Photovoltaic panel tracking sun control system and method based on image recognition

Abstract

The present invention relates to a control system and method for photovoltaic panels to track the sun based on image recognition, and relates to the technical field of photovoltaic power generation, and is applied to a photovoltaic system, wherein the photovoltaic system includes multiple photovoltaic panels, and the system includes: a light sensing module, including multiple light sensing units and a light uniformity judgment unit, wherein the multiple light sensing units are respectively arranged on the multiple photovoltaic panels, and the light uniformity judgment unit is used to determine the light uniformity of the light sensing unit based on the output of the light sensing unit; an image acquisition module, used to determine whether to perform image acquisition based on the light uniformity of each light sensing unit, and when it is determined to perform image acquisition, the image of the light sensing unit is acquired; a sun following module, used to determine the optimal sun following angle of the photovoltaic panel based on the image of the light sensing unit, and adjust the angle of the photovoltaic panel based on the optimal sun following angle, which has the advantage of improving the efficiency of photovoltaic power generation.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation, and in particular to a control system and method for photovoltaic panels to track the sun based on image recognition. Background Art

[0002] As environmental pollution worsens, people are beginning to increase the construction of new energy power generation systems. Solar photovoltaic power generation systems, a typical representative of the cleanest energy, have high investment costs, so it is very important to ensure their maximum power generation. Photovoltaic panel components are a type of power generation device that generates direct current when exposed to sunlight. They are composed of solid photovoltaic panels made almost entirely of semiconductor materials. Photovoltaic panel components can be made into different shapes, and the components can be connected to generate more electricity. Photovoltaic panel components can be used on rooftops and building surfaces, and can even be used as part of windows, skylights or shading devices. These photovoltaic facilities are usually called building-attached photovoltaic systems.

[0003] Since the current solar panels used for power generation are directly installed on the roof of the building through a positioning frame, the adjustment flexibility is extremely poor. As the angle of the sun changes, the angle between the sunlight and the solar panel also changes, resulting in large fluctuations in the power generation efficiency of the solar panel and low power generation efficiency.

[0004] Therefore, it is necessary to provide a control system and method for photovoltaic panels to track the sun based on image recognition, so as to improve the efficiency of photovoltaic power generation. Summary of the invention

[0005] The present invention provides a control system for photovoltaic panels tracking the sun based on image recognition, which is applied to a photovoltaic system, wherein the photovoltaic system includes multiple photovoltaic panels, including: a light sensing module, including multiple light sensing units and a light uniformity judgment unit, wherein the multiple light sensing units are respectively arranged on the multiple photovoltaic panels, and the light uniformity judgment unit is used to determine the light uniformity of the light sensing unit based on the output of the light sensing unit; an image acquisition module, used to determine whether to perform image acquisition based on the light uniformity of each of the light sensing units, and when it is determined to perform image acquisition, the image of the light sensing unit is acquired; a sun following module, used to determine the optimal sun following angle of the photovoltaic panel based on the image of the light sensing unit, and adjust the angle of the photovoltaic panel based on the optimal sun following angle.

[0006] Furthermore, the light sensing unit includes a light sensing panel arranged on one side of the photovoltaic panel, the light sensing panel and the photovoltaic panel are located in the same plane, a plurality of light sensing devices are arranged on the light sensing panel, a projection is arranged on the light sensing panel, and the projection is perpendicular to the light sensing panel.

[0007] Furthermore, the illumination uniformity judgment unit determines the illumination uniformity of the illumination sensing unit based on the output of the illumination sensing unit, including: for each of the illumination sensing devices, performing empirical mode decomposition on the output voltage signal of the illumination sensing device at multiple consecutive time points to generate an eigenmode function and a residual, and generating a corrected output voltage signal at multiple consecutive time points based on the eigenmode function and the residual through a denoising model; and determining the illumination uniformity of the illumination sensing unit based on the corrected output voltage signal at multiple consecutive time points corresponding to each of the illumination sensing devices.

[0008] Furthermore, the light uniformity judgment unit determines the light uniformity of the light sensing unit based on the corrected output voltage signals at multiple consecutive time points corresponding to each of the light sensing devices, including: for each of the light sensing devices, sampling the corrected output voltage signals at multiple consecutive time points corresponding to the light sensing device to obtain the output voltage signals of the light sensing device at multiple sampling time points; for each of the sampling time points, calculating the single-point light uniformity corresponding to the sampling time point based on the output voltage signal of each of the light sensing devices at the sampling time point; and determining the light uniformity of the light sensing unit based on the single-point light uniformity corresponding to each of the sampling time points.

[0009] Furthermore, the image acquisition module determines whether to perform image acquisition based on the illumination uniformity of each of the illumination sensing units, including: for any two of the illumination sensing units, calculating the illumination uniformity difference between the two illumination sensing units; based on the illumination uniformity difference between any two of the illumination sensing units, calculating the illumination uniformity difference parameter; when the illumination uniformity difference parameter is less than a preset illumination uniformity difference parameter, and there is at least one of the illumination sensing units whose illumination uniformity is greater than the preset illumination uniformity, determining to perform image acquisition.

[0010] Furthermore, a plurality of markers are provided on the light sensing panel; the image acquisition module acquires the image of the light sensing unit, including: acquiring the image of the light sensing unit under a plurality of image acquisition postures; the sun following module determines the optimal sun following angle of the photovoltaic panel based on the image of the light sensing unit, including: for each of the image acquisition postures, based on the image of the light sensing unit acquired under the image acquisition posture, determining the relative position relationship between the projection endpoint and the plurality of markers; based on the relative position relationship between the projection endpoint corresponding to each of the image acquisition postures and the plurality of markers, determining the second relative position relationship between the projection endpoint and the plurality of markers; based on the second relative position relationship between the projection endpoint and the plurality of markers, determining the optimal sun following angle of the photovoltaic panel.

[0011] Furthermore, the solar following module determines the optimal solar following angle of the photovoltaic panel based on the second relative position relationship between the projection endpoint and the multiple markers, including: determining multiple candidate solar following angles based on the second relative position relationship between the projection endpoint and the multiple markers; obtaining battery temperature information of the photovoltaic panel at multiple consecutive time points; obtaining ambient temperature information at multiple consecutive time points; based on the battery temperature information of the photovoltaic panel at multiple consecutive time points and the ambient temperature information at multiple consecutive time points, screening the multiple candidate solar following angles to determine the optimal solar following angle.

[0012] Furthermore, it also includes a temperature control module for controlling the temperature of the photovoltaic panel based on the battery temperature information of the photovoltaic panel at multiple consecutive time points, the ambient temperature information at multiple consecutive time points and the optimal sun following angle.

[0013] Furthermore, the solar following module adjusts the angle of the photovoltaic panel based on the optimal solar following angle, including: generating a preset solar following strategy based on historical solar following information, wherein the preset solar following strategy includes preset photovoltaic panel angles in multiple time periods; when the angle difference between the optimal solar following angle and the corresponding preset photovoltaic panel angle is greater than the preset difference, adjusting the angle of the photovoltaic panel and the preset solar following strategy according to the optimal solar following angle.

[0014] The present invention provides a control method for photovoltaic panels tracking the sun based on image recognition, which is executed by a control system for photovoltaic panels tracking the sun based on image recognition, and includes: setting multiple light sensing units on multiple photovoltaic panels; determining the light uniformity of the light sensing units based on outputs of the light sensing units; judging whether to perform image acquisition based on the light uniformity of each of the light sensing units; when it is determined to perform image acquisition, acquiring images of the light sensing units; determining the optimal sun-following angle of the photovoltaic panel based on the images of the light sensing units; and adjusting the angle of the photovoltaic panel based on the optimal sun-following angle.

[0015] Compared with the prior art, the control system and method for photovoltaic panel tracking the sun based on image recognition provided in this specification have at least the following beneficial effects:

[0016] 1. Through multiple light sensing units, the position and angle changes of the sun can be tracked more accurately. This ensures that the photovoltaic panels can always receive the maximum amount of direct sunlight, thereby improving the efficiency of photoelectric conversion. Based on the data of the light sensing unit, the optimal sun-following angle can be calculated and the angle of the photovoltaic panel can be adjusted in real time. This dynamic adjustment can ensure that the photovoltaic panels can maintain the best power generation state in different time periods and weather conditions;

[0017] 2. For multiple light sensing devices, empirical mode decomposition and denoising can be performed separately, and the corrected output voltage signal can be used to evaluate the light uniformity, which can eliminate the influence of noise on the evaluation results and improve the accuracy of the evaluation. In this way, the light uniformity of the entire photovoltaic panel can be more comprehensively understood, providing a basis for the subsequent adjustment and optimization of the angle of the photovoltaic panel;

[0018] 3. Combined with the battery temperature information and ambient temperature information of the photovoltaic panel at different time points, the candidate sun following angles can be screened. Considering the impact of temperature on the power generation efficiency of photovoltaic panels, such as the increase in temperature will lead to a decrease in the voltage of the photovoltaic panel, an increase in current but a decrease in overall power, choosing a following angle that can keep the battery at a lower temperature or the optimal operating temperature can significantly improve the power generation efficiency of the photovoltaic panel;

[0019] 4. The preset sun-following strategy is based on historical data and can reflect the approximate position of the sun at different times of the day. However, due to changes in weather, seasons and geographical locations, the actual position of the sun may deviate. By calculating the optimal sun-following angle in real time and adjusting the angle of the photovoltaic panel as needed, it can be ensured that the photovoltaic panel can always receive the maximum amount of direct sunlight, thereby improving power generation efficiency. When the difference between the optimal sun-following angle and the preset angle exceeds the preset difference, adjustments can be made quickly. This rapid response capability helps reduce power generation losses caused by angle deviations and ensures that the photovoltaic panels can maintain optimal working conditions at all times. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0021] Figure 1 It is a module diagram of a control system for photovoltaic panels tracking the sun based on image recognition shown in an embodiment of the present application;

[0022] Figure 2 is a schematic diagram of a projection shown in an embodiment of the present application;

[0023] Figure 3 It is a flow chart of a control method for photovoltaic panels tracking the sun based on image recognition shown in one embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly introduces the drawings required for describing the embodiments.

[0025] Figure 1 is a module diagram of a control system for photovoltaic panels tracking the sun based on image recognition shown in an embodiment of the present application, such as Figure 1 As shown, a control system for photovoltaic panels tracking the sun based on image recognition is applied to a photovoltaic system, wherein the photovoltaic system includes multiple photovoltaic panels and may include a light sensing module, an image acquisition module, a sun following module and a temperature control module.

[0026] The light sensing module may include multiple light sensing units and a light uniformity judgment unit. The multiple light sensing units are respectively arranged on multiple photovoltaic panels. The light uniformity judgment unit is used to determine the light uniformity of the light sensing units based on the output of the light sensing units. For example, the photovoltaic system includes five photovoltaic panels, and light sensing units may be arranged on three photovoltaic panels. The light sensing units may be arranged on the side of the photovoltaic panels to avoid blocking any photovoltaic panel, and when the angle of the photovoltaic panel changes, the angle of the light sensing unit changes in the same angle as the photovoltaic panel under the drive of the photovoltaic panel.

[0027] In some embodiments, the light sensing unit includes a light sensing panel disposed on one side of the photovoltaic panel, the light sensing panel and the photovoltaic panel are located in the same plane, a plurality of light sensing devices are disposed on the light sensing panel, a projection is disposed on the light sensing panel, and the projection is perpendicular to the light sensing panel. A plurality of markers are also disposed on the light sensing panel, the positions of the plurality of markers are fixed, and the relative positional relationship of the plurality of markers is predetermined. As an example only, the projection may be a rod-shaped object.

[0028] Specifically, the output voltage of the light sensing device is affected by the light intensity. For example, the stronger the light intensity, the higher the output voltage of the light sensing device.

[0029] In some embodiments, the illumination uniformity determination unit determines the illumination uniformity of the illumination sensing unit based on the output of the illumination sensing unit, including:

[0030] For each light sensing device, performing empirical mode decomposition on the output voltage signal of the light sensing device at multiple consecutive time points to generate an eigenmode function and a residual, and generating a corrected output voltage signal at multiple consecutive time points based on the eigenmode function and the residual by using a denoising model, wherein the denoising model may be a convolutional neural network (CNN) model;

[0031] The illumination uniformity of the illumination sensing unit is determined based on the corrected output voltage signals corresponding to each illumination sensing device at a plurality of consecutive time points.

[0032] Specifically, the output voltage signal of the light sensing device is collected at a plurality of consecutive time points, and these time points can be set according to specific needs, such as collecting once per second or once every few minutes.

[0033] The output voltage signal of the light sensing device at multiple consecutive time points can be subjected to empirical mode decomposition based on the following process to generate eigenmode functions and residuals:

[0034] S11, finding extreme value points: finding all the maximum value points and minimum value points in the output voltage signal of the light sensing device at multiple consecutive time points;

[0035] S12, constructing envelopes: using a cubic spline interpolation function to fit the maximum and minimum points respectively, to obtain an upper envelope and a lower envelope;

[0036] S13, calculating the mean: calculating the mean of the upper envelope and the lower envelope to obtain a local mean line;

[0037] S14, extracting the intrinsic mode function: subtracting the local mean line from the original signal to obtain the first intrinsic mode function, and checking whether the intrinsic mode function meets the two conditions of the intrinsic mode function: the difference between the number of extreme points and the number of zero-crossing points does not exceed 1, and the mean of the upper and lower envelopes is zero. If it meets the conditions, it is the first intrinsic mode function; if it does not meet the conditions, it is regarded as a new original signal, and S11-S13 are repeated until the conditions are met;

[0038] S15, repeat decomposition: subtract the extracted intrinsic mode function from the original signal to obtain a residual signal, and repeatedly perform S11-S14 on the residual signal until the residual signal becomes a monotonic function or is smaller than a preset threshold.

[0039] In some embodiments, the illumination uniformity determination unit determines the illumination uniformity of the illumination sensing unit based on the corrected output voltage signals corresponding to each illumination sensing device at a plurality of consecutive time points, including:

[0040] For each light sensing device, sampling the corrected output voltage signal corresponding to the light sensing device at a plurality of consecutive time points to obtain the output voltage signal of the light sensing device at a plurality of sampling time points;

[0041] For each sampling time point, based on the output voltage signal of each light sensing device at the sampling time point, calculate the single-point light uniformity corresponding to the sampling time point;

[0042] Based on the single-point illumination uniformity corresponding to each sampling time point, the illumination uniformity of the illumination sensing unit is determined.

[0043] The single-point illumination uniformity corresponding to the sampling time point can be calculated according to the following formula:

[0044]

[0045] Among them, U (i,t) is the single-point illumination uniformity corresponding to the i-th illumination sensing unit at the t-th sampling time point, P1 is the preset parameter, and P1>0, V ((n,t),i) is the output voltage signal of the nth light sensing device of the ith light sensing unit at the tth sampling time point, and N is the total number of light sensing devices included in the ith light sensing unit.

[0046] The illumination uniformity of the illumination sensing unit can be calculated according to the following formula:

[0047]

[0048] Among them, U (i,T) is the illumination uniformity of the ith illumination sensing unit, and T is the total number of sampling time points.

[0049] The image acquisition module can be used to determine whether to perform image acquisition based on the illumination uniformity of each illumination sensing unit. When it is determined that image acquisition is to be performed, the image of the illumination sensing unit is acquired.

[0050] In some embodiments, the image acquisition module determines whether to perform image acquisition based on the illumination uniformity of each illumination sensing unit, including:

[0051] For any two light sensing units, calculate the light uniformity difference between the two light sensing units;

[0052] Based on the illumination uniformity difference between any two illumination sensing units, a illumination uniformity difference parameter is calculated;

[0053] When the illumination uniformity difference parameter is less than the preset illumination uniformity difference parameter, and there is at least one illumination sensing unit whose illumination uniformity is greater than the preset illumination uniformity, it is determined to perform image acquisition.

[0054] Specifically, the illumination uniformity difference parameter can be calculated according to the following formula:

[0055]

[0056] D (i,j) =(U (i,T) -U (j,T) ) 2

[0057] Among them, D all is the illumination uniformity difference parameter, D (i,j) is the difference in illumination uniformity between the i-th illumination sensing unit and the j-th illumination sensing unit, I is the total number of illumination sensing units, and U (j,T) is the illumination uniformity of the jth illumination sensing unit.

[0058] It is understandable that when there is an angle between the photovoltaic panel and the sun, the projection casts a shadow on the light sensing panel, and the light sensing devices at different positions on the light sensing panel sense different light intensities, which causes differences in the output voltage signals of the light sensing devices of the light sensing units. When the light uniformity difference parameter is less than the preset light uniformity difference parameter, and there is at least one light sensing unit with a light uniformity greater than the preset light uniformity, it is determined that multiple light sensing units have detected projections, and the length of the projection exceeds the preset requirement, that is, the angle between the photovoltaic panel and the sun is greater than the preset deviation angle, and the photovoltaic panel needs to be adjusted in angle.

[0059] When the illumination uniformity difference parameter is greater than or equal to the preset illumination uniformity difference parameter, it indicates that at least one illumination sensing unit has failed or is partially blocked, resulting in a false alarm. Based on the illumination uniformity difference between any two illumination sensing units, the abnormal value of each illumination sensing unit is determined, and the illumination sensing unit with an abnormal value greater than the preset abnormal value is taken as an abnormal illumination sensing unit. Based on the illumination uniformity difference between any two non-abnormal illumination sensing units, the illumination uniformity difference parameter is recalculated. When the recalculated illumination uniformity difference parameter is less than the preset illumination uniformity difference parameter, and there is at least one non-abnormal illumination sensing unit whose illumination uniformity is greater than the preset illumination uniformity, it is determined that image acquisition is to be performed.

[0060] The abnormal value of the light sensing unit can be calculated based on the following formula:

[0061]

[0062] Among them, A i is the abnormal value of the i-th light sensing unit.

[0063] In some embodiments, the image acquisition module acquires images of the light sensing unit, including: acquiring images of the light sensing unit in multiple image acquisition postures.

[0064] Specifically, each light sensing unit may be pre-set with a corresponding plurality of image acquisition postures, wherein the image acquisition device for acquiring images of the light sensing unit is pre-calibrated under the plurality of image acquisition postures. When the illumination uniformity difference parameter is less than the preset illumination uniformity difference parameter, and there is at least one light sensing unit whose illumination uniformity is greater than the preset illumination uniformity, the light sensing unit whose illumination uniformity is greater than the preset illumination uniformity is taken as the target light sensing unit, the plurality of image acquisition postures corresponding to the target light sensing unit are queried, and the image of the target light sensing unit is acquired under the plurality of image acquisition postures corresponding to the target light sensing unit.

[0065] The sun-following module can be used to determine the optimal sun-following angle of the photovoltaic panel based on the image of the light sensing unit, and adjust the angle of the photovoltaic panel based on the optimal sun-following angle.

[0066] In some embodiments, the sun following module determines the optimal sun following angle of the photovoltaic panel based on the image of the light sensing unit, including:

[0067] For each image acquisition posture, determining a relative position relationship between a projection endpoint and a plurality of markers based on an image of the light sensing unit acquired in the image acquisition posture;

[0068] Determine the sun position information based on the relative position relationship between the projection endpoints corresponding to each image acquisition posture and multiple markers;

[0069] Based on the solar position information, the optimal sun-following angle of the photovoltaic panel is determined.

[0070] Specifically, Figure 2 is a schematic diagram of a projection shown in an embodiment of the present application, such as Figure 2 , P represents the position of the sun, A represents the position of the projected object, A' represents the projection endpoint, R1, R2, and R3 represent three markers respectively. For each image acquisition posture, the image of the light sensing unit acquired under the image acquisition posture can be used to determine the projection endpoint pixel coordinates of the projection of the projection object projected onto the light sensing panel in the pixel coordinate system and the marker pixel coordinates of each marker in the pixel coordinate system. Based on the projection endpoint pixel coordinates and the marker pixel coordinates, the relative position relationship between the projection endpoint and the multiple markers is determined.

[0071] The sun's azimuth information is determined through an azimuth determination model based on the relative position relationship between the projection endpoints corresponding to each image acquisition posture and multiple markers, wherein the azimuth determination model can be a convolutional neural network (CNN) model.

[0072] In some embodiments, the sun following module determines the optimal sun following angle of the photovoltaic panel based on the sun position information, including:

[0073] Based on the sun position information, determining multiple candidate sun following angles;

[0074] Obtain the battery temperature information of the photovoltaic panel at multiple consecutive time points;

[0075] Obtaining ambient temperature information at multiple consecutive time points;

[0076] Based on the battery temperature information of the photovoltaic panel at multiple consecutive time points and the ambient temperature information at multiple consecutive time points, multiple candidate sun-following angles are screened to determine the optimal sun-following angle.

[0077] Specifically, the candidate sun-following angle may be a cell angle whose angle difference with the cell angle at which the photovoltaic panel is perpendicular to the sun at the current position is less than a preset angle difference.

[0078] A plurality of candidate sun following angles may be determined based on the sun position information through an angle determination model, wherein the angle determination model may be a convolutional neural network (CNN) model.

[0079] The impact of temperature on photovoltaic panels is mainly reflected in the following aspects:

[0080] 1. The intrinsic performance of photovoltaic panel batteries decreases: The maximum power point of photovoltaic panel batteries moves with the change of temperature, and this movement direction causes the output voltage to decrease, thus affecting the intrinsic performance of the battery.

[0081] 2. The current of the photovoltaic panel decreases: As the temperature of the photovoltaic panel increases, the resistivity of the conductor material increases and the current becomes smaller, resulting in a decrease in the output current of the photovoltaic panel.

[0082] 3. Increased losses in photovoltaic panels: The resistance of photovoltaic panels increases with rising temperature, which leads to increased energy losses in the system.

[0083] Therefore, during the operation of a photovoltaic power station, in order to maintain a high power generation efficiency, some measures need to be taken to regulate the operating temperature of the solar panels.

[0084] For each candidate solar following angle, the battery temperature information of the photovoltaic panel at multiple consecutive time points after the photovoltaic panel is adjusted to the candidate solar following angle can be predicted by a temperature prediction model based on the battery temperature information of the photovoltaic panel at multiple consecutive time points and the ambient temperature information at multiple consecutive time points. Among them, the temperature prediction model can be an LSTM (Long short-term memory) model. Based on the predicted battery temperature information of the photovoltaic panel at multiple consecutive time points, the matching value corresponding to the candidate solar following angle is calculated. The candidate solar following angle with the largest matching value can be used as the optimal solar following angle.

[0085] The matching value corresponding to the candidate sun following angle can be calculated based on the following formula:

[0086]

[0087] Among them, M i is the matching value corresponding to the i-th candidate sun following angle, P2 is the preset parameter, and P2>0, T (i,t) is the predicted battery temperature of the photovoltaic panel at time t under the i-th candidate sun following angle, T0 is the preset battery temperature, and T is the total number of predicted time points.

[0088] The temperature control module can be used to control the temperature of the photovoltaic panel based on the battery temperature information of the photovoltaic panel at multiple consecutive time points, the ambient temperature information at multiple consecutive time points and the optimal sun following angle.

[0089] Specifically, a temperature control strategy can be determined by a temperature control strategy determination model based on the battery temperature information of the photovoltaic panel at multiple consecutive time points, the ambient temperature information at multiple consecutive time points, and the optimal sun following angle, to generate a temperature control strategy, and based on the temperature control strategy, control the temperature control device to perform temperature control on the photovoltaic panel. The temperature control strategy determination model can be a GRU (Gate Recurrent Unit) model.

[0090] In some embodiments, the sun following module adjusts the angle of the photovoltaic panel based on the optimal sun following angle, including:

[0091] Based on the historical sun-following information, generating a preset sun-following strategy, wherein the preset sun-following strategy includes preset photovoltaic panel angles in multiple time periods;

[0092] When the angle difference between the optimal sun-following angle and the corresponding preset photovoltaic panel angle is greater than the preset difference, the angle of the photovoltaic panel and the preset sun-following strategy are adjusted according to the optimal sun-following angle.

[0093] Specifically, the sun's position change information for each day in a time period adjacent to the current time (e.g., a week, a month, etc.) can be obtained, wherein the sun's position change information can include the sun's position information at multiple time points in a day. The similarity determination model is based on the sun's position change information for each day in the time period adjacent to the current time and the sun's position change information for each day in the historical time period, and the similarity of the sun's position change between the adjacent time period of the current time and the historical time period is calculated, and the historical time period whose sun's position change similarity is greater than the preset sun's position change similarity is taken as a similar historical time period, wherein the time lengths of the adjacent time period and the historical time period are equal, and the similarity determination model can be an LSTM (Long short-term memory) model.

[0094] A preset sun following strategy can be generated by a following strategy generation model based on historical sun following information of similar historical time periods, wherein the historical sun following information of the historical time period may include sun following angles of multiple historical time points of similar historical time periods, and the following strategy generation model may be a GRU (Gate Recurrent Unit) model.

[0095] When the angle difference between the optimal sun-following angle and the corresponding preset photovoltaic panel angle is greater than the preset difference, the angle of the photovoltaic panel and the preset sun-following strategy can be adjusted according to the optimal sun-following angle through the strategy adjustment model, wherein the strategy adjustment model can be a GRU (Gate Recurrent Unit) model.

[0096] In some embodiments, the control system for photovoltaic panels tracking the sun based on image recognition can also be applied to scenarios such as energy storage systems and grid connection.

[0097] For example, in the energy storage system scenario, through the control system of photovoltaic panels tracking the sun based on image recognition, photovoltaic panels can be aligned with the sun in real time, so as to maximize the capture of solar energy and convert it into electrical energy. When the power generation exceeds the immediate demand, the excess electrical energy can be efficiently stored in energy storage devices (such as lithium-ion batteries, lead-acid batteries, etc.). Since the control system of photovoltaic panels tracking the sun based on image recognition improves the power generation efficiency of photovoltaic panels, it also indirectly improves the energy storage efficiency of energy storage systems. In the energy storage system, combined with the real-time power generation data of photovoltaic panels, the charging and discharging process of energy storage devices can be more accurately predicted and scheduled. As an example only, when there is sufficient sunlight and large power generation, the energy storage device is charged first; when there is insufficient sunlight or peak power consumption, the electrical energy in the energy storage device is released to meet the power demand. The combination of the control system of photovoltaic panels tracking the sun based on image recognition and the energy storage system helps to improve the stability of the entire power system. Through the rapid response capability of the energy storage device, when the photovoltaic power generation fluctuates greatly, stable power output can be provided to reduce the impact on the power grid. In the energy storage system, the control system for photovoltaic panels tracking the sun based on image recognition can work together with components such as energy storage inverters and battery management systems to achieve dynamic power matching. When the grid load is low or the photovoltaic power generation is large, the energy storage system automatically increases the charging power of the energy storage device. When the grid load is peak or the photovoltaic power generation is insufficient, the energy storage system will release the electrical energy in the energy storage device to supplement the grid power supply. This dynamic adjustment mechanism helps to balance the supply and demand relationship of the grid and improve energy utilization efficiency.

[0098] For another example, in the grid-connected scenario, the photovoltaic panel tracking sun control system based on image recognition can ensure that the photovoltaic panel is always aligned with the sun at the best angle, thereby outputting stable and high-quality electric energy. This electric energy is converted into AC power that meets the requirements of the grid by the grid-connected inverter and then connected to the grid, which helps to improve the power quality of the grid. During the peak load period of the grid, the photovoltaic panel tracking sun control system based on image recognition can release the stored electric energy through the cooperation of energy storage equipment to relieve the pressure on the grid. The high power generation efficiency of the photovoltaic panel tracking sun control system based on image recognition allows more electric energy to be stored and reserved, enhancing the peak load regulation capacity of the grid. With the large-scale access of new energy to the grid, the problem of absorption is becoming increasingly prominent. The combined application of the photovoltaic panel tracking sun control system based on image recognition and the energy storage system can smooth the output power curve of photovoltaic power generation, reduce the impact of its volatility on the grid, and thus promote the absorption of new energy. Under the framework of the smart grid, the photovoltaic panel tracking sun control system based on image recognition can realize information sharing and collaborative control with the grid dispatching system. By real-time monitoring of the power generation status of photovoltaic power stations and the charging and discharging status of energy storage equipment, the grid dispatching system can more flexibly adjust the grid operation strategy to achieve the optimal configuration and coordinated operation of new energy and conventional energy. The control system for photovoltaic panels tracking the sun based on image recognition can achieve information sharing and coordinated control with the grid dispatching system. The dispatching system can comprehensively formulate power generation plans and dispatching instructions based on information such as grid load forecasts, photovoltaic power generation forecasts, and energy storage system status. The control system for photovoltaic panels tracking the sun based on image recognition adjusts the tracking angle of photovoltaic panels and the output power of inverters in real time according to these instructions to ensure a balance between photovoltaic power generation and grid demand.

[0099] Figure 2 FIG. 1 is a flow chart of a control method for photovoltaic panels tracking the sun based on image recognition in an embodiment of the present application. Figure 2 As shown, the control method of photovoltaic panel tracking the sun based on image recognition may include the following steps.

[0100] Step 210, setting a plurality of light sensing units on a plurality of photovoltaic panels;

[0101] Step 220, determining the illumination uniformity of the illumination sensing unit based on the output of the illumination sensing unit;

[0102] Step 230, determining whether to perform image acquisition based on the illumination uniformity of each illumination sensing unit;

[0103] Step 240, when it is determined to perform image acquisition, an image of the light sensing unit is acquired;

[0104] Step 250, determining the optimal sun-following angle of the photovoltaic panel based on the image of the light sensing unit;

[0105] Step 260, adjusting the angle of the photovoltaic panel based on the optimal sun-following angle.

[0106] The control method of photovoltaic panels tracking the sun based on image recognition can be executed by a control system of photovoltaic panels tracking the sun based on image recognition. For more descriptions of the control method of photovoltaic panels tracking the sun based on image recognition, please refer to the relevant description of the control system of photovoltaic panels tracking the sun based on image recognition, which will not be repeated here.

[0107] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A control system for photovoltaic panels tracking the sun based on image recognition, applied to photovoltaic systems, where: The photovoltaic system comprises a plurality of photovoltaic panels, and is characterized in that it comprises: A light sensing module, comprising a plurality of light sensing units and a light uniformity determination unit, wherein the plurality of light sensing units are respectively arranged on the plurality of photovoltaic panels, the light uniformity determination unit is used to determine the light uniformity of the light sensing unit based on the output of the light sensing unit, the light sensing unit comprises a light sensing panel arranged on one side of the photovoltaic panel, the light sensing panel and the photovoltaic panel are located in the same plane, a plurality of light sensing devices are arranged on the light sensing panel, and a projection object is arranged on the light sensing panel; An image acquisition module, used for determining whether to perform image acquisition based on the illumination uniformity of each of the illumination sensing units, and when it is determined that image acquisition is to be performed, acquiring the image of the illumination sensing unit; A sun following module, used to determine an optimal sun following angle of the photovoltaic panel based on the image of the light sensing unit, and adjust the angle of the photovoltaic panel based on the optimal sun following angle; The image acquisition module determines whether to perform image acquisition based on the illumination uniformity of each of the illumination sensing units, including: For any two of the light sensing units, calculating the light uniformity difference between the two light sensing units; Calculating a light uniformity difference parameter based on a light uniformity difference between any two of the light sensing units; When the illumination uniformity difference parameter is less than the preset illumination uniformity difference parameter, and the illumination uniformity of at least one of the illumination sensing units is greater than the preset illumination uniformity, it is determined to perform image acquisition; The light sensing panel is also provided with a plurality of markers; The image acquisition module acquires the image of the light sensing unit, including: acquiring the image of the light sensing unit in a plurality of image acquisition postures; The sun following module determines the optimal sun following angle of the photovoltaic panel based on the image of the light sensing unit, including: For each of the image acquisition postures, determining the relative positional relationship between the projection endpoint and the plurality of markers based on the image of the light sensing unit acquired in the image acquisition posture; Determine a second relative position relationship between the projection endpoint and the plurality of markers based on the relative position relationship between the projection endpoint corresponding to each of the image acquisition postures and the plurality of markers; Determining an optimal sun-following angle of the photovoltaic panel based on a second relative position relationship between the projection endpoint and the plurality of markers; The sun following module determines the optimal sun following angle of the photovoltaic panel based on the second relative position relationship between the projection endpoint and the plurality of markers, including: Determining a plurality of candidate sun-following angles based on a second relative position relationship between the projection endpoint and the plurality of markers; Acquiring battery temperature information of the photovoltaic panel at multiple consecutive time points; Obtaining ambient temperature information at multiple consecutive time points; Based on the battery temperature information of the photovoltaic panel at multiple consecutive time points and the ambient temperature information at multiple consecutive time points, the multiple candidate sun-following angles are screened to determine the optimal sun-following angle.

2. The photovoltaic panel sun tracking control system based on image recognition according to claim 1 is characterized in that: The illumination uniformity determination unit determines the illumination uniformity of the illumination sensing unit based on the output of the illumination sensing unit, including: For each of the light sensing devices, performing empirical mode decomposition on the output voltage signal of the light sensing device at multiple consecutive time points to generate an eigenmode function and a residual, and generating a corrected output voltage signal at multiple consecutive time points based on the eigenmode function and the residual by using a denoising model; The illumination uniformity of the illumination sensing unit is determined based on the corrected output voltage signals at a plurality of consecutive time points corresponding to each of the illumination sensing devices.

3. The photovoltaic panel sun tracking control system based on image recognition according to claim 2, characterized in that: The illumination uniformity judgment unit determines the illumination uniformity of the illumination sensing unit based on the corrected output voltage signals at multiple consecutive time points corresponding to each of the illumination sensing devices, including: for each of the illumination sensing devices, sampling the corrected output voltage signals at multiple consecutive time points corresponding to the illumination sensing device to obtain the output voltage signals of the illumination sensing device at multiple sampling time points; for each of the sampling time points, calculating the single-point illumination uniformity corresponding to the sampling time point based on the output voltage signal of each of the illumination sensing devices at the sampling time point; and determining the illumination uniformity of the illumination sensing unit based on the single-point illumination uniformity corresponding to each of the sampling time points.

4. The photovoltaic panel sun tracking control system based on image recognition according to claim 1, characterized in that: It also includes a temperature control module for controlling the temperature of the photovoltaic panel based on the battery temperature information of the photovoltaic panel at multiple consecutive time points, the ambient temperature information at multiple consecutive time points and the optimal sun following angle.

5. The photovoltaic panel sun tracking control system based on image recognition according to claim 4 is characterized in that: The sun following module adjusts the angle of the photovoltaic panel based on the optimal sun following angle, including: Based on the historical sun-following information, generating a preset sun-following strategy, wherein the preset sun-following strategy includes preset photovoltaic panel angles in multiple time periods; When the angle difference between the optimal sun-following angle and the corresponding preset photovoltaic panel angle is greater than the preset difference, the angle of the photovoltaic panel and the preset sun-following strategy are adjusted according to the optimal sun-following angle.

6. A control method for photovoltaic panels tracking the sun based on image recognition, which is executed by the control system for photovoltaic panels tracking the sun based on image recognition according to any one of claims 1 to 5, characterized in that: include: Arrange multiple light sensing units on multiple photovoltaic panels; Determining the illumination uniformity of the illumination sensing unit based on the output of the illumination sensing unit; Determining whether to perform image acquisition based on the illumination uniformity of each of the illumination sensing units; When it is determined to perform image acquisition, acquiring an image of the light sensing unit; Determining an optimal sun-following angle of a photovoltaic panel based on the image of the light sensing unit; The angle of the photovoltaic panel is adjusted based on the optimal sun-following angle.

Images