene comprises total irradiance data and environmental data, the total irradiance data representing total irradiance, the total irradiance being the total amount of direct irradiance, diffuse irradiance and reflected irradiance, and the environmental data comprising the ground material of the reflected irradiance corresponding to the photovoltaic panel and the distance of the reflected irradiance; determining the orientation of the photovoltaic panel under which the total irradiance meets a preset condition based on the photovoltaic panel data under the preset scene and the trained photovoltaic panel orientation determination model; and controlling the orientation of the photovoltaic panel under the preset scene based on the determined orientation of the photovoltaic panel.

[0006] In some embodiments, the photovoltaic panel orientation determination model can comprise a first model and a second model, wherein the output of the first model can comprise the proportion of the diffuse irradiance in the horizontal total irradiance, the input of the second model can comprise the output of the first model, and the output of the second model can comprise the angle of the orientation of the photovoltaic panel.

[0007] In some embodiments, the first model can be obtained by pre-training.

[0008] In some embodiments, the training sample of the first model can include time information, geographical location information and the historical horizontal total irradiance data, and the training label of the first model can include the proportion of the diffuse irradiance in the horizontal total irradiance.

[0009] In some embodiments, the pre-trained first model can be used to perform offline training on the second model under a plurality of simulated preset scenes, and the plurality of simulated preset scenes can include scenes of a plurality of different ground materials.

[0010] In some embodiments, the training sample of the second model can include the proportion of the diffuse irradiance in the horizontal total irradiance output by the first model, the distance of the reflected irradiance, the ground material of the reflected irradiance and the photovoltaic panel inclined surface irradiance, and the training label of the second model can include the angle of the photovoltaic panel orientation corresponding to the highest total irradiance.

[0011] In some embodiments, the training process of the second model can include determining a first angle of the photovoltaic panel orientation based on the output of the first model and the current irradiance data, and optimizing the first angle to obtain a second angle of the photovoltaic panel orientation as the output of the second model.

[0012] In some embodiments, a plurality of candidate angles can be obtained by rotating a preset angle from the first angle and iterating a plurality of times, and the second angle can be determined as the candidate angle corresponding to the highest total irradiance.

[0013] In some embodiments, the distance of the reflected irradiance can be determined based on the height of the main beam.

[0014] One of the embodiments of the present specification provides a system for determining the orientation of a photovoltaic panel, which comprises a model training module, a photovoltaic panel data acquisition module, a photovoltaic panel orientation determination module and a photovoltaic panel orientation control module; the model training module is configured to acquire historical data and train a photovoltaic panel orientation determination model based on the historical data, wherein the historical data comprises historical horizontal total irradiance data; the photovoltaic panel data acquisition module is configured to acquire photovoltaic panel data under a preset scene, wherein the photovoltaic panel data under the preset scene comprises total irradiance data and environmental data, the total irradiance data represents the total irradiance amount, the total irradiance amount is the total amount of direct irradiance, diffuse irradiance and reflected irradiance, and the environmental data comprises the ground material of the reflected irradiance corresponding to the photovoltaic panel and the distance of the reflected irradiance; the photovoltaic panel orientation determination module is configured to determine the photovoltaic panel orientation under which the total irradiance amount meets a preset condition based on the photovoltaic panel data under the preset scene and the trained photovoltaic panel orientation determination model; and the photovoltaic panel orientation control module is configured to control the orientation of the photovoltaic panel under the preset scene based on the determined photovoltaic panel orientation.

[0015] One of the embodiments of the present specification provides a device for determining the orientation of a photovoltaic panel, comprising a processor configured to perform the method for determining the orientation of a photovoltaic panel.

[0016] One of the embodiments of the present specification provides a computer readable storage medium storing computer instructions, when the computer reads the computer instructions in the storage medium, the computer performs the method for determining the orientation of a photovoltaic panel. BRIEF DESCRIPTION OF DRAWINGS

[0017] The present specification will be further illustrated in the manner of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same numbers represent the same structures, wherein:

[0018] FIG. 1 is a schematic diagram of an application scenario of a system for determining the orientation of a photovoltaic panel according to some embodiments of the present specification;

[0019] FIG. 2 is a schematic diagram of a system for determining the orientation of a photovoltaic panel according to some embodiments of the present specification;

[0020] FIG. 3 is an exemplary flowchart of a method for determining the orientation of a photovoltaic panel according to some embodiments of the present specification;

[0021] FIG. 4 is a schematic diagram of the relationship between diffuse irradiance and the horizontal direction of a photovoltaic panel according to some embodiments of the present specification;

[0022] FIG. 5 is a schematic diagram of a tracking algorithm according to some embodiments of the present specification;

[0023] FIG. 6 is a schematic diagram of a model for determining the orientation of a photovoltaic panel according to some embodiments of the present specification;

[0024] FIG. 7A is a schematic diagram of an irradiance model according to some embodiments of the present specification;

[0025] FIG. 7B is a schematic diagram of obtaining the proportion of diffuse irradiance using the irradiance model according to some embodiments of the present specification;

[0026] FIG. 8 is a schematic diagram of astronomical irradiance according to some embodiments of the present specification. DETAILED DESCRIPTION

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some examples or embodiments of the present specification, and for those skilled in the art, the present specification can also be applied to other similar scenarios without creative labor on the basis of these drawings. Unless it is clear from the language context or otherwise stated, the same reference numbers in the drawings represent the same structures or operations.

[0028] It should be understood that the "system", "device", "unit" and / or "module" used herein is a method for distinguishing different components, elements, parts, sections or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0029] As shown in the specification and claims, unless the context clearly indicates otherwise, the words "one", "a", "an", and / or "the" do not necessarily refer to the singular, but can also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements explicitly identified, and these steps and elements do not constitute an exclusive list, and the method or device can also include other steps or elements.

[0030] Flowcharts are used in the present specification to illustrate the operations performed by the system according to the embodiments of the present specification. It should be understood that the preceding or subsequent operations are not necessarily performed in sequence. On the contrary, each step can be processed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or more steps of the operation can be removed from these processes.

[0031] Figure 1 is a schematic diagram of an application scenario of a system for determining the orientation of a photovoltaic panel according to some embodiments of the present specification.

[0032] In the following of the present specification, the system 100 for determining the orientation of a photovoltaic panel will be referred to as system 100. In some embodiments, the system 100 can include a photovoltaic device 110, a first computing device 120, a second computing device 130, a user terminal 140, a storage device 150 and a network 160.

[0033] The photovoltaic device 110 refers to a device that converts solar energy into direct current electric energy by using the photovoltaic effect of photovoltaic semiconductor materials. In some embodiments, the photovoltaic device 110 can include a photovoltaic panel for converting solar energy into electric energy and a main beam for supporting the photovoltaic panel. In some embodiments, the photovoltaic device 110 can include a driving mechanism for driving the photovoltaic panel to rotate, and the photovoltaic panel can rotate around a rotation axis under the driving of the driving mechanism. For example, in FIG. 5, the photovoltaic panel 510 is installed on the main beam 520, and the photovoltaic panel 510 in the left subgraph can be rotated to the position of the photovoltaic panel 510 in the right subgraph under the driving of the driving mechanism. The photovoltaic device 110 provided above is only for illustrative purposes, and is not intended to limit the scope thereof. In some embodiments, the photovoltaic device 110 can receive instructions and / or signals sent by the first computing device 120, and rotate the photovoltaic panel according to the instructions and / or signals to obtain the maximum irradiance.

[0034] The first computing device 120 and the second computing device 130 are systems with computing and processing capabilities, which can include various computers such as servers, personal computers, and can also be computing platforms composed of multiple computers connected in various structures. In some embodiments, the first computing device 120 and the second computing device 130 can be implemented on a cloud platform. For example, the cloud platform can include one or a combination of private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, cross-cloud, multi-cloud, etc. In some embodiments, the first computing device 120 and the second computing device 130 can be the same device or different devices.

[0035] The first computing device 120 and the second computing device 130 can include one or more sub-processing devices (e.g., single-core processing devices or multi-core multi-core processing devices), and the processing devices can execute program instructions. For example only, the processing devices can include various common general-purpose central processing units (CPUs), graphics processing units (GPUs), microprocessors, application-specific integrated circuits (ASICs), or other types of integrated circuits.

[0036] The first computing device 120 can process information and data related to the photovoltaic panel. In some embodiments, the first computing device 120 can perform a method of determining a photovoltaic panel orientation as shown in some embodiments of the present specification, to obtain a determined photovoltaic panel orientation, for example, a photovoltaic panel orientation in which the total irradiance meets a preset condition, etc. In some embodiments, the second computing device 130 can obtain photovoltaic panel data of the photovoltaic device 110 under a preset scenario, for example, total irradiance data including direct irradiance and reflected irradiance, environmental data including ground material and beam height, etc. In some embodiments, the first computing device 120 can include a photovoltaic panel orientation determination model, and the first computing device 120 can obtain a determined photovoltaic panel orientation by the photovoltaic panel orientation determination model based on the photovoltaic panel data under the preset scenario. In some embodiments, the first computing device 120 can obtain a trained photovoltaic panel orientation determination model from the second computing device 130. In some embodiments, the first computing device 120 can control the orientation of the photovoltaic panel based on the determined photovoltaic panel orientation by instructions and / or signals. In some embodiments, the first computing device 120 can exchange information and data with the network 160 and / or other components in the system 100 (for example, the photovoltaic device 110, the second computing device 130, the user terminal 140, the storage device 150). In some embodiments, the first computing device 120 can be directly connected to the second computing device 130 and exchange information and / or data.

[0037] The second computing device 130 can be used for model training. In some embodiments, the second computing device 130 can perform a training method of a photovoltaic panel orientation determination model as shown in some embodiments of the present specification, to obtain a trained photovoltaic panel orientation determination model. In some embodiments, the second computing device 130 can obtain training samples such as historical data, photovoltaic panel data under a preset scenario, etc. for training the photovoltaic panel orientation determination model. In some embodiments, the first computing device 120 and the second computing device 130 can also be the same computing device.

[0038] The user terminal 140 can interact with the user. In some embodiments, the user terminal 140 can instruct the first computing device 120 and / or the second computing device 130 to perform a method of determining a photovoltaic panel orientation as shown in some embodiments of the present specification. In some embodiments, the user terminal 140 can control the orientation of the photovoltaic panel of the photovoltaic device 110 by instructions, etc. based on the photovoltaic panel orientation determined by the first computing device 120. In some embodiments, the user terminal 140 can be one or any combination of a mobile device 140-1, a tablet computer 140-2, a laptop computer 140-3, a desktop computer, etc. other devices with input and / or output functions.

[0039] The storage device 150 can store data or information generated by other devices. In some embodiments, the storage device 150 can store historical data related to irradiance. In some embodiments, the storage device 150 can store data and / or information processed by the first computing device 120 and / or the second computing device 130, such as a trained photovoltaic panel orientation determination model, a photovoltaic panel orientation at which the total irradiance meets a preset condition, etc. The storage device 150 can include one or more storage components, each of which can be a standalone device or a part of other devices. The storage device can be local or implemented through the cloud.

[0040] The network 160 can connect components of the system and / or connect the system with external resources. The network 160 enables communication between the components and / or between the system and other parts outside the system, facilitating exchange of data and / or information. In some embodiments, one or more components in the system 100 (e.g., the photovoltaic device 110, the first computing device 120, the second computing device 130, the user terminal 140, the storage device 150) can send data and / or information to other components through the network 160. In some embodiments, the network 160 can be any one or more of wired networks or wireless networks.

[0041] It should be noted that the foregoing description is provided for illustrative purposes only and is not intended to limit the scope of the present specification. Various changes and modifications can be made to the present specification under the guidance of one of ordinary skill in the art. The features, structures, methods, and other characteristics of the exemplary embodiments described in the present specification can be combined in various ways. For example, the first computing device 120 and / or the second computing device 130 can each have an independent storage device or can share a storage device. However, these changes and modifications do not depart from the scope of the present specification.

[0042] FIG. 2 is a schematic diagram of a system for determining photovoltaic panel orientation, according to some embodiments of the present specification.

[0043] As shown in FIG. 2, in some embodiments, the system 200 for determining photovoltaic panel orientation can include a model training module 210, a photovoltaic panel data acquisition module 220, a photovoltaic panel orientation determination module 230, and a photovoltaic panel orientation control module 240. In some embodiments, the model training module 210 can be implemented by the second computing device 130, and the photovoltaic panel data acquisition module 220, the photovoltaic panel orientation determination module 230, and the photovoltaic panel orientation control module 240 can be implemented by the first computing device 120.

[0044] In some embodiments, the model training module 210 can be configured to obtain historical data, and train the photovoltaic panel orientation determination model based on the historical data. The historical data can include historical horizontal total irradiance data, etc.

[0045] In some embodiments, the photovoltaic panel orientation determination model can include a first model and a second model. The output of the first model can include a proportion of diffuse irradiance in the horizontal total irradiance, the input of the second model can include the output of the first model, and the output of the second model can include an angle of the photovoltaic panel orientation.

[0046] In some embodiments, the first model can be pre-trained.

[0047] In some embodiments, the training samples of the first model can include time information, geographic location information, and historical horizontal total irradiance data, etc., and the training labels of the first model can include the proportion of diffuse irradiance in the horizontal total irradiance.

[0048] In some embodiments, the pre-trained first model can be used to train the second model under a plurality of simulated preset scenarios. The simulated preset scenarios can include a plurality of scenarios of different ground materials.

[0049] In some embodiments, the training samples of the second model can include the proportion of diffuse irradiance in the horizontal total irradiance output by the first model, the distance of the reflected irradiance, the ground material of the reflected irradiance, and the photovoltaic panel inclined surface irradiance, etc., and the training labels of the second model can include the angle of the photovoltaic panel orientation corresponding to the highest total irradiance.

[0050] In some embodiments, the model training module 210 can train the second model by the following steps: determining a first angle of the photovoltaic panel orientation based on the output of the first model and the current irradiance data; optimizing the first angle to obtain a second angle of the photovoltaic panel orientation as the output of the second model.

[0051] In some embodiments, the model training module 210 can start from the first angle, rotate by a preset angle and iterate multiple times to obtain a plurality of candidate angles, and determine the candidate angle corresponding to the highest total irradiance as the second angle.

[0052] In some embodiments, the distance of the reflected irradiance can be determined based on the height of the main beam.

[0053] In some embodiments, the photovoltaic panel data acquisition module 220 can be configured to acquire photovoltaic panel data of the photovoltaic panel under a preset scene. The photovoltaic panel data under the preset scene can include total irradiance data and environmental data, the total irradiance data representing total irradiance, and the environmental data can include ground material and distance of reflected irradiance corresponding to the photovoltaic panel.

[0054] In some embodiments, the photovoltaic panel orientation determination module 230 can be configured to determine a photovoltaic panel orientation under which the total irradiance meets a preset condition based on the photovoltaic panel data under the preset scene and the trained photovoltaic panel orientation determination model.

[0055] In some embodiments, the photovoltaic panel orientation control module 240 can be configured to control the orientation of the photovoltaic panel under the preset scene based on the determined photovoltaic panel orientation.

[0056] FIG. 3 is an exemplary flowchart of a method for determining a photovoltaic panel orientation according to some embodiments of the present specification.

[0057] A photovoltaic system receives solar irradiation through a photovoltaic panel and converts solar energy into electrical energy. Solar irradiation, also known as insolation, solar radiation, or solar exposure, is the cumulative energy power per unit area in a certain time in the form of electromagnetic radiation received from the sun within the wavelength range of the measuring instrument. Solar radiation includes direct normal irradiation (DNI) or beam radiation, diffuse horizontal irradiance (DHI) or diffuse sky radiation, and reflected radiation, in which the solar radiation propagates through the atmosphere and reaches the ground in the form of direct light and diffuse light, and then produces reflection after reaching the ground. In a photovoltaic system, the total irradiance on the photovoltaic panel is the sum of the direct irradiance, the diffuse irradiance, and the reflected irradiance, as shown in the following formula: t = H b + H d + H R (1),

[0058] wherein H t is the total irradiance, H b is the direct irradiance, H d is the diffuse irradiance, and H R is the reflected irradiance.

[0059] Direct irradiance, also referred to as direct irradiance, is the amount of solar radiation measured on a plane perpendicular to the direction of the incoming sunlight, excluding the diffuse solar radiation.

[0060] Reflected irradiance, also referred to as reflected irradiance, is the amount of solar radiation reflected by the ground. Since the proportion of reflected irradiance in total irradiance is usually small and is affected by many factors, it is difficult to obtain an accurate value, therefore, in some embodiments, the reflected irradiance can be excluded from the total irradiance, and only the direct irradiance and the diffuse irradiance are counted. For a photovoltaic system, the reflected irradiance will be affected by the height of the main beam, the higher the main beam, the less the reflected irradiance. The reflected irradiance will also be affected by the ground material, for example, the reflected irradiance of water is greater than that of grass.

[0061] Diffuse irradiance, also referred to as diffuse irradiance, is the amount of ground radiation scattered by the atmosphere, which is usually measured on a horizontal surface, and its direction can be considered as vertically downward, and the diffuse irradiance is the largest when the photovoltaic panel is placed horizontally. In some embodiments, the relationship between the diffuse irradiance and the angle between the photovoltaic panel and the horizontal direction can be shown in the following formula: R_d=(1+cos β) / 2 (2)

[0062] wherein R_d is the diffuse irradiance received by the photovoltaic panel, and β is the angle between the photovoltaic panel and the horizontal direction. As shown in FIG. 4, it is a schematic diagram of the relationship between the diffuse irradiance and the angle between the photovoltaic panel and the horizontal direction, wherein the abscissa is the angle between the photovoltaic panel and the horizontal direction, and the ordinate is the diffuse irradiance.

[0063] The orientation of the photovoltaic panel refers to the direction of the surface of the photovoltaic panel receiving solar radiation, which can be represented by the azimuth angle of the photovoltaic panel and the angle between the photovoltaic panel and the horizontal direction. In order to obtain as much total irradiance as possible to improve the power output, in some embodiments, the orientation of the photovoltaic panel can be adjusted according to the tracking algorithm. The commonly used tracking algorithm is the sun tracking algorithm, which assumes that the maximum direct irradiance can be obtained when the photovoltaic panel is perpendicular to the sun, and the sun elevation angle and the azimuth angle are calculated by calculating the position of the sun, and the photovoltaic panel is oriented to the sun based on the sun elevation angle and the azimuth angle, which is essentially tracking the sun. However, as described above, the maximum direct irradiance does not mean the maximum total irradiance, and the total irradiance is also affected by the diffuse irradiance and the reflected irradiance. In some embodiments, the tracking algorithm can be improved from tracking the sun to tracking the energy, and the orientation of the photovoltaic panel can be adjusted to a position that can obtain the maximum total irradiance by performing the method for determining the orientation of the photovoltaic panel shown in process 300.

[0064] The tracking algorithm is described by taking FIG. 5 as an example. As shown in FIG. 5, in the left subgraph, in sunny weather, the photovoltaic panel orientation is set according to the daily algorithm, so that the photovoltaic panel 510 is oriented to face the sun, i.e., the photovoltaic panel plane is perpendicular to the direct sunlight direction. In the right subgraph, in cloudy weather, due to the influence of clouds, the proportion of diffuse radiation in the total radiation increases, and the orientation of the photovoltaic panel 510 in sunny weather makes the photovoltaic panel unable to obtain the maximum total radiation. Therefore, the photovoltaic panel 510 can be controlled to rotate from the dashed line position in the right subgraph (i.e., the position of the photovoltaic panel 510 in the left subgraph) to the position of the photovoltaic panel 510 in the right subgraph according to the method for determining the photovoltaic panel orientation shown in the flow 300, at which time the orientation of the photovoltaic panel 510 is the direction in which the photovoltaic panel can receive the maximum total radiation.

[0065] As shown in FIG. 3, the flow 300 includes the following steps. In some embodiments, the flow 300 can be executed by the first computing device 120 and the second computing device 130.

[0066] At step 310, historical data is obtained, and a photovoltaic panel orientation determination model is trained based on the historical data. In some embodiments, step 310 can be executed by the model training module 210.

[0067] The historical data refers to historical data related to radiation, for example, time, location, and corresponding historical radiation data, etc. In some embodiments, the historical data can include time information, geographic location information, historical radiation data corresponding to the time and geographic location, etc. The historical radiation data can include historical horizontal total radiation, historical direct radiation, and historical diffuse radiation, etc. In some embodiments, the historical data can also include historical cloud cover information corresponding to the time and geographic location.

[0068] In some embodiments, the second computing device 130 can obtain the historical data in various ways, for example, by querying from a third-party institution or obtaining from a local database. In some embodiments, the second computing device can train a photovoltaic panel orientation determination model (for example, the photovoltaic panel orientation determination model 600) according to the obtained historical data to obtain a trained photovoltaic panel orientation determination model.

[0069] In some embodiments, the photovoltaic panel orientation determination model can include various machine learning models, for example, neural network models, decision tree models, etc.

[0070] In some embodiments, the input of the photovoltaic panel orientation determination model can include photovoltaic panel data, time information, and geographic location information, etc. in a preset scenario, and the output can include the photovoltaic panel orientation in which the total radiation meets a preset condition. For more information about the photovoltaic panel orientation determination model and how to train the model, please refer to the related description of FIG. 6, which will not be described here.

[0071] At step 320, the photovoltaic panel data under the preset scene is acquired. In some embodiments, step 320 can be performed by the photovoltaic panel data acquisition module 220.

[0072] The preset scene refers to the working scene of the photovoltaic device, which can include various scenes, such as city, desert, grassland, mountain, ocean, etc. The photovoltaic panel data under the preset scene refers to the data under the preset scene related to the photovoltaic panel power generation effect. In some embodiments, the photovoltaic panel data under the preset scene can include total irradiance data and environmental data, etc.

[0073] The total irradiance data represents the total irradiance. In some embodiments, the total irradiance can be the total amount of direct irradiance, diffuse irradiance and reflected irradiance. Since the reflected irradiance accounts for a small proportion in the total irradiance and has too many influencing factors, it is difficult to count, and in some embodiments, the reflected irradiance can be omitted, and the total irradiance can be the total amount of direct irradiance and diffuse irradiance. The horizontal total irradiance refers to the total solar irradiance incident on the horizontal plane. In some embodiments, the total irradiance can be measured by a horizontal irradiance meter installed on the photovoltaic device or the ground, etc., wherein when the horizontal irradiance meter is installed on the photovoltaic device, the irradiance data can be acquired by a sensor installed on the front of the photovoltaic panel, and at this time, the horizontal total irradiance can be considered as the total irradiance.

[0074] The environmental data refers to the data of the surrounding environment related to the irradiance obtained by the photovoltaic panel, such as at least one of the installation height of the photovoltaic panel, the cloud amount, the ground material, etc. In some embodiments, the environmental data can include the ground material of the reflected irradiance corresponding to the photovoltaic panel and the distance of the reflected irradiance, etc. The ground material of the reflected irradiance refers to the material of the ground at the installation position of the photovoltaic panel, which can include various types, such as any one of grassland, cement ground, land, desert, water surface, etc. The distance of the reflected irradiance refers to the distance between the photovoltaic panel and the ground at the installation position of the photovoltaic panel. In some embodiments, the distance of the reflected irradiance can be determined based on the main beam height of the photovoltaic device, and the main beam height refers to the distance from the ground to the center of the rotating shaft of the photovoltaic panel. In some embodiments, the main beam height can be a preset value, such as any one of 2 meters, 3 meters, etc.

[0075] In some embodiments, the first computing device 120 can acquire the photovoltaic panel data under the preset scene, i.e. the photovoltaic panel data when the photovoltaic device is working, in various ways. For example, the horizontal total irradiance when the photovoltaic device is working is acquired by real-time measurement through a horizontal irradiance meter installed on the photovoltaic device. For another example, the main beam height is preset. For another example, the ground material can be set by manual or detected in real time by a detection device installed on the photovoltaic device. For another example, real-time atmospheric data is acquired from a meteorological platform, etc., and the cloud amount is extracted therefrom.

[0076] The cloud cover has a greater impact on the proportion of diffuse radiation in the horizontal total radiation. In some embodiments, the cloud cover of the sky can be observed by installing a fisheye camera on the photovoltaic panel, and then performing binaryzation processing on the obtained image to obtain a binary image. For example, the closer to the position of the sun, the higher the weight; the farther away from the position of the sun, the lower the weight. In some embodiments, the first computing device 120 can perform image recognition on the binary image to obtain the cloud cover.

[0077] In some embodiments, the first computing device 120 can also obtain time information and geographic location information corresponding to the preset scene, i.e., time information and geographic location information when the photovoltaic device is working, wherein the geographic location information can be represented by latitude and longitude. In some embodiments, the time information when the photovoltaic device is working can be provided by a timing device on the photovoltaic device or externally, and the geographic location information can be provided by a positioning device on the photovoltaic device.

[0078] Astronomical irradiance refers to the amount of solar radiation reaching the upper boundary of the earth's atmosphere determined by the astronomical position of the sun to the earth. In some embodiments, the first computing device 120 can also obtain the astronomical irradiance under the preset scene, which can be calculated based on the time information and the geographic location information, for example. FIG. 8 is a comparison diagram of the calculated values of the astronomical irradiance calculated in different ways, wherein the horizontal coordinate is different times of a day, and the vertical coordinate is the value of the astronomical irradiance. JS astronomical irradiance correction is the astronomical irradiance calculated using the above-mentioned method. It can be seen from the figure that the astronomical irradiance calculated by the NASA astronomical irradiance, the Meteonorm astronomical irradiance and the JS astronomical irradiance correction basically coincides, which indicates that the accuracy of the astronomical irradiance calculated using the above-mentioned method is higher.

[0079] In step 330, the photovoltaic panel orientation when the total irradiance meets the preset condition is determined based on the photovoltaic panel data under the preset scene and the trained photovoltaic panel orientation determination model. In some embodiments, step 330 can be performed by the photovoltaic panel orientation determination module 230.

[0080] In some embodiments, the first computing device 120 can input at least part of the photovoltaic panel data under the preset scene into the trained photovoltaic panel orientation determination model to obtain the output photovoltaic panel orientation angle as the photovoltaic panel orientation when the total irradiance meets the preset condition.

[0081] Specifically, the first computing device 120 can input the time information when the photovoltaic device is working, the geographic location information, the horizontal total irradiance, and the astronomical irradiance into the first model to obtain the proportion of the diffuse irradiance in the horizontal total irradiance; and then input the proportion of the diffuse irradiance in the horizontal total irradiance, the distance of the reflected irradiance, and the ground material into the second model to obtain the orientation of the photovoltaic panel at which the total irradiance meets the preset condition, which is represented by an angle. In some embodiments, the preset condition can include that the total irradiance obtained by the photovoltaic panel is maximum at the orientation. In some embodiments, the proportion of the diffuse irradiance in the horizontal total irradiance can be referred to as the diffuse irradiance proportion.

[0082] In some embodiments, the first computing device 120 can also input the cloud amount into the first model, so that a more accurate result of the proportion of the diffuse irradiance in the horizontal total irradiance can be obtained.

[0083] At step 340, the orientation control of the photovoltaic panel under the preset scene is performed based on the determined orientation of the photovoltaic panel. In some embodiments, step 340 can be performed by the photovoltaic panel orientation control module 240.

[0084] In some embodiments, the first computing device 120 can control and adjust the orientation of the photovoltaic panel under the preset scene to the orientation of the photovoltaic panel obtained at step 330, so that the total irradiance obtained by the photovoltaic panel during working can be maximum, thereby improving the power generation.

[0085] In some embodiments of the present specification, by changing the tracking algorithm from tracking the sun to tracking the energy, the orientation of the photovoltaic panel is controlled based on the improved tracking algorithm, so that the photovoltaic panel can obtain more total irradiance, the power generation efficiency is higher, and the economic benefit is better; the orientation of the photovoltaic panel at which the total irradiance is maximum is determined by the machine learning model, which is simple, fast, efficient, and accurate; and the improved tracking algorithm only needs to install a horizontal irradiance instrument, which is simple to deploy and has low cost.

[0086] It should be noted that the above description of the process 300 is only for example and illustration, and does not limit the scope of the present specification. Those skilled in the art can make various modifications and changes to the process 300 under the guidance of the present specification. However, these modifications and changes are still within the scope of the present specification. For example, steps 320-340 can be periodically executed, for example, every 30 minutes; or can be executed after the preset scene changes, for example, when the detected change in the horizontal total irradiance is greater than a threshold value.

[0087] FIG. 6 is a schematic diagram of a photovoltaic panel orientation determination model according to some embodiments of the present specification.

[0088] In some embodiments, the photovoltaic panel orientation determination model can include one or more models, for example, two models. In some embodiments, the photovoltaic panel orientation determination model can include a first model and a second model. As shown in FIG. 6, the photovoltaic panel orientation determination model can include a first model 610 and a second model 650.

[0089] In some embodiments, the first model can include various machine learning models, for example, neural network models, decision tree models, etc.

[0090] In some embodiments, the input of the first model can include time information, geographic location information, horizontal total irradiance and astronomical irradiance corresponding to a preset scene, and the output of the first model can include an irradiance proportion. The irradiance proportion can be a proportion of at least one of diffuse irradiance and reflected irradiance in total irradiance, for example, a proportion of diffuse irradiance in total irradiance, a proportion of a sum of diffuse irradiance and reflected irradiance in total irradiance. In some embodiments, the reflected irradiance can be excluded from the total irradiance, i.e., the total irradiance includes direct irradiance and diffuse irradiance, and the output of the first model can include a proportion of diffuse irradiance in horizontal total irradiance. In this way, the complexity of model calculation can be reduced, and the efficiency of model calculation can be improved. As shown in FIG. 6, the input of the first model 610 can include time 621, latitude and longitude 622, horizontal total irradiance 623 and astronomical irradiance 624, and the output can include a diffuse irradiance proportion 640.

[0091] In some embodiments, the reflected irradiance can be included in the total irradiance, i.e., the total irradiance includes direct irradiance, diffuse irradiance and reflected irradiance, and the output of the first model can include a proportion of a sum of diffuse irradiance and reflected irradiance in horizontal total irradiance. In this way, the accuracy of the model output can be improved.

[0092] In some embodiments, the input of the first model can further include cloud amount under a preset scene. For details about how to obtain time information, geographic location information, cloud amount, horizontal total irradiance and astronomical irradiance corresponding to a preset scene, please refer to the related description of step 320, which will not be repeated here.

[0093] In some embodiments, the astronomical irradiance can not be included in the input of the first model, and the first model can obtain the astronomical irradiance based on time information and geographic location information corresponding to a preset scene.

[0094] The altitude angle, also referred to as the solar altitude, is the angle between the sun's rays and the surface tangent at the location and time. When the sun is directly overhead, the altitude angle is 90°, and the solar radiation intensity is the strongest. When the sun rises and sets, the altitude angle is 0°. The smaller the altitude angle, the greater the degree of sun oblique incidence on the ground, and the smaller the solar radiation intensity. The clearness factor, also referred to as the clearness index, is the ratio of the total solar radiation incident on a horizontal surface to the extraterrestrial radiation, and the value is between 0 and 1. The clearness factor is a comprehensive parameter for describing the influence of the atmosphere on solar shortwave radiation, and can not only reflect the solar radiation intensity reaching the ground, but also reflect the atmospheric conditions and the influence of solar radiation when the solar radiation penetrates the atmosphere. In some embodiments, the input of the first model can include the altitude angle and the clearness factor. The altitude angle can be calculated based on the time information and the geographic location information, and the clearness factor can be the ratio of the horizontal total radiation to the extraterrestrial radiation. In some embodiments, the first model can calculate the altitude angle according to the input time information and the geographic location information, and calculate the clearness factor according to the input horizontal total radiation and the extraterrestrial radiation. As shown in FIG. 6, the first computing device 120 can obtain the altitude angle 631 according to the time 621 and the latitude and longitude 622, and obtain the clearness factor 632 according to the horizontal total radiation 623 and the extraterrestrial radiation 624, and input the altitude angle 631 and the clearness factor 632 into the first model 610.

[0095] In some embodiments, the first computing device 120 can obtain the proportion of diffuse radiation in the horizontal total radiation in other ways, for example, through the OLSETH-SKARTVEIT radiation model, etc. As shown in FIG. 7A is a schematic diagram of the OLSETH-SKARTVEIT radiation model, through which the corresponding proportion of diffuse radiation under different altitude angles and clearness factors can be calculated. The left vertical coordinate in the figure is the proportion of diffuse radiation, denoted as where H d is the diffuse radiation, and H g is the horizontal total radiation; the right vertical coordinate is the altitude angle, and the horizontal coordinate is the clearness factor, denoted as k. As shown in FIG. 7B is a comparison diagram of the calculated value and the actual value of the proportion of diffuse radiation obtained using the OLSETH-SKARTVEIT radiation model. The horizontal coordinate is the clearness factor, and the vertical coordinate is the proportion of diffuse radiation. 710 is the actual value of the proportion of diffuse radiation corresponding to different clearness factors under the same altitude angle in the historical radiation data, and 720 is the calculated value of the proportion of diffuse radiation corresponding to different clearness factors under a certain altitude angle using the OLSETH-SKARTVEIT radiation model. It can be seen that the calculated value is a series of discrete points, and the fitting degree with the curve representing the actual value is not high.

[0096] In some embodiments, the second model can include various machine learning models, such as neural network models, decision tree models, etc.

[0097] In some embodiments, the input of the second model can include the output of the first model, the distance of the reflected irradiance, and the ground material of the reflected irradiance, etc., and the output of the second model can include the second angle, which is the angle of the orientation of the photovoltaic panel when the total irradiance is the highest. As shown in FIG. 6, the input of the second model 650 includes the diffuse irradiance proportion 640 output by the first model 610, the main beam height 662, and the ground material 663, and the output includes the second angle 680, wherein the main beam height 662 represents the distance of the reflected irradiance, the ground material 663 represents the ground material of the reflected irradiance, and the second angle 680 is the angle of the orientation of the photovoltaic panel when the total irradiance is the highest, corresponding to the time 621, the latitude and longitude 622, the horizontal total irradiance 623, the main beam height 662, and the ground material 663.

[0098] In some embodiments, the first model can be obtained by pre-training, i.e., a trained first model is obtained by training, which can be performed before training the second model. In some embodiments, the pre-training of the first model can be performed completely separately from the photovoltaic device, without waiting for the photovoltaic device to be installed before training.

[0099] In some embodiments of the present specification, the model for obtaining the output irradiance proportion is obtained by a pre-training mechanism, which reduces the cost of model training and greatly reduces the deployment time of intelligent algorithms.

[0100] In some embodiments, the training samples of the first model can include time information, geographic location information, and historical data such as historical horizontal total irradiance data, and the training label can include the proportion of diffuse irradiance in the horizontal total irradiance. The second computing device 130 can input the training samples such as time information, geographic location information, and historical horizontal total irradiance data into the first model to obtain the output value of the proportion of diffuse irradiance in the horizontal total irradiance, compare the output value with the training label, adjust the parameters of the first model according to the difference between the two, and perform the above operations in a loop until a preset condition is met. The preset condition can include at least one of the difference between the output value and the training label being less than a preset threshold, reaching a preset number of cycles, etc.

[0101] In some embodiments, the historical data can further include historical cloud cover data, and the training samples of the first model can further include the historical cloud cover data. In some embodiments, the historical data can further include historical astronomical irradiance data, wherein the historical astronomical irradiance data can be calculated based on the time information and the geographic location information as the training samples.

[0102] As shown in FIG. 6, the second computing device 130 can input historical irradiation data 611, time 612, and latitude and longitude 613 as training samples into the first model 610 for pre-training to obtain the trained first model 610, where the historical irradiation data 611 can include historical horizontal total irradiation data, and the time 612 and the latitude and longitude 613 are time information and geographic information corresponding to the historical irradiation data 611, respectively.

[0103] In some embodiments, the historical data can further include other environment-related data, for example, historical temperature and humidity, etc., which can affect the atmospheric refractive index and cause different degrees of influence on direct irradiation, diffuse irradiation, and reflected irradiation. The training samples of the first model can further include these data. As shown in FIG. 6, the training samples of the first model 610 can further include temperature and humidity 614.

[0104] In some embodiments of the present specification, the model outputting the irradiation proportion can be obtained by using the historical data including the historical irradiation data, without the need to collect data in real time or collect data for the actual application of the photovoltaic panel, which is simple, efficient, and adaptable; by adding at least one of the historical cloud cover and temperature and humidity into the training samples of the model, the accuracy of the output of the trained model is improved, and a more accurate output of the irradiation proportion can be obtained when used.

[0105] In some embodiments, the pre-trained first model can be used to perform offline training of the second model in a plurality of simulated preset scenes, where the preset scenes can include a plurality of scenes of different ground materials. The offline training of the second model means that the training is performed when the photovoltaic system is not working, and the preset scenes are simulated by various random parameters, for example, different girder heights, ground materials, irradiation amounts, and angles of the photovoltaic panel can be randomly combined to obtain a plurality of training samples for training of the second model.

[0106] In some embodiments, the training samples of the second model can include the output of the first model (for example, the proportion of diffuse irradiation in the horizontal total irradiation, the proportion of the sum of diffuse irradiation and reflected irradiation in the horizontal total irradiation), the distance of reflected irradiation, the ground material of reflected irradiation, and the inclined surface irradiation of the photovoltaic panel, etc., and the training label of the second model can include the angle of the photovoltaic panel corresponding to the highest total irradiation amount. As shown in FIG. 6, the second computing device 130 can perform offline training of the second model by using the diffuse irradiation proportion 640, the inclined surface irradiation 651, the sample girder height 652, and the sample ground material 653 as training samples.

[0107] The tilted surface irradiance of the photovoltaic panel includes the irradiance received by the front and back of the photovoltaic panel, which can be collected by the sensor. Since the photovoltaic panel can receive reflected irradiance, the range of the tilted surface irradiance is larger than that of the horizontal irradiance data. In some embodiments, the tilted surface irradiance measurement can be obtained by a separate horizontal irradiance meter, which is only used during the training of the second model. The horizontal irradiance meter obtains solar radiation through sensors installed on the front and back of the photovoltaic panel. By using a separate horizontal irradiance meter to measure the tilted surface irradiance only during training, the cost of using the model is reduced.

[0108] In some embodiments, the second computing device 130 can input the training samples such as the proportion of diffuse irradiance in the horizontal total irradiance, the distance of reflected irradiance, the ground material of reflected irradiance, and the tilted surface irradiance of the photovoltaic panel into the first model to obtain the output value of the angle of the photovoltaic panel orientation. The output value is compared with the training label, and the parameters of the second model are adjusted according to the difference between the two. The above operations are repeated until a preset condition is met. The preset condition can include at least one of the difference between the output value and the training label being less than a preset threshold, reaching a preset number of cycles, and the like.

[0109] If the output of the first model is the proportion of diffuse irradiance in the horizontal total irradiance, although the direct irradiance and the diffuse irradiance account for a large proportion of the total irradiance, the angle obtained according to the first model is not necessarily the best angle because the reflected irradiance is not considered. Therefore, the influence of the reflected irradiance can be added to the second model.

[0110] In some embodiments, during the training process of the second model, the second computing device 130 can determine a first angle of the photovoltaic panel orientation based on the output of the first model and the current irradiance data. The first angle is determined based on the proportion of diffuse irradiance in the horizontal total irradiance, which is used as a reference for the angle of the photovoltaic panel orientation. The angle only considers the direct irradiance and the diffuse irradiance. As shown in FIG. 6, the second computing device 130 can determine the first angle 661 as a reference based on the proportion of diffuse irradiance 640.

[0111] In some embodiments, the second computing device 130 can optimize the first angle to obtain a second angle of the photovoltaic panel orientation as the output of the second model.

[0112] In some embodiments, the second computing device 130 can start from the first angle, rotate by a preset angle and iterate multiple times to obtain multiple candidate angles, i.e., increase or decrease the first angle by a preset step to obtain multiple candidate angles, wherein each candidate angle corresponds to an irradiation of an inclined surface. In some embodiments, the range of candidate angles can be a preset value, for example, no more than 10° above and below the first angle. In some embodiments, the preset step can be a fixed value, for example, 0.5°, 1°, 2°, etc. In some embodiments, the preset step can be greater than or equal to 1°, which can avoid too many adjustments caused by too small step size. In some embodiments, the preset step can be a variable step, for example, the step size is a larger angle (e.g., 2°, etc.) at the beginning, and when the difference between the current and previous two output angles tends to be stable, the step size is changed to a smaller angle (e.g., 1°, etc.) to improve the accuracy of the final output.

[0113] In some embodiments, the second computing device 130 can determine the candidate angle corresponding to the highest total irradiance from the multiple candidate angles as the second angle, i.e., the output of the second model. Specifically, for each candidate angle, the second computing device 130 can obtain the inclined surface irradiation corresponding to the candidate angle, and then determine the maximum value from all inclined surface irradiations, and determine the candidate angle corresponding to the maximum value as the second angle. If the first model is not pre-trained to obtain a trained first model before training the second model, and the first angle output by the first model is determined as the basis for adjustment, it is possible that the first angle and the angle corresponding to the highest total irradiance are too far apart, i.e., even after training, the angle corresponding to the highest total irradiance cannot be found.

[0114] As shown in FIG. 6, when training the second model 650, the second computing device 130 can start from the first angle 661, and through step 671, the first angle 661 is operated by ± a (a is a preset step) to obtain multiple turning angles 654 as candidate angles, wherein the range of candidate angles can be 10° above and below the first angle 661, each turning angle 654 corresponds to an inclined surface irradiation 651, and the maximum value among all inclined surface irradiations 651 is determined, and the turning angle 654 corresponding to the maximum value is determined as the output value of the second model, i.e., the angle of the photovoltaic panel corresponding to the highest total irradiance.

[0115] In some embodiments, if the first model outputs the proportion of the sum of diffuse irradiance and reflected irradiance in total irradiance, the second model can directly determine the photovoltaic panel orientation angle according to the output, and take the angle as the output of the second model. At this time, the first model and the second model can be combined into one model, or the second model is omitted, and the first model directly outputs the angle of the photovoltaic panel orientation corresponding to the highest total irradiance. Taking FIG. 6 as an example, assuming that the output of the first model 610 is the proportion of the sum of diffuse irradiance and reflected irradiance in total irradiance, and the second model 650 obtains the first angle 661 according to the proportion, then the first angle 661 can be taken as the output of the second model 650, that is, the second angle 680.

[0116] In some embodiments, the photovoltaic panel orientation determination model can only include the second model. The second model can determine a plurality of candidate angles according to the input time information, geographic location information, horizontal total irradiance data, distance of reflected irradiance, and ground material of reflected irradiance of the photovoltaic device in operation, determine the inclined surface irradiance corresponding to each candidate angle, and then determine the candidate angle corresponding to the maximum value of all inclined surface irradiance as the output of the second model, that is, the angle of the photovoltaic panel orientation corresponding to the highest total irradiance. Among them, the candidate angle can be determined in various ways. For example, a plurality of angles are randomly selected as candidate angles within the value range of the photovoltaic panel orientation angle. For another example, a plurality of angles are determined as candidate angles by taking a preset angle (for example, 0°, 10 degrees, etc.) as a starting point and increasing by a preset step within the value range of the photovoltaic panel orientation angle.

[0117] The beneficial effects that the embodiments of the present specification can bring include but are not limited to: (1) by changing the tracking algorithm from tracking the sun to tracking the energy, the improved tracking algorithm is used to control the orientation of the photovoltaic panel, so that the photovoltaic panel can obtain a larger total irradiance, the power generation efficiency is higher, and the economic benefit is better; (2) the orientation of the photovoltaic panel when the total irradiance is maximum is determined through the machine learning model, which is simple, fast, efficient and accurate; (3) the improved tracking algorithm described above only needs to install one horizontal irradiation instrument, which is simple to deploy and has low cost; (4) the model outputting the irradiation proportion is obtained through the pre-training mechanism, which reduces the cost of model training and greatly reduces the deployment time of the intelligent algorithm; (5) the model outputting the irradiation proportion can be obtained through the historical data including the historical irradiation data, without the need for real-time data collection or data collection for the actual application photovoltaic panel, which is simple, efficient and adaptable; (6) by adding at least one of historical cloud cover and temperature and humidity into the training sample of the model outputting the irradiation proportion, the accuracy of the output of the trained model is improved, and a more accurate output irradiation proportion can be obtained when used. It should be noted that different embodiments can have different beneficial effects, and in different embodiments, the beneficial effects can be any one or a combination of the above, or any other beneficial effects that can be obtained.

[0118] The above detailed description has described the basic concepts, and it is obvious that the above detailed description is only used as an example and does not limit the present specification. Although the skilled in the art can make various modifications, improvements and corrections to the present specification, such modifications, improvements and corrections are suggested in the present specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of the present specification.

[0119] At the same time, the present specification uses specific words to describe the embodiments of the present specification. As "one embodiment", "an embodiment", and / or "some embodiments" means a certain feature, structure or characteristic related to at least one embodiment of the present specification. Therefore, it should be emphasized and noted that the "an embodiment" or "one embodiment" or "one alternative embodiment" mentioned in different places in the present specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present specification can be properly combined.

[0120] Furthermore, the order of the processing elements and sequences described in this specification are not intended to be construed as a limitation, unless specifically stated, but are included to provide a complete description of one or more embodiments of the present specification. Regardless of the particular sequence of processing elements and sequences, however, the description herein of a process should be understood to include any and all combinations of one or more elements, and sequences that can be perceived as either open-ended or specific.

[0121] It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting, save for any specifically set out limitation of the claims. It must be noted that as used herein, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise. The terms "comprises" and / or "comprising," or "includes" and / or "including" when used in this specification, specify the presence of stated features, regions, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, regions, integers, steps, operations, elements, components, and / or groups thereof.

[0122] Some embodiments use numerals to describe components, quantities of attributes. It should be understood that such numerals used in the description of embodiments are, in some examples, modified by the adjectives "about", "approximately", or "generally". Unless otherwise stated, "about", "approximately", or "generally" indicates that the stated numerical value allows for a ±20% variation. Accordingly, numerical parameters in the description and claims are approximations, and can vary depending upon the requirements of a given embodiment. In some embodiments, numerical parameters are determined by the use of common rounding techniques. Although the numerical ranges and parameters setting forth the broad scope of the embodiments of the specification are approximations, unless otherwise indicated, these numerical values are to be understood as being modified in all instances by a term selected from the terms of "about", "approximately", and "generally".

[0123] Every patent, patent application, publication, document, article, book, specification, publication, or other material cited in this specification is hereby incorporated by reference in its entirety for all purposes to the same extent as if each individual publication, document, article, book, or other material was specifically and individually indicated to be incorporated by reference in its entirety for its full scope of teaching and disclosure. For contrary or inconsistent disclosure between this specification and the incorporated-by-reference material, this specification controls. Note that if there is a discrepancy between the definitions, descriptions, and / or terminology used in this specification and those used in the incorporated-by-reference material, the definitions, descriptions, and / or terminology used in this specification control.

[0124] Finally, it should be understood that the embodiments described herein are only given by way of example and that other modifications can occur to persons skilled in the art. Therefore, the scope of the present description is not intended to be limited to the embodiments described herein but is only limited by the claims that follow.

Claims

1. A method of determining the orientation of a photovoltaic panel, characterized in that, The method comprises: acquiring historical data, and training a photovoltaic panel orientation determination model based on the historical data, wherein the historical data comprises historical horizontal total irradiance data; acquiring photovoltaic panel data under a preset scene, wherein the photovoltaic panel data under the preset scene comprises total irradiance data and environmental data, the total irradiance data represents a total irradiance amount, the total irradiance amount is a total amount of direct irradiance, diffuse irradiance and reflected irradiance, and the environmental data comprises a ground material of the reflected irradiance and a distance of the reflected irradiance corresponding to the photovoltaic panel; determining a photovoltaic panel orientation in which the total irradiance amount meets a preset condition based on the photovoltaic panel data under the preset scene and the trained photovoltaic panel orientation determination model; controlling the orientation of the photovoltaic panel under the preset scene based on the determined photovoltaic panel orientation.

2. The method of claim 1, wherein, The photovoltaic panel orientation determination model comprises a first model and a second model, wherein the output of the first model comprises a proportion of the diffuse irradiance in the horizontal total irradiance, the input of the second model comprises the output of the first model, and the output of the second model comprises an angle of a photovoltaic panel orientation.

3. The method of claim 2, wherein, The first model is obtained through pre-training.

4. The method of claim 3, wherein, The training sample of the first model comprises time information, geographical position information and the historical horizontal total irradiance data, and the training label of the first model comprises the proportion of the diffuse irradiance in the horizontal total irradiance.

5. The method of claim 3, wherein, The pre-trained first model is used for offline training of the second model under a plurality of simulated preset scenes, and the plurality of simulated preset scenes comprise a plurality of scenes with different ground materials.

6. The method of claim 2, wherein, The training sample of the second model comprises the proportion of the diffuse irradiance in the horizontal total irradiance output by the first model, the distance of the reflected irradiance, the ground material of the reflected irradiance and photovoltaic panel inclined surface irradiance, and the training label of the second model comprises an angle of the photovoltaic panel orientation corresponding to the highest total irradiance amount.

7. The method of claim 6, wherein, The training process of the second model comprises: determining a first angle of a photovoltaic panel orientation based on the output of the first model and current irradiance data; optimizing the first angle to obtain a second angle of a photovoltaic panel orientation as the output of the second model.

8. The method of claim 7, wherein, The optimization of the first angle to obtain a second angle of a photovoltaic panel orientation comprises: starting from the first angle, rotating by a preset angle and iterating multiple times to obtain a plurality of candidate angles, and determining the highest total irradiance amount corresponding to the plurality of candidate angles as the second angle.

9. The method of claim 1, wherein, The distance of the reflected irradiance is determined based on the height of a main beam.

10. A system for determining the orientation of a photovoltaic panel, characterized by, The method comprises a model training module, a photovoltaic panel data acquisition module, a photovoltaic panel orientation determination module and a photovoltaic panel orientation control module; the model training module is configured to acquire historical data, and train a photovoltaic panel orientation determination model based on the historical data, wherein the historical data comprises historical horizontal total irradiance data; The photovoltaic panel data acquisition module is configured to acquire photovoltaic panel data of the photovoltaic panel under a preset scene, wherein the photovoltaic panel data under the preset scene includes total irradiance data and environmental data, the total irradiance data represents total irradiance, the total irradiance is a total amount of direct irradiance, diffuse irradiance and reflected irradiance, and the environmental data includes ground material of the reflected irradiance corresponding to the photovoltaic panel and distance of the reflected irradiance; The photovoltaic panel orientation determination module is configured to determine a photovoltaic panel orientation in which the total irradiance meets a preset condition based on the photovoltaic panel data under the preset scene and the trained photovoltaic panel orientation determination model. The photovoltaic panel orientation control module is configured to control orientation of the photovoltaic panel under the preset scene based on the determined photovoltaic panel orientation.

11. The system of claim 10, wherein, The photovoltaic panel orientation determination model includes a first model and a second model, wherein an output of the first model includes a proportion of the diffuse irradiance in horizontal total irradiance, an input of the second model includes the output of the first model, and an output of the second model includes an angle of a photovoltaic panel orientation.

12. The method of claim 11, wherein, The first model is obtained through pre-training.

13. The system of claim 12, wherein, A training sample of the first model includes time information, geographic location information and the historical horizontal total irradiance data, and a training label of the first model includes the proportion of the diffuse irradiance in the horizontal total irradiance.

14. The system of claim 12, wherein, The pre-trained first model is configured to train the second model under a plurality of simulated preset scenes, and the plurality of simulated preset scenes include a plurality of scenes with different ground materials.

15. The system of claim 11, wherein, A training sample of the second model includes the proportion of the diffuse irradiance in the horizontal total irradiance output by the first model, distance of the reflected irradiance, ground material of the reflected irradiance and photovoltaic panel inclined surface irradiance, and a training label of the second model includes an angle of the photovoltaic panel orientation corresponding to the highest total irradiance.

16. The system of claim 15, wherein, The training process of the second model includes: determining a first angle of a photovoltaic panel orientation based on the output of the first model and current irradiance data; optimizing the first angle to obtain a second angle of the photovoltaic panel orientation as an output of the second model.

17. The system of claim 16, wherein, The optimization of the first angle to obtain the second angle of the photovoltaic panel orientation includes: starting from the first angle, rotating a preset angle and iterating multiple times to obtain a plurality of candidate angles, and determining the highest total irradiance corresponding to the plurality of candidate angles as the second angle.

18. The system of claim 10, wherein, The distance of the reflected irradiance is determined based on a main beam height.

19. An apparatus for determining a photovoltaic panel orientation, comprising a processor configured to perform the method of any one of claims 1-9.

20. A computer-readable storage medium, the storage medium storing computer instructions, when a computer reads the computer instructions in the storage medium, the computer executes the method of any one of claims 1-9.

Images