Prediction system and method applied to offshore photovoltaic generating capacity

Abstract

The invention discloses a prediction system and method applied to offshore photovoltaic power generation capacity, relates to the technical field of photovoltaic power generation capacity prediction, and solves the problem that in the prior art, when the offshore photovoltaic power generation capacity is predicted, it is not considered that a salt film possibly exists on a photovoltaic panel before prediction, prediction is directly carried out through the initial photovoltaic power generation efficiency, and the prediction efficiency is poor. And the prediction precision is not high. The method comprises the following steps: acquiring an image and salt mist concentration of a photovoltaic panel of an offshore photovoltaic platform; analyzing the image of the photovoltaic panel to obtain salt film data; calculating a salt film power generation efficiency loss factor based on the salt film data; calculating the initial power generation efficiency of the current photovoltaic panel based on the salt film power generation efficiency loss factor; calculating the photovoltaic power generation efficiency based on the salt mist concentration and the initial power generation efficiency; the offshore photovoltaic power generation amount is predicted based on the photovoltaic power generation efficiency, and the technical problem is solved.

Description

Technical Field

[0001] The present invention belongs to the field of photovoltaic power generation prediction, and in particular relates to a prediction system and method for offshore photovoltaic power generation. Background Art

[0002] With the continuous growth of global demand for clean energy, offshore photovoltaic power generation, as a renewable energy utilization method with huge potential, has attracted widespread attention.

[0003] The prior art (invention patent with application number 2022106044366) discloses a method and system for predicting the power generation of photovoltaic panels on offshore platforms; the method includes: obtaining relevant data for predicting the power generation of photovoltaic panels; the relevant data include: the solar radiation on the predicted date and the predicted sea area, the photovoltaic panel area, the photovoltaic panel efficiency, the power supply efficiency, the temperature influence coefficient of the sea water temperature on the photovoltaic panel power generation efficiency, the wave influence coefficient of the waves on the power generation of photovoltaic panels, the cloud cover coefficient of the clouds on the power generation of photovoltaic panels, and the solar radiation attenuation coefficient of the sea water covering the photovoltaic panels on their power generation; based on the relevant data, a custom photovoltaic panel power generation prediction model is used to predict the power generation of photovoltaic panels, which solves the problem of inaccurate power generation prediction of the photovoltaic panels on the existing offshore platforms due to factors such as the angle of the photovoltaic panels being affected by waves and the surface being covered by sea water, effectively improving the accuracy of the power generation prediction of the photovoltaic panels on the offshore platforms. However, existing technologies do not consider the dynamic impact of salt film on power generation efficiency. Salt film is a covering layer formed by salt crystallization on the surface of photovoltaic panels after seawater evaporates. It is mainly caused by the evaporation of water after the deposition of sea spray. When predicting offshore photovoltaic power generation, existing technologies do not take into account the possibility that salt film may already exist on the photovoltaic panels before the prediction. Therefore, the prediction is directly based on the initial photovoltaic power generation efficiency, which obviously has low prediction accuracy.

[0004] Therefore, the present invention proposes a prediction system and method for offshore photovoltaic power generation to solve the above problems. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a prediction system and method for offshore photovoltaic power generation, which is used to solve the technical problem that the prior art does not take into account the possible existence of salt film on the photovoltaic panels before the prediction when predicting offshore photovoltaic power generation, and directly predicts based on the initial photovoltaic power generation efficiency, resulting in low prediction accuracy.

[0006] To achieve the above-mentioned object, a first aspect of the present invention provides a prediction system for offshore photovoltaic power generation, comprising: a data acquisition module, a data analysis module and a power generation prediction module;

[0007] Data acquisition module: used to obtain images of photovoltaic panels on offshore photovoltaic platforms and salt spray concentration;

[0008] Data analysis module: Analyze the images of photovoltaic panels to obtain salt film data; the salt film data includes: salt film coverage area and salt film thickness; and

[0009] The salt film power generation efficiency loss factor is calculated based on the salt film data; the initial power generation efficiency of the current photovoltaic panel is calculated based on the salt film power generation efficiency loss factor; the photovoltaic power generation efficiency is calculated based on the salt fog concentration and the initial power generation efficiency;

[0010] Power generation prediction module: predicts the power generation of offshore photovoltaics based on photovoltaic power generation efficiency.

[0011] Preferably, the acquiring of images of photovoltaic panels of an offshore photovoltaic platform and salt spray concentration includes:

[0012] Collect images of the current photovoltaic panels through an image acquisition device;

[0013] The salt spray concentration is collected in real time through data sensors.

[0014] Preferably, the analyzing the image of the photovoltaic panel includes:

[0015] The photovoltaic panel image is converted into a black and white binary image by a binarization method, and the salt film coverage area in the black and white binary image is obtained; wherein the binarization method includes: a global threshold or an adaptive threshold;

[0016] The salt film coverage ratio is obtained by dividing the number of pixels in the salt film coverage area in the black and white binary image by the total number of pixels in the black and white binary image.

[0017] The salt film coverage area is obtained by calculating the product of the salt film coverage ratio and the area of the total photovoltaic panels;

[0018] Convert the photovoltaic panel image into a grayscale image; map the salt film coverage area in the black and white binary image to the grayscale image, and obtain the average grayscale value of the salt film coverage area in the grayscale image;

[0019] The average grayscale value of the salt film covered area is identified by the pre-trained salt film thickness recognition model to obtain the corresponding salt film thickness.

[0020] It should be noted that the salt film is a covering layer formed by the crystallization of salt (such as sodium chloride, calcium sulfate, etc.) on the surface of the photovoltaic panel after the evaporation of seawater. It is mainly caused by the evaporation of water after the deposition of seawater droplets. It reduces the transmittance (blocks incident light), increases surface reflection and scattering, and causes electrochemical corrosion (such as oxidation of metal frames and peeling of glass coatings), resulting in a significant decrease in the power generation efficiency of photovoltaic panels.

[0021] Preferably, the training method of the salt film thickness recognition model includes:

[0022] Based on the historical database, images of several salt film thicknesses and the average grayscale value of the salt film coverage area in the images are obtained;

[0023] The average grayscale value of the salt film covered area in the image is integrated as the standard input data, and the salt film thickness corresponding to the average grayscale value of the salt film covered area in the image is integrated as the standard output data;

[0024] An artificial intelligence model is trained based on standard input data and standard output data to obtain a salt film thickness recognition model; wherein the artificial intelligence model includes: a convolutional neural network or a deep belief network.

[0025] It should be noted that the historical database stores images of different salt film thicknesses obtained by manual measurement.

[0026] Preferably, the calculation of the salt film power generation efficiency loss factor based on the salt film data includes:

[0027] The salt film coverage area is marked as YS, and the salt film thickness is marked as YH;

[0028] The transmittance T corresponding to the current salt film thickness is calculated using the formula T = e^(-α×YH); where α is the light absorption coefficient of the salt film;

[0029] The overall average transmittance of the photovoltaic panels on the current photovoltaic platform is calculated using the formula Tavg = (YS / S) × T + (1-(YS / S)) × T0; where Tavg is the overall average transmittance, S is the total area of the photovoltaic panels on the photovoltaic platform, and T0 is the initial transmittance.

[0030] The salt film power generation efficiency loss factor is calculated by the formula Δη=η0×(1-(Tavg / T0)); where Δη is the salt film power generation efficiency loss factor and η0 is the initial power generation efficiency.

[0031] It should be noted that the light absorption coefficient α of the salt film is determined experimentally. For example, the transmittance (T1, T2) of the salt film at different thicknesses (H1, H2) is measured by a spectrometer, i.e., T1 = e^(-α×H1), T2 = e^(-α×H2), and thus it can be concluded that: α = ln(T1 / T2) / (H2-H1);

[0032] The initial light transmittance refers to the proportion of incident light that passes through the surface glass or protective layer of the photovoltaic panel and reaches the solar cell when the photovoltaic panel is completely clean and pollution-free; the initial power generation efficiency refers to the power generation efficiency of the photovoltaic panel on the photovoltaic platform in the absence of a salt film.

[0033] Preferably, the calculation of the initial power generation efficiency of the current photovoltaic panel based on the salt film power generation efficiency loss factor includes:

[0034] By calculating the difference between the initial power generation efficiency and the salt membrane power generation efficiency loss factor, the current photovoltaic starting power generation efficiency is obtained.

[0035] Preferably, the photovoltaic power generation efficiency is calculated based on the salt spray concentration and the initial power generation efficiency, including:

[0036] The current salt spray concentration is identified through the pre-trained salt spray deposition rate recognition model to obtain the corresponding salt spray deposition rate;

[0037] Obtain a preset time period for the required predicted power generation; divide the preset time period into a number of time periods i according to preset time intervals; where i = {1, 2, 3, ..., N}, and N is the total number of divided time periods;

[0038] By the formula ηi=η a ×e^(-β×R×ti) to calculate the photovoltaic power generation efficiency of each time period; where ηi refers to the photovoltaic power generation efficiency of the i-th time period, β is the proportional coefficient, η a is the initial power generation efficiency of photovoltaics, R is the salt spray deposition rate, and ti refers to the length of the i-th time period.

[0039] It should be noted that the preset time period refers to the overall time range for which the photovoltaic power generation needs to be predicted, such as the next 24 hours. If the preset time interval is 1 hour, that is, the time interval is divided once every hour, then the next 24 hours will be divided into 24 1-hour time periods, N=24;

[0040] The preset time interval is set according to the actual accuracy required. The smaller the preset time interval, the higher the accuracy of the predicted power generation;

[0041] The proportional coefficient β is set by those skilled in the art based on experience or experimental calibration, and reflects the degree to which the photovoltaic power generation efficiency is affected by the salt mist deposition rate.

[0042] Preferably, the training method of the salt spray deposition rate identification model includes:

[0043] Obtain several salt spray concentrations and corresponding salt spray deposition rates based on the historical salt spray database;

[0044] An artificial intelligence model is trained based on training data to obtain a salt spray deposition rate identification model; wherein the training data includes: training input data and training output data; the training input data is salt spray concentration, and the training output data is the salt spray deposition rate corresponding to the salt spray concentration.

[0045] It should be noted that the salt spray deposition rate corresponding to the salt spray concentration in the historical salt spray database is obtained manually.

[0046] Preferably, the predicting of offshore photovoltaic power generation based on photovoltaic power generation efficiency includes:

[0047] Predict the power generation of offshore photovoltaics through a preset prediction model;

[0048] The preset prediction model is: E=∑(ηi×Gi×S×ti); wherein ∑ is a summation factor, the summation range is (1, N), and Gi is the light intensity in the i-th time period.

[0049] It should be noted that the light intensity at different times is obtained through the weather forecast platform.

[0050] A second aspect of the present invention provides a method for predicting offshore photovoltaic power generation, comprising:

[0051] Obtain images of photovoltaic panels on offshore photovoltaic platforms and salt spray concentrations;

[0052] By analyzing the images of photovoltaic panels, salt film data is obtained;

[0053] The salt film power generation efficiency loss factor was calculated based on the salt film data;

[0054] The initial power generation efficiency of the current photovoltaic panel is calculated based on the salt film power generation efficiency loss factor;

[0055] The photovoltaic power generation efficiency is calculated based on the salt spray concentration and the initial power generation efficiency;

[0056] Predicting offshore photovoltaic power generation based on photovoltaic power generation efficiency.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] 1. When predicting offshore photovoltaic power generation, the existing technology does not take into account the possibility that salt film may already exist on the photovoltaic panels before the prediction, and directly predicts based on the initial photovoltaic power generation efficiency, resulting in low prediction accuracy. The present invention uses an image acquisition device to acquire photovoltaic panel images in real time, and uses binarization and grayscale analysis techniques to accurately extract the salt film coverage area and the average grayscale value of the coverage area. Combined with a pre-trained salt film thickness recognition model, the current salt film thickness is dynamically calculated, and then the attenuation of the salt film on the transmittance is quantified through a formula. The overall average transmittance is finally calculated, and the salt film power generation efficiency loss factor is calculated based on the overall average transmittance, thereby correcting the initial power generation efficiency to obtain the initial power generation efficiency. The power generation prediction module dynamically predicts the photovoltaic power generation efficiency in different time periods based on factors such as the real-time salt fog deposition rate, the corrected initial power generation efficiency, and the light intensity, and accumulates the total power generation. This effectively solves the problem of the existing technology ignoring the dynamic impact of salt film on power generation efficiency. It not only improves the prediction accuracy, but also better reflects the performance of the photovoltaic system under actual operating conditions, providing solid data support for optimizing the operation and maintenance strategy of offshore photovoltaic power stations.

[0059] 2. The existing technology ignores the impact of salt fog environment on photovoltaic power generation efficiency. In a salt fog environment, salt will deposit on the surface of the photovoltaic panel to form a salt film. As the salt fog concentration increases and time passes, the thickness of the salt film gradually increases, resulting in a decrease in light transmittance and enhanced surface reflection and scattering, thereby significantly reducing the actual power generation efficiency of the photovoltaic panel. The present invention uses a pre-trained artificial intelligence model to accurately identify the corresponding salt fog deposition rate according to the real-time salt fog concentration, so that the growth rate of the salt film can be quantified. The overall prediction time period is subdivided into multiple small time periods (such as one hour), and the photovoltaic power generation efficiency of each time period is dynamically calculated through a formula based on the salt fog deposition rate, the corrected initial power generation efficiency and the time length in each small time period. This method can more accurately reflect the impact of the salt film growth over time on the power generation efficiency. Combined with the light intensity and the area of the photovoltaic panel, an accumulation model is used to comprehensively predict the power generation during the entire prediction period, ensuring that the prediction result not only takes into account the power generation efficiency loss in the initial state, but also fully incorporates the continued impact of the future salt fog deposition rate on the power generation efficiency. This effectively solves the problem that the existing technology ignores the dynamic changes of the salt film during actual operation and the impact of the salt fog environment on the photovoltaic power generation efficiency, resulting in inaccurate predictions. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0061] Figure 1 Schematic diagram of system modules according to an embodiment of the present invention;

[0062] Figure 2 Schematic diagram of a method for acquiring salt film data according to an embodiment of the present invention;

[0063] Figure 3 Schematic diagram of the method steps of an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0065] See also Figure 1 , a first aspect of the present invention provides a prediction system for offshore photovoltaic power generation, comprising: a data acquisition module, a data analysis module and a power generation prediction module;

[0066] Data acquisition module: used to obtain images of photovoltaic panels on offshore photovoltaic platforms and salt spray concentration;

[0067] Data analysis module: Analyze the images of photovoltaic panels to obtain salt film data; the salt film data includes: salt film coverage area and salt film thickness; and

[0068] The salt film power generation efficiency loss factor is calculated based on the salt film data; the initial power generation efficiency of the current photovoltaic panel is calculated based on the salt film power generation efficiency loss factor; the photovoltaic power generation efficiency is calculated based on the salt fog concentration and the initial power generation efficiency;

[0069] Power generation prediction module: predicts the power generation of offshore photovoltaics based on photovoltaic power generation efficiency.

[0070] Acquire images of photovoltaic panels on offshore photovoltaic platforms and salt spray concentrations, including:

[0071] Collect images of the current photovoltaic panels through an image acquisition device;

[0072] The salt spray concentration is collected in real time through data sensors.

[0073] See Figure 2 , by analyzing the images of photovoltaic panels, including:

[0074] The photovoltaic panel image is converted into a black and white binary image by a binarization method, and the salt film coverage area in the black and white binary image is obtained; wherein the binarization method includes: a global threshold or an adaptive threshold;

[0075] The salt film coverage ratio is obtained by dividing the number of pixels in the salt film coverage area in the black and white binary image by the total number of pixels in the black and white binary image.

[0076] The salt film coverage area is obtained by calculating the product of the salt film coverage ratio and the area of the total photovoltaic panels;

[0077] Convert the photovoltaic panel image into a grayscale image; map the salt film coverage area in the black and white binary image to the grayscale image, and obtain the average grayscale value of the salt film coverage area in the grayscale image;

[0078] The average grayscale value of the salt film covered area is identified by the pre-trained salt film thickness recognition model to obtain the corresponding salt film thickness.

[0079] The training method of the salt film thickness recognition model includes:

[0080] Based on the historical database, images of several salt film thicknesses and the average grayscale value of the salt film coverage area in the images are obtained;

[0081] The average grayscale value of the salt film covered area in the image is integrated as the standard input data, and the salt film thickness corresponding to the average grayscale value of the salt film covered area in the image is integrated as the standard output data;

[0082] An artificial intelligence model is trained based on standard input data and standard output data to obtain a salt film thickness recognition model; wherein the artificial intelligence model includes: a convolutional neural network or a deep belief network.

[0083] The salt membrane power generation efficiency loss factor is calculated based on the salt membrane data, including:

[0084] The salt film coverage area is marked as YS, and the salt film thickness is marked as YH;

[0085] The transmittance T corresponding to the current salt film thickness is calculated using the formula T = e^(-α×YH); where α is the light absorption coefficient of the salt film;

[0086] The overall average transmittance of the photovoltaic panels on the current photovoltaic platform is calculated using the formula Tavg = (YS / S) × T + (1-(YS / S)) × T0; where Tavg is the overall average transmittance, S is the total area of the photovoltaic panels on the photovoltaic platform, and T0 is the initial transmittance.

[0087] The salt film power generation efficiency loss factor is calculated by the formula Δη=η0×(1-(Tavg / T0)); where Δη is the salt film power generation efficiency loss factor and η0 is the initial power generation efficiency.

[0088] The initial power generation efficiency of the current photovoltaic panel is calculated based on the salt film power generation efficiency loss factor, including:

[0089] By calculating the difference between the initial power generation efficiency and the salt membrane power generation efficiency loss factor, the current photovoltaic starting power generation efficiency is obtained.

[0090] The photovoltaic power generation efficiency is calculated based on the salt spray concentration and the initial power generation efficiency, including:

[0091] The current salt spray concentration is identified through the pre-trained salt spray deposition rate recognition model to obtain the corresponding salt spray deposition rate;

[0092] Obtain a preset time period for the required predicted power generation; divide the preset time period into a number of time periods i according to preset time intervals; where i = {1, 2, 3, ..., N}, and N is the total number of divided time periods;

[0093] By the formula ηi=η a ×e^(-β×R×ti) to calculate the photovoltaic power generation efficiency of each time period; where ηi refers to the photovoltaic power generation efficiency of the i-th time period, β is the proportional coefficient, η a is the initial power generation efficiency of photovoltaics, R is the salt spray deposition rate, and ti refers to the length of the i-th time period.

[0094] The training method of the salt spray deposition rate identification model includes:

[0095] Obtain several salt spray concentrations and corresponding salt spray deposition rates based on the historical salt spray database;

[0096] An artificial intelligence model is trained based on training data to obtain a salt spray deposition rate identification model; wherein the training data includes: training input data and training output data; the training input data is salt spray concentration, and the training output data is the salt spray deposition rate corresponding to the salt spray concentration.

[0097] Predicting offshore photovoltaic power generation based on photovoltaic power generation efficiency, including:

[0098] Predict the power generation of offshore photovoltaics through a preset prediction model;

[0099] The preset prediction model is: E=∑(ηi×Gi×S×ti); wherein ∑ is a summation factor, the summation range is (1, N), and Gi is the light intensity in the i-th time period.

[0100] For example, in a certain offshore floating photovoltaic power station, the area of a single photovoltaic panel is 10m 2 The total number of photovoltaic panels is 50, and the total area of photovoltaic panels on the entire platform is S = 500m 2 The 24-hour power generation of offshore photovoltaic panels is now predicted as follows:

[0101] 1. System parameter setting;

[0102] Photovoltaic panel parameters:

[0103] Initial power generation efficiency η0 = 20% (without salt membrane).

[0104] Initial light transmittance T0 = 90% (clean state).

[0105] Salt film light absorption coefficient α=0.05μm -1 (Experimentally determined value).

[0106] Environmental parameters:

[0107] Prediction time period: the next 24 hours, divided into each hour (N=24).

[0108] Light intensity by time period (Gi): obtained through the weather forecast platform, assuming typical values for sunny days:

[0109] 0-6 am (night): Gi = 0 W / m 2 (no light);

[0110] 7-10 a.m. (morning): Gi = 300 W / m 2 ;

[0111] 11:00-13:00 (noon): Gi=1000W / m 2 ;

[0112] 2:00 PM to 5:00 PM: Gi = 800 W / m 2 ;

[0113] 18:00 to 24:00 (evening to night): Gi = 0 W / m 2 ;

[0114] Salt spray related parameters:

[0115] Salt spray concentration: real-time monitoring value is 10mg / m 3 ;

[0116] Salt spray deposition rate identification model: input salt spray concentration 10mg / m 3 , output salt spray deposition rate R = 0.5mg / (m 2 h);

[0117] Proportional coefficient β: experimentally calibrated to 0.002h -1 mg -1 ·m 2 .

[0118] 2. Data acquisition module;

[0119] Image acquisition and salt film analysis:

[0120] Use drones to capture images of photovoltaic panel surfaces;

[0121] The salt film coverage area YS = 95.25m was obtained by binarization method. 2 ;

[0122] The image is converted into a grayscale image. Assuming that the average grayscale value of the salt film area in the grayscale image is 120 (the clean area is 200), the salt film thickness YH = 20 μm is output through pre-trained CNN model mapping.

[0123] 3. Data analysis module;

[0124] Calculation of salt membrane power generation efficiency loss factor:

[0125] Calculate the transmittance T corresponding to the current salt film thickness using the formula T = e^(-α×YH);

[0126] Substituting the data, we get T≈36.79%;

[0127] The overall average light transmittance of the photovoltaic panels on the current photovoltaic platform is calculated using the formula Tavg = (YS / S) × T + (1-(YS / S)) × T0;

[0128] Substituting the data, we get Tavg≈79.86%;

[0129] The salt membrane power generation efficiency loss factor is calculated by the formula Δη=η0×(1-(Tavg / T0));

[0130] Substituting the data, we obtain Δη≈1.99%;

[0131] Initial power generation efficiency η a =η0-Δη=18.01%;

[0132] Dynamic photovoltaic power generation efficiency calculation (time period calculation):

[0133] Hourly efficiency ηi:

[0134] Using the formula ηi=η a ×e^(-β×R×ti).

[0135] Calculation example:

[0136] First hour (7 o'clock): η1≈17.98%.

[0137] 12th hour (12 noon): After 5 hours of salt spray deposition, η12≈17.68%.

[0138] 24th hour (24 hours at night): η24≈17.24%.

[0139] 4. Power generation prediction module;

[0140] Calculation of power generation by time period: E = ∑(ηi×Gi×S×ti);

[0141] The final calculation is E = 651.8 kWh.

[0142] Comparison with existing methods:

[0143] If the effects of salt film and salt spray are ignored and the initial efficiency η0 = 20% is used directly, then: E ≈ 740 kWh;

[0144] Error analysis: The predicted value of 651.8 kWh by the present invention is about 12% lower than that of the existing method, and the prediction accuracy is higher (salt film and salt mist cause efficiency loss).

[0145] See Figure 3 The second embodiment of the present invention provides a method for predicting offshore photovoltaic power generation, comprising:

[0146] Obtain images of photovoltaic panels on offshore photovoltaic platforms and salt spray concentrations;

[0147] By analyzing the images of photovoltaic panels, salt film data is obtained;

[0148] The salt film power generation efficiency loss factor was calculated based on the salt film data;

[0149] The initial power generation efficiency of the current photovoltaic panel is calculated based on the salt film power generation efficiency loss factor;

[0150] The photovoltaic power generation efficiency is calculated based on the salt spray concentration and the initial power generation efficiency;

[0151] Predicting offshore photovoltaic power generation based on photovoltaic power generation efficiency.

[0152] Some of the data in the above formula are calculated by removing the dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0153] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A prediction system for offshore photovoltaic power generation, characterized in that: include: Data acquisition module, data analysis module and power generation prediction module; Data acquisition module: used to obtain images of photovoltaic panels on offshore photovoltaic platforms and salt spray concentration; Data analysis module: Analyze the images of photovoltaic panels to obtain salt film data; the salt film data includes: salt film coverage area and salt film thickness; and The salt film power generation efficiency loss factor is calculated based on the salt film data; the initial power generation efficiency of the current photovoltaic panel is calculated based on the salt film power generation efficiency loss factor; the photovoltaic power generation efficiency is calculated based on the salt fog concentration and the initial power generation efficiency; Power generation prediction module: predicts the power generation of offshore photovoltaics based on photovoltaic power generation efficiency.

2. The prediction system for offshore photovoltaic power generation according to claim 1, characterized in that: The obtaining of images of photovoltaic panels of the offshore photovoltaic platform and salt spray concentration includes: Collect images of the current photovoltaic panels through an image acquisition device; The salt spray concentration is collected in real time through data sensors.

3. The prediction system for offshore photovoltaic power generation according to claim 1, characterized in that: The analysis of the photovoltaic panel image includes: The photovoltaic panel image is converted into a black and white binary image by a binarization method, and the salt film coverage area in the black and white binary image is obtained; wherein the binarization method includes: a global threshold or an adaptive threshold; The salt film coverage ratio is obtained by dividing the number of pixels in the salt film coverage area in the black and white binary image by the total number of pixels in the black and white binary image. The salt film coverage area is obtained by calculating the product of the salt film coverage ratio and the area of the total photovoltaic panels; Convert the photovoltaic panel image into a grayscale image; map the salt film coverage area in the black and white binary image to the grayscale image, and obtain the average grayscale value of the salt film coverage area in the grayscale image; The average grayscale value of the salt film covered area is identified by the pre-trained salt film thickness recognition model to obtain the corresponding salt film thickness.

4. The prediction system for offshore photovoltaic power generation according to claim 3, characterized in that: The training method of the salt film thickness recognition model includes: Based on the historical database, images of several salt film thicknesses and the average grayscale value of the salt film coverage area in the images are obtained; The average grayscale value of the salt film covered area in the image is integrated as the standard input data, and the salt film thickness corresponding to the average grayscale value of the salt film covered area in the image is integrated as the standard output data; An artificial intelligence model is trained based on standard input data and standard output data to obtain a salt film thickness recognition model; wherein the artificial intelligence model includes: a convolutional neural network or a deep belief network.

5. The prediction system for offshore photovoltaic power generation according to claim 1, characterized in that: The salt film power generation efficiency loss factor calculated based on the salt film data includes: The salt film coverage area is marked as YS, and the salt film thickness is marked as YH; The transmittance T corresponding to the current salt film thickness is calculated using the formula T = e^(-α×YH); where α is the light absorption coefficient of the salt film; The overall average transmittance of the photovoltaic panels on the current photovoltaic platform is calculated using the formula Tavg = (YS / S) × T + (1-(YS / S)) × T0; where Tavg is the overall average transmittance, S is the total area of the photovoltaic panels on the photovoltaic platform, and T0 is the initial transmittance. The salt film power generation efficiency loss factor is calculated by the formula Δη=η0×(1-(Tavg / T0)); where Δη is the salt film power generation efficiency loss factor and η0 is the initial power generation efficiency.

6. The prediction system for offshore photovoltaic power generation according to claim 1, characterized in that: The calculation based on the salt film power generation efficiency loss factor to obtain the current photovoltaic panel's initial power generation efficiency includes: By calculating the difference between the initial power generation efficiency and the salt membrane power generation efficiency loss factor, the current photovoltaic starting power generation efficiency is obtained.

7. The prediction system for offshore photovoltaic power generation according to claim 1, characterized in that: The photovoltaic power generation efficiency is calculated based on the salt spray concentration and the initial power generation efficiency, including: The current salt spray concentration is identified through the pre-trained salt spray deposition rate recognition model to obtain the corresponding salt spray deposition rate; Obtain a preset time period for the required predicted power generation; divide the preset time period into a number of time periods i according to preset time intervals; where i = {1, 2, 3, ..., N}, and N is the total number of divided time periods; By the formula ηi=η a ×e^(-β×R×ti) to calculate the photovoltaic power generation efficiency of each time period; where ηi refers to the photovoltaic power generation efficiency of the i-th time period, β is the proportional coefficient, η a is the initial power generation efficiency of photovoltaics, R is the salt spray deposition rate, and ti refers to the length of the i-th time period.

8. The prediction system for offshore photovoltaic power generation according to claim 7, characterized in that: The training method of the salt spray deposition rate identification model includes: Obtain several salt spray concentrations and corresponding salt spray deposition rates based on the historical salt spray database; An artificial intelligence model is trained based on training data to obtain a salt spray deposition rate identification model; wherein the training data includes: training input data and training output data; the training input data is salt spray concentration, and the training output data is the salt spray deposition rate corresponding to the salt spray concentration.

9. The prediction system for offshore photovoltaic power generation according to claim 7, characterized in that: The method of predicting offshore photovoltaic power generation based on photovoltaic power generation efficiency includes: Predict the power generation of offshore photovoltaics through a preset prediction model; The preset prediction model is: E=∑(ηi×Gi×S×ti); wherein ∑ is a summation factor, the summation range is (1, N), and Gi is the light intensity in the i-th time period.

10. A method for predicting offshore photovoltaic power generation, applied to a system for predicting offshore photovoltaic power generation according to any one of claims 1 to 9, characterized in that: include: Obtain images of photovoltaic panels on offshore photovoltaic platforms and salt spray concentrations; By analyzing the images of photovoltaic panels, salt film data is obtained; The salt film power generation efficiency loss factor was calculated based on the salt film data; The initial power generation efficiency of the current photovoltaic panel is calculated based on the salt film power generation efficiency loss factor; The photovoltaic power generation efficiency is calculated based on the salt spray concentration and the initial power generation efficiency; Predicting offshore photovoltaic power generation based on photovoltaic power generation efficiency.

Images