A method and device for generating a synthetic image of a photovoltaic installation at sea

Abstract

The application is suitable for the technical field of image processing, and provides a generation method and device of a marine photovoltaic composite image, which are used for improving the authenticity and reliability of the marine photovoltaic composite image. The method comprises the following steps: acquiring environmental parameters of a target sea area and a marine photovoltaic initial image, the environmental parameters comprising chlorophyll concentration, suspended matter concentration and wind speed vector, and the marine photovoltaic initial image being a three-dimensional geometric image of a marine photovoltaic device arranged in the target sea area; inputting the environmental parameters into a preset four-stream radiative transfer model to output underwater radiance distribution and water path radiance distribution; calculating a sea surface specular reflection component according to the wind speed vector, and calculating a water body volume scattering component according to the chlorophyll concentration and the suspended matter concentration; and performing fusion processing on the marine photovoltaic initial image by using the underwater radiance distribution, the water path radiance distribution, the sea surface specular reflection component and the water body volume scattering component to obtain the marine photovoltaic composite image.

Description

Technical Field

[0001] This application belongs to the field of image processing technology, and in particular relates to a method and apparatus for generating composite images of marine photovoltaic systems. Background Technology

[0002] Offshore photovoltaic (PV) power has become an important development direction in the field of new energy development due to its advantages such as small footprint, proximity to load centers, and complementarity with offshore wind power. However, because offshore PV equipment is far from the coast, traditional manual inspections and drone-based visible light monitoring are difficult to obtain stable quality images due to the influence of weather, sunlight, and complex sea conditions.

[0003] Currently, to obtain stable images, 3D modeling and rendering of the geometric projection and ideal lighting model of offshore photovoltaic (PV) equipment are typically performed to generate synthetic images. However, when the PV panels in offshore PV equipment are installed underwater, the water body alters some characteristics of light. For example, the water body absorbs or scatters incident sunlight, affecting its radiation intensity or incident angle. The absorption and reflection processes of light by the PV panels will change under the influence of the water body. The synthetic images generated based on the current ideal lighting model cannot reproduce these changes, resulting in significant differences between the generated synthetic images and the actual offshore PV operation scenario. Consequently, the realism and reliability of the synthetic images cannot meet the operational monitoring needs of offshore PV. Summary of the Invention

[0004] This application provides a method and apparatus for generating composite images of marine photovoltaic systems, which can improve the realism and reliability of composite images of marine photovoltaic systems.

[0005] The first aspect of this application provides a method for generating composite images of marine photovoltaic systems, comprising: The environmental parameters and initial images of the marine photovoltaic system in the target sea area are obtained. The environmental parameters include chlorophyll concentration, suspended matter concentration and wind speed vector. The initial images of the marine photovoltaic system are three-dimensional geometric images of the marine photovoltaic equipment installed in the target sea area. The environmental parameters are input into a preset four-flow radiative transfer model, and the underwater radiance distribution and the water body path radiance distribution are output. The sea surface specular reflection component is calculated based on the wind speed vector, and the water body volume scattering component is calculated based on the chlorophyll concentration and the suspended matter concentration. The initial marine photovoltaic image is fused using the underwater radiance distribution, the water path radiance distribution, the sea surface specular reflection component, and the water volume scattering component to obtain a composite marine photovoltaic image.

[0006] Optionally, after fusing the initial marine photovoltaic image using the underwater radiance distribution, the water path radiance distribution, the sea surface specular reflection component, and the water volume scattering component to obtain a composite marine photovoltaic image, the method further includes: The first shadow topology of the initial image of the marine photovoltaic system and the second shadow topology of the composite image of the marine photovoltaic system are extracted respectively. Determine whether the shadow matching degree between the first shadow topology and the second shadow topology is less than a preset matching degree threshold; When the shadow matching degree is less than the preset matching degree threshold, the preset four-stream radiative transfer model is updated based on the deviation between the shadow matching degree and the preset matching degree threshold. The underwater radiance distribution and water path radiance distribution output by the updated preset four-stream radiative transfer model are used to correct the marine photovoltaic composite image. This process is repeated until the latest shadow matching degree is greater than or equal to the preset matching threshold.

[0007] Optionally, the step of fusing the initial marine photovoltaic image using the underwater radiance distribution, the water path radiance distribution, the sea surface specular reflection component, and the water volume scattering component to obtain a composite marine photovoltaic image includes: Based on the underwater radiance distribution and the reflectivity of the photovoltaic panels in the marine photovoltaic equipment, a first radiation budget equation is established for the front side of the photovoltaic panel; A second radiation budget equation for the back of the photovoltaic panel is established based on the water body path radiance distribution and the water body volume scattering component. Spectral curves were plotted based on the first radiation budget equation, the second radiation budget equation, and the sea surface specular reflection component. The initial image of the marine photovoltaic system and the spectral curve are fused using a standard color matching function to obtain a composite image of the marine photovoltaic system.

[0008] Optionally, calculating the specular reflection component of the sea surface based on the wind speed vector includes: Calculate the root mean square error of the slope along the wind direction and the root mean square error of the slope perpendicular to the wind direction based on the wind speed vector; The joint probability density function of wave slope driven by the wind speed vector is calculated based on the mean square error of the slope along the wind direction and the mean square error of the slope perpendicular to the wind direction. Calculate the set of microfacet normal elements required for specular reflection that satisfies the solar incident direction and the observation direction, and calculate the corresponding Fresnel reflection coefficient based on the zenith angle of each microfacet normal vector in the set of microfacet normal elements in the solar incident direction and the water reflectivity of the target sea area. The sea surface specular reflection component is obtained by integrating the joint probability density function of wave slope, the Fresnel reflection coefficient corresponding to each micro-surface normal vector, and the geometric projection parameter corresponding to each micro-surface normal vector.

[0009] Optionally, calculating the water volume scattering component based on the chlorophyll concentration and the suspended matter concentration includes: The first total scattering intensity and the first scattering intensity distribution of phytoplankton are determined based on the chlorophyll concentration, and the second total scattering intensity and the second scattering intensity distribution of suspended matter are determined based on the suspended matter concentration. The first volume scattering function value of the phytoplankton and the second volume scattering function value of the suspended matter are obtained by calculating the product of the first total scattering intensity and the first scattering intensity distribution, and the product of the second total scattering intensity and the second scattering intensity distribution, respectively. The water body scattering component is obtained by summing the values ​​of the first and second body scattering functions.

[0010] Optionally, the environmental parameters also include the water body optical absorption coefficient, water body scattering coefficient, solar zenith angle, and solar azimuth angle. The step of inputting the environmental parameters into a preset four-flow radiative transfer model and outputting the underwater radiance distribution and the water body path radiance distribution includes: The water body category of the target sea area is determined based on the chlorophyll concentration and the suspended matter concentration. The vertical discrete layer structure and asymmetric phase function of the preset four-stream radiative transfer model are determined based on the water body category. Each vertical discrete layer in the vertical discrete layer structure adopts the four-stream discrete ordinate format. The radiation intensity of each vertical discrete layer in the vertical discrete layer structure is iteratively calculated based on the water body optical absorption coefficient, the water body scattering coefficient, and the asymmetric phase function. The downward and upward radiance fields are determined based on the radiation intensity at the convergence of each vertical discrete layer. Acquire aerosol optical thickness data of the target sea area, calculate the downward sky radiance distribution based on the aerosol optical thickness data, the solar zenith angle and the solar azimuth angle, and determine the underwater radiance distribution based on the downward radiance field and the downward sky radiance distribution; The water body path radiance distribution is determined based on the upward radiance field.

[0011] Optionally, the iterative calculation of the radiation intensity of each vertical discrete layer in the vertical discrete layer structure based on the water body optical absorption coefficient, the water body scattering coefficient, and the asymmetric phase function includes: The wind speed vector is decomposed into wind speed magnitude and wind direction, and the mean square slope of the sea surface in the target sea area is calculated based on the wind speed magnitude and wind direction obtained from the decomposition. The effective transmittance of the water-air interface in the target sea area is calculated based on the mean square slope of the sea surface. The radiation intensity of each vertical discrete layer in the vertical discrete layer structure is iteratively calculated based on the water body optical absorption coefficient, the water body scattering coefficient, and the asymmetric phase function. During the iterative calculation, the effective transmittance is used to correct the radiation intensity calculated for each vertical discrete layer.

[0012] A second aspect of this application provides an apparatus for generating composite images of marine photovoltaic data, comprising: The acquisition unit is used to acquire environmental parameters of the target sea area and initial images of marine photovoltaic systems. The environmental parameters include chlorophyll concentration, suspended matter concentration and wind speed vector. The initial images of marine photovoltaic systems are three-dimensional geometric images of marine photovoltaic equipment installed in the target sea area. The input unit is used to input the environmental parameters into a preset four-flow radiative transfer model and output the underwater radiance distribution and the water body path radiance distribution. The calculation unit is used to calculate the specular reflection component of the sea surface based on the wind speed vector, and to calculate the volume scattering component of the water body based on the chlorophyll concentration and the suspended matter concentration. The fusion unit is used to fuse the initial marine photovoltaic image using the underwater radiance distribution, the water path radiance distribution, the sea surface specular reflection component, and the water volume scattering component to obtain a composite marine photovoltaic image.

[0013] A third aspect of this application provides an electronic device, comprising: Processor, memory, input / output units, and bus; The processor is connected to the memory, the input / output unit, and the bus; The memory stores a program that the processor calls to execute a method for generating a marine photovoltaic composite image, as described in the first aspect and any optional method in the first aspect.

[0014] A fourth aspect of this application provides a computer-readable storage medium comprising: instructions that, when executed on a computer, cause the computer to perform a method for generating a marine photovoltaic composite image as described in the first aspect and any optional method of the first aspect.

[0015] As can be seen from the above technical solutions, this application has the following effects: The method for generating a composite image of marine photovoltaic (PV) equipment provided in this application first acquires environmental parameters of the target sea area and an initial image of the PV system. The environmental parameters include chlorophyll concentration, suspended matter concentration, and wind speed vector. The initial image of the PV system is a three-dimensional geometric image of the PV equipment located in the target sea area. Then, the environmental parameters are input into a preset four-flow radiative transfer model, and the underwater radiance distribution and the water body path radiance distribution are output. Next, the specular reflection component of the sea surface is calculated based on the wind speed vector, and the volume scattering component of the water body is calculated based on the chlorophyll concentration and suspended matter concentration. Finally, the initial image of the PV system is fused using the underwater radiance distribution, the water body path radiance distribution, the specular reflection component of the sea surface, and the volume scattering component of the water body to obtain a composite image of the PV system. Based on this, the underwater radiance distribution and water path radiance distribution output by the preset four-flow radiative transfer model can be used to restore the water radiative transfer characteristics of the target sea area. At the same time, the two major optical components, the sea surface specular reflection component and the water body volume scattering component, are quantitatively calculated. This allows the fused marine photovoltaic composite image to restore the real optical transfer characteristics of marine photovoltaic equipment underwater, thereby improving the fit between the marine photovoltaic composite image and the actual marine photovoltaic operation scenario, and thus enhancing the authenticity and reliability of the marine photovoltaic composite image. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for generating a composite image of marine photovoltaic data according to this application. Figure 2 This is a structural schematic diagram of the initial image of the marine photovoltaic system in this application from a first-view perspective; Figure 3 This is a structural schematic diagram of the initial image of marine photovoltaic in this application from a second-view perspective; Figure 4 This is a schematic diagram of the structure of a device for generating composite images of marine photovoltaic data according to this application; Figure 5 This is a schematic diagram of the electronic device in this application. Detailed Implementation

[0017] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0018] It should be understood that, when used in this application specification, the term "comprising" indicates the presence of the described feature, integral, step, operation, element, and / or component, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof.

[0019] It should also be understood that the term “and / or” as used in this application specification means any combination of one or more of the associated listed items, as well as all possible combinations, and includes such combinations.

[0020] As used in this application specification, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [the described condition or event] is detected" may be interpreted, depending on the context, as "once determined," "in response to determination," "once [the described condition or event] is detected," or "in response to detection of [the described condition or event]."

[0021] Furthermore, in the description of this application, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0022] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0023] Offshore photovoltaic (PV) power has become an important development direction in the field of new energy development due to its advantages such as small footprint, proximity to load centers, and complementarity with offshore wind power. However, because offshore PV equipment is far from the coast, traditional manual inspections and drone-based visible light monitoring are difficult to obtain stable quality images due to the influence of weather, sunlight, and complex sea conditions.

[0024] Currently, to obtain stable images, 3D modeling and rendering of the geometric projection and ideal lighting model of offshore photovoltaic (PV) equipment are typically performed to generate synthetic images. However, when the PV panels in offshore PV equipment are installed underwater, the water body alters some characteristics of light. For example, the water body absorbs or scatters incident sunlight, affecting its radiation intensity or incident angle. The absorption and reflection processes of light by the PV panels will change under the influence of the water body. The synthetic images generated based on the current ideal lighting model cannot reproduce these changes, resulting in significant differences between the generated synthetic images and the actual offshore PV operation scenario. Consequently, the realism and reliability of the synthetic images cannot meet the operational monitoring needs of offshore PV.

[0025] To address the aforementioned issues, this application provides a method and apparatus for generating composite images of marine photovoltaic systems, which enhances the realism and reliability of such images.

[0026] Please see Figure 1 As shown in the embodiment of this application, a method for generating a composite image of marine photovoltaic data includes: 101. Obtain environmental parameters and initial images of marine photovoltaic systems in the target sea area. The environmental parameters include chlorophyll concentration, suspended matter concentration, and wind speed vector. The initial images of marine photovoltaic systems are three-dimensional geometric images of marine photovoltaic equipment located in the target sea area.

[0027] Specifically, the target sea area is the marine area where the offshore photovoltaic equipment is located. The definition of the target sea area can be set according to the actual coverage area of ​​the offshore photovoltaic equipment, and is not limited here.

[0028] Specifically, a joint observation array of anchored optical buoys and wave buoys can be deployed in the target sea area. The optical buoys are equipped with downlink and uplink hyperspectral radiometers, which continuously collect irradiance spectral sequences of the water surface and multiple underwater depths within a preset sampling period. The collected irradiance spectral sequences are then inverted using inherent optical characteristic inversion algorithms (such as semi-analysis algorithms) to obtain chlorophyll concentration and suspended matter concentration. The wave buoys are equipped with three-dimensional ultrasonic anemometers to detect wind speed and direction blowing towards the sea surface, obtaining the wind speed vector. It is understood that the environmental parameters of the target sea area include, but are not limited to, chlorophyll concentration, suspended matter concentration, and wind speed vector; for example, they may also include the water body optical absorption coefficient or water body scattering coefficient, which are not limited here.

[0029] It should be noted that before obtaining the initial image of the offshore photovoltaic (PV) system, the Building Information Modeling (BIM) parameters of the PV equipment can be acquired first. These BIM parameters are then used to generate the initial image. Specifically, the IFC format BIM file of the PV equipment is read from a pre-defined database, and all components within the BIM file are parsed. Based on predefined IFC types or custom attribute sets, component objects that are similar to or related to the PV panels, such as supports, beams, and connectors, are selected. This allows for the construction of a 3D geometric model of the PV equipment based on spatial location information such as the PV panel installation tilt angle, PV panel orientation angle, and the spatial layout of the support structure. Simultaneously, the spectral reflectance curves of materials used in the PV equipment components, such as anti-reflective coated glass, ethylene-vinyl acetate copolymer (EVA), and solar cells, are extracted from the material library within the BIM file. Finally, the spectral reflectance curves are mapped onto the 3D geometric model of the PV equipment to generate the initial image. Understandably, since the 3D geometric model is built based on BIM files, different virtual camera positions, viewing angles, and light source directions can be set to generate initial images of marine photovoltaic systems from different perspectives according to actual needs, such as... Figure 2 and Figure 3 These are initial images of marine photovoltaic systems from different perspectives. Furthermore, since all radiation contributions in these initial images originate from the surface reflection of the photovoltaic panels and support structures themselves, the calculations of all radiation contributions are based on the spectral reflectivity of the materials used in the photovoltaic equipment and the intensity of incident light. Therefore, these initial images do not include optical interference such as water absorption, scattering, or path radiation; they directly reflect the bulk geometric projection and surface reflection characteristics of the photovoltaic panels and support structures in an ideal vacuum or air environment.

[0030] 102. Input the environmental parameters into the preset four-flow radiative transfer model, and output the underwater radiance distribution and the water body path radiance distribution.

[0031] Specifically, the pre-defined four-flow radiative transfer model is a model that approximates the radiative transfer equation with four components: downflow direct radiation, downflow diffuse radiation, upflow diffuse radiation, and upflow direct radiation. In the vertical direction, its transmission and scattering are described by a set of coupled differential equations. After obtaining the environmental parameters of the target sea area, these parameters are input into the pre-defined four-flow radiative transfer model, which calculates the underwater radiance distribution and the radiance distribution along the water path.

[0032] In some embodiments, the environmental parameters further include the water body optical absorption coefficient, water body scattering coefficient, solar zenith angle, and solar azimuth angle. The environmental parameters are input into a preset four-flow radiative transfer model, and the output includes underwater radiance distribution and water path radiance distribution: The water body category of the target sea area is determined based on chlorophyll concentration and suspended matter concentration. The vertical discrete layer structure and asymmetric phase function of the preset four-flow radiative transfer model are determined based on the water body category. Each vertical discrete layer in the vertical discrete layer structure adopts the four-flow discrete ordinate format. The radiation intensity of each vertical discrete layer in the vertical discrete layer structure is iteratively calculated based on the water body optical absorption coefficient, water body scattering coefficient and asymmetric phase function. The downward and upward radiance fields are determined based on the radiation intensity at the convergence of each vertical discrete layer. Acquire aerosol optical thickness data for the target sea area, calculate the downward sky radiance distribution based on the aerosol optical thickness data, solar zenith angle and solar azimuth angle, and determine the underwater radiance distribution based on the downward radiance field and downward sky radiance distribution. The radiance distribution along the water path is determined based on the upward radiance field.

[0033] Among them, the optical absorption coefficient of water is used to represent the ability of the water in the target sea area to absorb light energy, and the water scattering coefficient is used to represent the ability of the water in the target sea area to deflect the light transmission path. When obtaining chlorophyll concentration and suspended matter concentration, similar methods can be used to obtain the optical absorption coefficient and water scattering coefficient of water. That is, by continuously collecting irradiance spectral sequences of the water surface and multiple underwater depths of the target sea area using an optical buoy within a preset sampling period, and then using an inherent optical characteristic inversion algorithm (such as a semi-analysis algorithm) to inversely deduce the collected irradiance spectral sequences to obtain the optical absorption coefficient and water scattering coefficient of water. The solar zenith angle and solar azimuth angle are both position angles of the sun, used to represent the spatial position of the sun relative to the observation plane at any given time. Both can be obtained from monitoring data of geostationary ocean color satellites.

[0034] Specifically, the water body category of the target sea area is determined based on chlorophyll concentration and suspended matter concentration. For example, when both chlorophyll concentration and suspended matter concentration are below a first preset value, the target sea area can be identified as an oligotrophic water body; when both chlorophyll concentration and suspended matter concentration are between the first and second preset values, the target sea area can be identified as a mesotrophic water body; and when both chlorophyll concentration and suspended matter concentration are above the second preset value, the target sea area can be identified as a eutrophic water body. Different types of water bodies exhibit different patterns of optical property variation with water depth. Therefore, the corresponding vertical discrete layer structure and asymmetric phase function can be obtained based on the determined water body category. For example, the vertical discrete layer structure can be a surface mixed layer, an intermediate cascade layer, or a bottom homogeneous layer, and the asymmetric phase function can be the Fournier-Forand function or the Petzold function. The four-stream discrete ordinate scheme within each vertical discrete layer is a numerical implementation of the pre-defined four-stream radiative transfer model based on the discrete ordinate method, used to calculate the radiation intensity in four different discrete directions: downlink direct radiation, downlink diffuse radiation, uplink diffuse radiation, and uplink direct radiation.

[0035] After determining the vertical discrete layer structure and asymmetric phase function, a radiative transfer equation is derived using a four-stream discrete ordinate scheme coupled with the asymmetric phase function. The water body's optical absorption coefficient and scattering coefficient are substituted into the radiative transfer equation to solve for the radiation intensity of light entering the water body after absorption and scattering. During iterative calculations, the process proceeds layer by layer downwards, starting from the first layer at the water surface. For the currently calculated vertical discrete layer, the following loop is executed: Based on the radiation intensity of the incident light entering the top of the vertical discrete layer, the initial radiation intensity distribution of the vertical discrete layer in four discrete directions is calculated as the starting point for iteration. Based on the radiation intensity in the current four discrete directions, combined with the water body's optical absorption coefficient and scattering coefficient, the total amount of incident light scattered from each direction to other directions by suspended particles and phytoplankton in the water body, as well as the total amount of incident light absorbed by suspended particles and phytoplankton in the water body, are calculated. Substituting the calculated total scattering and absorption into the radiative transfer equation, the radiation intensity in the four discrete directions within the vertical discrete layer is recalculated, resulting in a new set of radiation intensity distributions. The rate of change of radiation intensity between the latest and initial radiation intensity distributions is calculated. If the rate of change exceeds a preset convergence threshold, the convergence condition is not met. The total scattering and absorption are recalculated, and the latest radiation intensity distribution is updated. This process is repeated until the rate of change is less than or equal to the preset convergence threshold, at which point the iteration ends. The radiation intensities obtained after convergence in the downward direct and downward diffuse directions are used as the incident light rays at the top of the water body in the next vertical discrete layer for iterative calculation of the next vertical discrete layer. The solution is applied layer by layer downwards until the depth of the photovoltaic panels in the offshore photovoltaic equipment is reached, or the radiation intensity decays to a negligible level. Then, starting from the depth of the photovoltaic panels in the offshore photovoltaic equipment, the radiation intensities in the upward diffuse and upward direct directions are calculated layer by layer in reverse order upwards. The aforementioned iterative solution method is used within each vertical discrete layer until convergence. After convergence of all vertical discrete layers, the downward radiance field and the upward radiance field are output. The downward radiance field includes the downward radiance intensity at each depth of each vertical discrete layer, each light wavelength, each incident zenith angle, and each azimuth angle; the upward radiance field includes the upward radiance intensity at each depth of each vertical discrete layer, each light wavelength, each exit zenith angle, and each azimuth angle.

[0036] Specifically, aerosol optical thickness data is used to represent the aerosol extinction capacity of the atmosphere in the target sea area, and it can be obtained from monitoring data of geostationary ocean color satellites. The atmosphere is divided into several layers vertically, with aerosols mainly concentrated near the ground. Starting from the top of the atmosphere, the known boundary condition is that only the solar direction receives incident light radiation at the top, with zero incident light in other directions. Then, layer by layer downwards, within each atmospheric layer, based on the aerosol optical thickness data and scattering characteristics of that layer, the change in radiation intensity after passing through that atmospheric layer is calculated based on the solar zenith angle and solar azimuth angle. During the calculation of each atmospheric layer, the change in radiation intensity is iteratively calculated until it is less than a preset threshold. After completing the calculation for all atmospheric layers, the diffuse radiation intensity of the lowest atmospheric layer (i.e., at water surface height) in each discrete direction is obtained. The direct solar radiation intensity is added to the diffuse radiation intensity in the same direction to obtain the downward sky radiance in that direction. The downward sky radiance distribution is obtained by summing up the downward sky radiance in all discrete directions. It is understandable that the downward sky radiance distribution represents the radiant intensity distribution of sunlight incident on the water surface, while the downward radiance field represents the radiant intensity of sunlight propagating downwards at various depths underwater. Therefore, the downward sky radiance distribution and the downward radiance field can be combined to generate the underwater radiance distribution, which represents the radiant intensity distribution of the complete transmission path of sunlight from the sky to the water. The upward radiance field represents the radiant intensity of reflected light propagating upwards at various depths underwater. Based on the radiant intensity at different depths, the radiant intensity distribution of the path of reflected light from the bottom to the water surface can be determined, thus generating the water body path radiance distribution.

[0037] In some embodiments, the radiation intensity of each vertical discrete layer in the vertical discrete layer structure is iteratively calculated based on the water body optical absorption coefficient, water body scattering coefficient, and asymmetric phase function, including: The wind speed vector is decomposed into wind speed magnitude and direction, and the mean square slope of the sea surface in the target sea area is calculated based on the wind speed magnitude and wind direction obtained from the decomposition. The effective transmittance of the water-air interface in the target sea area is calculated based on the mean square slope of the sea surface. The radiation intensity of each vertical discrete layer in the vertical discrete layer structure is iteratively calculated based on the water body optical absorption coefficient, water body scattering coefficient and asymmetric phase function. During the iterative calculation, the effective transmittance is used to correct the radiation intensity calculated for each vertical discrete layer.

[0038] Specifically, the wind speed magnitude and direction, decomposed from the wind speed vector, are input into an isotropic empirical model or an anisotropic model, outputting the mean square slope of the sea surface in the target sea area. This mean square slope reflects the roughness of the sea surface. Understandably, the higher the wind speed, the steeper the waves, and the rougher the sea surface. When light passes through the water-air interface, Fresnel reflection and refraction occur. Calm sea surfaces have high light transmittance, while rough sea surfaces have reduced transmittance due to wave reflection. In other words, the mean square slope of the sea surface affects the effective transmittance of the water-air interface. Therefore, the calculated mean square slope of the sea surface can be input into a preset rough sea surface effective transmittance correction model to obtain the effective transmittance of the water-air interface. The effective transmittance is used as a correction value for the radiation intensity of the incident light. That is, the calculated effective transmittance is multiplied by the radiation intensity of the incident light entering the first vertical discrete layer, and the resulting product is used as the actual radiation intensity of the first vertical discrete layer. This allows the calculated underwater radiance distribution and water path radiance distribution to change in real time with the roughness of the sea surface, improving adaptability to complex sea conditions and further enhancing the realism and reliability of the subsequently generated marine photovoltaic composite image.

[0039] 103. Calculate the specular reflection component of the sea surface based on the wind speed vector, and calculate the volume scattering component of the water body based on the chlorophyll concentration and suspended matter concentration.

[0040] The specular reflection component represents the radiation intensity from sunlight directly hitting the sea surface or reflected by a specific wave slope in the preset observation direction. The volume scattering component represents the intensity of light scattered by the water body.

[0041] In some embodiments, calculating the specular reflection component of the sea surface based on the wind speed vector includes: Calculate the root mean square error of the slope along the wind direction and the root mean square error of the slope perpendicular to the wind direction based on the wind speed vector; The joint probability density function of wave slope driven by wind speed vector is calculated based on the mean square error of slope along the wind direction and the mean square error of slope perpendicular to the wind direction. Calculate the set of micro-element normal vectors required for specular reflection that satisfies the solar incident direction and the observation direction, and calculate the corresponding Fresnel reflection coefficient based on the zenith angle of each micro-element normal vector in the solar incident direction and the water reflectivity of the target sea area. The specular reflection component of the sea surface is obtained by integrating the joint probability density function of wave slope, the Fresnel reflection coefficient corresponding to the normal vector of each micro-element, and the geometric projection parameter corresponding to the normal vector of each micro-element.

[0042] Specifically, the wind speed vector is decomposed into the main wave component of large-scale waves and the high-frequency component of wind-generated gravity waves of small-scale waves. The main wave component and the high-frequency component of wind-generated gravity waves are substituted into the Cox-Munk empirical formula to calculate the root mean square error of the slope along the wind direction and the root mean square error of the slope perpendicular to the wind direction. Simultaneously, based on the wind speed magnitude and wind direction angle corresponding to the main wave component and the high-frequency component of wind-generated gravity waves, the slope along the wind direction and the slope perpendicular to the wind direction are calculated. The joint probability density function of the wave slope is calculated based on the slope along the wind direction, the root mean square error of the slope along the wind direction, and the slope perpendicular to the wind direction. The specific calculation formula is as follows:

[0043] in, Let be the joint probability density function of wave slope. For the slope along the wind direction, The slope is perpendicular to the wind direction. , The mean squared deviation of the slope along the wind direction. Mean squared deviation of vertical wind slope.

[0044] Calculate the first average of the zenith angles of the solar incidence direction and the observation direction, and the second average of the azimuth angles of the solar incidence direction and the observation direction. Determine the micro-surface element normal vectors with the first average zenith angle and the second average azimuth angle as the micro-surface element normal vectors required for specular reflection. Substitute the zenith angle of the solar incidence direction and the water reflectivity of the target sea area into the polarization Fresnel formula to calculate the Fresnel reflection coefficient of the micro-surface element normal vector required for each specular reflection. Transform the micro-surface element normal vectors to slope space to obtain the micro-surface element slope integral domain. Finally, calculate the sea surface specular reflection components according to the following formula: in, This represents the specular reflection component of the sea surface. The zenith angle is the direction of the sun's incidence. The zenith angle in the direction of observation. For Fresnel reflection coefficient, For water reflectivity, For the slope integration domain of the micro-surface element, These are the geometric projection parameters.

[0045] In some embodiments, calculating the volume scattering component of the water body based on chlorophyll concentration and suspended matter concentration includes: The first total scattering intensity and the first scattering intensity distribution of phytoplankton were determined based on chlorophyll concentration, and the second total scattering intensity and the second scattering intensity distribution of suspended matter were determined based on suspended matter concentration. The first volume scattering function value of phytoplankton and the second volume scattering function value of suspended matter are obtained by calculating the product of the first total scattering intensity and the first scattering intensity distribution, and the product of the second total scattering intensity and the second scattering intensity distribution, respectively. The water body scattering component is obtained by summing the values ​​of the first and second body scattering functions.

[0046] Specifically, higher chlorophyll concentrations, more phytoplankton, and higher total scattering intensity of phytoplankton are directly proportional. The relationship between chlorophyll concentration and total scattering intensity of phytoplankton can be determined using historical data. After obtaining real-time chlorophyll concentration data, the total scattering intensity of phytoplankton is determined based on this relationship. Because phytoplankton exhibits strong forward scattering, meaning most light is scattered within a small angle very close to its original direction, while lateral and backward scattering is very weak, the scattering intensity of phytoplankton at various angles can be obtained to determine the distribution of the first scattering intensity. Multiplying the total scattering intensity by the first scattering intensity distribution yields the first volume scattering function value of phytoplankton at each angle.

[0047] Specifically, the relationship between suspended solids concentration and the second total scattering intensity of suspended solids can be obtained based on historical experience. After obtaining the real-time suspended solids concentration, the second total scattering intensity of suspended solids is determined according to the relationship between the two. It is understandable that the scattering properties of suspended solids differ from those of phytoplankton. While the forward scattering of suspended solids is also strong, its backward scattering is significantly stronger because the inorganic particles in suspended solids are usually larger and have a higher refractive index. Based on this, the scattering intensity of suspended solids at various angles can be obtained, thereby determining the distribution of the second scattering intensity. Multiplying the second total scattering intensity by the second scattering intensity distribution yields the second volume scattering function values ​​of suspended solids at various angles. Finally, the first volume scattering function value and the second volume scattering function value are added together, and the sum is the volume scattering component of the water body.

[0048] In another feasible approach, since pure water may also exhibit inherent scattering, its third total scattering intensity is a known constant, and the distribution of the third scattering intensity at various angles is essentially symmetrical. Therefore, the third total scattering intensity of pure water can be multiplied by its third scattering intensity distribution to obtain the third-body scattering function value of pure water. Adding the first-body scattering function value, the second-body scattering function value, and the third-body scattering function value yields the actual volume scattering component of the water. By increasing the scattering function value based on the amount of pure water, the accuracy of the volume scattering component of the water can be improved, as well as its adaptability to different water bodies can be enhanced.

[0049] 104. The initial images of marine photovoltaic systems are fused using underwater radiance distribution, water path radiance distribution, sea surface specular reflection component, and water volume scattering component to obtain a composite image of marine photovoltaic systems.

[0050] Specifically, the multi-dimensional marine optical influence components are fused and corrected with the initial marine photovoltaic image at the pixel level. This corrects problems such as light and shadow deviation, color distortion, and lack of detail caused by water absorption, sea surface reflection, and water scattering in the initial marine photovoltaic image. The optical characteristics of the real marine environment are superimposed on the three-dimensional geometric image of the marine photovoltaic device to generate a marine photovoltaic composite image that fits the real environmental conditions of the target sea area.

[0051] In some embodiments, an initial image of marine photovoltaic systems is fused using underwater radiance distribution, water path radiance distribution, sea surface specular reflection component, and water volume scattering component to obtain a composite image of marine photovoltaic systems, including: The first radiation budget equation for the front of the photovoltaic panel is established based on the underwater radiance distribution and the reflectivity of the photovoltaic panel in the marine photovoltaic equipment. A second radiation budget equation for the back of a photovoltaic panel is established based on the water path radiance distribution and the water volume scattering component. Spectral curves were plotted based on the first radiation budget equation, the second radiation budget equation, and the specular reflection component of the sea surface. A composite image of marine photovoltaic power generation was obtained by fusing the initial image and spectral curve of the marine photovoltaic power generation system using a standard color matching function.

[0052] Specifically, the front of the photovoltaic panel directly receives sunlight incident light after it has passed through the atmosphere and sea surface. Its light absorption and reflection are mainly affected by water absorption, water scattering, and the photovoltaic panel's own reflectivity. By establishing a first radiation budget equation, the balance of light incident, reflection, and radiation on the front of the photovoltaic panel can be quantified, accurately simulating the optical imaging characteristics of the front of the photovoltaic panel. The back of the photovoltaic panel receives no direct sunlight, only light radiation and reflection from the water environment. Based on the water path radiance distribution, a background radiation substrate is obtained. The full-domain scattered irradiance of the water body scattering component is superimposed, establishing a second radiation budget equation for the back of the photovoltaic panel to simulate the imaging enhancement effect of water scattering light on the back of the photovoltaic panel and its bottom support. Based on the solution results of the first and second radiation budget equations, the specular reflection component of the sea surface is used as an interference component. Wavelength-by-wavelength irradiance fitting is performed within a preset visible light band to generate a spectral curve adapted to the current sea conditions of the target sea area. A standard color matching function (such as the CIE function) is used to perform pixel-level spectral mapping on the initial image of the marine photovoltaic system. The brightness, chromaticity, and saturation features corresponding to the spectral curves are then superimposed pixel-by-pixel onto the initial image to generate a composite image of the marine photovoltaic system. Based on this, the contribution of each pixel in the generated composite image (such as specular flare, water scattering, and target self-reflection) can be traced back to real physical parameters, avoiding the "black box" problem in traditional image stitching or empirical models. Furthermore, it can realistically reproduce the "backlighting" effect caused by the upward irradiance of the water on the back of the photovoltaic panel, as well as the gradual shadow gradation caused by the support structure blocking the back.

[0053] In some embodiments, after fusing the initial marine photovoltaic image using underwater radiance distribution, water path radiance distribution, sea surface specular reflection component, and water volume scattering component to obtain a composite marine photovoltaic image, the method for generating the composite marine photovoltaic image in this application embodiment further includes: The first shadow topology of the initial image of marine photovoltaic power and the second shadow topology of the composite image of marine photovoltaic power are extracted respectively; Determine whether the shadow matching degree between the first shadow topology and the second shadow topology is less than a preset matching degree threshold; When the shadow matching degree is less than the preset matching degree threshold, the preset four-stream radiative transfer model is updated based on the deviation between the shadow matching degree and the preset matching degree threshold. The underwater radiance distribution and water path radiance distribution output by the updated preset four-stream radiative transfer model are used to correct the marine photovoltaic composite image. This process is repeated until the latest shadow matching degree is greater than or equal to the preset matching degree threshold.

[0054] Specifically, the shadow matching degree between the first and second shadow topologies can be determined by calculating quantitative indicators such as the structural similarity index, edge preservation coefficient, or Hausdorff distance of the shadow contour. Specifically, the geometric occlusion relationship corresponding to the spatial layout of the support structure can be extracted from the BIM file of the offshore photovoltaic equipment to determine the first shadow topology of the initial image of the offshore photovoltaic system. Specifically, the shadow topology features at the connection between the back of the photovoltaic panel and the support structure can be extracted from the composite image of the offshore photovoltaic system to obtain the second shadow topology. When the shadow matching degree is greater than or equal to a preset matching degree threshold, no adjustment is needed to the composite image of the offshore photovoltaic system. However, when the shadow matching degree is less than the preset matching degree threshold, the shadow boundaries of the composite image of the offshore photovoltaic system may appear jagged. In this case, the jagged shadow boundaries can be eliminated by adjusting the ray tracing sampling precision. Additionally, when the shadow matching degree is less than the preset matching degree threshold, the shadow area of ​​the composite image of the offshore photovoltaic system may be too wide. In this case, the number of vertical discrete layers or the number of angular discrete layers in the preset four-flow radiative transfer model can be adjusted to avoid overly smooth or insufficient shadow edges due to excessively coarse layering. Based on this, the impact of excessively large shadow area estimation bias can be avoided, thereby improving the accuracy of power generation prediction or fault identification based on marine photovoltaic composite images.

[0055] In this embodiment, environmental parameters and an initial image of the marine photovoltaic system are first acquired for the target sea area. The environmental parameters include chlorophyll concentration, suspended matter concentration, and wind speed vector. The initial image of the marine photovoltaic system is a three-dimensional geometric image of the marine photovoltaic equipment located in the target sea area. Then, the environmental parameters are input into a preset four-flow radiative transfer model, and the underwater radiance distribution and water body path radiance distribution are output. Next, the specular reflection component of the sea surface is calculated based on the wind speed vector, and the volume scattering component of the water body is calculated based on the chlorophyll concentration and suspended matter concentration. Finally, the initial image of the marine photovoltaic system is fused using the underwater radiance distribution, the water body path radiance distribution, the specular reflection component of the sea surface, and the volume scattering component of the water body to obtain a composite image of the marine photovoltaic system. Based on this, the underwater radiance distribution and water path radiance distribution output by the preset four-flow radiative transfer model can be used to reconstruct the water radiative transfer characteristics of the target sea area. Simultaneously, the two major optical components—the specular reflection component of the sea surface and the volume scattering component of the water body—are quantified and calculated. This allows the fused marine photovoltaic composite image to accurately reproduce the actual underwater optical transfer characteristics of marine photovoltaic equipment, improving the fit between the composite image and the actual marine photovoltaic operation scenario, thereby enhancing the realism and reliability of the composite image. Please see Figure 4 As shown, another embodiment of this application provides an apparatus for generating composite images of marine photovoltaic data, comprising: The acquisition unit 401 is used to acquire environmental parameters of the target sea area and initial images of marine photovoltaic systems. The environmental parameters include chlorophyll concentration, suspended matter concentration and wind speed vector. The initial images of marine photovoltaic systems are three-dimensional geometric images of marine photovoltaic equipment installed in the target sea area. Input unit 402 is used to input environmental parameters into a preset four-flow radiative transfer model and output the underwater radiance distribution and the water body path radiance distribution. The calculation unit 403 is used to calculate the specular reflection component of the sea surface based on the wind speed vector, and to calculate the volume scattering component of the water body based on the chlorophyll concentration and suspended matter concentration. The fusion unit 404 is used to fuse the initial marine photovoltaic image using the underwater radiance distribution, the water path radiance distribution, the sea surface specular reflection component, and the water volume scattering component to obtain a composite marine photovoltaic image.

[0056] In this embodiment, the acquisition unit 401 acquires environmental parameters of the target sea area and an initial image of the marine photovoltaic system. The environmental parameters include chlorophyll concentration, suspended matter concentration, and wind speed vector. The initial image of the marine photovoltaic system is a three-dimensional geometric image of the marine photovoltaic equipment located in the target sea area. The input unit 402 inputs the environmental parameters into a preset four-flow radiative transfer model and outputs the underwater radiance distribution and the water body path radiance distribution. The calculation unit 403 calculates the specular reflection component of the sea surface based on the wind speed vector and the water body volume scattering component based on the chlorophyll concentration and suspended matter concentration. The fusion unit 404 uses the underwater radiance distribution, the water body path radiance distribution, the specular reflection component of the sea surface, and the water body volume scattering component to fuse the initial image of the marine photovoltaic system to obtain a composite image of the marine photovoltaic system. Based on this, the underwater radiance distribution and water path radiance distribution output by the preset four-flow radiative transfer model can be used to restore the water radiative transfer characteristics of the target sea area. At the same time, the two major optical components, the sea surface specular reflection component and the water body volume scattering component, are quantitatively calculated. This allows the fused marine photovoltaic composite image to restore the real optical transfer characteristics of marine photovoltaic equipment underwater, thereby improving the fit between the marine photovoltaic composite image and the actual marine photovoltaic operation scenario, and thus enhancing the authenticity and reliability of the marine photovoltaic composite image.

[0057] Please see Figure 5 As shown, another embodiment of this application provides an electronic device including: Processor 501, memory 502, input / output unit 503, and bus 504; The processor 501 is connected to the memory 502, the input / output unit 503, and the bus 504; The memory 502 stores a program, which the processor 501 calls to execute the method for generating marine photovoltaic composite images in any of the aforementioned optional embodiments.

[0058] This application also provides a computer-readable storage medium, including: instructions that, when executed on a computer, cause the computer to perform the method for generating a marine photovoltaic composite image in any of the foregoing alternative embodiments.

[0059] In the description of this specification, the references to terms such as "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0060] Furthermore, the above are merely preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for generating a composite image of marine photovoltaic power, characterized in that, include: The environmental parameters and initial images of the marine photovoltaic system in the target sea area are obtained. The environmental parameters include chlorophyll concentration, suspended matter concentration and wind speed vector. The initial images of the marine photovoltaic system are three-dimensional geometric images of the marine photovoltaic equipment installed in the target sea area. The environmental parameters are input into a preset four-flow radiative transfer model, and the underwater radiance distribution and the water body path radiance distribution are output. The sea surface specular reflection component is calculated based on the wind speed vector, and the water body volume scattering component is calculated based on the chlorophyll concentration and the suspended matter concentration. The initial marine photovoltaic image is fused using the underwater radiance distribution, the water path radiance distribution, the sea surface specular reflection component, and the water volume scattering component to obtain a composite marine photovoltaic image.

2. The method for generating a composite image of marine photovoltaic power according to claim 1, characterized in that, After fusing the initial marine photovoltaic image using the underwater radiance distribution, the water path radiance distribution, the sea surface specular reflection component, and the water volume scattering component to obtain a composite marine photovoltaic image, the method further includes: The first shadow topology of the initial image of the marine photovoltaic system and the second shadow topology of the composite image of the marine photovoltaic system are extracted respectively. Determine whether the shadow matching degree between the first shadow topology and the second shadow topology is less than a preset matching degree threshold; When the shadow matching degree is less than the preset matching degree threshold, the preset four-stream radiative transfer model is updated based on the deviation between the shadow matching degree and the preset matching degree threshold. The underwater radiance distribution and water path radiance distribution output by the updated preset four-stream radiative transfer model are used to correct the marine photovoltaic composite image. This process is repeated until the latest shadow matching degree is greater than or equal to the preset matching threshold.

3. The method for generating a composite image of marine photovoltaic power according to claim 1, characterized in that, The process of fusing the initial marine photovoltaic image using the underwater radiance distribution, the water path radiance distribution, the sea surface specular reflection component, and the water volume scattering component to obtain a composite marine photovoltaic image includes: Based on the underwater radiance distribution and the reflectivity of the photovoltaic panels in the marine photovoltaic equipment, a first radiation budget equation is established for the front side of the photovoltaic panel; A second radiation budget equation for the back of the photovoltaic panel is established based on the water body path radiance distribution and the water body volume scattering component. Spectral curves were plotted based on the first radiation budget equation, the second radiation budget equation, and the sea surface specular reflection component. The initial image of the marine photovoltaic system and the spectral curve are fused using a standard color matching function to obtain a composite image of the marine photovoltaic system.

4. The method for generating a composite image of marine photovoltaic power according to claim 1, characterized in that, The calculation of the sea surface specular reflection component based on the wind speed vector includes: Calculate the root mean square error of the slope along the wind direction and the root mean square error of the slope perpendicular to the wind direction based on the wind speed vector; The joint probability density function of wave slope driven by the wind speed vector is calculated based on the mean square error of the slope along the wind direction and the mean square error of the slope perpendicular to the wind direction. Calculate the set of microfacet normal elements required for specular reflection that satisfies the solar incident direction and the observation direction, and calculate the corresponding Fresnel reflection coefficient based on the zenith angle of each microfacet normal vector in the set of microfacet normal elements in the solar incident direction and the water reflectivity of the target sea area. The sea surface specular reflection component is obtained by integrating the joint probability density function of wave slope, the Fresnel reflection coefficient corresponding to each micro-surface normal vector, and the geometric projection parameter corresponding to each micro-surface normal vector.

5. The method for generating a composite image of marine photovoltaic power according to claim 1, characterized in that, The calculation of the water volume scattering component based on the chlorophyll concentration and the suspended matter concentration includes: The first total scattering intensity and the first scattering intensity distribution of phytoplankton are determined based on the chlorophyll concentration, and the second total scattering intensity and the second scattering intensity distribution of suspended matter are determined based on the suspended matter concentration; The first volume scattering function value of the phytoplankton and the second volume scattering function value of the suspended matter are obtained by calculating the product of the first total scattering intensity and the first scattering intensity distribution, and the product of the second total scattering intensity and the second scattering intensity distribution, respectively. The water body scattering component is obtained by summing the values ​​of the first and second body scattering functions.

6. The method for generating a composite image of marine photovoltaic power according to claim 1, characterized in that, The environmental parameters also include the water body optical absorption coefficient, water body scattering coefficient, solar zenith angle, and solar azimuth angle. The process of inputting these environmental parameters into a preset four-flow radiative transfer model and outputting the underwater radiance distribution and the water body path radiance distribution includes: The water body category of the target sea area is determined based on the chlorophyll concentration and the suspended matter concentration. The vertical discrete layer structure and asymmetric phase function of the preset four-stream radiative transfer model are determined based on the water body category. Each vertical discrete layer in the vertical discrete layer structure adopts the four-stream discrete ordinate format. The radiation intensity of each vertical discrete layer in the vertical discrete layer structure is iteratively calculated based on the water body optical absorption coefficient, the water body scattering coefficient, and the asymmetric phase function. The downward and upward radiance fields are determined based on the radiation intensity at the convergence of each vertical discrete layer. Acquire aerosol optical thickness data of the target sea area, calculate the downward sky radiance distribution based on the aerosol optical thickness data, the solar zenith angle and the solar azimuth angle, and determine the underwater radiance distribution based on the downward radiance field and the downward sky radiance distribution; The water body path radiance distribution is determined based on the upward radiance field.

7. The method for generating a composite image of marine photovoltaic power according to claim 6, characterized in that, The iterative calculation of the radiation intensity of each vertical discrete layer in the vertical discrete layer structure based on the water body's optical absorption coefficient, the water body's scattering coefficient, and the asymmetric phase function includes: The wind speed vector is decomposed into wind speed magnitude and wind direction, and the mean square slope of the sea surface in the target sea area is calculated based on the wind speed magnitude and wind direction obtained from the decomposition. The effective transmittance of the water-air interface in the target sea area is calculated based on the mean square slope of the sea surface. The radiation intensity of each vertical discrete layer in the vertical discrete layer structure is iteratively calculated based on the water body optical absorption coefficient, the water body scattering coefficient, and the asymmetric phase function. During the iterative calculation, the effective transmittance is used to correct the radiation intensity calculated for each vertical discrete layer.

8. A device for generating composite images of marine photovoltaic data, characterized in that, include: The acquisition unit is used to acquire environmental parameters of the target sea area and initial images of marine photovoltaic systems. The environmental parameters include chlorophyll concentration, suspended matter concentration and wind speed vector. The initial images of marine photovoltaic systems are three-dimensional geometric images of marine photovoltaic equipment installed in the target sea area. The input unit is used to input the environmental parameters into a preset four-flow radiative transfer model and output the underwater radiance distribution and the water body path radiance distribution. The calculation unit is used to calculate the specular reflection component of the sea surface based on the wind speed vector, and to calculate the volume scattering component of the water body based on the chlorophyll concentration and the suspended matter concentration. The fusion unit is used to fuse the initial marine photovoltaic image using the underwater radiance distribution, the water path radiance distribution, the sea surface specular reflection component, and the water volume scattering component to obtain a composite marine photovoltaic image.

9. An electronic device, characterized in that, include: Processor, memory, input / output units, and bus; The processor is connected to the memory, the input / output unit, and the bus; The memory stores a program, which the processor calls to execute the method for generating a marine photovoltaic composite image as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, include: Instructions, when executed on a computer, cause the computer to perform the method for generating a marine photovoltaic composite image as described in any one of claims 1 to 7.

Images