Solar power generation analysis method and system for light tracking

Abstract

The invention discloses a solar power generation analysis method and system for light tracking, and relates to the field of power generation analysis, and the method comprises the steps: obtaining track feature points related to the motion of the sun; introducing time interval information of variable step size according to the key features, constructing a sunlight motion track model, and capturing variable travel track points between adjacent track feature points; key point identification is carried out on the change travel trajectory points, a target trajectory calculation model is constructed, and sun moving trajectory points are output; analyzing the light tracking area of the photovoltaic module under the condition of the sun moving trajectory point, and determining the optimal installation angle of the photovoltaic module; and acquiring a sunlight angle height and a motion azimuth angle, constructing a power generation output power model on the premise of the optimal installation angle of the photovoltaic module, and acquiring a sunlight generating capacity prediction result. According to the method, the light tracking response capability is dynamically solved based on the actual geographical environment condition, the power generation output power model under the optimal installation angle condition is determined, and the quality of solar power generation analysis is improved.

Description

Technical Field

[0001] This invention relates to the field of power generation analysis, and more particularly to a method and system for analyzing solar power generation for light tracking. Background Technology

[0002] Solar power generation refers to the process of converting solar radiation energy into electrical energy through a solar photovoltaic system. It has significant advantages such as being green, renewable, and having low carbon emissions. Its core relies on the photovoltaic modules' ability to absorb solar radiation and output electrical energy. Solar power generation prediction is a comprehensive task that combines astrophysical modeling, meteorological data analysis, and power system behavior simulation. Its core purpose is to estimate how much electrical energy a solar system can output within a given time frame.

[0003] From the perspective of power system operation, solar energy is highly random and intermittent, and is easily affected by factors such as weather, season, and solar intensity. Predicting power generation is the key to ensuring the stable operation of the power grid and the balance between supply and demand. Furthermore, accurate power generation forecasting is the basis for formulating electricity sales plans, participating in electricity price settlement, and asset valuation.

[0004] However, existing solar power generation analysis methods fail to fully integrate the solar trajectory points with the installation environment of photovoltaic modules during implementation, resulting in an inability to accurately analyze changes in the light tracking area of ​​photovoltaic modules during operation, thus affecting the accuracy of solar power generation analysis results.

[0005] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0006] To address the aforementioned problems, this invention proposes a method and system for analyzing solar power generation using light tracking, thereby achieving the goal of analyzing solar power generation over a continuous time dimension.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for analyzing solar power generation for light tracking, comprising: Historical astronomical and temporal parameters within the area to be installed with photovoltaic modules are used as solar tracking data. The maximum information coefficient is used to filter out trajectory feature points related to solar motion in the solar tracking data. Based on key features, time interval information with variable step size is introduced and combined with time series prediction technology to construct a solar motion trajectory model, capturing the changing trajectory points between adjacent trajectory feature points; Key points are identified for the changing trajectory points, and a target trajectory calculation model is constructed based on the key point identification results to output the solar trajectory points; The terrain conditions of the area to be installed are collected, and dynamic simulation technology is used to analyze the light tracking area of ​​the photovoltaic module under the condition of the sun's orbital point, so as to determine the optimal installation angle of the photovoltaic module. Obtain the solar angle height and azimuth angle of the target time period in the area to be installed, construct a power output model under the premise of the optimal installation angle of the photovoltaic module, and obtain the predicted results of solar power generation.

[0008] Preferably, the terrain conditions within the installation area are collected, and dynamic simulation technology is used to analyze the light tracking area of ​​the photovoltaic modules under the conditions of the sun's orbital path, determining the optimal installation angle of the photovoltaic modules, including: The terrain conditions in the area to be installed are collected using remote sensing technology, and the terrain conditions are combined with the geometric parameters of the photovoltaic modules to establish a three-dimensional state model of the relative position of the photovoltaic modules and the terrain conditions. By using dynamic simulation technology to run a three-dimensional state model, the effective light-receiving area of ​​the photovoltaic module at different installation angles under the condition of the sun's orbit point is analyzed, and the light tracking area is determined. Based on the optimal incident angle of sunlight under the condition of searching the solar trajectory point using the light tracking area and multidimensional optimization algorithm, the optimal installation angle of photovoltaic modules is determined based on the optimal incident angle of sunlight and boundary conditions.

[0009] Preferably, the solar radiation angle height and azimuth angle of the area to be installed during the target time period are obtained, and a power output model is constructed under the premise of the optimal installation angle of the photovoltaic module. The predicted results of solar power generation include: Acquire solar radiation angle height and azimuth data at various time points within the target time period of the area to be installed, and analyze the solar trajectory based on the solar radiation angle height and azimuth data to construct an irradiance receiving model; Based on the irradiance receiving model, the effective irradiance of the sun at each time point is analyzed, the total incident irradiance projected onto the photovoltaic module is analyzed, and a temperature coefficient is introduced to correct the conversion efficiency of the photovoltaic module at each time point. A power output model is constructed based on the photovoltaic module conversion efficiency and the optimal installation angle of the photovoltaic module. The instantaneous power output curve is calculated, and the solar power generation prediction result is obtained by integrating the power output within the target time period.

[0010] Secondly, the present invention also provides a solar power generation analysis system for light tracking, the system comprising: The trajectory feature point acquisition module is used to use historical astronomical and temporal parameters within the area to be installed of the photovoltaic module as solar tracking data, and to filter out trajectory feature points related to solar motion from the solar tracking data using the maximum information coefficient; The trajectory point analysis module is used to introduce time interval information with variable step size based on key features, and combine it with time series prediction technology to construct a solar motion trajectory model, capturing the changing trajectory points between adjacent trajectory feature points; The attribution status acquisition module is used to identify key points of the changing trajectory points, construct the target trajectory calculation model based on the key point identification results, and output the solar trajectory points. The photovoltaic installation angle determination module is used to collect the terrain conditions in the area to be installed, and to analyze the light tracking area of ​​the photovoltaic module under the condition of the sun's orbital point using dynamic simulation technology, so as to determine the optimal installation angle of the photovoltaic module. The solar power generation prediction module is used to obtain the solar angle height and azimuth angle of the target time period in the area to be installed, construct a power output model under the premise of the optimal installation angle of the photovoltaic module, and obtain the solar power generation prediction results.

[0011] The beneficial effects of this invention are as follows: 1. This invention uses the maximum information coefficient for feature selection based on historical astronomical and time parameters, effectively extracting trajectory feature points most strongly correlated with solar motion from time series, greatly improving the sensitivity and explanatory power of key variables. Combined with the Linde evolutionary Buzogre algorithm for key change point extraction, it can accurately identify non-stationary abrupt changes in the trajectory and avoid feature misjudgments caused by noise interference, thus ensuring the stability and accuracy of solar trajectory point extraction. Furthermore, the target trajectory calculation model constructed using the minimum distortion criterion and Welch spectral estimation algorithm further optimizes the trajectory spectrum energy structure, resulting in the final output solar trajectory points possessing good continuity and physical consistency. Simultaneously, by integrating terrain data input and using dynamic simulation technology to calculate the changes in the solar projection area of ​​photovoltaic modules in real time, it dynamically solves the light tracking response capability based on actual geographical conditions to determine the power output model under optimal installation angle conditions, improving the quality of analysis.

[0012] 2. This invention utilizes the mean of the feature vectors of candidate key points to construct initial cluster centers, and continuously updates the category boundaries through total distortion calculation and a splitting mechanism, which can improve clustering accuracy and key point discrimination stability. Furthermore, by combining the minimum distortion criterion with the Welch algorithm, the frequency component change patterns in motion signals can be extracted, enabling the target trajectory model to maintain spatial continuity while possessing frequency domain response consistency. This provides a high-quality input basis for subsequent photovoltaic module installation angle selection, illumination simulation, and dynamic power generation prediction.

[0013] 3. This invention acquires solar angular height and azimuth data for a target time period within a target area, and combines this with time series analysis to form a high-resolution solar trajectory model. This ensures the spatial accuracy and temporal integrity of the illumination angle simulation. Furthermore, it constructs an irradiance receiving model to reflect the impact of solar radiation on photovoltaic modules, thereby accurately analyzing the effective irradiance values ​​at different time points. It also calculates the actual usable total incident irradiance per unit area using projection modeling, and introduces a temperature coefficient to dynamically correct the conversion efficiency of the photovoltaic modules, effectively solving the problem of efficiency fluctuations caused by temperature rise in the photovoltaic modules. This makes the power model more closely reflect real-world operating conditions. Based on this, a power output model is established by combining the optimal installation angle of the photovoltaic modules, calculating the instantaneous power generation curve at each time point. Finally, by integrating the power within the target time period, the predicted solar power generation over a continuous time dimension can be obtained. Attached Figure Description

[0014] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0015] Figure 1 This is a flowchart of a solar power generation analysis method for light tracking according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a solar power generation analysis system for light tracking according to an embodiment of the present invention.

[0016] In the picture: 1. Trajectory feature point acquisition module; 2. Travel trajectory point analysis module; 3. Attribution status acquisition module; 4. Photovoltaic installation angle determination module; 5. Solar power generation prediction module. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0018] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. Furthermore, it should be understood that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0021] Please see Figure 1 This invention provides a method for analyzing solar power generation for light tracking, comprising: Step S1: Use historical astronomical and temporal parameters within the area to be installed on the photovoltaic module as solar tracking data, and use the maximum information coefficient to filter out trajectory feature points related to solar motion in the solar tracking data.

[0022] In one embodiment, when using the maximum information coefficient to filter out trajectory feature points related to solar motion in solar tracking data, two sets of random variable sequences, altitude and azimuth, can be filtered out from the solar tracking data. A scatter plot of data is generated using the random variable sequences, and a grid is drawn based on the scatter plot. The mutual information of the grid represents the state of the random variable sequences, and the maximum value of the mutual information is selected as the maximum information coefficient. The maximum information coefficient is used to analyze the random variable sequences. The importance ranking of angle change features is obtained based on the analysis results, and the average value of the maximum information coefficient of the angle change features is used as the screening threshold to filter out angle change features with a maximum information coefficient greater than the threshold. The screening results are used as key features of trajectory change, and the key features of trajectory change are reconstructed to filter out trajectory feature points related to solar motion from the key features of trajectory change.

[0023] It needs to be explained that in the process of acquiring trajectory feature points, solar tracking data is obtained from the historical astronomical data of the area where the photovoltaic modules are to be installed. This includes the solar altitude angle and azimuth angle. The altitude angle and azimuth angle determine the trajectory of the sun relative to the ground and the photovoltaic modules, and are used as the raw input to form two sets of random variable sequences, namely the solar altitude angle sequence and the azimuth angle sequence. Then, the random variable sequences are used to generate a data scatter plot to reflect the correlation of variables. After generating the scatter plot, the relationships between the data are plotted in a grid, which helps to refine the understanding of the distribution of the data space, and allows us to view the concentration and trend of the data in different intervals.

[0024] Mutual information represents the amount of information about one variable contained in another variable. It is a non-linear correlation measure. Through mutual information analysis, we can identify which intervals of solar motion information are most critical, that is, which time periods of solar motion characteristics have the greatest impact on subsequent analysis and prediction. Furthermore, by selecting the maximum mutual information value as the maximum information coefficient, we can further determine which part of the trajectory is the most important.

[0025] Using the obtained maximum information coefficient, a detailed analysis of the random variable sequence is conducted to evaluate the contribution of different solar trajectory points to the overall power generation prediction. Based on the analysis results, the importance of angle change features is ranked, identifying which solar angle changes are most representative or have the greatest impact on the light-receiving area of ​​photovoltaic modules. This allows for the selection of the most informative angle change features to further guide the accurate prediction of solar power generation. The selection threshold is set based on the average value of the maximum information coefficient, ensuring that only angle change features exhibiting strong correlation are retained, while other irrelevant features are filtered out, thereby improving the efficiency and accuracy of subsequent analysis. The selected angle change features are then used as key features for trajectory change reconstruction, retaining only those that significantly influence changes in the solar illumination angle. Using these reconstructed key features, the trajectory feature points most relevant to solar motion are further selected. These trajectory feature points will become the core input of the subsequent solar power generation prediction model, ensuring higher accuracy and practical significance in the prediction results.

[0026] Furthermore, by accurately extracting key feature points from the sun's trajectory, this process solves common problems in traditional solar power generation prediction methods, such as information overload and redundant data interference, making the light tracking and power generation prediction of photovoltaic modules more accurate.

[0027] Specifically, assuming the proposed photovoltaic power station is located at latitude 34.34°N and longitude 108.94°E, historical astronomical data of this region will be used as input to analyze the sun's trajectory, thereby predicting solar tracking and power generation of the photovoltaic modules. The specific process of selecting trajectory feature points related to solar motion from the solar tracking data is as follows: Step 1: Data Acquisition and Historical Astronomical Data Processing; Historical solar data for the region is obtained using a meteorological data management platform, including the solar altitude and azimuth angles. At the same time, the time parameters for the day are also obtained, including the date and time. The solar altitude and azimuth angles are recorded daily according to time periods (e.g., one data point every 5 minutes).

[0028] Suppose the collected data for a certain day is as shown in Table 1 (taking hourly data as an example): Table 1: Collected Data Assuming a year has 365 days, with 24 data points recorded each day, the total data volume is 365 × 24 = 8760 data points.

[0029] Step 2: Screening solar motion features using the maximum information coefficient; Construct a solar tracking dataset: The key features of the solar trajectory are the solar elevation angle and azimuth angle. Each pair (elevation angle, azimuth angle) can be regarded as the solar position at a certain moment. The number of data points is 8760.

[0030] Data processing: Historical astronomical data are organized into two sets of random variable sequences (solar altitude angle sequence and azimuth angle sequence). For each moment, the changing trends of altitude angle and azimuth angle are recorded as input features for training the model.

[0031] Maximum Information Coefficient (MIC) Calculation: The maximum information coefficient (MIC) algorithm is used to analyze the nonlinear relationship between the solar altitude angle and azimuth angle, and to calculate the mutual information of the altitude angle and azimuth angle at each moment.

[0032] Suppose that the changes in solar altitude angle and azimuth angle between 10:00 AM and 11:00 AM on June 1st, XXX year are as follows: At 10:00 AM on June 1, XXX, the altitude is 50.9° and the azimuth is 115.4°; at 11:00 AM on June 1, XXX, the altitude is 65.1° and the azimuth is 125.6°. The maximum information coefficient of the data points is calculated, assuming a result of 0.75. This maximum information coefficient value indicates a strong non-linear correlation between the elevation angle and azimuth angle within that time period. This method allows us to identify the time periods most critical to changes in the light-receiving angle of photovoltaic modules.

[0033] Step 3: Trajectory Feature Filtering and Reconstruction; Filtering key trajectory feature points: Based on the calculation results of the maximum information coefficient, select the trajectory feature points with the most information. For each data segment (e.g., by hour or by day), select the time period with MIC>0.5.

[0034] Reconstruction of key trajectory features: For the selected key time periods, the changes in elevation and azimuth are reconstructed, for example, the point of maximum change or inflection point of the sun's position (such as the point of abrupt change in maximum elevation or azimuth) is extracted.

[0035] For example, at 11:00 on June 1, XXX, the altitude angle is 65.1° and the azimuth angle is 125.6°; at 12:00 on June 1, XXX, the altitude angle is 75.4° and the azimuth angle is 135.7°; and at 13:00 on June 1, XXX, the altitude angle is 80.1° and the azimuth angle is 145.8°. From this, we can extract the key time point (12:00-13:00), that is, the angle of sunlight changes the most at this time, which will become the key input for subsequent solar power generation prediction models.

[0036] Furthermore, by filtering and reconstructing features based on the maximum information coefficient, irrelevant features are reduced, prediction accuracy is improved, and the data overload problem in traditional photovoltaic power generation prediction methods is solved, providing strong support for accurate prediction of solar power generation.

[0037] Step S2: Based on key features, introduce time interval information with variable step size, and combine it with time series prediction technology to construct a solar motion trajectory model, capturing the changing trajectory points between adjacent trajectory feature points.

[0038] It should be explained that, in the process of capturing points on the changing trajectory, the time step can be dynamically adjusted according to the rate of change of trajectory characteristics and the fluctuation of sunlight intensity. This allows for the precise capture of rapid changes in the sun's trajectory. For example, when the sun is just rising or about to set, the trajectory changes more dramatically, and using a shorter time step at this time can improve the accuracy of trajectory modeling. Conversely, during periods when the sun's trajectory is relatively flat, a larger time step can be used to reduce computational load while maintaining model efficiency. Introducing a variable time step not only optimizes computational efficiency but also makes the model's response to the sun's trajectory more flexible, enabling it to automatically adjust computational accuracy at different times, thereby improving the adaptability and accuracy of the power generation prediction model.

[0039] Furthermore, by combining this with time series forecasting techniques, advanced time series modeling methods such as ARIMA (Autoregressive Integrated Moving Average) and LSTM (Long Short-Term Memory) are employed. Historical solar angle data and relevant environmental information are used to predict future solar trajectories, capturing long-term dependencies and temporal patterns in the trajectory data. This allows for the prediction of future solar motion states and accurate calculation of future solar angle changes. Furthermore, periodic variations in the time series (such as day-night cycles and seasonal changes) can be considered. Using these prediction results, the changing trajectory points between adjacent trajectory feature points can be further captured, identifying the changing trajectory between two adjacent solar trajectory points. These changing trajectory points provide a better understanding of the nonlinear characteristics of solar radiation, such as speed variations and trajectory curvature, offering crucial information for subsequent power generation forecasting and the selection of optimal installation angles for photovoltaic modules.

[0040] Step S3: Identify key points of the changing trajectory points, construct the target trajectory calculation model based on the key point identification results, and output the solar trajectory points.

[0041] In one embodiment, when identifying key points of the changing trajectory points and constructing a target trajectory calculation model based on the key point identification results to output the solar trajectory points, the instantaneous velocity of the trajectory can be determined based on the time difference between any two trajectory points within the changing trajectory points. Feature vectors are extracted from the changing trajectory points to generate vectors in the form of a combination of normalized velocity and position as feature vectors for candidate key points. Cluster centers are set based on the mean of the feature vectors of the candidate key points, and the total distortion is calculated. Based on the calculation results, splitting and initializing the total distortion are performed to determine the relative distortion improvement. The difference between the relative distortion improvement and the distortion threshold is compared, and when the calculated relative distortion improvement is greater than the threshold, the attribution status of the cluster centers and the feature vectors of the candidate key points is output. The optimal trajectory is calculated based on the cluster centers using the minimum distortion criterion. The target trajectory calculation model is constructed using the Welch algorithm to screen key points, and the probability attribution status is calculated to output the solar trajectory points.

[0042] In this process, when calculating the optimal trajectory based on the cluster center using the minimum distortion criterion, constructing a target trajectory calculation model using the Welch algorithm to screen key points, and calculating the probability of attribution states to output the solar trajectory points, the minimum distortion criterion can be used to randomly select variable trajectory points and determine the corresponding cluster centers. The Linde evolutionary Buzogre algorithm is then executed to obtain candidate key points generated by the variable trajectory points. A Hidden Markov Chain is constructed using the cluster centers as the observation sequence and the candidate key points as the hidden states. The initialization parameters of the Hidden Markov Model are calculated based on the attribution hidden states and observation values ​​of the trajectory points in the variable trajectory points. The observation sequence and distortion threshold are used as inputs to the Welch algorithm, and initialization and iterative processing are performed to complete parameter estimation and output the parameter information of the Hidden Markov Model, which serves as the target trajectory calculation model. The target trajectory calculation model is used to screen the variable trajectory point set corresponding to the variable trajectory points, and the Viterbi algorithm is used to analyze the key point set to obtain the attribution state with the highest probability. Maximum likelihood estimation is used to recalculate the state parameters of the key points with the attribution states, and the state parameter calculation results are combined with Gaussian mixture regression to obtain the reconstructed trajectory, outputting the solar trajectory points.

[0043] It should be explained that during the output of the solar trajectory points, the Linde Evolutionary Boozogray algorithm is used to identify key points of the changing trajectory. The core of the Linde Evolutionary Boozogray algorithm is to gradually optimize the division of trajectory points through evolution, and select the most representative changing points in the trajectory. In the implementation process, its function is to accurately identify key nodes in the solar trajectory. These nodes directly affect the calculation of the photovoltaic module's illumination and the prediction of power generation, ensuring that subsequent calculations only focus on key data and avoid interference from noise points, thereby improving calculation efficiency and accuracy.

[0044] Furthermore, by analyzing the time difference, the instantaneous velocity between adjacent trajectory points can be determined. By calculating the time difference and spatial distance between trajectory points with varying travel distances, the instantaneous velocity of the trajectory can be obtained. This is a key physical characteristic of the movement of sunlight. Based on this, feature vectors are extracted from trajectory points with varying travel distances. The resulting feature vector combination includes normalized velocity and position. This vector form enables candidate key points to have sufficient spatiotemporal feature information. The normalization of feature vectors helps to eliminate scale differences caused by different time spans.

[0045] Simultaneously, cluster centers are set based on the mean of the feature vectors of candidate key points, and the total distortion is calculated. Then, the distribution of key points is determined through clustering methods, and the quality of each cluster is measured by the distortion. The lower the distortion, the better the clustering effect. Based on this, splitting and initialization of total distortion processing are performed, and the relative distortion improvement is judged. At this stage, by comparing the distortion changes between different clusters, the position of the cluster centers is automatically optimized, the optimal cluster structure is selected, and the difference between the relative distortion improvement and the distortion threshold is compared to determine whether to output the belonging status of the cluster centers and candidate key points. In this way, it can be determined which candidate points' features best match the actual changes of the solar trajectory, thereby ensuring more accurate trajectory analysis.

[0046] Based on cluster centers, the optimal trajectory is calculated using the minimum distortion criterion, and combined with the Welch algorithm to construct the target trajectory calculation model. The minimum distortion criterion aims to find trajectory points that minimize trajectory variation errors, while the Welch algorithm optimizes the trajectory point calculation results by maximizing the model's likelihood. The combination of these two algorithms can output a precisely optimized solar trajectory, further providing a reliable basis for solar power generation prediction. The minimum distortion criterion is used to randomly select variable trajectory points and determine the relationship between the trajectory points and their corresponding cluster centers. The Linde evolutionary Buzogre algorithm is then executed to obtain candidate key points generated by the variable trajectory points. The algorithm identifies the most representative trajectory points through a stepwise optimization process, thereby refining the extraction of trajectory features.

[0047] Using cluster centers as the observation sequence and candidate keypoints as hidden states, a Hidden Markov Model (HMM) is constructed. The purpose of the HMM is to handle the temporal dependencies of changing trajectory points through time series modeling, ensuring that the dynamic changes of trajectory points are fully captured. Using this model, the initialization parameters of the trajectory points are calculated, and initialization and iterative processing are performed based on the observation sequence and distortion threshold as input to complete parameter estimation, thereby optimizing the trajectory point calculation.

[0048] The introduction of the Welch algorithm enables the adjustment of model parameters by maximizing the likelihood function of the data, ensuring optimal matching between the output trajectory points and the actual solar trajectory. This algorithm can address uncertainties and dynamic changes in the solar trajectory, optimize power generation prediction results, and analyze the key point set using the Viterbi algorithm to obtain the most probable state of the key points. Maximum likelihood estimation is then used to recalculate the state parameters of the key points based on their state. Maximum likelihood estimation, by maximizing the likelihood function of the parameters, estimates the most likely state parameters based on existing data, further improving the accuracy of trajectory point identification. Combined with Gaussian Mixture Regression, the trajectory is reconstructed, and the solar trajectory points are output, ensuring accurate capture and analysis of the solar trajectory. This results in more accurate predictions of solar power generation and provides reliable data support for practical applications, further optimizing the operation and management of photovoltaic modules.

[0049] Step S4: Collect the terrain conditions in the area to be installed, and use dynamic simulation technology to analyze the light tracking area of ​​the photovoltaic module under the condition of the sun's orbital point, and determine the optimal installation angle of the photovoltaic module.

[0050] In one embodiment, during the process of collecting terrain conditions within the installation area and analyzing the light-tracking area of ​​the photovoltaic module under the conditions of the solar trajectory point using dynamic simulation technology to determine the optimal installation angle of the photovoltaic module, the terrain conditions within the installation area can be collected based on remote sensing technology. The terrain conditions are then combined with the geometric parameters of the photovoltaic module to establish a three-dimensional state model of the relative position of the photovoltaic module and the terrain conditions. The three-dimensional state model is then run using dynamic simulation technology to analyze the effective light-receiving area of ​​the photovoltaic module at different installation angles under the conditions of the solar trajectory point, thus determining the light-tracking area. Based on the light-tracking area and a multi-dimensional optimization algorithm, the optimal incident angle of sunlight under the conditions of the solar trajectory point is searched. Finally, the optimal installation angle of the photovoltaic module is determined based on the optimal incident angle of sunlight and boundary conditions.

[0051] Specifically, in the process of searching for the optimal incident angle of sunlight based on the light-tracking area and the multidimensional optimization algorithm, and determining the optimal installation angle of the photovoltaic module based on the optimal incident angle and boundary conditions, different installation angles and corresponding light-tracking areas can be obtained. A search map of the incident angle of sunlight can be generated using triangulation technology, and the tilt angle limit and azimuth angle limit of the photovoltaic module are selected as the influencing indicators of the incident angle of sunlight based on the allocation theory. Based on the search map of the incident angle of sunlight and the influencing indicators of the incident angle of sunlight, a quantitative model of search capability is established to analyze the changing law of the influencing indicators of the incident angle of sunlight under the changing conditions of the incident angle of sunlight. Based on the changing law, a normal vector is constructed to determine the effective illumination projection coefficient, and the optimal incident angle of sunlight is searched in combination with the multidimensional optimization algorithm. The optimal installation angle of the photovoltaic module is determined based on the search results.

[0052] The process involves several steps: First, a normal vector is constructed based on the changing patterns to determine the effective illumination projection coefficient. Then, a multi-dimensional optimization algorithm is used to search for the optimal illumination incident angle. Based on the search results, the optimal installation angle for the photovoltaic module is determined. This includes: constructing a normal vector for the photovoltaic module that includes tilt and azimuth constraints based on the changing patterns; analyzing the angle between the normal vector and the direction of sunlight at any given time to determine the effective illumination projection coefficient; combining the effective illumination projection coefficient with the light tracking area at the corresponding installation angle to construct an illumination objective function; and using collaborative swarm intelligence technology to search for the optimal illumination incident angle under the condition of the sun's orbital trajectory. Finally, the optimal illumination incident angle is used as a benchmark to initially determine the optimal installation angle for the photovoltaic module, and boundary conditions are used as influencing factors to verify the feasibility of the initial determination and optimize the installation angle.

[0053] It should be explained that in determining the optimal installation angle of photovoltaic modules, remote sensing technology is used to collect the terrain conditions of the area to be installed, obtaining spatial information including surface elevation, slope, and orientation. This spatial information forms the basic background of the photovoltaic module deployment environment. The terrain data is combined with the geometric parameters of the photovoltaic modules (such as length, width, height, and tilt angle range) to establish a three-dimensional state model that truly reflects the relative relationship between the modules and the terrain. This provides a dynamic and interactive simulation platform that can realistically present the geometric adaptation relationship between the modules and the terrain environment under different postures, ensuring that the analysis results have spatial logical consistency. Dynamic simulation technology is used to run the three-dimensional model and calculate the effective light-receiving area of ​​the photovoltaic modules at different installation angles under different solar trajectory points. Based on this, the light tracking area at each moment is determined, which helps to identify the loss of light-receiving efficiency caused by terrain shading or shadow changes, thereby eliminating inefficient deployment schemes.

[0054] Based on the collected light-tracking areas corresponding to different installation angles, a light incident angle search map is generated using triangulation technology. This is a key step in mapping the light-receiving area and installation angle into a continuous searchable surface in two-dimensional space. To quantify the impact of each angle combination on illumination efficiency, tilt and azimuth constraints are selected as constraint indicators based on the distribution theorem, so that the model can adapt to the mechanical installation conditions and design specifications of the site. Based on the search map and influence indicators, a search capability quantification model is constructed. This model is used to analyze the variation law of the light-receiving area under different light incident angles, thereby predicting the actual light-receiving change trend caused by angle changes.

[0055] Based on the above principles, a normal vector is constructed to represent the orientation of the module surface. The angle between this normal vector and the solar incidence direction at any given time is calculated to determine the effective illumination projection coefficient, which reflects the vertical component of the sunlight on the photovoltaic surface and serves as a direct input indicator for estimating power generation. This coefficient is combined with the light tracking area to construct the illumination objective function. A collaborative swarm intelligence algorithm (such as particle swarm optimization or ant colony optimization) is introduced to search globally for the optimal illumination incident angle that maximizes the objective function. The optimal illumination incident angle obtained from the search is used as the basis for preliminary judgment. Combined with the boundary conditions of the actual site (such as the upper limit of the support angle, terrain feasibility, and safety requirements), the feasibility of the preliminary results is verified and the parameters are fine-tuned. Finally, the optimal installation angle of the photovoltaic module is output. This installation angle can maximize the illumination projection area under given terrain and solar trajectory conditions, thereby improving the power generation efficiency of the entire photovoltaic system. It realizes dynamic response and global search under terrain constraints, ultimately providing an interpretable, achievable, and high-power-generating-potential optimal solution for photovoltaic design.

[0056] Step S5: Obtain the solar angle height and azimuth angle of the target time period in the area to be installed, construct a power output model of the photovoltaic module under the premise of the optimal installation angle, and obtain the predicted results of solar power generation.

[0057] In one embodiment, during the process of obtaining the solar angular height and azimuth angle of the target time period in the area to be installed, constructing a power output model under the premise of the optimal installation angle of the photovoltaic module, and obtaining the solar power generation prediction result, the solar angular height and azimuth angle data at each time point in the target time period of the area to be installed can be obtained. Based on the solar angular height and azimuth angle data, the solar trajectory can be analyzed to construct an irradiance receiving model. Based on the irradiance receiving model, the effective irradiance value of the sun at each time point can be analyzed, the total incident irradiance projected by the sun onto the photovoltaic module can be analyzed, and a temperature coefficient can be introduced to correct the conversion efficiency of the photovoltaic module at each time point. Based on the conversion efficiency of the photovoltaic module and the optimal installation angle of the photovoltaic module, a power output model can be constructed, the instantaneous power generation curve can be calculated, and the solar power generation prediction result can be obtained by integrating the power generation within the target time period.

[0058] The expression for the power generation output model is as follows: ; In the formula, P ( t ) indicates in t The output power of photovoltaic modules at all times G POA ( t ) indicates in t The total incident irradiance on the surface of the photovoltaic module at any given time. η ref This indicates the conversion efficiency of the photovoltaic module. α Indicates the temperature coefficient. T amb ( t ) indicates in t The ambient temperature at any given time T NOCT Indicates the operating cell temperature of the photovoltaic module. T ref Indicates standard temperature. Cos n ( β ( t )) indicates in t The incident angle correction factor at time 10:00. n Indicates an index. δ Indicates the effective area loss rate of photovoltaic modules. A shadow / A total This represents the ratio between the current shaded area and the total area of ​​the photovoltaic module.

[0059] It should be explained that in obtaining the predicted solar power generation results, it is necessary to obtain the solar altitude angle (i.e., solar elevation angle) and azimuth angle at each time point within the target period. The sampling interval is usually every 5 to 15 minutes to ensure the continuity of the trajectory. These angular data form the basis of the sun's trajectory in the sky and are used for subsequent irradiance modeling. Based on the solar altitude angle and azimuth angle, an irradiance receiving model is constructed. This model comprehensively considers direct solar irradiance, diffuse irradiance, and surface reflected irradiance. According to the relative angle between the sun and the photovoltaic module, the effective irradiance value at each time point is calculated, and the total incident irradiance on the photovoltaic module surface at each time point is further analyzed, that is, the influence of the solar projection angle on the light intensity is considered. For example, the closer the incident angle is to vertical, the higher the energy per unit area.

[0060] To enhance the model's practical adaptability, a temperature coefficient is introduced to correct the component efficiency. The geometric reception conditions of the photovoltaic module at the optimal installation angle are combined with the corrected conversion efficiency to construct a power output model, resulting in a complete instantaneous power generation curve. This curve reflects the dynamic changes in the conversion of solar energy into electrical energy within the target time period. By integrating this power curve over time, the predicted solar power generation for the target time period (typically in kWh) can be obtained. Numerical integration techniques such as the trapezoidal rule and Simpson's rule can be used to ensure calculation accuracy. This prediction result can be used in scenarios such as grid dispatching, economic assessment, and energy storage planning, and is a core indicator in the design and operation management of photovoltaic systems.

[0061] Please see Figure 2 The present invention also provides a solar power generation analysis system for light tracking, the system comprising: The trajectory feature point acquisition module 1 is used to use historical astronomical and time parameters within the area to be installed of the photovoltaic module as solar tracking data, and to filter out trajectory feature points related to solar motion from the solar tracking data using the maximum information coefficient. The trajectory point analysis module 2 is used to introduce time interval information with variable step size based on key features, and combine it with time series prediction technology to construct a solar motion trajectory model, capturing the changing trajectory points between adjacent trajectory feature points; The attribution status acquisition module 3 is used to identify key points of the changing trajectory points, and to construct a target trajectory calculation model based on the key point identification results, and output the solar trajectory points; The photovoltaic installation angle determination module 4 is used to collect the terrain conditions in the area to be installed, and to analyze the light tracking area of ​​the photovoltaic module under the condition of the sun's orbital point using dynamic simulation technology, so as to determine the optimal installation angle of the photovoltaic module. The solar power generation prediction module 5 is used to obtain the solar angle height and azimuth angle of the target time period in the area to be installed, construct a power output model under the premise of the optimal installation angle of the photovoltaic module, and obtain the solar power generation prediction results.

[0062] The trajectory feature point acquisition module 1, the travel trajectory point analysis module 2, the attribution status acquisition module 3, the photovoltaic installation angle determination module 4, and the solar power generation prediction module 5 are connected in sequence.

[0063] Those skilled in the art will recognize that the units and algorithm steps described in conjunction with the embodiments herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0064] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for analyzing solar power generation for light tracking, characterized in that, include: Historical astronomical and temporal parameters within the area to be installed with photovoltaic modules are used as solar tracking data. The maximum information coefficient is used to filter out trajectory feature points related to solar motion in the solar tracking data. Based on key features, time interval information with variable step size is introduced and combined with time series prediction technology to construct a solar motion trajectory model, capturing the changing trajectory points between adjacent trajectory feature points; Key points are identified for the changing trajectory points, and a target trajectory calculation model is constructed based on the key point identification results to output the solar trajectory points; The terrain conditions of the area to be installed are collected, and dynamic simulation technology is used to analyze the light tracking area of ​​the photovoltaic module under the condition of the sun's orbital point, so as to determine the optimal installation angle of the photovoltaic module. Obtain the solar angle height and azimuth angle of the target time period in the area to be installed, construct a power output model under the premise of the optimal installation angle of the photovoltaic module, and obtain the predicted results of solar power generation.

2. The method for analyzing solar power generation for light tracking according to claim 1, characterized in that, The method of using the maximum information coefficient to filter out trajectory feature points related to solar motion in solar tracking data includes: Two sets of random variable sequences, altitude and azimuth, are extracted from the solar tracking data. A scatter plot is generated using the random variable sequences, and a grid is drawn based on the scatter plot. The mutual information of the grid is used to characterize the state of the random variable sequence, and the maximum value of the mutual information is selected as the maximum information coefficient. The maximum information coefficient is then used to analyze the random variable sequence. Based on the analysis results, the importance ranking of angle change features is obtained, and the average value of the maximum information coefficient of the angle change features is used as the screening threshold to screen out angle change features with a maximum information coefficient greater than the threshold. The screening results are used as key features of trajectory changes, and these key features are reconstructed to select trajectory feature points related to solar motion.

3. The method for analyzing solar power generation for light tracking according to claim 2, characterized in that, The process of identifying key points in the changing trajectory and constructing a target trajectory calculation model based on the key point identification results, outputting the solar trajectory points, includes: The instantaneous velocity of the trajectory is determined based on the time difference between any two trajectory points within the variable travel trajectory points. Feature vectors are extracted from the variable travel trajectory points to generate a vector form that combines normalized velocity and position as the feature vector of candidate key points. Cluster centers are set based on the mean of the feature vectors of candidate key points, and the total distortion is calculated. Based on the calculation results, splitting and initialization of total distortion are performed, and the relative distortion improvement is determined. The difference between the relative distortion improvement and the distortion threshold is compared, and the attribution status of the feature vectors of the cluster center and the candidate key point is output when the relative distortion improvement is greater than the threshold. The optimal trajectory is calculated based on the cluster center using the minimum distortion criterion. The target trajectory calculation model is constructed by combining the Welch algorithm to screen key points and calculate the probability attribution state to output the solar trajectory points.

4. The method for analyzing solar power generation for light tracking according to claim 3, characterized in that, The process of calculating the optimal trajectory based on the cluster center using the minimum distortion criterion, constructing a target trajectory calculation model using the Welch algorithm to screen key points, and calculating the probability attribution state to output the solar trajectory points includes: The minimum distortion criterion is used to randomly select points on the changing travel trajectory, and the corresponding cluster centers are determined. The Linde evolutionary Buzogre algorithm is then executed to obtain candidate key points generated by the corresponding changing travel trajectory points. A hidden Markov chain is constructed with cluster centers as the observation sequence and candidate key points as the hidden states. The initialization parameters of the hidden Markov model are calculated based on the hidden states and observation values ​​of the trajectory points in the changing travel trajectory points. The observation sequence and distortion threshold are used as inputs to the Welch algorithm. Initialization and iterative processing are performed to complete parameter estimation and output the parameter information of the Hidden Markov Model, which is used as the target trajectory calculation model. The target trajectory calculation model is used to filter out the set of key points corresponding to the changing travel trajectory points, and the Viterbi algorithm is used to analyze the set of key points to obtain the ownership status of the key points with the highest probability. The maximum likelihood estimation technique is used to re-obtain the state parameters of the key points of the assigned state. The state parameter calculation results are combined with Gaussian mixture regression to obtain the reconstructed trajectory and output the solar orbit trajectory points.

5. The method for analyzing solar power generation for light tracking according to claim 1, characterized in that, The process of collecting terrain conditions within the installation area and using dynamic simulation technology to analyze the light tracking area of ​​the photovoltaic modules under the conditions of the sun's orbital path, and determining the optimal installation angle of the photovoltaic modules, includes: The terrain conditions in the area to be installed are collected using remote sensing technology, and the terrain conditions are combined with the geometric parameters of the photovoltaic modules to establish a three-dimensional state model of the relative position of the photovoltaic modules and the terrain conditions. By using dynamic simulation technology to run a three-dimensional state model, the effective light-receiving area of ​​the photovoltaic module at different installation angles under the condition of the sun's orbit point is analyzed, and the light tracking area is determined. Based on the optimal incident angle of sunlight under the condition of searching the solar trajectory point using the light tracking area and multidimensional optimization algorithm, the optimal installation angle of photovoltaic modules is determined based on the optimal incident angle of sunlight and boundary conditions.

6. The method for analyzing solar power generation for light tracking according to claim 5, characterized in that, The process of determining the optimal installation angle of the photovoltaic module based on the optimal incident angle of sunlight under the condition of searching the solar trajectory point using the light tracking area and a multi-dimensional optimization algorithm, and based on the optimal incident angle of sunlight and boundary conditions, includes: Obtain different installation angles and corresponding light tracking areas, generate a light incident angle search map using triangulation technology, and select the photovoltaic module tilt angle limit and azimuth angle limit as the light incident angle influence indicators based on the allocation theory. Based on the search map of the incident angle of light and the index of the influence of the incident angle of light, a quantitative model of search capability is established to analyze the changing law of the index of the influence of the incident angle of light under the condition of changing incident angle of light. Based on the variation pattern, a normal vector is constructed to determine the effective illumination projection coefficient. Then, a multi-dimensional optimization algorithm is used to search for the optimal illumination incident angle. Based on the search results, the optimal installation angle of the photovoltaic module is determined.

7. The method for analyzing solar power generation for light tracking according to claim 6, characterized in that, The process of constructing a normal vector based on the variation law to determine the effective illumination projection coefficient, and combining it with a multi-dimensional optimization algorithm to search for the optimal illumination incident angle, and determining the optimal installation angle of the photovoltaic module based on the search results, includes: Based on the changing patterns, a normal vector containing tilt and azimuth constraints is constructed for the photovoltaic module. The angle between the normal vector and the direction of sunlight at any time is analyzed to determine the effective light projection coefficient. The effective illumination projection coefficient is combined with the light tracking area at the corresponding installation angle to construct the illumination objective function, and the optimal illumination incident angle is searched under the condition of the sun's orbital trajectory point using collaborative swarm intelligence technology. The optimal installation angle of photovoltaic modules was initially determined by using the optimal incident angle of sunlight as a benchmark, and the feasibility of the initial determination was verified by using boundary conditions as influencing factors, thereby optimizing the installation angle.

8. The method for analyzing solar power generation for light tracking according to claim 1, characterized in that, The process of obtaining the solar radiation angle height and motion azimuth angle of the target time period in the area to be installed, constructing a power output model under the premise of the optimal installation angle of the photovoltaic module, and obtaining the solar power generation prediction results includes: Acquire solar radiation angle height and azimuth data at various time points within the target time period of the area to be installed, and analyze the solar trajectory based on the solar radiation angle height and azimuth data to construct an irradiance receiving model; Based on the irradiance receiving model, the effective irradiance of the sun at each time point is analyzed, the total incident irradiance projected onto the photovoltaic module is analyzed, and a temperature coefficient is introduced to correct the conversion efficiency of the photovoltaic module at each time point. A power output model is constructed based on the photovoltaic module conversion efficiency and the optimal installation angle of the photovoltaic module. The instantaneous power output curve is calculated, and the solar power generation prediction result is obtained by integrating the power output within the target time period.

9. A method for analyzing solar power generation for light tracking according to claim 8, characterized in that, The expression for the power generation output model is: ； In the formula, P ( t ) indicates in t The output power of photovoltaic modules at all times G POA ( t ) indicates in t The total incident irradiance on the surface of the photovoltaic module at any given time. η ref This indicates the conversion efficiency of the photovoltaic module. α Indicates the temperature coefficient. T amb ( t ) indicates in t The ambient temperature at any given time T NOCT Indicates the operating cell temperature of the photovoltaic module. T ref Indicates standard temperature. Cos n ( β ( t )) indicates in t The incident angle correction factor at time 10:

00. n Indicates an index. δ Indicates the effective area loss rate of photovoltaic modules. A shadow / A total This represents the ratio between the current shaded area and the total area of ​​the photovoltaic module.

10. A solar power generation analysis system for light tracking, used to implement the solar power generation analysis method for light tracking as described in any one of claims 1-9, characterized in that, The system includes: The trajectory feature point acquisition module is used to use historical astronomical and temporal parameters within the area to be installed of the photovoltaic module as solar tracking data, and to filter out trajectory feature points related to solar motion from the solar tracking data using the maximum information coefficient; The trajectory point analysis module is used to introduce time interval information with variable step size based on key features, and combine it with time series prediction technology to construct a solar motion trajectory model, capturing the changing trajectory points between adjacent trajectory feature points; The attribution status acquisition module is used to identify key points of the changing trajectory points, construct the target trajectory calculation model based on the key point identification results, and output the solar trajectory points. The photovoltaic installation angle determination module is used to collect the terrain conditions in the area to be installed, and to analyze the light tracking area of ​​the photovoltaic module under the condition of the sun's orbital point using dynamic simulation technology, so as to determine the optimal installation angle of the photovoltaic module. The solar power generation prediction module is used to obtain the solar angle height and azimuth angle of the target time period in the area to be installed, construct a power output model under the premise of the optimal installation angle of the photovoltaic module, and obtain the solar power generation prediction results.

Images