Efficiency optimization method, device and system of photovoltaic photo-thermal system

Abstract

The invention relates to the technical field of solar power generation, in particular to an efficiency optimization method, equipment and system of a photovoltaic photo-thermal system. According to real-time weather conditions, a primary tracking device of a four-side step photoelectric sensor is started, primary adjustment is carried out on an azimuth angle and an elevation angle of sunlight incidence, photovoltaic power generation or photo-thermal power generation is alternately selected, and real-time weather is obtained through a cloud picture and is corrected through a meteorological radar; in addition, the actual generating capacity of one day is compared with the generating capacity predicted according to the weather of one day, a secondary tracking device of a four-side stepped photoelectric sensor is started according to the comparison result, secondary adjustment is conducted on the azimuth angle and the elevation angle of sunlight incidence, and the alternating selection rule of photovoltaic power generation or photo-thermal power generation is corrected, so that the generation capacity is improved. Therefore, photovoltaic and photo-thermal efficiency optimization is achieved, the service life of the photovoltaic panel and the photo-thermal device is prolonged, and solar power generation efficiency optimization is achieved according to real-time weather conditions every day.

Description

Technical Field

[0001] This invention relates to the field of solar power generation technology, and in particular to a method, equipment and system for optimizing the efficiency of a photovoltaic thermal system. Background Technology

[0002] As a renewable energy source, solar energy utilization technology is attracting increasing attention and is developing rapidly. Solar power generation includes various forms such as photovoltaic (PV) power generation and concentrated solar power (CSP). PV power generation directly converts light energy into electrical energy using semiconductor materials and has become the mainstream technology for solar energy utilization. CSP power generation converts solar energy into heat energy through a concentrating system and then generates electricity through a thermal cycle, combining power generation and energy storage functions.

[0003] However, photovoltaic power generation is still significantly affected by day and night cycles and weather fluctuations. In particular, the photoelectric conversion efficiency of photovoltaic power generation will decrease significantly as the temperature of photovoltaic panels rises due to continuous solar radiation. On the other hand, solar thermal power generation has high investment costs, and solar thermal collectors are difficult to further improve the utilization efficiency of solar energy due to heat loss and other reasons. Moreover, the power generation efficiency of solar thermal power generation is not as high as that of photovoltaic power generation. Summary of the Invention

[0004] In view of the shortcomings of existing solar photovoltaic and solar thermal power generation technologies, this solution proposes a combined solar power generation method, equipment and system using photoelectric sensors to optimize the efficiency of photovoltaic and solar thermal power generation, extend the service life of photovoltaic panels and solar thermal generators, and achieve the highest efficiency of solar power generation based on daily real-time weather conditions.

[0005] On the one hand, such as Figure 1 As shown, the present invention provides a method for optimizing the efficiency of a photovoltaic-thermal system, comprising:

[0006] S1: Obtain real-time weather information, which includes first weather information. The first weather information is obtained by collecting ground-based cloud images in real time, identifying and segmenting cloud shapes in the ground-based cloud images, and calculating cloud cover.

[0007] S2: Based on the first weather information, activate the first-level tracking device of the four-sided stepped photoelectric sensor to make a first-level adjustment to the azimuth and elevation angles of the sunlight incident.

[0008] S3: Obtain the second weather information, which is a weather forecast. The second weather information is obtained once when the first weather information changes.

[0009] S4: Determine the duration of the real-time weather based on the first weather information and the second weather information; compare the duration with a first set value based on the first weather information and the duration; and select whether to use photovoltaic or solar thermal power generation based on the comparison result.

[0010] S5: Calculate the difference between the actual power generation and the predicted power generation, activate the secondary tracking device of the four-sided stepped photoelectric sensor based on the difference, perform secondary adjustments to the azimuth and elevation angles of the sunlight incident, adjust the first set value to obtain the second set value based on the difference, and adjust the comparison result based on the second set value.

[0011] Specifically, such as Figure 2 As shown, S5 includes:

[0012] S501: The actual power generation includes a first power generation and a second power generation, the first power generation is the actual photovoltaic power generation, the second power generation is the actual solar thermal power generation, and the predicted power generation includes a first predicted power and a second predicted power, the first predicted power is the photovoltaic predicted power, and the second predicted power is the solar thermal predicted power.

[0013] S502: The difference includes a first difference and a second difference. The first difference is the difference between the first power generation and the first predicted power. The second difference is the difference between the second power generation and the second predicted power. When the first difference is greater than the first threshold and / or the second difference is greater than the second threshold, return to S3, adjust the first setting value to obtain the second setting value, adjust the comparison result according to the second setting value, and select to use photovoltaic or solar thermal power generation according to the comparison result.

[0014] Preferably, such as Figure 3 As shown, the real-time weather information also includes:

[0015] S601: Third weather information, wherein the third weather information is obtained by acquiring the original signal through weather radar, and then performing ground clutter identification and ground clutter removal. The ground clutter identification is performed by using a segmented entropy value feature identification algorithm to identify ground clutter, and the ground clutter removal is performed by using an adaptive Gaussian filtering algorithm to remove ground clutter from the identified ground clutter signal to obtain the third weather information.

[0016] S602: When the first weather information is not sunny, the first weather information is corrected using the third weather information.

[0017] Specifically, it also includes:

[0018] When the first weather information indicates a sunny day, photovoltaic or solar thermal power generation is selected based on the comparison results.

[0019] When the first weather information is not sunny, humidity information is obtained. When the humidity information is greater than a third threshold, photovoltaic power generation is used. When the humidity information is less than the third threshold, solar thermal power generation is used.

[0020] Specifically, the cloud-shaped identification of the ground cloud map includes: performing pixel point calculation on the ground cloud map to obtain a first pixel value and a second pixel value, wherein the first pixel value is the number of cloud map pixels, the second pixel value is the number of non-sky pixels, and the first pixel value is subtracted from the second pixel value to obtain a third pixel value;

[0021] The cloud segmentation includes: extracting features from the ground cloud map, upsampling the extracted data to obtain the segmented ground cloud map, calculating the third pixel value of the segmented ground cloud map, and the fourth pixel value being the number of cloud pixels;

[0022] The first pixel value, the second pixel value, the third pixel value, and the fourth pixel value are substituted into the cloud cover calculation formula to obtain cloud cover information, and the real-time weather is determined by the cloud cover information.

[0023] Specifically, the first predicted power includes:

[0024] The first weather information is acquired every M time intervals, and N first weather information items are obtained as a historical weather set, where M and N are positive integers greater than 0;

[0025] The historical weather set is used to determine the weather type using historical GHI sequences, and the historical weather set after type determination is used to classify the weather using a Gaussian mixture model.

[0026] The first predicted power is obtained by inputting the weather classification data into a multi-head attention LSTM model.

[0027] Specifically, the second predicted power includes:

[0028] The photothermal efficiency coefficient is calculated based on the parameters of the photothermal device, including pipe spacing, pipe outer diameter, pipe inner diameter, fluid heat transfer coefficient, pipe and fin conductivity, and fin efficiency.

[0029] The average mass flow rate, specific heat of the fluid, heat sink area, and total conductivity of the photothermal device are obtained, and the heat dissipation coefficient is calculated based on the photothermal efficiency coefficient.

[0030] The predicted effective heat gain is calculated based on the heat dissipation coefficient, the overall heat transfer coefficient, and the average temperature of the heat absorber plate.

[0031] The first weather information is acquired every M time intervals, and N first weather information are obtained as a historical weather set, where M and N are positive integers greater than 0. The predicted incident irradiance for that day is calculated based on the historical weather information set.

[0032] The second predicted power is calculated based on the predicted effective heat gain and the predicted incident irradiance.

[0033] Specifically, it also includes:

[0034] The four-sided stepped photoelectric sensor includes a primary tracking device and a secondary tracking device;

[0035] The primary tracking device includes an eastward sensor, a westward sensor, a southward sensor, a northward sensor, and a dual-angle adjustment platform. The eastward, westward, southward, and northward sensors are respectively distributed along the four stepped directions of the photoelectric sensor, with the eastward and westward sensors facing each other, and the southward and northward sensors facing each other. The dual-angle adjustment platform includes an azimuth adjustment platform and an elevation adjustment platform. The azimuth adjustment platform is adjusted according to the magnitude of the electrical signals output by the eastward and westward sensors after receiving sunlight, and the elevation adjustment platform is adjusted according to the magnitude of the electrical signals output by the southward and northward sensors after receiving sunlight.

[0036] The secondary tracking device includes two square Fresnel lenses, four photosensitive elements, and a rotating shaft. After the primary tracking device adjusts the azimuth and elevation angles, the azimuth and elevation angles are adjusted in a secondary manner through the dual-angle adjustment platform based on the difference. When the square wave amplitudes output by the four photosensitive elements are equal in pairs, the azimuth angle adjustment is completed. The elevation angle is then adjusted, and when the square wave amplitudes output by the four photosensitive elements are all equal, the elevation angle adjustment is completed.

[0037] On the one hand, the present invention also provides an efficiency optimization device for a photovoltaic and solar thermal system, comprising:

[0038] The first weather information acquisition unit is used to acquire real-time weather information, which includes first weather information. The first weather information is obtained by collecting ground-based cloud images in real time, identifying and segmenting cloud shapes in the ground-based cloud images, and calculating cloud cover.

[0039] The primary tracking unit is used to activate the primary tracking device of the four-sided stepped photoelectric sensor based on the first weather information, and to make primary adjustments to the azimuth and elevation angles of the incident sunlight.

[0040] The second weather information acquisition unit is used to acquire the second weather information, which is weather forecast information. The second weather information is acquired once when the first weather information changes.

[0041] The third weather information acquisition unit obtains the third weather information by acquiring the original signal through meteorological radar, and then performing ground clutter identification and ground clutter removal. The ground clutter identification is performed by using a segmented entropy value feature identification algorithm to identify ground clutter, and the ground clutter removal is performed by using an adaptive Gaussian filtering algorithm to remove ground clutter from the identified ground clutter signal to obtain the third weather information.

[0042] The judgment unit is used to determine the duration of real-time weather based on the first weather information and the second weather information, compare the duration with a first set value based on the first weather information and the duration, and select to use photovoltaic or solar thermal power generation based on the comparison result.

[0043] The first adjustment unit is used to calculate the difference between the actual power generation and the predicted power generation, adjust the first set value according to the difference to obtain a second set value, and adjust the comparison result according to the second set value.

[0044] The secondary tracking unit is used to activate the secondary tracking device of the four-sided stepped photoelectric sensor based on the difference between the calculated actual power generation and the predicted power generation, and to make secondary adjustments to the azimuth and elevation angles of the incident sunlight.

[0045] The second adjustment unit is used to correct the first weather information using the third weather information when the first weather information is not sunny.

[0046] Preferably, the efficiency optimization device for the photovoltaic-thermal system further includes:

[0047] The real-time weather judgment unit includes a ground-based cloud image recognition unit and a cloud shape segmentation unit. The real-time weather judgment unit is used to input the calculation data of the ground-based cloud image recognition unit and the cloud shape segmentation unit into the cloud volume calculation formula to obtain cloud volume information, and to judge the real-time weather through the cloud volume information.

[0048] The first prediction power unit is used to determine the weather type after obtaining the historical weather set, then use a Gaussian mixture model to classify the weather, and finally input the weather classification into a multi-head attention LSTM model to obtain the first prediction power.

[0049] The second prediction power unit is used to calculate the second prediction power based on the predicted effective heat gain and predicted incident irradiance of the photothermal device.

[0050] On the one hand, the present invention also provides an efficiency optimization system for a photovoltaic-thermal system, including a photovoltaic power generation device and a solar thermal power generation device, and further including:

[0051] Imaging system, weather radar, humidity sensor, temperature sensor, solar tracking device, processor and memory;

[0052] The imaging system is an all-sky imaging device with a fisheye lens, used to capture ground-based cloud image information of the sky panorama;

[0053] The weather radar is an X-band weather radar used to acquire real-time weather information;

[0054] The humidity sensor is used to acquire ambient humidity information;

[0055] The temperature sensor is used to acquire the device temperature and ambient temperature of the solar thermal power generation equipment.

[0056] The solar tracking device is used to track sunlight in real time and adjust the position of the photovoltaic and solar thermal power generation equipment relative to the sunlight based on the tracking results;

[0057] The memory is coupled to the processor, and the memory stores executable program modules. The processor is used to run the program modules stored in the memory to implement the efficiency optimization method of the photovoltaic and solar thermal system as described above.

[0058] The above scheme activates the primary tracking device of the four-sided stepped photoelectric sensor based on real-time weather conditions. This device performs primary adjustments to the azimuth and elevation angles of sunlight incidence and alternates between photovoltaic (PV) and solar thermal (CSP) power generation. Real-time weather data is obtained from cloud images and corrected using weather radar. Furthermore, the scheme compares the actual daily power generation with the predicted power generation based on the day's weather forecast. Based on the comparison results, the secondary tracking device of the four-sided stepped photoelectric sensor is activated to perform secondary adjustments to the azimuth and elevation angles of sunlight incidence and correct the alternation rules for PV and CSP power generation. This optimizes the efficiency of both PV and CSP, extends the lifespan of PV panels and CSP units, and achieves optimal solar power generation efficiency based on daily real-time weather conditions. Attached Figure Description

[0059] Figure 1 Flowchart of efficiency optimization method for photovoltaic and solar thermal systems;

[0060] Figure 2 Another flowchart for efficiency optimization methods of photovoltaic and solar thermal systems;

[0061] Figure 3 Another flowchart for efficiency optimization methods of photovoltaic and solar thermal systems;

[0062] Figure 4 A block diagram of equipment for optimizing the efficiency of photovoltaic and solar thermal systems. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in sequences other than those illustrated or described herein.

[0065] It should be understood that in the various embodiments of the present invention, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0066] It should be understood that in this invention, "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product, or device.

[0067] It should be understood that in this invention, "multiple" refers to two or more. "And / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, "and / or B" can represent: A existing alone, A and B existing simultaneously, and B existing alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "Contains A, B, and C", "Contains A, B, and C" means that all three A, B, and C are contained; "Contains A, B, or C" means that one of A, B, and C is contained; "Contains A, B, and / or C" means that any one, two, or three of A, B, and C are contained.

[0068] It should be understood that in this invention, "B corresponding to A", "B corresponding to A", "A and B correspond", or "B and A correspond" means that B is associated with A, and B can be determined based on A. Determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information. Matching A and B is defined as a similarity between A and B that is greater than or equal to a preset threshold.

[0069] Depending on the context, "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection."

[0070] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0071] Example 1

[0072] This embodiment provides a method for optimizing the efficiency of a photovoltaic thermal system, including:

[0073] S1: Obtain real-time weather information, which includes first weather information. The first weather information is obtained by collecting ground-based cloud images in real time, identifying and segmenting cloud shapes in the ground-based cloud images, and calculating the cloud cover.

[0074] S2: Based on the first weather information, activate the first-level tracking device of the four-sided stepped photoelectric sensor to make a first-level adjustment to the azimuth and elevation angles of the sunlight incident.

[0075] S3: Obtain the second weather information, which is the weather forecast information. The second weather information is obtained once when the first weather information changes.

[0076] S4: Determine the duration of real-time weather based on the first weather information and the second weather information. Compare the duration with the first set value based on the first weather information and the duration. Select whether to use photovoltaic or solar thermal power generation based on the comparison result.

[0077] S5: Calculate the difference between the actual power generation and the predicted power generation, activate the secondary tracking device of the four-sided stepped photoelectric sensor based on the difference, perform secondary adjustments to the azimuth and elevation angles of the sunlight incident, adjust the first set value to obtain the second set value based on the difference, and adjust the comparison result based on the second set value.

[0078] It should be noted that, regarding step S1, clouds in the sky are an important reference indicator for meteorological observation. The amount of cloud cover can be used to determine whether the weather is sunny or cloudy, overcast or rainy, etc. Cloud image monitoring is generally used in flight, ship navigation, and weather forecasting. This step applies the calculation of cloud cover from ground-based cloud images to determine real-time weather conditions in the field of solar power generation, providing an important basis for the selection of solar power generation methods. Ground-based cloud images are acquired using an all-sky imaging device integrated with a fisheye lens. The acquired ground-based cloud images first need to be identified by cloud formation, then segmented, and finally the amount of cloud cover is calculated using a formula. By comparing the correspondence between cloud cover and weather data in the database, real-time weather information can be obtained. Cloud pattern recognition requires the establishment of a cloud pattern recognition model. This model first performs image preprocessing on the ground-based cloud image, followed by feature extraction. Image preprocessing includes segmenting the ground-based cloud image into blocks, flattening each block, adding coordinate information, and then inputting the image blocks with added coordinate information into a multi-head attention mechanism and a multilayer perceptron designed with a convolutional neural network. The multi-head attention mechanism obtains the basic information of the ground-based cloud image and the distance relationships between the various image blocks, which is then input into the multilayer perceptron for linear transformation, thereby extracting the deeper information of the ground-based cloud image. The system identifies 11 cloud states: altocumulus, stratocumulus, cirrocumulus, altocumulus, cirrostratus, nimbostratus, cumulonimbus, cirrus, cumulus, clear sky, and stratus. The identified cloud states are then segmented using a Transformer cloud image segmentation model with encoding and decoding. The ground-based cloud image with extracted deep information is first deconvolved, then input into a ReLU activation function, followed by bilinear interpolation, and finally outputting the segmented image. Finally, cloud cover is calculated, and the relationship between cloud cover and weather in the database is compared to obtain real-time weather data.

[0079] Regarding step S2, after obtaining the cloud cover from real-time weather information collected from the ground-based cloud map, it is determined whether to activate the four-sided stepped photoelectric sensor to track sunlight based on the weather conditions. In non-sunny weather conditions, such as cloudy, rainy, or snowy days, sunlight is obviously insufficient. Therefore, in such weather conditions, it is not necessary to activate the solar tracking device; the existing photovoltaic thermal equipment can be used directly for power generation. In sunny weather conditions, the angle and direction of the photovoltaic thermal device are adjusted to better align it with sunlight. Combined with the photovoltaic thermal selection strategy, this allows for more efficient utilization of sunlight for power generation.

[0080] For step S3, weather forecast information issued by the meteorological bureau is obtained online. This information is only acquired when changes in real-time weather, as determined by the captured ground-based cloud image, occur. The purpose of this is that while the immediate weather is determined, its duration is difficult to predict using ground-based cloud images; therefore, it is necessary to combine this information with the weather forecast information issued by the meteorological bureau to determine the duration.

[0081] Regarding step S4, the first weather information, namely the real-time weather information obtained through ground-based cloud maps, is used to determine the real-time weather at a certain moment. The second weather information, namely the weather forecast information, is used to determine how long the real-time weather at that moment will last. Since the power generation efficiency of photovoltaic panels decreases as the temperature rises to a certain level, in the case of a sunny day, if the sunny weather lasts for a certain period of time, causing the temperature of the photovoltaic panels to rise to the maximum power generation efficiency, the continuous solar radiation will continue to heat the photovoltaic panels, leading to a decrease in the power generation efficiency of the photovoltaic panels. Therefore, a first set value is set, which is a time duration. When the sunny weather lasts for the first set value, the photovoltaic panels are stopped from generating electricity, and instead, a solar thermal generator, which has a higher power generation efficiency at high temperatures, is used. This achieves the effect of alternating the use of photovoltaic panels and solar thermal generators to extend the service life of both devices, and also optimizes the power generation efficiency.

[0082] For step S5, the predicted power generation is obtained by predicting the power generation based on the real-time weather data recorded throughout the day. This predicted power generation is then compared with the actual power generation of the day, achieved through the alternating selection of photovoltaic panels and solar thermal generators in steps S1-S4. This comparison reveals whether the actual power generation is above or below the predicted power generation. If the actual power generation is above the predicted power generation, it indicates that the alternation of photovoltaic and solar thermal power generation is effective that day, meaning the selection result obtained by comparing the duration using the first setpoint meets the requirements. If the actual power generation is below the predicted power generation, it indicates that the alternation of photovoltaic and solar thermal power generation is ineffective that day, meaning the selection result obtained by comparing the duration using the first setpoint does not meet the requirements. In this case, the first setpoint needs to be adjusted to obtain an adjusted second setpoint. This second setpoint is then used as the benchmark setpoint for comparing the duration of photovoltaic and solar thermal power generation on the following day. This allows the system to adaptively adjust the alternation rules of photovoltaic and solar thermal devices, thereby optimizing power generation efficiency. In addition, the actual power generation differs significantly from the predicted power generation, which is below the predicted power generation. This may be due to poor solar tracking. Therefore, in order to further optimize power generation efficiency and improve the utilization of sunlight, a secondary tracking device was added to adjust the incident azimuth and elevation angles of sunlight in two stages. This adjusts the angle and direction of the photovoltaic thermal device to better align it with the sunlight and further optimize power generation efficiency.

[0083] Specifically, S5 includes:

[0084] S501: Actual power generation includes first power generation and second power generation. The first power generation is the actual photovoltaic power generation and the second power generation is the actual solar thermal power generation. Predicted power generation includes first predicted power and second predicted power. The first predicted power is the photovoltaic predicted power and the second predicted power is the solar thermal predicted power.

[0085] S502: The difference includes a first difference and a second difference. The first difference is the difference between the first power generation and the first predicted power, and the second difference is the difference between the second power generation and the second predicted power. When the first difference is greater than the first threshold and / or the second difference is greater than the second threshold, return to S3, adjust the first setting value to obtain the second setting value, adjust the comparison result according to the second setting value, and select to use photovoltaic or solar thermal power generation according to the comparison result.

[0086] It should be noted that the calculation standards for the actual and predicted power generation of photovoltaic (PV) and solar thermal power are different. Therefore, in step S5, the comparison between the actual and predicted power generation must be treated separately. A large absolute value of the first difference likely indicates that the PV usage time is too short, while a large absolute value of the second difference likely indicates that the solar thermal usage time is too short. In either case, if the actual PV power generation differs significantly from the predicted PV power generation, or the actual solar thermal power generation differs significantly from the predicted solar thermal power generation, it means that the rule of alternating PV and solar thermal usage needs to be changed. Therefore, a threshold judgment for the difference is added to this step, that is, a first threshold for PV comparison and a second threshold for solar thermal comparison are set to obtain the second set value and comparison result more accurately.

[0087] Preferably, the real-time weather information also includes:

[0088] S601: Third weather information. The third weather information is obtained by acquiring the original signal through weather radar, and then performing ground clutter identification and removal. Ground clutter identification is performed by using a segmented entropy value feature identification algorithm to identify ground clutter. Ground clutter removal is performed by using an adaptive Gaussian filtering algorithm to remove ground clutter from the identified ground clutter signal to obtain the third weather information.

[0089] S602: When the first weather information is not sunny, the third weather information is used to correct the first weather information.

[0090] It should be noted that ground-based cloud imagery has its limitations in determining real-time weather. For certain weather conditions, such as cloudy skies, especially heavy snow, or abundant clouds without rain, the assessment may be inaccurate. Therefore, this step incorporates the use of X-band weather radar to correct the real-time weather assessment from the ground-based cloud imagery. Because of its shorter wavelength, X-band weather radar experiences significant signal attenuation in the air, which greatly affects the radar echo and thus the assessment results. Therefore, when using X-band weather radar for complementary correction, ground clutter removal is also necessary to ensure the cleanliness and integrity of the radar signal, which is more conducive to correcting the real-time weather from the ground-based cloud imagery. Ground clutter removal involves first identifying ground clutter, then removing it, and finally obtaining the third-party weather information. First, a segmented entropy feature identification algorithm is used to identify ground clutter. Ground clutter exists at zero frequency, requiring a normal distribution based on zero frequency. The radar spectral data is sorted according to amplitude, and all spectral data is segmented. The segmentation is based on the absolute value of the difference between the maximum and minimum amplitudes, corresponding to the number of segments. The number of spectra in each segment is calculated, and the sum of the probabilities of each spectra occurrence in each segment is obtained. This is then calculated using the information entropy calculation formula:

[0091] F(a)=-∑S(a i log2S(a i )

[0092] Where S(a) i The probability of each frequency band appearing in each segment is multiplied by the number of frequency points in the interval.

[0093] The spectral entropy value for the entire interval is then obtained as follows:

[0094]

[0095] Where N represents the number of segments corresponding to the absolute value of the difference between the maximum and minimum amplitudes. Then, a threshold judgment is performed on the calculated SE(a) value. The judgment threshold is selected based on the simulation data in the laboratory, which shows that a threshold of 0.35 is more suitable. Therefore, 0.35 is selected as the judgment threshold. If the calculated spectral entropy value of the entire interval is greater than 0.35, it is determined that ground clutter exists and the next step of removal operation is required.

[0096] First, the radar data window is filtered. Since the removal effect is related to the selection of the data window, a smaller cone-shaped window width results in a smaller calculated variance, which can lead to misjudgment of ground clutter signals and reduce the removal effect. Conversely, a smaller cone-shaped window width can produce the opposite result. Therefore, it is necessary to select an appropriate data window width by calculating the clutter-to-noise ratio, based on the formula for calculating the clutter-to-noise ratio:

[0097]

[0098] Where N represents noise power, T is the number of samples, and P(t) is the radar echo signal. Different window functions are selected based on the calculated clutter-to-noise ratio (CNR). These window functions include rectangular windows, Blackman windows, and Haining windows. The Haining window is used when the calculated CNR indicates no ground clutter in the echo signal; the Blackman window is used when the calculated CNR indicates mixed ground clutter.

[0099] Next, the acceleration spectral density is calculated and the clutter range is determined. The calculation method for the acceleration spectral density is a common algorithm and will not be elaborated here. Using the calculated acceleration spectral density, the spectral coefficients near the zero Doppler velocity where ground clutter contamination occurs can be determined. Because the acceleration spectral density power amplitude of clutter signals changes rapidly, it will attenuate at the peak, differing from the expected value. In contrast, the clean radar signal, due to the slow change in acceleration spectral density power amplitude, shows almost no deviation. Finally, an adaptive Gaussian filtering algorithm is used to remove the ground clutter signal. The radar signal after removing the ground clutter signal becomes the third weather information. This ground clutter-removed radar signal is used to correct the real-time weather information only when the first weather information, i.e., the real-time weather information, is not clear.

[0100] Specifically, it also includes:

[0101] S6011: When the first weather information is sunny, select either photovoltaic or solar thermal power generation based on the comparison results;

[0102] S6012: When the first weather information is not sunny, obtain the humidity information. When the humidity information is greater than the third threshold, use photovoltaic power generation. When the humidity information is less than the third threshold, use solar thermal power generation.

[0103] It's important to note that when the real-time weather is sunny, the power generation efficiency of photovoltaic (PV) panels decreases as the temperature rises to a certain level. Specifically, in a prolonged sunny day, the continuous solar radiation heats the PV panels, causing their temperature to rise continuously. Once the temperature reaches a certain point, the PV panel's power generation efficiency decreases, and prolonged use under high temperatures reduces its lifespan. Continuing to use PV power generation in this situation obviously doesn't achieve optimal efficiency. However, such sunny weather with continuous high solar radiation is actually beneficial for the power generation efficiency of solar thermal collectors (SPPGs). Therefore, the above describes using real-time weather assessment and its duration to determine when to choose between PV panels and SPPGs for operation. When the real-time weather is not sunny, such as cloudy or rainy weather... In general, weather conditions vary, with some solar panels being more suitable than others, such as cloudy days, after rain, or extremely dry weather. For example, on cloudy days, the power generation efficiency of solar thermal pipes is significantly lower than that of solar panels. However, there is no issue of solar panels' efficiency decreasing due to high temperatures. Therefore, if the initial weather forecast is cloudy and persistent, solar panels should be used for power generation. On rainy days and after rain, the rainwater washes away impurities from the solar panels, improving their efficiency. However, the lower temperatures on rainy days and after rain are unsuitable for solar thermal pipes. Therefore, if the initial weather forecast is rainy or after rain, solar panels should be used for power generation. In extremely dry weather, solar thermal pipes will actually have a better power generation efficiency than solar panels. Therefore, if the initial weather forecast is not sunny and the humidity sensor shows a humidity level below 40%, solar thermal pipes should be used for power generation.

[0104] Specifically, it also includes:

[0105] S101: Cloud identification of the ground cloud map includes: calculating the number of pixels in the ground cloud map to obtain the first pixel value and the second pixel value. The first pixel value is the number of pixels in the cloud map, the second pixel value is the number of non-sky pixels, and the third pixel value is obtained by subtracting the first pixel value from the second pixel value.

[0106] S102: Cloud segmentation includes: extracting features from the ground cloud map, upsampling the extracted data to obtain the segmented ground cloud map, and calculating the fourth pixel value of the segmented ground cloud map, where the fourth pixel value is the number of cloud pixels.

[0107] S103: Substitute the first pixel value, the second pixel value, the third pixel value, and the fourth pixel value into the cloud cover calculation formula to obtain cloud cover information, and use the cloud cover information to determine the real-time weather.

[0108] It should be noted that when all-sky imaging devices with fisheye lenses acquire ground-based cloud image information, the four corners of the acquired ground-based cloud image contain redundant images, i.e., invalid images that are neither clouds nor sky. When calculating cloud cover, these redundant images need to be removed first to obtain the total area of ​​the valid cloud and sky region. Finally, the cloud area is calculated to obtain the cloud cover value of the ground-based cloud image. First, the number of pixels in the ground-based cloud image is calculated to determine the number of cloud image pixels and the number of non-sky pixels. The third pixel value is obtained by subtracting the number of cloud image pixels from the number of non-sky pixels. Then, the segmentation model used is the Transformer cloud image segmentation model with encoding and decoding. The ground-based cloud image with extracted deep information is first deconvolved and then input into the ReLU activation function, followed by bilinear interpolation to obtain the fourth pixel value. Finally, the first, second, third, and fourth pixel values ​​calculated above are substituted into the cloud cover calculation formula to calculate the cloud cover.

[0109]

[0110] Where Y is cloud cover, Ji is the correction coefficient, K1 is the first pixel value, K2 is the second pixel value, K3 is the third pixel value, and K4 is the fourth pixel value.

[0111] Specifically, the first predicted power includes:

[0112] S50111: Obtain the first weather information every M time intervals, and obtain N first weather information as a historical weather set, where M and N are positive integers greater than 0;

[0113] S50112: Use historical GHI sequences to determine weather types from historical weather sets, and use Gaussian mixture models to classify weather from the determined historical weather sets.

[0114] S50113: The weather classification input is used to obtain the first prediction power in a multi-head attention LSTM model.

[0115] It should be noted that after sunset, the predicted power of photovoltaic power generation throughout the day is statistically analyzed to verify whether the selection rules for photovoltaic panels and solar thermal pipes are optimal. The first weather information, obtained through ground-based cloud maps and corrected using third-party weather information, is sampled every M time intervals. For example, if M is 1, the first weather information is sampled every hour. After sampling, a set of N sets of first weather information is obtained, revealing all the weather conditions for the day. Then, historical weather sets are used to determine the weather type using historical GHI sequences. The input step size is determined based on the historical GHI sequences. The training set data is classified according to the weather clustering results and the input step size. A backpropagation (BP) classifier is trained using the training set data, and the trained BP classifier is used to classify the test set data to determine the weather type. Next, a Gaussian mixture model is used to classify the weather from the historical weather set after type determination. The test set data is classified according to the weather clustering results and the input step size to obtain the weather classification. Finally, the weather classification is input into a multi-head attention LSTM model to obtain the first predicted power.

[0116] Specifically, the second predicted power includes:

[0117] S50121: The photothermal efficiency coefficient is calculated based on the parameters of the photothermal device. The parameters of the photothermal device include pipe spacing, pipe outer diameter, pipe inner diameter, fluid heat transfer coefficient, radiation heat transfer coefficient, pipe and fin conductivity, heat transfer area to heat collection aperture ratio, fin efficiency, fin thermal conductivity and fin thickness.

[0118] S50122: Obtain the average mass flow rate, specific heat of the fluid, heat sink area, and total conductivity of the photothermal device, and calculate the heat dissipation coefficient based on the photothermal efficiency coefficient;

[0119] S50123: Calculate the predicted effective heat gain based on the heat dissipation coefficient, the overall heat transfer coefficient, and the average temperature of the heat absorber plate;

[0120] S50124: Obtain the first weather information every M time intervals, obtain N first weather information as a historical weather set, where M and N are positive integers greater than 0, and calculate the predicted incident irradiance for that day based on the historical weather information set;

[0121] S50125: Calculate the second predicted power based on the predicted effective heat gain and the predicted incident irradiance.

[0122] It should be noted that, firstly, the photothermal efficiency coefficient E' is calculated based on the parameters of the photothermal device, including the pipe spacing D, the pipe outer diameter D0, and the pipe inner diameter D. i Fluid heat transfer coefficient l wf Conductivity B of tubes and fins C The specific formulas for calculating fin efficiency E and overall conductivity U are:

[0123]

[0124] Then, the average mass flow rate N of the fluid in the photothermal device is obtained. v Fluid specific heat Q P Radiator area A c And based on the photothermal efficiency coefficient E', the heat dissipation coefficient E is calculated. R The formula for calculating the heat dissipation coefficient is:

[0125]

[0126] Next, based on the heat dissipation coefficient E R Overall heat transfer coefficient U, average temperature of the absorber plate T PM and average ambient temperature T a Calculate the predicted effective heat gain Q U The formula for calculating effective heat gain is:

[0127]

[0128] Then, the first weather information is obtained every M time interval, and N first weather information are obtained as a historical weather set, where M and N are positive integers greater than 0. The predicted incident irradiance for that day is calculated based on the historical weather information set. Similar to the calculation of the first predicted power, after sunset, the predicted power of photovoltaic power generation during the day is statistically analyzed to verify whether the rules for selecting photovoltaic panels and solar thermal pipes are optimal. The first weather information obtained through ground-based cloud maps and corrected by the third weather information is sampled every M time interval. For example, if M is 1, the first weather information is sampled every 1 hour. After sampling, a set of N first weather information will be obtained, and all the weather conditions for that day will be known. The predicted incident irradiance for that day is calculated based on the first weather information set.

[0129] Finally, based on the predicted effective heat gain Q U And the predicted incident irradiance R, and the second predicted power P is calculated. r The second predicted power calculation formula is:

[0130] Specifically, the four-sided stepped photoelectric sensor includes a primary tracking device and a secondary tracking device;

[0131] The primary tracking device includes an eastward sensor, a westward sensor, a southward sensor, a northward sensor, and a dual-angle adjustment platform. The eastward, westward, southward, and northward sensors are respectively distributed along the four stepped directions of the photoelectric sensor, with the eastward and westward sensors facing each other, and the southward and northward sensors facing each other. The dual-angle adjustment platform includes an azimuth adjustment platform and an elevation adjustment platform. The azimuth adjustment platform is adjusted according to the magnitude of the electrical signals output by the eastward and westward sensors after receiving sunlight, and the elevation adjustment platform is adjusted according to the magnitude of the electrical signals output by the southward and northward sensors after receiving sunlight.

[0132] The secondary tracking device includes four circular Fresnel lenses, four photosensitive elements, and a rotating shaft. After the primary tracking device adjusts the azimuth and elevation angles, the azimuth and elevation angles are adjusted in a secondary manner through the dual-angle adjustment platform based on the difference. When the square wave amplitudes output by the four photosensitive elements are equal in pairs, the azimuth angle adjustment is completed. The elevation angle is then adjusted, and when the square wave amplitudes output by the four photosensitive elements are all equal, the elevation angle adjustment is completed.

[0133] It should be noted that for the primary tracking device, there are four sensors in each direction of the four-sided steps; that is, there are four sensors each for the east, west, south, and north directions, for a total of sixteen. These sixteen photoelectric sensors, distributed across the four-sided steps, detect the intensity of sunlight incident from each direction in real time. By comparing the peak intensities of the four sets of signals (east, west, south, and north), the spatial azimuth of the sun is calculated. Based on the azimuth calculation results, the stepper motor of the dual-angle adjustment platform is controlled to perform large-angle rotations in both the azimuth (east-west axis) and altitude (north-south axis) to complete the initial capture and tracking of the sun. It is important to note that the dual-angle adjustment platform is part of the primary tracking device and is only used to adjust the angle and direction of the solar tracking device. First, the azimuth angle needs to be adjusted. This requires eight photoelectric sensors along the east-west axis. The signals output by these eight sensors are differentially detected in real time. The photocurrent gradient function is used to calculate and compare the electrical signals of the eight photoelectric sensors to determine whether the dual-angle adjustment platform needs to be adjusted in the azimuth direction. This continues until the electrical signals output by the photoelectric sensors in the east and west directions are equal, i.e., the difference is zero. At this point, it can be determined that the platform is basically aligned with the sun in the east-west direction, i.e., the azimuth angle. Next, the elevation angle needs to be adjusted. The method for adjusting the elevation angle is the same as that for adjusting the azimuth angle. The difference between the photoelectric sensor signals in the north and south directions is compared using the same calculation method to determine whether the platform is basically aligned with the sun. If there is a difference, the dual-angle adjustment platform needs to be continuously adjusted; if there is no difference, the basic alignment is complete.

[0134] For the secondary tracking device, located on top of the primary tracking device, two cross-shaped Fresnel lenses, a rotating shaft, and four photosensitive elements located in the four quadrants are arranged from top to bottom. A secondary calibration is performed to correct the azimuth deviation after adjustment of the primary tracking device. First, the azimuth is calibrated again. The rotating shaft is driven by a stepper motor, and its surface is alternately coated with black and white coatings. The stepper motor controls the shaft to rotate at a constant speed. Sunlight is focused by a long, narrow Fresnel lens, forming a crosshair-shaped spot on the rotating axis. When the spot falls on the black light-absorbing area of ​​the axis, the light is completely absorbed, resulting in the four photosensitive elements below not receiving sunlight and thus outputting a zero electrical signal. When the axis rotates to the white reflective surface, the crosshair spot is reflected onto the four photosensitive elements below. At this point, the projection areas of the spot in the four quadrants are C1, C2, C3, and C4. The dual-angle adjustment platform is then adjusted so that the projection areas of the spot in the four quadrants are C1 = C2 and C3 = C4. This means that the square wave amplitudes output by the four photosensitive elements are equal in pairs, indicating that the angle between the sensor's principal optical axis and the incident sunlight in the azimuth direction is close to zero degrees, signifying that the azimuth direction has been precisely aligned. Then, the elevation angle is calibrated a second time, which is the same as the azimuth angle. At this time, it is only necessary to adjust the dual-angle adjustment platform so that the projection area of ​​the light spot in the four quadrants is equal. That is, when the square wave amplitude output by the four photosensitive elements is equal, it means that the angle between the main optical axis of the sensor and the incident sunlight in the elevation angle direction is close to zero degrees, indicating that the elevation angle direction has been accurately aligned.

[0135] The dual-angle adjustment platform, through the aforementioned mechanism and method, has basically achieved precise aiming at sunlight. At this point, the angle adjustment parameters of the dual-angle adjustment platform are synchronized to the photovoltaic thermal equipment, so that the photovoltaic thermal equipment also adjusts its orientation and height according to the angle of the dual-angle adjustment platform on the sensor side, thus accurately aligning the photovoltaic thermal equipment with sunlight.

[0136] By analyzing real-time weather conditions, the primary tracking device of the four-sided stepped photoelectric sensor is activated to adjust the azimuth and elevation angles of sunlight incidence and alternate between photovoltaic (PV) and solar thermal power generation. Real-time weather data is obtained from cloud images and corrected using weather radar. Furthermore, the actual daily power generation is compared with the predicted power generation based on the day's weather. Based on the comparison results, the secondary tracking device of the four-sided stepped photoelectric sensor is activated to adjust the azimuth and elevation angles of sunlight incidence and correct the alternation rules for PV and solar thermal power generation. This optimizes the efficiency of both PV and solar thermal power generation, extends the lifespan of PV panels and solar thermal generators, and achieves optimal solar power generation efficiency based on daily real-time weather conditions.

[0137] Example 2

[0138] This embodiment provides an efficiency optimization device for a photovoltaic thermal system, such as... Figure 4 As shown, it includes:

[0139] The first weather information acquisition unit 101 is used to acquire real-time weather information, which includes first weather information. The first weather information is obtained by collecting ground-based cloud images in real time, identifying and segmenting cloud shapes in the ground-based cloud images, and calculating cloud cover.

[0140] The primary tracking unit 401 is used to activate the primary tracking device of the four-sided stepped photoelectric sensor based on the first weather information, and to make primary adjustments to the azimuth and elevation angles of the incident sunlight.

[0141] The second weather information acquisition unit 102 is used to acquire the second weather information, which is weather forecast information. The second weather information is acquired once when the first weather information changes.

[0142] The third weather information acquisition unit 103 obtains the third weather information by acquiring the original signal through weather radar, and then performing ground clutter identification and ground clutter removal. Ground clutter identification is performed by using a segmented entropy value feature quantity identification algorithm to identify ground clutter, and ground clutter removal is performed by using an adaptive Gaussian filtering algorithm to remove ground clutter from the identified ground clutter signal to obtain the third weather information.

[0143] The judgment unit 201 is used to determine the duration of real-time weather through the first weather information and the second weather information, compare the duration with a first set value according to the first weather information and the duration, and select to use photovoltaic or solar thermal power generation according to the comparison result.

[0144] The first adjustment unit 301 is used to calculate the difference between the actual power generation and the predicted power generation, adjust the first set value according to the difference to obtain the second set value, and adjust the comparison result according to the second set value.

[0145] The secondary tracking unit 402 is used to activate the secondary tracking device of the four-sided stepped photoelectric sensor based on the difference between the calculated actual power generation and the predicted power generation, and to make secondary adjustments to the azimuth and elevation angles of the incident sunlight.

[0146] The second adjustment unit 302 is used to correct the first weather information using the third weather information when the first weather information is not sunny.

[0147] It should be noted that, for the first weather information acquisition unit, clouds in the sky are an important reference indicator for meteorological observation. The amount of cloud cover can be used to determine whether the weather is sunny or cloudy, overcast or rainy, etc. Cloud image monitoring is generally used in flight, ship navigation, and weather forecasting. This step applies the calculation of cloud cover from ground-based cloud images to determine real-time weather conditions in the field of solar power generation, providing an important basis for the selection of solar power generation methods. Ground-based cloud images are acquired using an all-sky imaging device integrated with a fisheye lens. The acquired ground-based cloud images first need to be identified by cloud formation, then segmented, and finally the amount of cloud cover is calculated using a formula. By comparing the correspondence between cloud cover and weather data in the database, real-time weather information can be obtained. Cloud pattern recognition requires the establishment of a cloud pattern recognition model. This model first performs image preprocessing on the ground-based cloud image, followed by feature extraction. Image preprocessing includes segmenting the ground-based cloud image into blocks, flattening each block, adding coordinate information, and then inputting the image blocks with added coordinate information into a multi-head attention mechanism and a multilayer perceptron designed with a convolutional neural network. The multi-head attention mechanism obtains the basic information of the ground-based cloud image and the distance relationships between the various image blocks, which is then input into the multilayer perceptron for linear transformation, thereby extracting the deeper information of the ground-based cloud image. The system identifies 11 cloud states: altocumulus, stratocumulus, cirrocumulus, altocumulus, cirrostratus, nimbostratus, cumulonimbus, cirrus, cumulus, clear sky, and stratus. The identified cloud states are then segmented using a Transformer cloud image segmentation model with encoding and decoding. The ground-based cloud image with extracted deep information is first deconvolved, then input into a ReLU activation function, followed by bilinear interpolation, and finally outputting the segmented image. Finally, cloud cover is calculated, and the relationship between cloud cover and weather in the database is compared to obtain real-time weather data.

[0148] For the primary solar tracking unit, after obtaining cloud cover from real-time weather information collected from ground-based cloud maps, the system determines whether to activate the four-sided stepped photoelectric sensor to track sunlight based on the weather conditions. In non-sunny weather conditions, such as cloudy, rainy, or snowy days, sunlight is obviously insufficient; therefore, in such weather conditions, it is not necessary to activate the solar tracking device, and the existing photovoltaic thermal equipment can be used directly for power generation. In sunny weather conditions, the angle and direction of the photovoltaic thermal equipment are adjusted to better align it with sunlight. Combined with the photovoltaic thermal selection strategy, this allows for more efficient utilization of sunlight for power generation.

[0149] For the second weather information acquisition unit, weather forecast information issued by the meteorological bureau is acquired via network connection. This acquisition only occurs when changes in real-time weather, as determined by the captured ground-based cloud imagery, occur. The purpose of this is that while the immediate weather can be determined, its duration is difficult to predict using ground-based cloud images; therefore, it is necessary to combine this information with the weather forecast information issued by the meteorological bureau to determine the duration.

[0150] For the third weather information acquisition unit and the second adjustment unit, ground-based cloud imagery has its own limitations in determining real-time weather. In some weather conditions, such as cloudy skies, especially heavy snow, or abundant clouds without rain, the judgments may be inaccurate. Therefore, this step incorporates the use of X-band weather radar to correct the real-time weather judgments from the ground-based cloud imagery. Because of its shorter wavelength, X-band weather radar suffers from severe signal attenuation in the air, significantly impacting the radar echo and thus the judgment results. Therefore, when using X-band weather radar for complementary correction, ground clutter removal is also necessary to ensure the cleanliness and integrity of the radar signal, which is more conducive to correcting the real-time weather from the ground-based cloud imagery. Ground clutter removal involves first identifying ground clutter, then removing it, and finally obtaining the third weather information. First, a segmented entropy feature identification algorithm is used to identify ground clutter. Ground clutter exists at zero frequency, requiring a normal distribution based on zero frequency. The radar spectral data is sorted according to amplitude, and all spectral data is segmented. The segmentation is based on the absolute value of the difference between the maximum and minimum amplitudes, corresponding to the number of segments. The number of spectra in each segment is calculated, and the sum of the probabilities of each spectra occurrence in each segment is obtained. This is then calculated using the information entropy calculation formula:

[0151] F(a)=-∑S(a i log2S(a i )

[0152] Where S(a) i The probability of each frequency band appearing in each segment is multiplied by the number of frequency points in the interval.

[0153] The spectral entropy value for the entire interval is then obtained as follows:

[0154]

[0155] Where N represents the number of segments corresponding to the absolute value of the difference between the maximum and minimum amplitudes. Then, a threshold judgment is performed on the calculated SE(a) value. The judgment threshold is selected based on the simulation data in the laboratory, which shows that a threshold of 0.35 is more suitable. Therefore, 0.35 is selected as the judgment threshold. If the calculated spectral entropy value of the entire interval is greater than 0.35, it is determined that ground clutter exists and the next step of removal operation is required.

[0156] First, the radar data window is filtered. Since the removal effect is related to the selection of the data window, a smaller cone-shaped window width results in a smaller calculated variance, which can lead to misjudgment of ground clutter signals and reduce the removal effect. Conversely, a smaller cone-shaped window width can produce the opposite result. Therefore, it is necessary to select an appropriate data window width by calculating the clutter-to-noise ratio, based on the formula for calculating the clutter-to-noise ratio:

[0157]

[0158] Where N represents noise power, T is the number of samples, and P(t) is the radar echo signal. Different window functions are selected based on the calculated clutter-to-noise ratio (CNR). These window functions include rectangular windows, Blackman windows, and Haining windows. The Haining window is used when the calculated CNR indicates no ground clutter in the echo signal; the Blackman window is used when the calculated CNR indicates mixed ground clutter.

[0159] Next, the acceleration spectral density is calculated and the clutter range is determined. The calculation method for the acceleration spectral density is a common algorithm and will not be elaborated here. Using the calculated acceleration spectral density, the spectral coefficients near the zero Doppler velocity where ground clutter contamination occurs can be determined. Because the acceleration spectral density power amplitude of clutter signals changes rapidly, it will attenuate at the peak, differing from the expected value. In contrast, the clean radar signal, due to the slow change in acceleration spectral density power amplitude, shows almost no deviation. Finally, an adaptive Gaussian filtering algorithm is used to remove the ground clutter signal. The radar signal after removing the ground clutter signal becomes the third weather information. This ground clutter-removed radar signal is used to correct the real-time weather information only when the first weather information, i.e., the real-time weather information, is not clear.

[0160] For the judgment unit, the first weather information, namely the real-time weather information obtained through ground-based cloud maps, is used to determine the real-time weather at a certain moment. The second weather information, namely the weather forecast information, is used to determine how long the real-time weather at that moment will last. Since the power generation efficiency of photovoltaic panels decreases as the temperature rises to a certain level, in the case of a sunny day, if the sunny weather lasts for a certain period of time, causing the temperature of the photovoltaic panels to rise to the maximum power generation efficiency, the continuous solar radiation will continue to heat the photovoltaic panels, leading to a decrease in the power generation efficiency of the photovoltaic panels. Therefore, a first set value is set, which is a time duration. When the sunny weather lasts for the first set value, the photovoltaic panels are stopped from generating electricity, and instead, a solar thermal generator, which has a higher power generation efficiency at high temperatures, is used. This achieves the effect of alternating the use of photovoltaic panels and solar thermal generators to extend the lifespan of both devices, and also optimizes the power generation efficiency.

[0161] For the first adjustment unit, the predicted power generation is obtained by predicting the power generation based on the real-time weather data recorded throughout the day. This predicted power generation is then compared with the actual power generation of the day, achieved through steps S1-S3 of alternating photovoltaic (PV) and solar thermal (SP) power generation. This comparison reveals whether the actual power generation is above or below the predicted power generation. If the actual power generation is above the predicted power generation, it indicates that the alternation of PV and SP power is effective that day, meaning the selection result obtained by comparing the duration using the first setpoint meets the requirements. If the actual power generation is below the predicted power generation, it indicates that the alternation of PV and SP power is ineffective that day, meaning the selection result obtained by comparing the duration using the first setpoint does not meet the requirements. In this case, the first setpoint needs to be adjusted to obtain an adjusted second setpoint. This second setpoint is then used as the benchmark setpoint for comparing the duration of the next day's PV and SP power generation. This allows the system to adaptively adjust the alternation rules of PV and SP power generation devices, thereby optimizing power generation efficiency. In addition, the actual power generation differs significantly from the predicted power generation, which is below the predicted power generation. This may be due to poor solar tracking. Therefore, in order to further optimize power generation efficiency and improve the utilization of sunlight, a secondary tracking device was added to adjust the incident azimuth and elevation angles of sunlight in two stages. This adjusts the angle and direction of the photovoltaic thermal device to better align it with the sunlight and further optimize power generation efficiency.

[0162] Preferably, the efficiency optimization device for the photovoltaic-thermal system further includes:

[0163] The real-time weather judgment unit includes a ground-based cloud image recognition unit and a cloud segmentation unit. The real-time weather judgment unit is used to input the calculation data of the ground-based cloud image recognition unit and the cloud segmentation unit into the cloud amount calculation formula to obtain cloud amount information, and to judge the real-time weather through the cloud amount information.

[0164] The first prediction power unit is used to determine the weather type after obtaining the historical weather set, then use the Gaussian mixture model to classify the weather, and finally input the weather classification into the multi-head attention LSTM model to obtain the first prediction power.

[0165] The second prediction power unit is used to calculate the second prediction power based on the predicted effective heat gain and predicted incident irradiance of the photothermal device.

[0166] It should be noted that, for real-time weather judgment units, all-sky imaging devices with fisheye lenses, when acquiring ground-based cloud image information, contain redundant images in the four corners of the acquired ground-based cloud images, i.e., invalid images that are neither clouds nor sky. When calculating cloud cover, these redundant images need to be removed first to obtain the total area of ​​the valid cloud-sky region. Finally, the cloud area is calculated to obtain the cloud cover value of the ground-based cloud image. First, the number of pixels in the ground-based cloud image is calculated, including the number of cloud image pixels and the number of non-sky pixels. The third pixel value is obtained by subtracting the number of cloud image pixels from the number of non-sky pixels. Then, the segmentation model used is the Transformer cloud image segmentation model using an encoding and decoding method. After deconvolution of the ground-based cloud image with extracted deep information, it is input into ReLU. The prime value and the fourth pixel value are substituted into the cloud cover calculation formula to calculate the cloud cover:

[0167] Where Y is cloud cover, Ji is the correction coefficient, K1 is the first pixel value, K2 is the second pixel value, K3 is the third pixel value, and K4 is the fourth pixel value.

[0168] For the first predicted power unit, after sunset, the predicted power of photovoltaic power generation throughout the day is statistically analyzed to verify whether the rules for selecting photovoltaic panels and solar thermal pipes are optimal. The first weather information, obtained through ground-based cloud maps and corrected using third-party weather information, is sampled every M time intervals (e.g., if M is 1, the first weather information is sampled every hour). After sampling, a set of N sets of first weather information is obtained, revealing all the weather conditions for the day. Then, historical GHI sequences are used to determine the weather type from the historical weather set. The input step size is determined based on the historical GHI sequence. The training set data is classified according to the weather clustering results and the input step size. A backpropagation (BP) classifier is trained using the training set data, and the trained BP classifier is used to classify the test set data to determine the weather type. Next, a Gaussian mixture model is used to classify the weather from the historical weather set after type determination. The test set data is classified according to the weather clustering results and the input step size to obtain the weather classification. Finally, the weather classification is input into a multi-head attention LSTM model to obtain the first predicted power.

[0169] For the second predicted power unit, firstly, the photothermal efficiency coefficient E' is calculated based on the parameters of the photothermal device, including the pipe spacing D, the pipe outer diameter D0, and the pipe inner diameter D. i Fluid heat transfer coefficient l wf Conductivity B of tubes and fins C The specific formulas for calculating fin efficiency E and overall conductivity U are:

[0170]

[0171] Then, the average mass flow rate N of the fluid in the photothermal device is obtained. v Fluid specific heat Q P Radiator area A c And based on the photothermal efficiency coefficient E', the heat dissipation coefficient E is calculated. R The formula for calculating the heat dissipation coefficient is:

[0172]

[0173] Next, based on the heat dissipation coefficient E R Overall heat transfer coefficient U, average temperature of the absorber plate T PM and average ambient temperature T a Calculate the predicted effective heat gain Q U The formula for calculating effective heat gain is:

[0174]

[0175] Then, the first weather information is obtained every M time interval, and N first weather information are obtained as a historical weather set, where M and N are positive integers greater than 0. The predicted incident irradiance for that day is calculated based on the historical weather information set. Similar to the calculation of the first predicted power, after sunset, the predicted power of photovoltaic power generation during the day is statistically analyzed to verify whether the rules for selecting photovoltaic panels and solar thermal pipes are optimal. The first weather information obtained through ground-based cloud maps and corrected by the third weather information is sampled every M time interval. For example, if M is 1, the first weather information is sampled every 1 hour. After sampling, a set of N first weather information will be obtained, and all the weather conditions for that day will be known. The predicted incident irradiance for that day is calculated based on the first weather information set.

[0176] Finally, based on the predicted effective heat gain Q U And the predicted incident irradiance R, and the second predicted power P is calculated. r The second predicted power calculation formula is:

[0177] By using real-time weather conditions, the system alternates between photovoltaic (PV) and solar thermal (CSP) power generation. Real-time weather data is obtained from cloud images and corrected using weather radar. In addition, the actual power generation for a day is compared with the power generation predicted based on the weather for that day. The rules for alternating between PV and CSP power generation are adjusted based on the comparison results, thereby optimizing the efficiency of PV and CSP, extending the lifespan of PV panels and CSP units, and achieving the highest efficiency of solar power generation based on daily real-time weather conditions.

[0178] Preferably, the efficiency optimization device for the photovoltaic-thermal system further includes a four-sided stepped photoelectric sensor, which includes a primary tracking device and a secondary tracking device;

[0179] The primary tracking device includes an eastward sensor, a westward sensor, a southward sensor, a northward sensor, and a dual-angle adjustment platform. The eastward, westward, southward, and northward sensors are respectively distributed along the four stepped directions of the photoelectric sensor, with the eastward and westward sensors facing each other, and the southward and northward sensors facing each other. The dual-angle adjustment platform includes an azimuth adjustment platform and an elevation adjustment platform. The azimuth adjustment platform is adjusted according to the magnitude of the electrical signals output by the eastward and westward sensors after receiving sunlight, and the elevation adjustment platform is adjusted according to the magnitude of the electrical signals output by the southward and northward sensors after receiving sunlight.

[0180] The secondary tracking device includes four circular Fresnel lenses, four photosensitive elements, and a rotating shaft. After the primary tracking device adjusts the azimuth and elevation angles, the azimuth and elevation angles are adjusted in a secondary manner through the dual-angle adjustment platform based on the difference. When the square wave amplitudes output by the four photosensitive elements are equal in pairs, the azimuth angle adjustment is completed. The elevation angle is then adjusted, and when the square wave amplitudes output by the four photosensitive elements are all equal, the elevation angle adjustment is completed.

[0181] It should be noted that for the primary tracking device, there are four sensors in each direction of the four-sided steps; that is, there are four sensors each for the east, west, south, and north directions, for a total of sixteen. These sixteen photoelectric sensors, distributed across the four-sided steps, detect the intensity of sunlight incident from each direction in real time. By comparing the peak intensities of the four sets of signals (east, west, south, and north), the spatial azimuth of the sun is calculated. Based on the azimuth calculation results, the stepper motor of the dual-angle adjustment platform is controlled to perform large-angle rotations in both the azimuth (east-west axis) and altitude (north-south axis) to complete the initial capture and tracking of the sun. It is important to note that the dual-angle adjustment platform is part of the primary tracking device and is only used to adjust the angle and direction of the solar tracking device. First, the azimuth angle needs to be adjusted. This requires eight photoelectric sensors along the east-west axis. The signals output by these eight sensors are differentially detected in real time. The photocurrent gradient function is used to calculate and compare the electrical signals of the eight photoelectric sensors to determine whether the dual-angle adjustment platform needs to be adjusted in the azimuth direction. This continues until the electrical signals output by the photoelectric sensors in the east and west directions are equal, i.e., the difference is zero. At this point, it can be determined that the platform is basically aligned with the sun in the east-west direction, i.e., the azimuth angle. Next, the elevation angle needs to be adjusted. The method for adjusting the elevation angle is the same as that for adjusting the azimuth angle. The difference between the photoelectric sensor signals in the north and south directions is compared using the same calculation method to determine whether the platform is basically aligned with the sun. If there is a difference, the dual-angle adjustment platform needs to be continuously adjusted; if there is no difference, the basic alignment is complete.

[0182] For the secondary tracking device, located on top of the primary tracking device, two cross-shaped Fresnel lenses, a rotating shaft, and four photosensitive elements located in the four quadrants are arranged from top to bottom. A secondary calibration is performed to correct the azimuth deviation after adjustment of the primary tracking device. First, the azimuth is calibrated again. The rotating shaft is driven by a stepper motor, and its surface is alternately coated with black and white coatings. The stepper motor controls the shaft to rotate at a constant speed. Sunlight is focused by a long, narrow Fresnel lens, forming a crosshair-shaped spot on the rotating axis. When the spot falls on the black light-absorbing area of ​​the axis, the light is completely absorbed, resulting in the four photosensitive elements below not receiving sunlight and thus outputting a zero electrical signal. When the axis rotates to the white reflective surface, the crosshair spot is reflected onto the four photosensitive elements below. At this point, the projection areas of the spot in the four quadrants are C1, C2, C3, and C4. The dual-angle adjustment platform is then adjusted so that the projection areas of the spot in the four quadrants are C1 = C2 and C3 = C4. This means that the square wave amplitudes output by the four photosensitive elements are equal in pairs, indicating that the angle between the sensor's principal optical axis and the incident sunlight in the azimuth direction is close to zero degrees, signifying that the azimuth direction has been precisely aligned. Then, the elevation angle is calibrated a second time, which is the same as the azimuth angle. At this time, it is only necessary to adjust the dual-angle adjustment platform so that the projection area of ​​the light spot in the four quadrants is equal. That is, when the square wave amplitude output by the four photosensitive elements is equal, it means that the angle between the main optical axis of the sensor and the incident sunlight in the elevation angle direction is close to zero degrees, indicating that the elevation angle direction has been accurately aligned.

[0183] The dual-angle adjustment platform, through the aforementioned mechanism and method, has basically achieved precise aiming at sunlight. At this point, the angle adjustment parameters of the dual-angle adjustment platform are synchronized to the photovoltaic thermal equipment, so that the photovoltaic thermal equipment also adjusts its orientation and height according to the angle of the dual-angle adjustment platform on the sensor side, thus accurately aligning the photovoltaic thermal equipment with sunlight.

[0184] By analyzing real-time weather conditions, the primary tracking device of the four-sided stepped photoelectric sensor is activated to adjust the azimuth and elevation angles of sunlight incidence and alternate between photovoltaic (PV) and solar thermal power generation. Real-time weather data is obtained from cloud images and corrected using weather radar. Furthermore, the actual daily power generation is compared with the predicted power generation based on the day's weather. Based on the comparison results, the secondary tracking device of the four-sided stepped photoelectric sensor is activated to adjust the azimuth and elevation angles of sunlight incidence and correct the alternation rules for PV and solar thermal power generation. This optimizes the efficiency of both PV and solar thermal power generation, extends the lifespan of PV panels and solar thermal generators, and achieves optimal solar power generation efficiency based on daily real-time weather conditions.

[0185] Example 3

[0186] This embodiment provides an efficiency optimization system for a photovoltaic-thermal system, including a photovoltaic power generation device and a solar thermal power generation device, and further including:

[0187] Imaging system, weather radar, humidity sensor, temperature sensor, solar tracking device, processor and memory;

[0188] The imaging system is an all-sky imaging device with a fisheye lens, used to capture ground-based cloud image information of the sky panorama;

[0189] The weather radar is an X-band weather radar used to obtain real-time weather information;

[0190] Humidity sensors are used to acquire information about the humidity of the environment;

[0191] Temperature sensors are used to acquire the device temperature and ambient temperature of solar thermal power generation equipment;

[0192] The solar tracking device is used to track sunlight in real time and adjust the position of the photovoltaic and solar thermal power generation equipment relative to the sunlight based on the tracking results;

[0193] The memory is coupled to the processor. The memory stores executable program modules, and the processor runs the program modules stored in the memory to implement the efficiency optimization method of the photovoltaic-thermal system as described in Example 1.

[0194] It should be noted that the dual-angle adjustment platform has basically achieved precise aiming at sunlight through the above-mentioned mechanism and method. At this time, the angle adjustment parameters of the dual-angle adjustment platform are synchronized to the photovoltaic thermal equipment, so that the photovoltaic thermal equipment also adjusts its orientation and height according to the angle of the dual-angle adjustment platform on the sensor side, thus accurately aligning the photovoltaic thermal equipment with sunlight.

[0195] Furthermore, the memory is a readable storage medium, which can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the ASIC can reside in a user device. Of course, the processor and the readable storage medium can also exist as discrete components in a communication device. The readable storage medium can be a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0196] The present invention also provides a program product including executable instructions stored in a readable storage medium. At least one processor of the device can read the executable instructions from the readable storage medium, and the at least one processor executes the executable instructions to cause the device to implement the methods provided in the various embodiments described above.

[0197] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing the efficiency of a photovoltaic-thermal system, characterized in that, include: S1: Obtain real-time weather information, which includes first weather information. The first weather information is obtained by collecting ground-based cloud images in real time, identifying and segmenting cloud shapes in the ground-based cloud images, and calculating cloud cover. S2: Based on the first weather information, activate the first-level tracking device of the four-sided stepped photoelectric sensor to make a first-level adjustment to the azimuth and elevation angles of the sunlight incident. S3: Obtain the second weather information, which is a weather forecast. The second weather information is obtained once when the first weather information changes. S4: Determine the duration of the real-time weather based on the first weather information and the second weather information; compare the duration with a first set value based on the first weather information and the duration; and select whether to use photovoltaic or solar thermal power generation based on the comparison result. S5: Calculate the difference between the actual power generation and the predicted power generation, activate the secondary tracking device of the four-sided stepped photoelectric sensor based on the difference, perform secondary adjustments to the azimuth and elevation angles of the sunlight incident, adjust the first set value to obtain the second set value based on the difference, and adjust the comparison result based on the second set value.

2. The efficiency optimization method for a photovoltaic-thermal system as described in claim 1, characterized in that, S5 includes: The actual power generation includes a first power generation and a second power generation, the first power generation being the actual photovoltaic power generation and the second power generation being the actual solar thermal power generation; the predicted power generation includes a first predicted power and a second predicted power, the first predicted power being the photovoltaic predicted power and the second predicted power being the solar thermal predicted power. The difference includes a first difference and a second difference. The first difference is the difference between the first power generation and the first predicted power, and the second difference is the difference between the second power generation and the second predicted power. When the first difference is greater than the first threshold and / or the second difference is greater than the second threshold, return to S3, adjust the first setting value to obtain the second setting value, adjust the comparison result according to the second setting value, and select to use photovoltaic or solar thermal power generation according to the comparison result.

3. The efficiency optimization method for a photovoltaic-thermal system as described in claim 2, characterized in that, The real-time weather information also includes: The third weather information is obtained by acquiring the original signal through weather radar, and then performing ground clutter identification and removal. The ground clutter identification is performed by using a segmented entropy feature identification algorithm to identify ground clutter, and the ground clutter removal is performed by using an adaptive Gaussian filtering algorithm to remove ground clutter from the identified ground clutter signal to obtain the third weather information. When the first weather information is not sunny, the third weather information is used to correct the first weather information.

4. The efficiency optimization method for a photovoltaic-thermal system as described in claim 3, characterized in that, include: When the first weather information indicates a sunny day, photovoltaic or solar thermal power generation is selected based on the comparison results. When the first weather information is not sunny, humidity information is obtained. When the humidity information is greater than a third threshold, photovoltaic power generation is used. When the humidity information is less than the third threshold, solar thermal power generation is used.

5. The efficiency optimization method for a photovoltaic-thermal system as described in claim 1, characterized in that, The cloud-like identification of the ground cloud map includes: performing pixel calculation on the ground cloud map to obtain a first pixel value and a second pixel value, wherein the first pixel value is the number of cloud map pixels, the second pixel value is the number of non-sky pixels, and the first pixel value is subtracted from the second pixel value to obtain a third pixel value; The cloud segmentation includes: extracting features from the ground cloud map, upsampling the extracted data to obtain the segmented ground cloud map, calculating the third pixel value of the segmented ground cloud map, and the fourth pixel value being the number of cloud pixels; The first pixel value, the second pixel value, the third pixel value, and the fourth pixel value are substituted into the cloud cover calculation formula to obtain cloud cover information, and the real-time weather is determined by the cloud cover information.

6. The efficiency optimization method for a photovoltaic-thermal system as described in claim 3, characterized in that, The first predicted power includes: The first weather information is acquired every M time intervals, and N first weather information items are obtained as a historical weather set, where M and N are positive integers greater than 0; The historical weather set is used to determine the weather type using historical GHI sequences, and the historical weather set after type determination is used to classify the weather using a Gaussian mixture model. The first predicted power is obtained by inputting the weather classification data into a multi-head attention LSTM model.

7. The efficiency optimization method for a photovoltaic-thermal system as described in claim 3, characterized in that, The second predicted power includes: The photothermal efficiency coefficient is calculated based on the parameters of the photothermal device, including pipe spacing, pipe outer diameter, pipe inner diameter, fluid heat transfer coefficient, pipe and fin conductivity, and fin efficiency. The average mass flow rate, specific heat of the fluid, heat sink area, and total conductivity of the photothermal device are obtained, and the heat dissipation coefficient is calculated based on the photothermal efficiency coefficient. The predicted effective heat gain is calculated based on the heat dissipation coefficient, the overall heat transfer coefficient, and the average temperature of the heat absorber plate. The first weather information is acquired every M time intervals, and N first weather information are obtained as a historical weather set, where M and N are positive integers greater than 0. The predicted incident irradiance for that day is calculated based on the historical weather information set. The second predicted power is calculated based on the predicted effective heat gain and the predicted incident irradiance.

8. The method for optimizing the efficiency of a photovoltaic-thermal system as described in any one of claims 1-7, characterized in that, include: The four-sided stepped photoelectric sensor includes a primary tracking device and a secondary tracking device; The primary tracking device includes an eastward sensor, a westward sensor, a southward sensor, a northward sensor, and a dual-angle adjustment platform. The eastward, westward, southward, and northward sensors are respectively distributed along the four stepped directions of the photoelectric sensor, with the eastward and westward sensors facing each other, and the southward and northward sensors facing each other. The dual-angle adjustment platform includes an azimuth adjustment platform and an elevation adjustment platform. The azimuth adjustment platform is adjusted according to the magnitude of the electrical signals output by the eastward and westward sensors after receiving sunlight, and the elevation adjustment platform is adjusted according to the magnitude of the electrical signals output by the southward and northward sensors after receiving sunlight. The secondary tracking device includes two square Fresnel lenses, four photosensitive elements, and a rotating shaft. After the primary tracking device adjusts the azimuth and elevation angles, the azimuth and elevation angles are adjusted in a secondary manner through the dual-angle adjustment platform based on the difference. When the square wave amplitudes output by the four photosensitive elements are equal in pairs, the azimuth angle adjustment is completed. The elevation angle is then adjusted, and when the square wave amplitudes output by the four photosensitive elements are all equal, the elevation angle adjustment is completed.

9. An efficiency optimization device for a photovoltaic-thermal system, characterized in that, include: The first weather information acquisition unit is used to acquire real-time weather information, which includes first weather information. The first weather information is obtained by collecting ground-based cloud images in real time, identifying and segmenting cloud shapes in the ground-based cloud images, and calculating cloud cover. The primary tracking unit is used to activate the primary tracking device of the four-sided stepped photoelectric sensor based on the first weather information, and to make primary adjustments to the azimuth and elevation angles of the incident sunlight. The second weather information acquisition unit is used to acquire the second weather information, which is weather forecast information. The second weather information is acquired once when the first weather information changes. The third weather information acquisition unit obtains the third weather information by acquiring the original signal through meteorological radar, and then performing ground clutter identification and ground clutter removal. The ground clutter identification is performed by using a segmented entropy value feature identification algorithm to identify ground clutter, and the ground clutter removal is performed by using an adaptive Gaussian filtering algorithm to remove ground clutter from the identified ground clutter signal to obtain the third weather information. The judgment unit is used to determine the duration of real-time weather based on the first weather information and the second weather information, compare the duration with a first set value based on the first weather information and the duration, and select to use photovoltaic or solar thermal power generation based on the comparison result. The first adjustment unit is used to calculate the difference between the actual power generation and the predicted power generation, adjust the first set value according to the difference to obtain a second set value, and adjust the comparison result according to the second set value. The secondary tracking unit is used to activate the secondary tracking device of the four-sided stepped photoelectric sensor based on the difference between the calculated actual power generation and the predicted power generation, and to make secondary adjustments to the azimuth and elevation angles of the incident sunlight. The second adjustment unit is used to correct the first weather information using the third weather information when the first weather information is not sunny.

10. An efficiency optimization system for a photovoltaic-thermal system, comprising photovoltaic power generation equipment and solar thermal power generation equipment, characterized in that, Also includes: Imaging system, weather radar, humidity sensor, temperature sensor, solar tracking device, processor and memory; The imaging system is an all-sky imaging device with a fisheye lens, used to capture ground-based cloud image information of the sky panorama; The weather radar is an X-band weather radar used to acquire real-time weather information; The humidity sensor is used to acquire ambient humidity information; The temperature sensor is used to acquire the device temperature and ambient temperature of the solar thermal power generation equipment. The solar tracking device is used to track sunlight in real time and adjust the position of the photovoltaic and solar thermal power generation equipment relative to the sunlight based on the tracking results; The memory is coupled to the processor, and the memory stores executable program modules. The processor is used to run the program modules stored in the memory to implement the method as described in any one of claims 1-8.

Images