A method and apparatus for optimizing the tilt angle of photovoltaic panels based on data fusion analysis

Abstract

This application provides a photovoltaic (PV) panel tilt angle optimization method and apparatus based on data fusion analysis. It assesses the power generation revenue and risk losses of a PV power station when the target PV panel is at its optimal tilt angle at different time intervals, based on the optimal tilt angle and strong wind forecast results at different time intervals, thereby ensuring the stability and safety of the system under harsh weather conditions. Then, based on the power generation revenue and risk losses of the PV power station when the target PV panel is at its optimal tilt angle at different time intervals, and the actual situation of the PV power station, it calculates the optimal tilt angle adjustment strategy that maximizes the overall revenue and minimizes the risk of the PV power station. This application, by dynamically adjusting the PV panel tilt angle, not only optimizes energy harvesting efficiency but also significantly enhances the structure's wind resistance, achieving dual optimization of energy and safety, and contributing to the development and application of sustainable energy technologies.

Description

Technical Field

[0001] This application relates to the field of photovoltaic power generation technology, and in particular to a method and apparatus for optimizing the tilt angle of photovoltaic panels based on data fusion analysis. Background Technology

[0002] Building a data-fusion-based low-voltage distributed photovoltaic (PV) situational awareness and decision-making system presents a series of complex technical challenges. To maximize solar energy capture efficiency, the angle of the solar panels must be adjusted in real time to adapt to changes in the sun's position. This requires precise monitoring of the sun's azimuth and altitude angles. However, the optimal tilt angle of the solar panels is not fixed but varies with the sun's position throughout the day. Furthermore, the angle adjustment of the solar panels must also take into account potential environmental risks, such as strong winds and other natural phenomena, which could threaten the structural safety of the PV panels. Summary of the Invention

[0003] The purpose of this application is to at least address one of the aforementioned technical deficiencies, particularly the lack of a method for adjusting the tilt angle of photovoltaic panels in the prior art that can minimize the risks of photovoltaic power plants while maximizing overall benefits.

[0004] This application provides a method for optimizing the tilt angle of photovoltaic panels based on data fusion analysis, the method comprising:

[0005] The solar trajectory and specification data of the target photovoltaic panel are obtained within a preset future time period. Based on the solar trajectory and specification data, the optimal tilt angle of the target photovoltaic panel at different time intervals within the preset future time period is determined.

[0006] The strong wind conditions in the area where the target photovoltaic panel is located are predicted in the preset future time period by using a pre-configured target strong wind prediction model, and the strong wind prediction results are obtained.

[0007] Based on the optimal tilt angle at different time intervals and the strong wind prediction results, evaluate the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals;

[0008] Based on the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, and the actual situation of the photovoltaic power station, calculate the optimal tilt angle adjustment strategy that maximizes the overall revenue and minimizes the risk of the photovoltaic power station.

[0009] Optionally, obtaining the solar trajectory within a preset future time period includes:

[0010] Obtain the pre-configured solar trajectory prediction model and the currently collected solar azimuth and solar altitude angles;

[0011] Calculate the current sun position data based on the solar azimuth angle and the solar altitude angle;

[0012] The current solar position data is input into the solar trajectory prediction model to obtain the solar trajectory within a preset future time period output by the solar trajectory prediction model.

[0013] Optionally, the training process of the target strong wind prediction model includes:

[0014] Hourly meteorological observation data of the area where the target photovoltaic panel is located within a preset historical period are obtained from the meteorological department, and the meteorological observation data is preprocessed to obtain a strong wind historical dataset.

[0015] The preset initial strong wind prediction model is trained using the historical strong wind dataset, and when the preset training stopping condition is met, the trained initial strong wind prediction model is used as the target strong wind prediction model.

[0016] Optionally, training the preset initial strong wind prediction model using the historical strong wind dataset includes:

[0017] The correlation between different types of data in the historical strong wind dataset is analyzed by correlation analysis method, and wind speed, wind direction and air pressure gradient features are selected from the historical strong wind dataset as training data based on the analysis results.

[0018] The training data is used to train a preset initial strong wind prediction model.

[0019] Optionally, the step of evaluating the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals based on the optimal tilt angle at different time intervals and the strong wind prediction results includes:

[0020] Establish a three-dimensional geometric model corresponding to the target photovoltaic panel;

[0021] Based on the three-dimensional geometric model and the strong wind prediction results, the structural stress distribution of the target photovoltaic panel at the optimal tilt angle under different time intervals is calculated, and the structural stability of the target photovoltaic panel is evaluated based on the structural stress distribution.

[0022] The Sandia model was used to calculate the theoretical power generation of the target photovoltaic panel at the optimal tilt angle under different time intervals, and the energy capture efficiency of the target photovoltaic panel was evaluated based on the theoretical power generation.

[0023] A comprehensive evaluation of the structural stability and energy capture efficiency of the target photovoltaic panel is conducted to determine the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals.

[0024] Optionally, the calculation of the optimal tilt angle adjustment strategy that maximizes the overall benefit and minimizes the risk of the photovoltaic power station based on the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, and the actual situation of the photovoltaic power station, includes:

[0025] Based on the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, high-risk photovoltaic panels whose strong wind loss is greater than their power generation revenue are identified.

[0026] Based on the actual situation of the photovoltaic power station, a photovoltaic array layout optimization model is constructed with minimizing the number of high-risk photovoltaic panels as the optimization objective and the spacing and arrangement of photovoltaic panels as optimization variables.

[0027] The high-risk photovoltaic panels are optimized using the photovoltaic array layout optimization model to obtain the optimal tilt angle adjustment strategy that maximizes the overall benefits and minimizes the risks of the photovoltaic power station.

[0028] Optionally, the method further includes:

[0029] Based on the optimal tilt angle of the target photovoltaic panel at different time intervals, analyze the changes in revenue and risk of adjusting the target photovoltaic panel to the optimal tilt angle at different time intervals;

[0030] The optimal tilt angle adjustment timeliness strategy is determined based on the changes in revenue and risk associated with adjusting the target photovoltaic panel to the optimal tilt angle at different time intervals.

[0031] This application also provides a photovoltaic panel tilt angle optimization device based on data fusion analysis, including:

[0032] The optimal tilt angle prediction module is used to acquire the solar motion trajectory and the specification data of the target photovoltaic panel within a preset future time period, and to determine the optimal tilt angle of the target photovoltaic panel at different time intervals within the preset future time period based on the solar motion trajectory and the specification data.

[0033] The strong wind prediction module is used to predict the strong wind conditions in the area where the target photovoltaic panel is located during the preset future time period using a pre-configured target strong wind prediction model, and obtain the strong wind prediction result.

[0034] The revenue and risk assessment module is used to assess the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, based on the optimal tilt angle at different time intervals and the strong wind prediction results.

[0035] The optimal tilt angle determination module is used to calculate the optimal tilt angle adjustment strategy that maximizes the overall benefits and minimizes the risks of the photovoltaic power station based on the power generation revenue and risk losses of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, as well as the actual situation of the photovoltaic power station.

[0036] This application also provides a computer-readable storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of a photovoltaic panel tilt angle optimization method based on data fusion analysis as described in any of the above embodiments.

[0037] This application also provides a computer device, including: one or more processors, and memory;

[0038] The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of a photovoltaic panel tilt angle optimization method based on data fusion analysis as described in any of the above embodiments.

[0039] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0040] This application provides a photovoltaic (PV) panel tilt angle optimization method and apparatus based on data fusion analysis. After obtaining the solar trajectory and target PV panel specifications for a preset future time period, the optimal tilt angle of the target PV panel at different time intervals within that preset future time period can be determined based on the solar trajectory and specifications, thereby improving solar energy capture efficiency. Next, this application can predict the strong wind conditions in the target PV panel's location within the preset future time period using a pre-configured target strong wind prediction model, obtaining strong wind prediction results. Then, based on the optimal tilt angle and strong wind prediction results at different time intervals, the power generation revenue and risk losses of the PV power station when the target PV panel is at the optimal tilt angle at different time intervals are evaluated, thus ensuring the stability and safety of the system under severe weather conditions. Finally, based on the power generation revenue and risk losses of the PV power station when the target PV panel is at the optimal tilt angle at different time intervals, and the actual situation of the PV power station, this application can calculate the optimal tilt angle adjustment strategy that maximizes the overall revenue and minimizes the risk of the PV power station. By dynamically adjusting the PV panel tilt angle, this application not only optimizes energy harvesting efficiency but also significantly enhances the structure's wind resistance, achieving dual optimization of energy and safety. Furthermore, the implementation of this application provides an intelligent and automated solution for the photovoltaic industry, which helps to promote the development and application of sustainable energy technologies. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 A flowchart illustrating a photovoltaic panel tilt angle optimization method based on data fusion analysis provided in this application embodiment;

[0043] Figure 2 A schematic diagram of a photovoltaic panel tilt angle optimization device based on data fusion analysis is provided in an embodiment of this application;

[0044] Figure 3 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0045] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0046] In one embodiment, such as Figure 1 As shown, Figure 1 This application provides a flowchart illustrating a method for optimizing the tilt angle of a photovoltaic panel, as illustrated in an embodiment of the present application. The present application also provides a method for optimizing the tilt angle of a photovoltaic panel based on data fusion analysis, which may include:

[0047] S110: Obtain the solar trajectory and the specification data of the target photovoltaic panel within a preset future time period, and determine the optimal tilt angle of the target photovoltaic panel at different time intervals within the preset future time period based on the solar trajectory and specification data.

[0048] In this step, to improve solar energy capture efficiency, the solar trajectory within a preset future time period can be obtained first. Then, combined with the specification data of the target photovoltaic panel, the optimal tilt angle of the target photovoltaic panel at different time intervals within the preset future time period can be calculated. This allows the generation of angle adjustment commands for the target photovoltaic panel. The target photovoltaic panel in this application can be a single panel or multiple panels; the specific number of panels can be set according to actual conditions and is not limited here.

[0049] Specifically, this application can first predict the solar trajectory over a future period of time, and then obtain the specification data of the target photovoltaic panel, including size, material, conversion efficiency, etc. Based on the duration of the solar trajectory, the optimal angle for the target photovoltaic panel to receive solar radiation in each time period is calculated in units of a certain time interval, forming a photovoltaic panel tilt angle data sequence.

[0050] This application analyzes solar radiation intensity sensor data to obtain the variation trend of solar radiation intensity over different time periods. This data is then combined with photovoltaic panel tilt angle data sequences. Using heuristic algorithms, including genetic algorithms or particle swarm optimization algorithms, a target function that maximizes power generation is set to optimize the photovoltaic panel angle adjustment strategy. Based on the optimized photovoltaic panel angle adjustment strategy, corresponding angle adjustment commands are generated and transmitted to the central processing unit to achieve real-time adjustment of the photovoltaic panel angle.

[0051] Furthermore, this application can continuously monitor the photovoltaic panel's power generation efficiency, temperature, and voltage parameters during the photovoltaic panel angle adjustment process. It uses algorithms to establish a performance prediction model for the photovoltaic panel, including models using multiple linear regression or principal component analysis (PCA). When actual data deviates from the model's predicted values, a fault is identified. Based on the specific parameters of the deviation, the fault type is inferred, and corresponding maintenance instructions are generated, including cleaning and component replacement. The photovoltaic panel's power generation data and angle adjustment data are aggregated and uploaded to a central processing unit. Machine learning algorithms, including decision trees and random forests, are used to continuously optimize the photovoltaic panel angle adjustment strategy and fault diagnosis model.

[0052] S120: Predict the strong wind conditions in the area where the target photovoltaic panel is located within a preset future time period using a pre-configured target strong wind prediction model, and obtain the strong wind prediction results.

[0053] In this step, after determining the optimal tilt angle of the target photovoltaic panel at different time intervals within a preset future period based on the solar trajectory and specification data in S110, this application can predict the strong wind conditions in the area where the target photovoltaic panel is located within a preset future period using a pre-configured target strong wind prediction model, and obtain the strong wind prediction results.

[0054] In detail, this application can obtain historical climate data for the target area, i.e., the area where the target photovoltaic panels are located, from meteorological departments within a preset historical period. The focus is on collecting data related to wind speed, wind direction, and strong winds. This data is then cleaned and formatted to construct a historical database of strong winds. This process can be achieved by deploying wind speed, wind direction, and air pressure environmental sensors at the photovoltaic power station site. These sensors collect current environmental parameters in real time and transmit the data to a central processing unit via a wireless communication network, forming a real-time environmental information data stream.

[0055] Next, this application inputs historical climate data and real-time environmental information data into a big data analysis platform, and uses the Apriori association rule mining algorithm and time series pattern mining algorithm to analyze the strong wind data. Specifically, the Apriori algorithm discovers association rules for strong wind occurrence by calculating the support and confidence of different combinations of meteorological parameters; the time series pattern mining algorithm discovers precursor patterns of strong wind occurrence by analyzing the time series changes of parameters such as wind speed and wind. Support Vector Machine (SVM) or Random Forest machine learning models are applied to establish a strong wind prediction model using wind speed, wind direction, and air pressure parameters, and the model performance is optimized through feature engineering and cross-validation methods to improve the accuracy of risk prediction. The optimized strong wind prediction model is used to predict strong wind conditions in the future, obtaining the probability distribution, intensity distribution, and wind direction distribution risk indicators of strong wind occurrence, and generating a strong wind occurrence risk report.

[0056] Based on the strong wind risk report and combined with the geographical location and photovoltaic panel layout information of the photovoltaic power station, this application uses the finite element analysis simulation method to simulate and calculate the stress and deformation of the photovoltaic panels under different angles and wind speeds, identify weak points, and assess the impact of strong winds on the photovoltaic panels to optimize the mechanical design of the photovoltaic panels.

[0057] For example, the following wind protection measures can be implemented for photovoltaic (PV) panels affected by strong winds: adjust PV panels equipped with angle adjustment mechanisms to the position with the minimum wind-exposed area in advance and lock the adjustment mechanisms; spray a hydrophobic and dust-proof coating on the surface of the PV panels to reduce dust and foreign object adsorption and lower wind resistance; set up windbreaks around the PV array to slow down the incoming wind speed; optimize the arrangement of PV panels, including using staggered arrangements to reduce the wake effect; activate the wind-resistant vibration mechanism of the PV panels, including increasing damping and using vibration damping devices, to reduce the destructive force of strong winds on the PV panels. Correlate strong wind risk reports with the power generation data of the PV power plant to assess the impact of strong winds on PV power generation efficiency. Utilize machine learning algorithms, including support vector regression and neural networks, to establish a PV power generation prediction model that incorporates strong wind factors. Through training with historical data and parameter optimization, improve the accuracy of power generation prediction and provide decision support for power plant energy management and grid dispatch.

[0058] S130: Based on the optimal tilt angle at different time intervals and the strong wind prediction results, evaluate the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals.

[0059] In this step, after obtaining the strong wind prediction result through S120, this application can evaluate the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals based on the optimal tilt angle at different time intervals and the strong wind prediction result.

[0060] Specifically, this application can obtain hourly wind speed and direction data for the next 24 hours from the target strong wind prediction model, and at the same time obtain the current optimal tilt angle data of the photovoltaic panel (such as 30°) as input to the risk assessment model. The risk assessment model can be constructed using a long short-term memory neural network (LSTM), and the risk level and the gains and losses of the photovoltaic panel with respect to the current optimal tilt angle data are used as the output of the risk assessment model.

[0061] Next, for photovoltaic panels where the loss from strong winds outweighs the power generation revenue, this application can re-optimize the layout of the photovoltaic panel array, adjust the spacing and arrangement between the photovoltaic panels to reduce the impact of strong winds on the photovoltaic panels, and re-analyze the revenue to calculate the number of photovoltaic panels that can continue to generate electricity under the influence of strong winds; if the re-analyzed revenue is higher than the loss from strong winds, then power generation can be carried out on the photovoltaic panel array after the re-optimization.

[0062] S140: Based on the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, and the actual situation of the photovoltaic power station, calculate the optimal tilt angle adjustment strategy that maximizes the comprehensive revenue and minimizes the risk of the photovoltaic power station.

[0063] In this step, after evaluating the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals through S130, this application can also calculate the optimal tilt angle adjustment strategy that maximizes the overall revenue and minimizes the risk of the photovoltaic power station based on the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, and the actual situation of the photovoltaic power station.

[0064] Specifically, this application addresses photovoltaic panels where strong wind losses outweigh power generation revenue by re-layouting and optimizing the photovoltaic panel array. This involves adjusting the spacing and arrangement of the photovoltaic panels to reduce the impact of strong winds on the panels, re-analyzing the revenue, and calculating the number of photovoltaic panels that can continue to generate electricity under the influence of strong winds. If the re-analyzed revenue is higher than the strong wind losses, then power generation is carried out on the re-layout optimized photovoltaic panel array.

[0065] For example, this application can obtain loss data and power generation revenue data of each photovoltaic panel under strong winds from a risk assessment model. Based on the ratio of loss to revenue, high-risk photovoltaic panels whose losses exceed power generation revenue under strong winds are identified and marked as photovoltaic panels to be optimized. Then, this application can use a genetic algorithm to optimize the layout of the photovoltaic array, with minimizing the number of high-risk photovoltaic panels as the optimization objective, and the spacing and arrangement of photovoltaic panels as optimization variables, to construct a photovoltaic array layout optimization model. This photovoltaic array layout optimization model is then used to calculate the optimal tilt angle adjustment strategy that maximizes the overall revenue and minimizes the risk of the photovoltaic power station, and the photovoltaic panel angle is adjusted using this optimal tilt angle adjustment strategy.

[0066] Furthermore, this application can send the optimal tilt angle adjustment strategy to the angle control device in the form of an instruction. After receiving the photovoltaic panel angle adjustment instruction from the central processing unit, the angle control device can determine the execution strategy based on the target angle value, adjustment speed, and adjustment accuracy parameters, since the photovoltaic panel angle adjustment instruction contains the target angle value, adjustment speed, and adjustment accuracy parameters. Next, this application can use a servo motor or stepper motor as the actuator for angle adjustment. By controlling the motor's speed and rotation angle, the photovoltaic panel angle is adjusted. During the angle adjustment process, the motor's speed and torque parameters are monitored in real time, and the motor's control signal is dynamically adjusted through a PID control algorithm. The proportional, integral, and derivative coefficients of the PID controller are set according to the angle deviation and adjustment speed requirements. The proportional coefficient determines the controller's response speed, the integral coefficient affects the steady-state accuracy of the motor's control system, and the derivative coefficient is used to suppress oscillations and overshoot of the motor's control system. In the design of the transmission mechanism, high-strength, lightweight materials, including aerospace aluminum alloy and carbon fiber, are used to reduce the inertia and friction of the transmission components and improve the flexibility and efficiency of angle adjustment. Meanwhile, by optimizing the transmission structure, including adopting planetary gear transmission to reduce backlash and increase the transmission ratio, transmission efficiency is improved and energy consumption is reduced. The angle control device incorporates an angle sensor to collect the actual angle of the photovoltaic panel in real time, comparing it with the target angle to generate an angle deviation signal. Through a feedback control mechanism, the angle deviation signal is input to a PID controller to calculate the motor control signal, driving the motor to adjust the photovoltaic panel angle until the actual angle matches the target angle. This closed-loop control method continuously tracks changes in the sun's position, dynamically adjusting the photovoltaic panel angle to ensure it always maintains the optimal light-receiving angle, maximizing solar energy capture efficiency. During the operation of the angle control device, real-time detection and diagnosis of abnormalities during angle adjustment are achieved by monitoring motor current and mechanical vibration parameters. In the event of a fault, a preset emergency handling strategy, including switching to a backup actuator and adjusting control parameters, ensures the continuity and reliability of the angle adjustment function. Simultaneously, fault information is reported to the central processing unit for remote maintenance and fault analysis.

[0067] In one specific implementation, after receiving an angle adjustment command with a target angle value of 30 degrees, an adjustment speed of 5 degrees / second, and an adjustment accuracy of 0.1 degrees, the angle control device can extract these parameters through the command parsing module. Based on a preset strategy table, it determines that a servo motor will be used as the actuator, setting the angle adjustment range to 0-60 degrees and the speed limit to within 10 degrees / second. The controller calculates an angle deviation of 25 degrees based on the target and current angle values. Using a PID control algorithm with proportional coefficient Kp = 2.5, integral coefficient Ki = 0.8, and derivative coefficient Kd = 0.2, it calculates the control signal for the servo motor, driving it to adjust the angle at a speed of 8 degrees / second. During the adjustment process, the angle sensor collects actual angle data every 0.1 seconds. Noise is reduced using a moving average filtering algorithm, and the filtered angle value is compared with the target angle to obtain the real-time deviation, which serves as the feedback input to the PID controller. Simultaneously, the current sensor monitors the servo motor's operating current. When the current value exceeds 1.2 times the rated current, it is determined that the motor is overloaded, triggering the overcurrent protection mechanism, cutting off the motor power supply, and reporting the fault information to the central processing unit. After 25 seconds of continuous adjustment, the actual angle reaches 29.9 degrees, with a deviation from the target angle of less than 0.1 degrees. The controller issues a completion command, the servo motor stops rotating, and the angle adjustment process ends. Data from this adjustment process is recorded, including the initial angle, target angle, adjustment time, average speed, and peak current, as a basis for optimizing control strategies and predicting faults.

[0068] In the above embodiments, after obtaining the solar trajectory and the specification data of the target photovoltaic panel within a preset future time period, the optimal tilt angle of the target photovoltaic panel at different time intervals within the preset future time period can be determined based on the solar trajectory and specification data, thereby improving the solar energy capture efficiency. Next, this application can predict the strong wind conditions in the area where the target photovoltaic panel is located within the preset future time period using a pre-configured target strong wind prediction model, obtaining strong wind prediction results. Then, based on the optimal tilt angle and strong wind prediction results at different time intervals, the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals are evaluated, thereby ensuring the stability and safety of the system under severe weather conditions. Finally, this application can calculate the optimal tilt angle adjustment strategy that maximizes the overall revenue and minimizes the risk of the photovoltaic power station based on the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, and the actual situation of the photovoltaic power station. By dynamically adjusting the photovoltaic panel tilt angle, this application not only optimizes energy collection efficiency but also significantly enhances the wind resistance of the structure, achieving dual optimization of energy and safety. Furthermore, the implementation of this application provides an intelligent and automated solution for the photovoltaic industry, helping to promote the development and application of sustainable energy technologies.

[0069] In one embodiment, obtaining the solar trajectory within a preset future time period in step S110 may include:

[0070] S111: Obtain the pre-configured solar trajectory prediction model and the currently collected solar azimuth and solar altitude angles.

[0071] S112: Calculate the current solar position data based on the solar azimuth angle and the solar altitude angle.

[0072] S113: Input the current solar position data into the solar trajectory prediction model to obtain the solar motion trajectory within a preset future time period output by the solar trajectory prediction model.

[0073] In this embodiment, when acquiring the sun's trajectory within a preset future time period, the azimuth and elevation angle data collected by sensors can be used. Then, this application can utilize the acquired sun position data to construct a sun trajectory prediction model based on a machine learning model using historical data, including a Support Vector Machine (SVM) or a Long Short-Term Memory Neural Network (LSTM). By training on historical sun position data, the sun's trajectory within a future period can be predicted.

[0074] In one specific implementation, the sensor collects solar azimuth and elevation angle data every 10 minutes, and calculates the real-time position coordinates (x, y, z) of the sun in the sky using spherical trigonometric functions, forming solar position data. Next, this application utilizes historical solar position data from the past 30 days and employs a Support Vector Machine (SVM) algorithm, using time as a feature and position coordinates as the target variable, to train a solar trajectory prediction model. Then, based on the photovoltaic panel's specifications, such as its 1m × 1.5m size, monocrystalline silicon material, and 18% conversion efficiency, and combined with the solar trajectory prediction results, at 30-minute intervals, the optimal tilt angle for the photovoltaic panel to receive solar radiation in each time period is calculated, forming a tilt angle data sequence containing 48 elements.

[0075] For example, this application analyzes the data variation trends of solar radiation intensity sensors over different time periods to obtain a radiation intensity sequence corresponding to the tilt angle data sequence. Then, this application uses a genetic algorithm, with maximizing power generation as the objective function, to optimize the photovoltaic panel angle adjustment strategy, resulting in an optimized tilt angle sequence, such as [10°, 15°, 20°, ..., 30°]. Based on this sequence, a corresponding angle adjustment command is generated and transmitted to the central processing unit via the CAN bus, adjusting the photovoltaic panel angle every 30 minutes.

[0076] In addition, during the operation of photovoltaic panels, the power generation efficiency, temperature, and voltage parameters of the panels can be continuously monitored. A performance prediction model is established using a multiple linear regression model, with power generation efficiency as the dependent variable and temperature and voltage as independent variables. When the actual power generation efficiency is 20% lower than the predicted value, a potential fault is identified. Based on the deviation of temperature and voltage parameters from normal values, the fault type is inferred to be dust accumulation or component damage, generating corresponding cleaning or component replacement maintenance instructions. The power generation data and angle adjustment data of the photovoltaic panels are uploaded to the central processing unit via the MQTT protocol. A random forest algorithm is used, with power generation as the target variable and photovoltaic panel tilt angle and radiation intensity as feature variables, to continuously optimize the photovoltaic panel angle adjustment strategy. Simultaneously, with fault type as the target variable and temperature and voltage parameters as feature variables, the fault diagnosis model is optimized to improve the accuracy and efficiency of fault detection.

[0077] In one embodiment, the training process of the target strong wind prediction model may include:

[0078] S210: Obtain hourly meteorological observation data of the area where the target photovoltaic panel is located within a preset historical period from the meteorological department, and obtain a strong wind historical dataset after preprocessing the meteorological observation data.

[0079] S220: The preset initial strong wind prediction model is trained using the historical strong wind dataset, and when the preset training stopping condition is reached, the trained initial strong wind prediction model is used as the target strong wind prediction model.

[0080] In this embodiment, when training the target strong wind prediction model, hourly wind speed and direction data for the target area from 2010 to 2020 can be obtained from the meteorological department. The data is cleaned, removing 2% of invalid or missing data, and the wind speed data is discretized to an accuracy of 0.1 m / s, while the wind direction data is discretized to an accuracy of 10°. Finally, multiple valid data points are obtained, and a historical strong wind dataset is constructed.

[0081] Specifically, this application deploys 20 wind speed sensors and 10 wind direction sensors at a photovoltaic power station site. The sensors sample at a frequency of 1Hz and transmit the collected data to the central processing unit every minute via the ZigBee wireless communication protocol, forming a real-time environmental information data stream. Using the Apriori association rule mining algorithm, with a minimum support of 0.05 and a minimum confidence of 0.8, a strong wind occurrence pattern with wind speeds greater than 20 m / s and a duration exceeding 10 minutes was identified. Furthermore, a time-series pattern mining algorithm was used to identify precursor patterns where the average wind speed increase rate exceeds 2 m / s / h within one hour prior to a strong wind event. A strong wind prediction model composed of 100 decision trees was trained, using wind speed, wind direction, and air pressure parameters as features, and the occurrence of strong winds as the objective variable. Through 10-fold cross-validation, the model achieved an average accuracy of 85%. Using this model to predict the probability of strong winds occurring within the next 24 hours, the time period with a probability exceeding 50% was found to be from 2 PM to 4 PM, with an average wind speed potentially reaching 25 m / s and a prevailing northwest wind direction. Based on the CAD layout of the photovoltaic power station, the stress distribution of the photovoltaic panels under different wind speeds and directions was calculated using the finite element analysis software ANSYS. It was found that under a northwest wind of 14 m / s, the stress in the supporting structure of some photovoltaic panels exceeded the yield strength of the material, posing a risk of damage. For these high-risk photovoltaic panels, their tilt angle can be adjusted to 5°, and a nano-level hydrophobic coating can be applied to achieve a water contact angle of 150°. Simultaneously, a 10-meter-high windbreak net can be installed around them to reduce the wind speed distribution within the photovoltaic array.

[0082] Furthermore, strong wind data analysis indicates that when the average wind speed exceeds 20 m / s, the total power generation of a photovoltaic power station may decrease by more than 20%. Therefore, this application utilizes a three-layer BP neural network to construct a photovoltaic power generation prediction model. The model takes wind speed, wind direction, and solar irradiance as inputs and the total power generation of the photovoltaic power station as the output. The training samples are historical data from the past year. The average relative error after network convergence is 6%. This model can provide relatively accurate power generation predictions for grid dispatch.

[0083] In one embodiment, training a preset initial strong wind prediction model using the historical strong wind dataset in step S220 may include:

[0084] S221: The correlation between different types of data in the historical strong wind dataset is analyzed by correlation analysis method, and wind speed, wind direction and air pressure gradient features are selected from the historical strong wind dataset as training data based on the analysis results.

[0085] S222: Use the training data to train the preset initial strong wind prediction model.

[0086] In this embodiment, when training the initial strong wind prediction model, hourly meteorological observation data of the target area over a preset period can be obtained from the meteorological department, including wind speed, wind direction, air pressure, temperature, and humidity parameters, forming a historical strong wind dataset. Next, this application can preprocess the historical strong wind dataset, including using linear interpolation and nearest neighbor interpolation to handle missing values, and using Z-score standardization and maximum-minimum normalization to normalize the data. Then, through correlation analysis and principal component analysis, wind speed, wind direction, and air pressure gradient features are selected as input features for the model, reducing the model's complexity. Furthermore, this application can also divide the preprocessed historical strong wind data into a training set, a validation set, and a test set according to time order. The training set is used for model training, the validation set is used for model hyperparameter tuning, and the test set is used to evaluate the final performance of the model.

[0087] Next, this application selects machine learning algorithms for processing time-series data, including Long Short-Term Memory (LSTM) neural networks and Gated Recurrent Unit (GRU) neural networks, to construct an initial strong wind prediction model. The model's hyperparameters are optimized using a combination of grid search and random search. During model training, early stopping and dropping regularization methods are employed to prevent overfitting. The early stopping method stops training when the validation set loss does not decrease for several consecutive epochs. The model's generalization performance is evaluated by calculating the mean squared error (MSE), mean absolute error (MAE), and loss function value on the validation set, and the model with the best performance is selected as the final target strong wind prediction model. Using the trained target strong wind prediction model, meteorological data for a preset future time is predicted. Combining predicted wind speed, wind direction, and duration factors, a strong wind judgment threshold is determined based on local terrain and building distribution, outputting the predicted wind speed and probability of strong wind occurrence for each hour. Based on the difference between the predicted wind speed and the threshold, and the duration of the strong wind, the risk level is classified into low risk, medium risk, high risk, and extremely high risk. Based on strong wind forecasts, a strong wind risk warning is generated for a preset time in the future, including the time period of occurrence, duration, average wind speed, and risk level. This warning is sent to the photovoltaic power station's operation and maintenance personnel in JSON or XML standard format via various channels, including SMS, email, application push notifications, and voice calls. Simultaneously, the strong wind forecast results are transmitted to the photovoltaic power station's intelligent control system, allowing for advance adjustment of the photovoltaic panel angles and activation of appropriate wind protection mechanisms. A risk management feedback mechanism is established to track the actual situation of the strong wind event for post-event assessment and model improvement.

[0088] Specifically, this application obtained hourly meteorological observation data for the target area from 1999 to 2019 for 20 years from the meteorological department, including 10 parameters such as wind speed, wind direction, air pressure, temperature, and humidity, forming 175,200 historical records of strong winds. After cleaning the raw data, it was found that the missing rate for wind speed data was 2%, while the missing rates for other data were all less than 0.5%. Nearest neighbor interpolation was used to fill in the missing values ​​in the wind speed data, and all data were normalized to their maximum and minimum values. The correlation between each parameter and wind speed was analyzed using the Spearman rank correlation coefficient method, finding that the correlation coefficients between wind direction, air pressure gradient, and wind speed were all greater than 0.7, while the correlation coefficients for temperature and humidity were less than 0.3. Therefore, wind speed, wind direction, and air pressure gradient were selected as three features to construct an intensity prediction model. An LSTM model was used, and the 20 years of data were divided into a training set (80%), a validation set (10%), and a test set (10%) according to time sequence. First, through 50 random searches, the number of hidden layers in the LSTM model was determined to be 2, the number of hidden units to be 128, and the learning rate to be 0.01. Then, through 50 grid searches, the optimal model hyperparameters were obtained as follows: 2 hidden layers, 64 hidden units, learning rate 0.005, and dropout rate 0.3. An early stopping callback function was set to stop training when the validation set loss did not decrease for 5 consecutive epochs. The final model achieved an MSE of 0.25 and a MAE of 0.35 on the test set, which is better than the baseline model's persistent prediction MSE (0.5) and MAE (0.6). The model was used to predict wind speed, wind direction, and pressure gradient for the next 24 hours. When the predicted wind speed exceeds 20 m / s, lasts for more than 1 hour, and the wind direction is northwest, it is classified as a high-risk level (orange warning). The final strong wind warning information included the warning time (13:00 to 16:00 on May 15, 2024), the probability level of strong wind occurrence (orange), the average wind speed (25 m / s), and the wind direction (northwest). This information was sent to relevant personnel via SMS, email, and app push notifications, and simultaneously transmitted to the intelligent photovoltaic control system in JSON format. Tracking the actual wind speed revealed a maximum wind speed of 23 m / s lasting for 2 hours, which largely matched the warning information, proving the predictive model's accuracy.

[0089] In one embodiment, S130, evaluating the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals based on the optimal tilt angle at different time intervals and the strong wind prediction results, may include:

[0090] S131: Establish a three-dimensional geometric model corresponding to the target photovoltaic panel.

[0091] S132: Based on the three-dimensional geometric model and the strong wind prediction results, calculate the structural stress distribution of the target photovoltaic panel at the optimal tilt angle at different time intervals, and evaluate the structural stability of the target photovoltaic panel according to the structural stress distribution.

[0092] S133: The theoretical power generation of the target photovoltaic panel at the optimal tilt angle under different time intervals is calculated using the Sandia model, and the energy capture efficiency of the target photovoltaic panel is evaluated based on the theoretical power generation.

[0093] S134: A comprehensive evaluation of the structural stability and energy capture efficiency of the target photovoltaic panel is conducted to obtain the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals.

[0094] In this embodiment, after obtaining the strong wind forecast results, this application can evaluate the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals based on the optimal tilt angle at different time intervals and the strong wind forecast results.

[0095] For example, this application can obtain hourly wind speed and direction data for a predetermined future time from a target strong wind prediction model, and simultaneously obtain the optimal tilt angle data of the current photovoltaic panel. These are used as input to a risk assessment model. Machine learning algorithms, including support vector regression and neural networks, are used to construct a risk assessment model, with the risk level and the photovoltaic panel's gains and losses based on the current optimal tilt angle data as outputs. Computational fluid dynamics (CFD) is used to establish a three-dimensional geometric model of the photovoltaic panel, incorporating its dimensions, materials, and support structure parameters. A three-dimensional geometric model is created based on the actual dimensions and structure of the photovoltaic panel, and the model is meshed, with mesh refinement in key areas. Boundary conditions and initial conditions for wind speed and direction are set, a turbulence model and discretization scheme are selected, and numerical iterative calculations are performed until the solution converges. The obtained velocity and pressure field results are processed to analyze the stress distribution on the photovoltaic panel surface. Based on the photovoltaic panel's material properties, including elastic modulus and yield strength, the structural stress distribution of the photovoltaic panel at different tilt angles is calculated to determine whether it exceeds the allowable stress of the material and to evaluate the structural stability of the photovoltaic panel. If the structural stress exceeds the allowable stress value determined based on the actual material properties and safety factor of the photovoltaic panel, it is determined that the structural stability has not reached the preset threshold.

[0096] Using the Sandia model, and comprehensively considering factors such as the photovoltaic (PV) panel's material properties, tilt angle, azimuth angle, irradiance, temperature, and dust obstruction, the theoretical power generation of the PV panel at different tilt angles is calculated, and the energy capture efficiency of the PV panel is evaluated. If the power generation is lower than a preset threshold for the optimal tilt angle under the same conditions, the energy capture efficiency is determined to have failed to reach the preset threshold. A comprehensive evaluation of the PV panel's structural stability and energy capture efficiency is conducted using a weighted average method, assigning weights to structural stability and energy capture efficiency respectively, to obtain the PV panel's comprehensive performance index at the current tilt angle. If the comprehensive performance index is lower than a preset threshold, the current tilt angle is determined to pose a risk. For PV panels determined to pose a risk, corresponding tilt angle adjustment strategies are adopted based on their structural stability and energy capture efficiency.

[0097] When determining the optimal tilt angle, this application can use the benefit-cost ratio (BCR) as an evaluation index. BCR = ΔE / (C_risk + C_adjust), where ΔE is the difference in power generation before and after tilt angle adjustment, C_risk is the risk cost reduced by tilt angle adjustment, and C_adjust is the actual cost of tilt angle adjustment, including energy consumption and labor. The tilt angle with the largest BCR is selected as the optimal tilt angle. If the structural stability is less than a preset threshold, the tilt angle is adjusted to the minimum angle to ensure the structural safety of the photovoltaic panel; if the energy capture efficiency is less than a preset threshold, the tilt angle is adjusted to the optimal angle to increase power generation while ensuring structural stability; if neither is ideal, the tilt angle is adjusted to the tilt angle with the largest BCR. The tilt angle adjustment strategy of the photovoltaic panel is formed into a decision scheme and transmitted to the control system of the photovoltaic power station to guide the actual adjustment process of the photovoltaic panel. Simultaneously, the risk assessment results and decision scheme of the photovoltaic panel are fed back to the risk assessment model for model optimization and improvement.

[0098] In one specific implementation, this application can obtain hourly wind speed and direction data for the next 24 hours from a target strong wind prediction model, and simultaneously obtain the current optimal tilt angle data of the photovoltaic panel (e.g., 30°), as input to a risk assessment model. A Long Short-Term Memory (LSTM) neural network is used to construct the risk assessment model, with the risk level and the photovoltaic panel's gains and losses related to the current optimal tilt angle data as outputs. A 3D geometric model of a 10×10 photovoltaic array is established using the CFD software ANSYS Fluent. The photovoltaic panel size is 1m×1.5m, the material is monocrystalline silicon, and the support is fixed. The model is unstructured and meshed, with mesh refinement in key areas at the edge of the photovoltaic panel. The minimum mesh size is 1cm, and a northwest wind direction with a wind speed of 25m / s is set as the boundary condition. A k-ε turbulence model and a second-order upwind discretization scheme are selected, with a residual convergence criterion of 0.0001. The maximum wind pressure on the photovoltaic panel surface is calculated to be 2500Pa, which is greater than the yield strength of monocrystalline silicon (1800Pa), indicating insufficient structural stability. Using the Sandia model, the theoretical power generation of the photovoltaic panel at a 30° tilt angle is calculated to be 180 kWh / day, while the power generation at a 0° tilt angle is 108 kWh / day, indicating poor energy capture efficiency. A comprehensive evaluation shows that the overall performance index of the current tilt angle is 0.65, which is less than the threshold of 0.8, indicating a risk associated with the current tilt angle. Adjusting the tilt angle to 10° would reduce the structural stress to 1500 Pa, meeting stability requirements, but the power generation would also decrease to 126 kWh / day. Adjusting the tilt angle to 20° would increase the power generation to 162 kWh / day, but the structural stress would increase to 2100 Pa, still exceeding the yield strength. A trade-off calculation shows that the BCR value is highest at a 15° tilt angle; therefore, the optimal tilt angle is determined to be 15°, with a corresponding structural stress of 1800 Pa and a power generation of 144 kWh / day. The optimal tilt angle of 15° and the structural stress threshold of 1800 Pa are sent to the photovoltaic power plant control system as target and monitoring values ​​for tilt angle adjustment. Meanwhile, the input data (wind speed 25m / s, wind direction northwest), assessment results (tilt angle 30°, comprehensive index 0.65) and decision results (optimal tilt angle 15°) of this risk assessment are fed back to the risk assessment model to update the model's training dataset.

[0099] In one embodiment, S140 calculates the optimal tilt angle adjustment strategy that maximizes the overall benefit and minimizes the risk of the photovoltaic power station based on the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, and the actual situation of the photovoltaic power station. This strategy may include:

[0100] S141: Based on the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, identify high-risk photovoltaic panels where the strong wind loss is greater than the power generation revenue.

[0101] S142: Based on the actual situation of the photovoltaic power station, with minimizing the number of high-risk photovoltaic panels as the optimization objective and the spacing and arrangement of photovoltaic panels as optimization variables, a photovoltaic array layout optimization model is constructed.

[0102] S143: The high-risk photovoltaic panels are optimized using the photovoltaic array layout optimization model to obtain the optimal tilt angle adjustment strategy that maximizes the overall benefits and minimizes the risks of the photovoltaic power station.

[0103] In this embodiment, when calculating the optimal tilt angle adjustment strategy that maximizes the overall benefit and minimizes the risk of the photovoltaic power station, loss data and power generation revenue data of each photovoltaic panel under strong winds can be obtained from the risk assessment model. Based on the ratio of loss to revenue, high-risk photovoltaic panels whose losses under strong winds exceed their power generation revenue are identified and marked as photovoltaic panels to be optimized. A genetic algorithm is used to optimize the layout of the photovoltaic array, with minimizing the number of high-risk photovoltaic panels as the optimization objective, and the spacing and arrangement of photovoltaic panels as optimization variables, to construct a photovoltaic array layout optimization model. In the genetic algorithm, real-number encoding is used to randomly generate the initial population. By conducting multiple experiments with different parameter combinations, the optimization effect and convergence speed are compared, and the optimal parameter settings are selected. In each generation of evolution, each individual in the population, i.e., a layout scheme, is evaluated. The CFD method is used to calculate the wind field distribution of the photovoltaic array under this layout, using a turbulence model and a second-order upwind discretization scheme, setting boundary conditions consistent with the actual environment, and performing structured meshing of the computational domain, with the mesh scale locally refined according to the geometric characteristics of the photovoltaic panels. After obtaining the wind speed and wind pressure data for each photovoltaic panel, the stress distribution under the current wind load is calculated using the finite element method based on the structural parameters and material properties of the photovoltaic panels. This stress distribution is then compared with the allowable stress limit to obtain the number and location distribution of high-risk photovoltaic panels in this layout. Based on the number and distribution of high-risk photovoltaic panels, the fitness of each layout scheme is evaluated. A multi-objective fitness function is constructed, comprehensively combining multiple evaluation indicators such as the number of high-risk photovoltaic panels, the power generation efficiency of the photovoltaic array, and the compactness of the layout. A weighted summation method is used to normalize each objective, yielding the comprehensive fitness value of the layout scheme. The population is sorted according to its fitness value; the higher the fitness value, the higher the ranking. Genetic operations of selection, crossover, and mutation are performed on the population to generate a new generation. The selection operation uses a roulette wheel selection strategy, where the selection probability of an individual is proportional to its fitness value. The crossover operation encodes and operates on two optimization variables: the spacing and arrangement of the photovoltaic panels. Arithmetic crossover is used for the spacing variable, and partial matching crossover is used for the arrangement variable. The crossover probability is adaptively adjusted according to the evolutionary state of the population. The mutation operation employs uniform mutation, and the mutation probability also adopts an adaptive strategy, being high in the early stages of evolution and gradually decreasing in later stages. After multiple generations of evolution, several individuals with the highest fitness values ​​are selected from the final population and cross-validated to evaluate their generalization performance under different wind conditions. Individuals whose performance exceeds a preset threshold under all operating conditions are selected as the optimal layout scheme. The optimal layout scheme is converted into photovoltaic array configuration parameters, including photovoltaic panel spacing, tilt angle, and azimuth angle, and sent to the photovoltaic power station control system to guide the actual reconfiguration process of the photovoltaic array. After the photovoltaic array reconfiguration is completed, the power generation revenue and risk losses of each photovoltaic panel are recalculated. A medium- and long-term economic benefit module for the photovoltaic power station is established, comprehensively considering factors such as the reconfigured power generation, electricity price subsidies, and operation and maintenance costs to evaluate the economics of the photovoltaic power station under different operating strategies.If the reconstructed photovoltaic power station can achieve break-even or even profitability under the current electricity price subsidies and operation and maintenance costs, it will continue to operate according to the current layout plan. However, if policy changes or market fluctuations prevent the reconstructed photovoltaic power station from achieving profitability, further optimization of the operation strategy is needed, including reducing peak power, adding energy storage equipment, or suspending the operation of some high-risk photovoltaic panels to reduce losses. The actual operating data of the reconstructed photovoltaic array will be fed back to the risk assessment model for model verification and iterative updates.

[0104] Specifically, the risk assessment model calculated that 8 out of 24 photovoltaic panels experienced wind losses exceeding their power generation revenue, with an average loss 1.5 times the revenue. These 8 panels were designated as optimization targets, and a photovoltaic array layout optimization model based on a genetic algorithm was constructed. Through 100 parameter combination experiments, the optimal parameter settings were determined: population size 100, crossover probability 0.8, mutation probability 0.1, and number of generations 50. During the layout optimization process, ANSYS Fluent software was used for CFD analysis, employing a Realizable k-ε turbulence model and a second-order upwind discretization scheme. The computational domain size was 200m × 200m × 50m, divided into approximately 2 million hexahedral structured meshes with a minimum mesh size of 0.05m. The calculated maximum wind speed on the photovoltaic panel surface was 25m / s, and the maximum wind pressure was 2500Pa. Finite element analysis using ANSYS Mechanical software revealed a maximum von Mises stress of 180MPa for the photovoltaic panel, exceeding its allowable stress (150MPa). The fitness function is set as the sum of the products of the objective weights and the normalized index, with the number of high-risk photovoltaic panels weighing 0.5, power generation efficiency weighing 0.3, and layout compactness weighing 0.2. The optimal layout scheme, obtained through 50 generations of genetic evolution, reduces the number of high-risk photovoltaic panels to 2, increases layout compactness by 20%, but slightly decreases power generation efficiency. After 10-fold cross-validation, this layout scheme reduces the average risk loss by 50% under various wind conditions. Based on the optimal layout scheme, the photovoltaic panel spacing is adjusted to 1.5 times the photovoltaic panel width, the tilt angle is adjusted to 15°, and the azimuth angle is adjusted to due south. The reconfigured photovoltaic power station is expected to generate an average annual power of 8,000,000 kWh. Based on the current electricity price of RMB 0.8 / kWh, the annual revenue is RMB 6,400,000, the operation and maintenance cost is RMB 800,000 / year, the investment payback period is 8 years, the project net present value (NPV) is RMB 12,000,000, and the internal rate of return (IRR) is 10%, meeting the economic requirements. Therefore, it was decided to continue operation according to the current layout plan. The actual operating data of the photovoltaic power station over the next 5 years was incorporated into the training set of the risk assessment model, and incremental learning was performed using a BP neural network algorithm, improving the model's risk prediction accuracy from 90% to [a higher level].

[0105] In addition, this application can also output a decision report on adjusting the angle of the photovoltaic panel based on the benefits and losses through a cost-benefit calculation unit. By adopting new materials, optimizing the support structure, and adding windproof devices, the wind resistance of the photovoltaic panel can be improved, so that the control system can dynamically adjust the angle of the photovoltaic panel according to the decision report, thereby optimizing energy capture and reducing the risk of strong winds.

[0106] Specifically, the expected power generation revenue and risk loss data of photovoltaic panels at different tilt angles are obtained from the risk assessment model to form a revenue-loss matrix. This matrix is ​​then input into a cost-benefit calculation unit to evaluate the economic feasibility of various tilt adjustment strategies by calculating their net present value (NPV), internal rate of return (IRR), and investment payback period. Based on the cost-benefit analysis and considering the actual conditions of the photovoltaic power plant, multi-objective optimization algorithms, including genetic algorithms and particle swarm optimization, are used to find the optimal tilt adjustment strategy that maximizes overall revenue and minimizes risk, generating a decision report. This optimal tilt adjustment strategy is embedded into the photovoltaic power plant's control system, dynamically adjusting the tilt angle of the photovoltaic panels based on real-time wind speed and direction data and the current state of the panels. Simultaneously, a risk warning mechanism is established. When the wind speed exceeds a certain threshold, the control system automatically adjusts the photovoltaic panel angle to a safe position with minimal wind exposure. Following the guidance in the decision report, corresponding wind protection measures are taken, including locking the supports and disconnecting circuits, to minimize the risk of strong wind disasters and optimize energy capture efficiency, ensuring the safe and stable operation of the photovoltaic power plant. The material properties of photovoltaic (PV) panels are analyzed. Finite element method (FEM) simulation is used to evaluate the impact of different materials, including high-strength steel and carbon fiber composites, on the wind resistance performance of PV panels. The material option with the best overall cost-effectiveness is selected and applied in the design and manufacturing of PV panels. The basic FEM process includes establishing a geometric model and defining material properties, meshing and setting boundary conditions, solving the governing equations, and performing error analysis. Topology optimization and parametric design methods are used to optimize the geometric structure of the PV panel support. While ensuring the strength and stiffness of the support meet requirements, the drag coefficient of the support is reduced to mitigate the adverse effects of wind loads on the PV panels. The optimization process includes defining the objective function and constraints, including minimizing flexibility and maximizing stiffness; using the SIMP method for topology optimization; and smoothing the optimization results to obtain the final structural design scheme. Passive wind protection devices, such as windbreak nets and windbreak panels, are installed around the PV array. Numerical wind tunnel simulation is used to optimize parameters such as the height, spacing, and porosity of the wind protection devices to minimize inflow wind speed and reduce shading of sunlight. Numerical wind tunnel simulation can employ computational fluid dynamics (CFD) methods. Specifically, this involves determining the scope and boundary conditions of the computational domain based on wind tunnel experiments or actual environments, dividing the computational domain into meshes, refining the mesh in key areas, selecting appropriate turbulence models and discretization schemes, setting control parameters, performing numerical solutions, obtaining the distribution of physical quantities such as velocity and pressure fields, visualizing and analyzing the calculation results, and optimizing the parameters of wind protection devices.

[0107] In one specific implementation, the risk assessment model predicts that when the photovoltaic panel tilt angle is 30°, the annual power generation revenue is 1 million yuan, and the risk loss is 200,000 yuan; when the tilt angle is 60°, the revenue is 1.2 million yuan, and the loss is 50 yuan. The revenue and loss data under different tilt angles are input into the cost-benefit calculation unit. If the project cycle is 20 years, the discount rate is 8%, and the initial investment is 5 million yuan, the NPV for the 30° tilt angle strategy is 8 million yuan, and the IRR is 12%; the NPV for the 60° tilt angle strategy is 6 million yuan, and the IRR is 10%. A genetic algorithm is used for multi-objective optimization, with the objective function being maximizing total revenue and minimizing total loss, and the photovoltaic panel tilt angle and tracking system power as decision variables, considering the constraints that the maximum allowable tilt angle of the photovoltaic panel does not exceed 70° and the tracking system power does not exceed 10kW. After 500 generations of evolution, the Pareto optimal solution set was obtained, including a strategy with a tilt angle of 45° and a tracking power of 8kW. This strategy yields an average annual power generation revenue of 1.1 million yuan and a risk loss of 300,000 yuan, achieving the best overall benefit. This strategy was incorporated into the decision report. When the measured wind speed exceeds 20 m / s, the control system adjusts the photovoltaic panel to a safe position of 10° and, according to the decision guidance, shuts down the tracking system and locks the support. Structural simulation analysis of a 1.5m × 1m monocrystalline silicon photovoltaic panel was performed using ANSYS software, employing both ordinary steel (elastic modulus 200 GPa) and carbon fiber reinforced composite material (elastic modulus 500 GPa). After meshing, 100,000 nodes and 200,000 elements were obtained. Applying a wind load of 25 m / s, the results show that using carbon fiber material reduces the maximum deformation of the photovoltaic panel by 50% and the stress by 30%. The photovoltaic panel support structure was topologically optimized using OptiStruct software to minimize flexibility while maintaining a volume fraction of at least 0.5. After 50 iterative optimization steps, the drag coefficient of the support structure decreased from 1.2 to 0.8, and the mass was reduced by 20%. A 6m high windbreak net was installed around the photovoltaic array. Numerical wind tunnel simulations were performed using Fluent software. The computational domain was 100m × 50m × 20m, divided into a hexahedral structured mesh with a total of 5 million pixels. The SSTk-ω turbulence model was used, with an inlet wind speed of 25m / s and an outlet pressure of 0. The remaining surfaces were non-slip walls. The calculation results show that the windbreak net can reduce the average wind speed at a height of 1m by 60% and the wind speed inside the array by 80%, while having a minimal impact on solar illumination.

[0108] In one embodiment, the method may further include:

[0109] S150: Based on the optimal tilt angle of the target photovoltaic panel at different time intervals, analyze the changes in revenue and risk of adjusting the target photovoltaic panel to the optimal tilt angle at different time intervals.

[0110] S151: Determine the optimal tilt angle adjustment timeliness strategy based on the changes in revenue and risk of adjusting the target photovoltaic panel to the optimal tilt angle at different time intervals.

[0111] In this embodiment, after determining the optimal tilt angle adjustment strategy that maximizes the overall benefits and minimizes the risks of the photovoltaic power station, this application can also combine the tilt angle data of the photovoltaic panels in the time series to analyze the changes in benefits and risks of adjusting the photovoltaic panels to the optimal tilt angle at different time intervals. Through a risk assessment model, the impact of adjusting the tilt angle at different time intervals on the structural stability and energy capture efficiency of the photovoltaic panels is evaluated. By weighing the changes in benefits and risks, the optimal tilt angle adjustment time-sensitive strategy is determined.

[0112] For example, this application can obtain photovoltaic panel tilt angle adjustment data and corresponding meteorological data for a preset time period from a photovoltaic power plant, forming a time series dataset. The data is preprocessed to remove outliers, interpolate missing values, and aggregate the data according to different time intervals. The frequency and amplitude of photovoltaic panel tilt angle adjustment at different time intervals are calculated to obtain a frequency-amplitude sequence. A mechanistic model is constructed by combining the physical relationship between photovoltaic panel tilt angle, meteorological parameters, and power generation efficiency. The model parameters are fitted using multiple regression or machine learning methods to analyze the relationship between frequency and amplitude and photovoltaic panel power generation efficiency. Based on the structural parameters and material properties of the photovoltaic panel, a finite element model is pre-established to obtain the response surfaces of stress and deformation with wind speed and tilt angle. A general polynomial response surface equation is adopted: y(x)=a_0+∑a_ix_i+∑a_ijx_ix_j, where y is the structural stress or deformation, x is the wind speed and tilt angle, and a is the fitting coefficient. Meteorological data is input into the risk assessment model to obtain photovoltaic panel structural stress and deformation data under different risk levels. This study analyzes the relationship between structural stress and deformation data and the frequency and amplitude of tilt adjustment, evaluating the impact of different tilt adjustment strategies over various time intervals on the structural stability of photovoltaic panels. Theoretical calculations based on astronomical parameters and atmospheric transmittance, including the Meinel model, are used: G = G_sc·cosθ_z·τ_M, where G is the surface irradiance, G_sc is the solar constant, θ_z is the zenith angle, and τ_M is the atmospheric transmittance. τ_M can be adjusted based on atmospheric composition and pollution levels. The theoretical power generation of the photovoltaic panels at different tilt angles is calculated, and the theoretical power generation is compared with the actual power generation to obtain the energy loss rate at different time intervals. The relationship between the energy loss rate and the frequency and amplitude of tilt adjustment is analyzed to evaluate the impact of different tilt adjustment strategies over various time intervals on energy capture efficiency. For each tilt adjustment strategy over a given time interval, the theoretical lifetime and power generation revenue of the photovoltaic panel under that strategy are calculated based on its corresponding structural stability and energy capture efficiency. A combination of the analytic hierarchy process (AHP) and the entropy weight method is used to comprehensively evaluate the risks and benefits of different strategies. Objective weights are obtained by calculating the coefficient of variation of the indicators. These objective weights are then combined with subjective weights to obtain combined weights, reducing the influence of subjectivity. Risk and return scoring matrices are constructed. Fuzzy comprehensive evaluation is used to perform a weighted summation of the risk and return scoring matrices. Triangular fuzzy numbers are used to represent linguistic variables, and membership functions are constructed. A weighted average operator is used for fuzzy synthesis: B = A·R = [b_1, b_2, ..., b_n], where A is the weight vector, R is the fuzzy evaluation matrix, and B is the comprehensive evaluation result. B is defuzzified using the λ-cut method to obtain the comprehensive score. The time interval with the highest comprehensive score is selected as the optimal tilt adjustment timeliness, forming the optimal tilt adjustment timeliness strategy.The tilt angle adjustment time-sensitive strategy is applied to the actual control system of a photovoltaic power plant, and the implementation effect of the strategy is tracked, monitored, and dynamically adjusted. Monitoring indicators are set to evaluate the actual effect of the strategy. When the deviation between the actual effect and the expected effect exceeds a preset deviation threshold, the strategy is corrected and optimized, including adjusting the tilt angle adjustment range and modifying the risk threshold. This forms a closed-loop feedback mechanism to continuously improve the overall efficiency of the photovoltaic power plant.

[0113] Specifically, this application obtains hourly tilt angle adjustment records and meteorological observation data of a photovoltaic power station from January to December 2022, totaling 8760 records. The data is cleaned and standardized to obtain a tilt angle adjustment frequency sequence (0.1 times / hour to 1 time / hour) and an adjustment amplitude sequence (1 degree to 10 degrees). Based on a physical mechanism model, the functional relationship between power generation efficiency and tilt angle, irradiance, and temperature is obtained, η = 0.18··cosθ. Through multiple linear regression, it is found that for every 0.1 times / hour increase in frequency, η decreases by 0.2%; for every 1 degree increase in amplitude, η increases by 0.5%. Using ANSYS finite element software, the stress distribution of the photovoltaic panel under wind speeds of 0-30 m / s and tilt angles of 0-60 degrees is analyzed, yielding the polynomial response surface equation σ = 0.5 + 2.1v + 1.2θ + 0.8vθ (σ is stress, v is wind speed, and θ is tilt angle). Calculations showed that for every 0.1 cycles / hour increase in frequency, the maximum stress increased by 1.5 MPa; for every 1 degree increase in amplitude, the maximum stress increased by 0.8 MPa. The theoretical power generation at different tilt angles was calculated using the Meiner model: G = 1367·cosθ_z·(0.75 + 0.00002h) (θ_z is the zenith angle, h is the altitude). Comparing the theoretical and actual values, it was found that for every 0.1 cycles / hour increase in frequency, the energy loss rate increased by 0.1%; for every 1 degree increase in amplitude, the energy loss rate decreased by 0.3%. The risks and benefits of adjustment strategies for six time intervals (0.5h, 1h, 2h, 4h, 8h, 12h) were evaluated. The analytic hierarchy process (AHP) was used to construct a risk matrix R_ij and a benefit matrix B_ij (i, j represent strategy numbers). The entropy weight W_j = 1 - H_j / lnn (where H_j is the entropy value of the j-th indicator) is calculated, resulting in the risk weight vector W_R = (0.3, 0.7) and the return weight vector W_B = (0.6, 0.4). A fuzzy evaluation matrix is ​​constructed and synthesized using a weighted average operator, yielding the comprehensive evaluation vector C = (0.2, 0.5, 0.8, 0.6, 0.4, 0.3), corresponding to time intervals of 0.5h, 1h, 2h, 4h, 8h, and 12h, respectively. Using a cutoff of λ = 0.8, the optimal time interval is determined to be 2h. Applying the 2h adjustment strategy to actual operation, continuous monitoring for one month resulted in a 2% increase in average power generation efficiency and a reduction of one risk event.

[0114] The photovoltaic panel tilt angle optimization device based on data fusion analysis provided in the embodiments of this application will be described below. The photovoltaic panel tilt angle optimization device based on data fusion analysis described below can be referred to in correspondence with the photovoltaic panel tilt angle optimization method based on data fusion analysis described above.

[0115] In one embodiment, such as Figure 2 As shown, Figure 2 This application provides a schematic diagram of a photovoltaic panel tilt angle optimization device based on data fusion analysis, which is an embodiment of the present application. The present application also provides a photovoltaic panel tilt angle optimization device based on data fusion analysis, including an optimal tilt angle prediction module 210, a strong wind condition prediction module 220, a benefit and risk assessment module 230, and an optimal tilt angle determination module 240, specifically including the following:

[0116] The optimal tilt angle prediction module 210 is used to acquire the solar motion trajectory and the specification data of the target photovoltaic panel within a preset future time period, and to determine the optimal tilt angle of the target photovoltaic panel at different time intervals within the preset future time period based on the solar motion trajectory and the specification data.

[0117] The strong wind prediction module 220 is used to predict the strong wind conditions in the area where the target photovoltaic panel is located during the preset future time period using a pre-configured target strong wind prediction model, and obtain the strong wind prediction result.

[0118] The revenue and risk assessment module 230 is used to assess the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, based on the optimal tilt angle at different time intervals and the strong wind prediction results.

[0119] The optimal tilt angle determination module 240 is used to calculate the optimal tilt angle adjustment strategy that maximizes the overall benefits and minimizes the risks of the photovoltaic power station based on the power generation revenue and risk losses of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, and the actual situation of the photovoltaic power station.

[0120] In the above embodiments, after obtaining the solar trajectory and the specification data of the target photovoltaic panel within a preset future time period, the optimal tilt angle of the target photovoltaic panel at different time intervals within the preset future time period can be determined based on the solar trajectory and specification data, thereby improving the solar energy capture efficiency. Next, this application can predict the strong wind conditions in the area where the target photovoltaic panel is located within the preset future time period using a pre-configured target strong wind prediction model, obtaining strong wind prediction results. Then, based on the optimal tilt angle and strong wind prediction results at different time intervals, the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals are evaluated, thereby ensuring the stability and safety of the system under severe weather conditions. Finally, this application can calculate the optimal tilt angle adjustment strategy that maximizes the overall revenue and minimizes the risk of the photovoltaic power station based on the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, and the actual situation of the photovoltaic power station. By dynamically adjusting the photovoltaic panel tilt angle, this application not only optimizes energy collection efficiency but also significantly enhances the wind resistance of the structure, achieving dual optimization of energy and safety. Furthermore, the implementation of this application provides an intelligent and automated solution for the photovoltaic industry, helping to promote the development and application of sustainable energy technologies.

[0121] In one embodiment, this application also provides a computer-readable storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of a photovoltaic panel tilt angle optimization method based on data fusion analysis as described in any of the above embodiments.

[0122] In one embodiment, this application also provides a computer device, including: one or more processors, and memory.

[0123] The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of a photovoltaic panel tilt angle optimization method based on data fusion analysis as described in any of the above embodiments.

[0124] Indicatively, such as Figure 3 As shown, Figure 3 This is a schematic diagram of the internal structure of a computer device 300 provided in an embodiment of this application. The computer device 300 can be provided as a server. (Refer to...) Figure 3The computer device 300 includes a processing component 302, which further includes one or more processors, and memory resources represented by memory 301 for storing instructions, such as application programs, that can be executed by the processing component 302. The application programs stored in memory 301 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 302 is configured to execute instructions to perform a photovoltaic panel tilt angle optimization method based on data fusion analysis according to any of the above embodiments.

[0125] The computer device 300 may also include a power supply component 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input / output (I / O) interface 305. The computer device 300 may operate on an operating system stored in memory 301, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0126] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0127] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0128] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0129] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for optimizing the tilt angle of photovoltaic panels based on data fusion analysis, characterized in that, The method includes: The solar trajectory and specification data of the target photovoltaic panel are obtained within a preset future time period. Based on the solar trajectory and specification data, the optimal tilt angle of the target photovoltaic panel at different time intervals within the preset future time period is determined. The strong wind conditions in the area where the target photovoltaic panel is located are predicted in the preset future time period by using a pre-configured target strong wind prediction model, and the strong wind prediction results are obtained. Based on the optimal tilt angle at different time intervals and the strong wind prediction results, evaluate the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals; Based on the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, and the actual situation of the photovoltaic power station, calculate the optimal tilt angle adjustment strategy that maximizes the overall revenue and minimizes the risk of the photovoltaic power station. The assessment of the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, based on the optimal tilt angle at different time intervals and the strong wind prediction results, includes: Establish a three-dimensional geometric model corresponding to the target photovoltaic panel; Based on the three-dimensional geometric model and the strong wind prediction results, the structural stress distribution of the target photovoltaic panel at the optimal tilt angle under different time intervals is calculated, and the structural stability of the target photovoltaic panel is evaluated based on the structural stress distribution. The Sandia model was used to calculate the theoretical power generation of the target photovoltaic panel at the optimal tilt angle under different time intervals, and the energy capture efficiency of the target photovoltaic panel was evaluated based on the theoretical power generation. A comprehensive evaluation of the structural stability and energy capture efficiency of the target photovoltaic panel is conducted to determine the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals.

2. The photovoltaic panel tilt angle optimization method based on data fusion analysis according to claim 1, characterized in that, The acquisition of the solar trajectory within a preset future time period includes: Obtain the pre-configured solar trajectory prediction model and the currently collected solar azimuth and solar altitude angles; Calculate the current sun position data based on the solar azimuth angle and the solar altitude angle; The current solar position data is input into the solar trajectory prediction model to obtain the solar trajectory within a preset future time period output by the solar trajectory prediction model.

3. The photovoltaic panel tilt angle optimization method based on data fusion analysis according to claim 1, characterized in that, The training process of the target strong wind prediction model includes: Hourly meteorological observation data of the area where the target photovoltaic panel is located within a preset historical period are obtained from the meteorological department, and the meteorological observation data is preprocessed to obtain a strong wind historical dataset. The preset initial strong wind prediction model is trained using the historical strong wind dataset, and when the preset training stopping condition is met, the trained initial strong wind prediction model is used as the target strong wind prediction model.

4. The photovoltaic panel tilt angle optimization method based on data fusion analysis according to claim 3, characterized in that, The step of training a pre-defined initial strong wind prediction model using the historical strong wind dataset includes: The correlation between different types of data in the historical strong wind dataset is analyzed by correlation analysis method, and wind speed, wind direction and air pressure gradient features are selected from the historical strong wind dataset as training data based on the analysis results. The training data is used to train a preset initial strong wind prediction model.

5. The photovoltaic panel tilt angle optimization method based on data fusion analysis according to claim 1, characterized in that, The optimal tilt angle adjustment strategy, which maximizes the overall benefit and minimizes the risk of the photovoltaic power station based on the power generation revenue and risk loss of the photovoltaic panel at the optimal tilt angle at different time intervals and the actual situation of the photovoltaic power station, includes: Based on the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, high-risk photovoltaic panels whose strong wind loss is greater than power generation revenue are identified. Based on the actual situation of the photovoltaic power station, a photovoltaic array layout optimization model is constructed with minimizing the number of high-risk photovoltaic panels as the optimization objective and the spacing and arrangement of photovoltaic panels as optimization variables. The high-risk photovoltaic panels are optimized using the photovoltaic array layout optimization model to obtain the optimal tilt angle adjustment strategy that maximizes the overall benefits and minimizes the risks of the photovoltaic power station.

6. The photovoltaic panel tilt angle optimization method based on data fusion analysis according to claim 1 or 5, characterized in that, The method further includes: Based on the optimal tilt angle of the target photovoltaic panel at different time intervals, analyze the changes in revenue and risk of adjusting the target photovoltaic panel to the optimal tilt angle at different time intervals; The optimal tilt angle adjustment timeliness strategy is determined based on the changes in revenue and risk associated with adjusting the target photovoltaic panel to the optimal tilt angle at different time intervals.

7. A photovoltaic panel tilt angle optimization device based on data fusion analysis, characterized in that, include: The optimal tilt angle prediction module is used to acquire the solar motion trajectory and the specification data of the target photovoltaic panel within a preset future time period, and to determine the optimal tilt angle of the target photovoltaic panel at different time intervals within the preset future time period based on the solar motion trajectory and the specification data. The strong wind prediction module is used to predict the strong wind conditions in the area where the target photovoltaic panel is located during the preset future time period using a pre-configured target strong wind prediction model, and obtain the strong wind prediction result. The revenue and risk assessment module is used to assess the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, based on the optimal tilt angle at different time intervals and the strong wind prediction results. The optimal tilt angle determination module is used to calculate the optimal tilt angle adjustment strategy that maximizes the overall benefits and minimizes the risks of the photovoltaic power station based on the power generation revenue and risk losses of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals, as well as the actual situation of the photovoltaic power station. The benefit and risk assessment module includes: Establish a three-dimensional geometric model corresponding to the target photovoltaic panel; Based on the three-dimensional geometric model and the strong wind prediction results, the structural stress distribution of the target photovoltaic panel at the optimal tilt angle under different time intervals is calculated, and the structural stability of the target photovoltaic panel is evaluated based on the structural stress distribution. The Sandia model was used to calculate the theoretical power generation of the target photovoltaic panel at the optimal tilt angle under different time intervals, and the energy capture efficiency of the target photovoltaic panel was evaluated based on the theoretical power generation. A comprehensive evaluation of the structural stability and energy capture efficiency of the target photovoltaic panel is conducted to determine the power generation revenue and risk loss of the photovoltaic power station when the target photovoltaic panel is at the optimal tilt angle at different time intervals.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the photovoltaic panel tilt angle optimization method based on data fusion analysis as described in any one of claims 1 to 6.

9. A computer device, characterized in that, include: One or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the photovoltaic panel tilt angle optimization method based on data fusion analysis as described in any one of claims 1 to 6.

Images