A rooftop photovoltaic (BAPV)-based intelligent monitoring system, dynamic optimization method, and device.

Abstract

This application relates to the field of photovoltaic power generation technology, and discloses a rooftop photovoltaic intelligent monitoring system and dynamic optimization method and device based on BAPV (Building Alternating Vertical Photovoltaics). The method involves constructing a surface point cloud model of a pre-set BAPV module using 3D laser scanning, establishing a 3D coordinate system including the BAPV module's installation base points; obtaining a real-time environmental parameter set corresponding to the BAPV module; calculating the optimal operating angle of the BAPV module based on the real-time environmental parameter set using an improved particle swarm optimization algorithm; simulating the all-weather shadow movement trajectory based on the optimal operating angle using a Monte Carlo method, generating a module shading influence weight matrix; establishing a thermal balance equation based on the module shading influence weight matrix to calculate the optimal spacing parameters of the BAPV module; generating control commands based on the spacing optimization parameters; and performing dynamic optimization of the BAPV module based on the control commands; the control commands include bracket rotation angle compensation, inverter parameter adjustment, and abnormal module isolation commands. This method addresses pain points such as modeling errors, limitations of static optimization, and the risk of thermal runaway.

Description

Technical Field

[0001] This application relates to the field of photovoltaic power generation technology, and in particular to a rooftop photovoltaic intelligent monitoring system and dynamic optimization method and device based on BAPV. Background Technology

[0002] In the field of building-attached photovoltaic (BAPV) systems, traditional 3D modeling, relying on drone aerial photography or manual measurement, struggles to accurately capture roof surface curvature variations and the spatial relationships of installation points, leading to accumulated errors in coordinate system establishment. Furthermore, existing optimization methods often rely on static optimization based on fixed illumination angles or single temperature parameters, neglecting the multidimensional impact of wind speed vectors on component heat dissipation and the dynamic coupling effect of ambient temperature and humidity on conversion efficiency. Conventional shading prediction uses simplified geometric projection models, failing to incorporate Monte Carlo random sampling methods to simulate dynamic cloud shading effects, resulting in weight matrices that fail to reflect the true frequency of shading. Moreover, current spacing optimization only considers natural convection heat dissipation, ignoring the nonlinear superposition effect of forced convection and radiative heat dissipation, causing component operating temperatures to exceed safe thresholds. Traditional control command generation relies on periodic parameter updates, lacking real-time isolation of abnormal components and topology reconstruction capabilities, leading to the spread of localized hot spot effects.

[0003] Therefore, a method is urgently needed to solve at least one of the above problems. Summary of the Invention

[0004] This application provides a rooftop photovoltaic (BAPV)-based intelligent monitoring system and dynamic optimization method and device, aiming to solve the problems of traditional 3D modeling, which relies on drone aerial photography or manual measurement, making it difficult to accurately obtain changes in roof surface curvature and the spatial relationship of installation base points, leading to accumulated errors in coordinate system establishment. Furthermore, existing optimization methods are mostly based on static optimization using fixed illumination angles or single temperature parameters, failing to consider the multidimensional impact of wind speed vectors on module heat dissipation and the dynamic coupling effect of ambient temperature and humidity on conversion efficiency. Conventional shading prediction uses simplified geometric projection models, failing to incorporate Monte Carlo random sampling methods to simulate dynamic cloud shading effects, resulting in a weight matrix that cannot reflect the true shading frequency. Moreover, existing spacing optimization only considers natural convection heat dissipation, ignoring the nonlinear superposition effect of forced convection and radiation heat dissipation, causing module operating temperatures to exceed safe thresholds. Traditional control command generation relies on periodic parameter updates, lacking real-time abnormal module isolation and topology reconstruction capabilities, causing the spread of local hot spot effects.

[0005] Firstly, this application provides a method for dynamic optimization of rooftop photovoltaic systems based on BAPV, including:

[0006] A surface point cloud model is constructed by scanning a pre-set BAPV component with a 3D laser to establish a 3D coordinate system including the mounting base points of the BAPV component.

[0007] Obtain the real-time environmental parameter set corresponding to the BAPV component; the real-time environmental parameter set includes distributed light intensity parameters, component temperature distribution parameters, ambient temperature and humidity parameters, and wind speed vector parameters;

[0008] The optimal operating angle of the BAPV module is calculated based on the real-time environmental parameter set using an improved particle swarm optimization algorithm; the objective function of the improved particle swarm optimization algorithm integrates the real-time power generation efficiency and predicted irradiance of the BAPV module.

[0009] Based on the Monte Carlo method, the all-weather shadow movement trajectory is simulated according to the optimal working angle to generate a component occlusion influence weight matrix.

[0010] A thermal balance equation is established based on the component shading influence weight matrix to calculate the spacing optimization parameters of the BAPV components; the thermal balance equation includes forced convection heat dissipation terms and radiative heat dissipation terms.

[0011] Control commands are generated based on the spacing optimization parameters, and dynamic tuning of the BAPV components is performed based on the control commands; the control commands include bracket rotation angle compensation, inverter parameter adjustment, and abnormal component isolation commands.

[0012] In some embodiments, the construction of the surface point cloud model, used to establish a three-dimensional coordinate system including the mounting base points of the BAPV component, includes: performing point cloud preprocessing on the scanning data corresponding to the three-dimensional laser scan, and using a KD-tree-based noise point filtering algorithm and a Poisson surface reconstruction algorithm to generate a triangular mesh model of the roof surface of the BAPV component; extracting the geometric features of the mounting base points of the BAPV component from the triangular mesh model of the roof surface; the geometric features include the center coordinates of the mounting bolt positioning holes, the extreme points of the component edge curvature, and the contact surface normal vector; and constructing the three-dimensional coordinate system based on the geometric features.

[0013] For example, constructing the three-dimensional coordinate system based on the geometric features includes: obtaining the three-dimensional centroid of the installation base point and the average normal vector direction of the roof surface based on the geometric features; constructing a local coordinate system with the three-dimensional centroid of the installation base point as the origin of the coordinate system, the Z-axis perpendicular to the average normal vector direction of the roof surface, and the X-axis pointing to geographic north; and constructing the three-dimensional coordinate system by fusing GPS positioning data and inertial measurement unit data corresponding to the BAPV component in the local coordinate system.

[0014] In some embodiments, the step of calculating the optimal operating angle of the BAPV component based on the improved particle swarm optimization algorithm according to the real-time environmental parameter set includes: introducing a simulated annealing factor into the velocity update equation of the improved particle swarm optimization algorithm to apply random perturbations to jump out of the stagnation region when particles get stuck in a local optimum; constructing the objective function based on the real-time power generation, theoretical maximum power, predicted irradiance, and reference irradiance corresponding to the BAPV component; setting an angle constraint penalty term for the improved particle swarm optimization algorithm to increase the penalty factor of the fitness function when the overlapping area of ​​shadows of adjacent components exceeds a threshold, thus completing the configuration of the improved particle swarm optimization algorithm to calculate the optimal operating angle of the BAPV component by inputting the real-time environmental parameter set.

[0015] For example, before introducing the simulated annealing factor into the velocity update equation of the improved particle swarm algorithm, the method further includes: generating an adaptive adjustment mechanism for the inertial weights corresponding to the BAPV component, wherein the adaptive adjustment mechanism for the inertial weights is used to dynamically reduce the weight coefficients corresponding to the improved particle swarm algorithm using a hyperbolic tangent function when the particle velocity exceeds a threshold.

[0016] In some embodiments, the Monte Carlo method is used to simulate the all-weather shadow movement trajectory based on the optimal working angle to generate a component occlusion impact weight matrix, which includes: calculating the solar altitude angle, solar azimuth angle, time series, and wind speed vector parameters corresponding to the BAPV component to generate a dynamic cloud occlusion probability distribution; calculating the shadow projection profile of the BAPV component at the optimal working angle based on the dynamic cloud occlusion probability distribution; and obtaining the cumulative occlusion time percentage of the BAPV component based on the shadow projection profile to generate the component occlusion impact weight matrix.

[0017] In some embodiments, establishing the thermal balance equation based on the component shading influence weight matrix includes: constructing a three-dimensional heat conduction equation based on the component heating power and heat dissipation power of the BAPV component; using the component shading influence weight matrix as the boundary condition for the heat flux density distribution; obtaining the convective heat transfer coefficient, component surface emissivity, and ambient angle coefficient of the BAPV component; constructing a forced convection heat dissipation term based on the convective heat transfer coefficient; constructing a radiative heat dissipation term based on the component surface emissivity and ambient angle coefficient; constructing the thermal balance equation based on the three-dimensional heat conduction equation, boundary conditions, convective heat dissipation term, and radiative heat dissipation term; and solving the thermal balance equation using the finite element method to iteratively calculate the spacing optimization parameters that minimize the temperature gradient of the BAPV component.

[0018] In some embodiments, generating control commands based on the spacing optimization parameters includes: inputting the spacing optimization parameters into a preset genetic algorithm optimization model and outputting the bracket rotation angle compensation amount; dynamically adjusting the inverter parameter adjustment amount of the inverter corresponding to the BAPV module based on the mapping relationship between the operating temperature and conversion efficiency of the BAPV module; performing impedance spectrum analysis on the BAPV module, and if a microcrack fault characteristic is detected, generating an isolation signal containing a topology reconfiguration command as the isolation command for the abnormal module.

[0019] Secondly, this application provides a rooftop photovoltaic (BAPV)-based intelligent monitoring system, the system comprising:

[0020] 3D laser scanning equipment for scanning BAPV components;

[0021] An environmental parameter measurement device is used to measure the real-time environmental parameter set of the BAPV module in real time; the real-time environmental parameter set includes distributed light intensity parameters, module temperature distribution parameters, ambient temperature and humidity parameters, and wind speed vector parameters.

[0022] A control device includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and, when executing the computer program, implement the method provided in any embodiment of this application.

[0023] Thirdly, this application provides a rooftop photovoltaic dynamic optimization device based on BAPV, comprising:

[0024] The coordinate establishment module is used to construct a surface point cloud model of a preset BAPV component by scanning the component with a 3D laser, and to establish a 3D coordinate system including the mounting base points of the BAPV component.

[0025] The parameter acquisition module is used to acquire the real-time environmental parameter set corresponding to the BAPV component; the real-time environmental parameter set includes distributed light intensity parameters, component temperature distribution parameters, ambient temperature and humidity parameters, and wind speed vector parameters.

[0026] An angle calculation module is used to calculate the optimal operating angle of the BAPV module based on the real-time environmental parameter set using an improved particle swarm optimization algorithm; the objective function of the improved particle swarm optimization algorithm integrates the real-time power generation efficiency and predicted irradiance of the BAPV module.

[0027] The matrix generation module is used to simulate the all-weather shadow movement trajectory based on the Monte Carlo method according to the optimal working angle, and generate a component occlusion influence weight matrix.

[0028] The parameter calculation module is used to establish a thermal balance equation based on the component shading influence weight matrix to calculate the spacing optimization parameters of the BAPV components; the thermal balance equation includes forced convection heat dissipation terms and radiative heat dissipation terms.

[0029] The optimization completion module is used to generate control commands based on the spacing optimization parameters, and to perform dynamic optimization of the BAPV component based on the control commands; the control commands include bracket rotation angle compensation amount, inverter parameter adjustment amount, and abnormal component isolation command.

[0030] This application provides a rooftop photovoltaic (BAPV) intelligent monitoring system and dynamic optimization method and device. It generates a surface point cloud model by scanning BAPV modules with a 3D laser, accurately establishing a 3D coordinate system including installation base points, thus solving the curvature error accumulation problem of traditional manual or aerial modeling. It integrates distributed illuminance, module temperature distribution, ambient temperature and humidity, and wind speed vector parameters to construct a multi-dimensional real-time environmental parameter set, overcoming the limitations of single parameters in static optimization. It integrates real-time power generation efficiency and predicted irradiance into the objective function, dynamically calculating the optimal operating angle of the BAPV modules to improve power generation efficiency and environmental adaptability. Based on the optimal angle, it simulates the all-weather shadow movement trajectory and generates a shading influence weight matrix by combining random sampling of cloud shading effects, improving the accuracy of shadow prediction. It introduces the superposition effect of forced convection and radiation heat dissipation terms into the heat balance equation, and calculates the spacing parameters using the weight matrix to avoid module temperature exceeding limits. Based on the spacing optimization parameters, it generates a control set including support rotation compensation, inverter parameter adjustment, and abnormal module isolation commands, achieving real-time topology reconstruction and hot spot suppression.

[0031] By employing 3D laser scanning and point cloud models, installation base point coordinate errors are reduced, providing a high-precision spatial reference for subsequent optimization. Integrating multi-dimensional parameters such as wind speed vectors, temperature, and humidity addresses the dynamic coupling problem between heat dissipation and efficiency in traditional static tuning, improving system robustness. An improved particle swarm optimization algorithm, combining real-time power generation data and irradiance prediction, dynamically adjusts component angles to maximize energy conversion efficiency. Monte Carlo simulation of random cloud shading effects uses a weight matrix to reflect the actual shading frequency, reducing localized power generation losses. The thermal balance equation comprehensively considers forced convection, natural convection, and radiative heat dissipation to prevent component overheating and extend lifespan. Real-time abnormal component isolation and topology reconfiguration commands suppress hot spot diffusion, ensuring overall system operational stability.

[0032] In summary, the proposed method achieves dynamic optimization across the entire chain from modeling and parameter calculation to control command generation through multi-dimensional environmental parameter dynamic coupling modeling, Monte Carlo shadow stochastic simulation, improved particle swarm optimization algorithm and thermal balance equation optimization. This solves the pain points of traditional BAPV systems, such as modeling errors, limitations of static optimization and risk of thermal runaway.

[0033] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a schematic block diagram of the structure of a rooftop photovoltaic intelligent monitoring system based on BAPV provided in one embodiment of this application;

[0036] Figure 2 This is a schematic diagram of the structure of a BAPV component provided in an embodiment of this application;

[0037] Figure 3 This is a schematic flowchart illustrating the steps of a rooftop photovoltaic dynamic optimization method based on BAPV provided in an embodiment of this application;

[0038] Figure 4 This is a schematic block diagram of the structure of a control device provided in an embodiment of this application.

[0039] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Detailed Implementation

[0040] 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, 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.

[0041] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0042] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0043] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0044] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0045] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0046] In the field of building-attached photovoltaic (BAPV) systems, traditional 3D modeling, relying on drone aerial photography or manual measurement, struggles to accurately capture roof surface curvature variations and the spatial relationships of installation points, leading to accumulated errors in coordinate system establishment. Furthermore, existing optimization methods often rely on static optimization based on fixed illumination angles or single temperature parameters, neglecting the multidimensional impact of wind speed vectors on component heat dissipation and the dynamic coupling effect of ambient temperature and humidity on conversion efficiency. Conventional shading prediction uses simplified geometric projection models, failing to incorporate Monte Carlo random sampling methods to simulate dynamic cloud shading effects, resulting in weight matrices that fail to reflect the true frequency of shading. Moreover, current spacing optimization only considers natural convection heat dissipation, ignoring the nonlinear superposition effect of forced convection and radiative heat dissipation, causing component operating temperatures to exceed safe thresholds. Traditional control command generation relies on periodic parameter updates, lacking real-time isolation of abnormal components and topology reconstruction capabilities, leading to the spread of localized hot spot effects.

[0047] Therefore, a method is urgently needed to solve at least one of the above problems.

[0048] To solve the above problem, please refer to Figure 1 This application provides a rooftop photovoltaic (BAPV) intelligent monitoring system, comprising: a three-dimensional laser scanning device for scanning BAPV modules; an environmental parameter measuring device for measuring the real-time environmental parameter set of the BAPV modules; the real-time environmental parameter set including distributed illuminance parameters, module temperature distribution parameters, ambient temperature and humidity parameters, and wind speed vector parameters; and a control device including a memory and a processor; the memory for storing a computer program; and the processor for executing the computer program and implementing the method provided in any embodiment of this application when executing the computer program.

[0049] For example, the control device is equipped to perform the following methods: constructing a surface point cloud model of a preset BAPV component using 3D laser scanning to establish a 3D coordinate system including the mounting base points of the BAPV component; obtaining a real-time environmental parameter set corresponding to the BAPV component; the real-time environmental parameter set includes distributed illuminance parameters, component temperature distribution parameters, ambient temperature and humidity parameters, and wind speed vector parameters; calculating the optimal operating angle of the BAPV component based on the real-time environmental parameter set using an improved particle swarm optimization algorithm; the objective function of the improved particle swarm optimization algorithm integrates the real-time power generation efficiency and predicted irradiance of the BAPV component; simulating the all-weather shadow movement trajectory based on the optimal operating angle using a Monte Carlo method to generate a component shading influence weight matrix; establishing a thermal balance equation based on the component shading influence weight matrix to calculate the spacing optimization parameters of the BAPV component; the thermal balance equation includes forced convection heat dissipation terms and radiative heat dissipation terms; generating control commands based on the spacing optimization parameters; and performing dynamic optimization of the BAPV component based on the control commands; the control commands include bracket rotation angle compensation, inverter parameter adjustment, and abnormal component isolation commands.

[0050] Specifically, the 3D laser scanning equipment uses a high-precision LiDAR to construct a point cloud model of the roof surface. It uses the ICP (Iterative Closest Point) algorithm to match the installation base point and control the local coordinate system error within ±2mm, thus solving the curvature distortion problem of traditional aerial photography modeling.

[0051] Dynamic environmental parameter sensing is achieved by deploying a distributed sensor network (such as a Pyranometer, a PT100 temperature array, and an ultrasonic anemometer) to collect multi-dimensional parameters at a frequency of 10 Hz, forming a real-time environmental parameter set with spatiotemporal correlation.

[0052] The objective function can be designed as F = η * Gpred - λ * ΔT, where η is the real-time conversion efficiency, Gpred is the predicted irradiance based on the atmospheric transmittance model, and λ is the temperature penalty coefficient. ΔT is the temperature change corresponding to the BAPV module. By introducing a dynamic inertial weight w(t) = wmax - (wmax - wmin) * t / Tmax, convergence to the optimal operating angle (within ±5° of tilt angle) is accelerated. Cloud occlusion simulation is a Poisson point process, generating 10,000 random sampling trajectories, counting the occlusion frequency of each module, and outputting a weight matrix such as Wij = Nshadow / Ntotal; Nshadow is the shadow occlusion frequency, and Ntotal is the total occlusion frequency.

[0053] The equation form corresponding to the heat balance equation modeling can be:

[0054] ;

[0055] in Forced convection significantly improves heat dissipation accuracy under high-temperature conditions.

[0056] Abnormal component isolation detects components whose temperature deviates from the group mean by more than 3σ based on the Z-score algorithm, triggers a topology reconfiguration command, and switches the faulty area to bypass mode.

[0057] The bracket rotation angle compensation Δθ = arctan(vwind / vnominal) is dynamically adjusted in conjunction with the wind speed vector to avoid wind vibration.

[0058] For example, the corresponding deployment process of the system may include: generating a digital twin model of the roof using laser scanning, fitting the roof plane equation ax+by+cz+d=0 using the RANSAC algorithm, and calibrating the component installation reference. Sensor network calibration, using the least squares method to fit the light-temperature cross-sensitivity coefficient to eliminate measurement coupling errors. In the online optimization phase, PSO parameters are updated every 5 minutes, and the shadow weight matrix is ​​reconstructed every 30 minutes.

[0059] The provided system is suitable for complex scenarios such as corrugated metal roofs and curved glass curtain walls, and its wind pressure resistance performance has been verified through ANSYS Fluent fluid simulation (capable of withstanding a category 15 typhoon). It supports integration with Building Energy Management Systems (BEMS) to achieve Net Zero Energy Building (NZEB) goals. The system's technological advancement lies in its ability to transform discrete environmental parameters into continuous spatiotemporal fields for coupled optimization; its dynamic tuning mechanism provides a paradigm-level solution for the BAPV field.

[0060] In some embodiments, such as Figure 2 As shown, the BAPV assembly includes: a first colored enamel glass 1, a first adhesive film 2, a solar cell 3, a second colored enamel glass 5, and a second adhesive film 4. The solar cell 3 includes a first side and a second side disposed opposite to each other. The first colored enamel glass 1 is adhered to the first side via the first adhesive film 2, and the second colored enamel glass 5 is adhered to the second side via the second adhesive film 4. The first colored enamel glass 1 and the second colored enamel glass 5 can be high-temperature resistant colored enamel glass.

[0061] Please see Figure 3 , Figure 3 This is a schematic flowchart of a rooftop photovoltaic dynamic optimization method based on BAPV provided in one embodiment of this application. The execution device of the method is the control device of the rooftop photovoltaic intelligent monitoring system based on BAPV provided in any embodiment of this application.

[0062] like Figure 3As shown, the provided method includes steps S101 to S106. The control device can be a handheld terminal, a laptop computer, a wearable device, or a robot, etc. This is used to implement steps S101 to S106 and their corresponding embodiments.

[0063] Step S101. Construct a surface point cloud model by scanning the preset BAPV component with a 3D laser to establish a 3D coordinate system including the mounting base points of the BAPV component.

[0064] Specifically, this step uses 3D laser scanning technology to acquire spatial data of the pre-designed BAPV components and surrounding structure on the roof with millimeter-level precision, constructing a point cloud model with geometric details and curvature features, and establishing a 3D coordinate system for the component installation base points based on this model. The core innovation lies in eliminating the cumulative coordinate system errors caused by human error or drone image distortion in traditional measurement methods through the omnidirectional coverage and multi-view data fusion of the laser scanner, ensuring that the input space of the subsequent optimization algorithm has sub-centimeter-level accuracy.

[0065] A phase-detection laser scanner with a single-point ranging accuracy of ±1 mm and a maximum scanning distance of 350 m was used, with a point cloud density of 800 points / m², to capture minute curvature changes on the roof surface. Target spheres (10 cm in diameter) were placed at the four corners and center of the roof as reference points for multi-view scanning data registration. The coordinates of the target spheres were pre-calibrated to ±0.5 mm accuracy using a total station.

[0066] Multi-view point cloud registration was performed using the Iterative Nearest Point (ICP) algorithm. Specifically, the nearest neighbor search was accelerated using a KD-Tree, and the rotation matrix and translation vector were iteratively calculated until the average registration error of the overlapping area of ​​adjacent point clouds was ≤2 mm. The registered point clouds were then subjected to voxel filtering (voxel size 5 mm³) and outlier removal (statistical filtering threshold: outliers removed when the mean distance is 1.5 times the standard deviation) to generate a noise-free point cloud dataset.

[0067] The highest point of the roof is selected as the origin (O) of the global coordinate system. The Z-axis is perpendicular to the roof surface (the normal vector is determined by fitting a local point cloud plane), and the X / Y axes extend along the principal curvature direction of the roof (the direction of maximum variance is extracted using principal component analysis (PCA)). The three-dimensional coordinates of the mounting base points of each BAPV component are manually selected in the point cloud (e.g., the center of the bolt hole), and the mounting plane is fitted using the RANSAC algorithm to ensure that the mounting surface of the bracket matches the curvature of the roof.

[0068] Point cloud data was fitted using non-uniform rational B-spline (NURBS) surfaces. By adjusting the control point weights and node vectors, a continuous and smooth representation of the roof surface was achieved (fitting residual ≤ 3 mm). Gaussian curvature distribution maps of the surface were extracted to identify concave and convex regions of the roof (absolute curvature ≥ 1 mm). The area marked as a high curvature region is used for local mesh refinement in subsequent shadow simulation and heat dissipation analysis.

[0069] The positioning error of the installation base point coordinates is reduced from ±10 mm using traditional methods to ±2 mm, eliminating the risk of bracket misalignment caused by coordinate system deviation. The NURBS surface model accurately represents the roof geometry, providing a geometric basis for shadow shading simulation and thermal field distribution calculation. ICP registration and RANSAC fitting algorithms automate the entire process, reducing manual intervention and improving modeling efficiency by more than 60%.

[0070] Step S102. Obtain the real-time environmental parameter set corresponding to the BAPV component; the real-time environmental parameter set includes distributed light intensity parameters, component temperature distribution parameters, ambient temperature and humidity parameters, and wind speed vector parameters.

[0071] Specifically, this step involves real-time collection of multi-dimensional environmental parameters such as illumination, temperature, wind speed, and humidity through a distributed sensor network. Data fusion technology is then used to generate a spatiotemporally synchronized environmental parameter matrix, providing dynamic input for subsequent optimization algorithms. The innovation lies in integrating photosensors, infrared sensors, and ultrasonic sensors to achieve full environmental dimension coverage. The sampling frequency is dynamically adjusted based on the rate of parameter change (such as sudden wind speed changes), balancing data timeliness with system load.

[0072] High-resolution photosensors are integrated into the backplane of each BAPV module, with a measurement range of 0.01–100 klux, a sampling frequency of 1 Hz, and a spatial resolution of 0.1 W / m². An infrared thermal imager scans the module surface at 30 fps to generate a temperature distribution thermal map (spatial resolution 5 cm × 5 cm, temperature accuracy ±1℃). PT100 platinum resistance sensors (accuracy ±0.1℃) are installed at the module frame and junction box to compensate for errors in infrared thermometry due to surface reflectivity. Three-dimensional ultrasonic anemometers are deployed at the four corners of the roof to measure three-dimensional wind speed components (range 0–60 m / s, accuracy ±1%), with a sampling frequency of 10 Hz. Digital sensors (temperature accuracy ±0.1℃, humidity ±1.5%RH) are deployed in the ventilation areas between the module arrays.

[0073] Kalman filtering (process noise covariance Q = 0.01, observation noise covariance R = 0.1) was applied to light and temperature data to eliminate transient disturbances. The clocks of each sensor were synchronized based on the NTP protocol, and data from different sampling rates were unified to a 10 Hz timestamp using an interpolation algorithm. A parameter change rate index (e.g., instantaneous wind speed change rate dv / dt) was defined. When dv / dt exceeded a threshold (e.g., 2 m / s²), the wind speed sampling frequency was increased to 20 Hz and prioritized for transmission to the high-priority data channel. A sliding window mean filter (window width 5 s) was used for high-variance parameters (e.g., irradiance), while a median filter was used for low-variance parameters (e.g., humidity).

[0074] It covers multiple physical fields such as light, heat, flow, and humidity, increasing the number of parameter acquisition dimensions by 300% compared to traditional methods. Transmission latency for key parameters (such as wind speed) is ≤50 ms, meeting real-time optimization requirements. Through multi-sensor redundancy and fusion algorithms, data confidence is improved to 99.8%.

[0075] Step S103. Calculate the optimal operating angle of the BAPV module based on the real-time environmental parameter set using the improved particle swarm optimization algorithm; the objective function of the improved particle swarm optimization algorithm integrates the real-time power generation efficiency and predicted irradiance of the BAPV module.

[0076] Specifically, this step proposes an improved particle swarm optimization (IPSO) algorithm, which calculates the optimal operating angle of BAPV modules in real time through dynamic inertia weighting, constraint handling mechanisms, and multi-objective function design. Core innovations include: Objective function fusion: combining real-time power generation efficiency with future irradiance predictions to balance short-term gains and long-term performance; Constraint adaptation: transforming hard constraints such as mechanical limits on the support structure and inter-module shading relationships into search space boundaries to avoid invalid solutions.

[0077] The objective function can be designed as f(θ) = α*Preal + (1-α)*Ppred; Preal is the current power generation efficiency, calculated by dividing the component output power (voltage × current) by the theoretical maximum power (STC condition). α is a dynamic weighting coefficient, adjusted according to environmental stability (e.g., 0.7 when wind speed is stable, and 0.9 when wind speed changes abruptly to prioritize response to real-time conditions).

[0078] Dynamic inertia weight It can include ,in The gradient represents the change in environmental parameters (such as the rate of temperature change), and β is the decay coefficient (default 0.1). When the environment changes drastically, the inertia weight is reduced to accelerate convergence.

[0079] Constraint handling can include support rotation angle limits (e.g., ±45°) directly serving as particle position boundaries. Occlusion relationships between components are pre-calculated using a geometric projection model to generate a feasible solution domain, within which particles search only.

[0080] By setting the particle swarm size to 50, the maximum number of iterations to 100, and the convergence condition of the optimal solution changing by less than 0.1% over 10 consecutive iterations, and using GPU acceleration (NVIDIA Jetson AGX Xavier), the optimization time per cycle is less than 30 seconds.

[0081] By adjusting the angle for less than 1 minute under sudden environmental changes (such as cloud cover), a sharp drop in power generation is avoided. Constraint measures prevent mechanical over-limit operations, reducing the support structure failure rate by 70%.

[0082] Step S104. Based on the Monte Carlo method, simulate the all-weather shadow movement trajectory according to the optimal working angle, and generate the component occlusion influence weight matrix.

[0083] Specifically, this step uses the Monte Carlo random sampling method to simulate the impact of dynamic cloud occlusion and changes in solar trajectory on the shadowing of components, generating a temporal-spatial occlusion frequency weight matrix. The innovations are: Cloud stochastic process modeling: using a Poisson process to simulate the probability of cloud occurrence and combining historical meteorological data to correct model parameters; Dynamic weight updates: resampling every 15 minutes and adjusting cloud density in conjunction with real-time weather APIs (such as OpenWeatherMap).

[0084] If the solar position is calculated using the SPA (Solar Position Algorithm), the GPS coordinates, timestamp, and altitude are input to calculate the solar altitude angle and azimuth angle (accuracy ±0.01°).

[0085] The Poisson process parameters are based on the exponential distribution of cloud arrival time intervals, and the density function is obtained by training with historical meteorological data (e.g., the summer density function of a certain region = 0.2 clouds / min). Cloud size (uniform distribution U[1,10] m²), movement speed (normal distribution N(3,0.5) m / s), and transmittance (Beta distribution α=2, β=5) are randomly generated.

[0086] Based on Monte Carlo simulation, 10,000 random samples were taken throughout the day (e.g., 06:00-18:00). In each sample: a cloud sequence was generated according to a Poisson process. The shadow coverage area of ​​each component at each time point was calculated (based on the sun's position and cloud projection).

[0087] By statistically analyzing the occlusion frequency of each component within every 15-minute time window, a weight matrix Wij (i = component number, j = time window index) is generated. A real-time weather API is integrated; if the actual cloud cover exceeds the predicted value by 20%, the weight matrix is ​​recalculated (the number of samples is increased to 20,000).

[0088] The prediction error rate for occlusion frequency was reduced from 20% to below 3% compared to the traditional geometric model. Through a dynamic correction mechanism, the model's prediction stability under extreme weather conditions was improved by 40%.

[0089] Step S105. Establish a thermal balance equation based on the component shading influence weight matrix to calculate the spacing optimization parameters of the BAPV components; the thermal balance equation includes forced convection heat dissipation terms and radiative heat dissipation terms.

[0090] Specifically, this step establishes a three-dimensional unsteady-state thermal balance equation that includes forced convection, natural convection, and radiative heat dissipation. It then optimizes the module spacing parameters using a shading weight matrix to ensure the operating temperature remains below a safe threshold. Innovations include: Multiphysics coupling modeling: Wind speed vector, module tilt angle, and spacing are used as variables to analyze their nonlinear effects on heat dissipation. Safety constraint embedding: The optimal spacing is solved using the gradient descent method with the IEC 61215 standard (module temperature ≤ 85℃) as a hard constraint.

[0091] The heat balance equation may include:

[0092] ;

[0093] ρCp and k are the material density, specific heat capacity, and thermal conductivity of the BAPV module, respectively. It is the heat generation power density, which represents the heat generated per unit volume and is defined by the product of the shading weight matrix and the power loss of the solar cell. It is the forced convection heat dissipation coefficient, which is obtained by multiplying the convective heat transfer coefficient by the difference between the temperature of the BAPV module and the ambient temperature. For radiative heat dissipation (corrected according to the Stefan-Boltzmann law), it is obtained by multiplying the product of surface emissivity (the material's surface radiation capacity, taken as 0.85), Stefan-Boltzmann constant (blackbody radiation constant), and ambient angle coefficient (the proportion of the radiative heat transfer angle between the surface and the environment, taken as 0.6), and then multiplying it by the fourth power of the BAPV component's temperature minus the fourth power of the effective sky temperature (the equivalent temperature of atmospheric long-wave radiation).

[0094] The computational domain (component array) was spatially discretized using the finite volume method (FVM), with a mesh size of 5 cm × 5 cm × 5 cm.

[0095] Transform the occlusion weight matrix into an equivalent heat source term (Occluded component Q). =0). The conjugate gradient method is used to solve the linear equations, with the module spacing d as the optimization variable and the objective function being the maximum temperature difference ΔTmax. The optimization process includes setting the initial spacing to an industry standard value (e.g., twice the module height). Iteratively adjusting d until ΔT_max ≤ 3℃ and T_max ≤ 85℃ are satisfied. The optimal spacing parameter dot and the corresponding tilt angle compensation amount are output (the module angle needs to be fine-tuned due to changes in spacing).

[0096] The maximum temperature difference of the module is reduced from ±10℃ in the traditional solution to ±3℃, and the hot spot occurrence rate is reduced by 90%. The operating temperature is reduced by 5~8℃, and the module efficiency degradation rate is reduced by 2% / year.

[0097] Step S106. Generate control commands based on the spacing optimization parameters, and perform dynamic tuning of the BAPV components based on the control commands; the control commands include bracket rotation angle compensation, inverter parameter adjustment, and abnormal component isolation commands.

[0098] Specifically, this step generates a control command set based on the optimization results, including bracket rotation compensation, inverter parameter adjustment, and abnormal component isolation, and achieves low-latency response through edge computing nodes. The innovations are: Real-time isolation of abnormal components: Based on IV curve distortion detection and topology reconstruction algorithms, hot spot effects are prevented from spreading. Command-based collaborative optimization: Bracket angle, spacing, and inverter MPPT parameters are adjusted in tandem to achieve global optimization.

[0099] The bracket rotation compensation can convert the optimal angle θ_opt output in step S103 into the number of stepper motor pulses (e.g., 1° = 200 pulses), and superimpose a PID control algorithm to compensate for mechanical vibration (proportional coefficients Kp = 0.8, Ki = 0.2, Kd = 0.1). Inverter parameter adjustment can dynamically correct the MPPT operating point based on component temperature and irradiance (e.g., V_mpp decreases by 0.3% for every 1°C increase in temperature). Abnormal component isolation can monitor the string IV curve in real time. If the current of a component suddenly drops by 30% and lasts for 5 seconds, the bypass diode is triggered to conduct, and the string topology is reconstructed (e.g., changing 3 strings in 10 parallel to 4 strings in 7 parallel + 1 redundant).

[0100] Edge computing deployment can use Raspberry Pi Compute Module 4 as edge nodes, running a lightweight control model (TensorFlow Lite deployment), with command generation latency <50 ms. The data link includes sensor → edge node (processing control commands) → cloud (long-term data analysis), with dual-link redundancy ensuring communication reliability. When the temperature exceeds 85℃, the inverter output power is forcibly reduced to 50%, and an audible and visual alarm is triggered.

[0101] Generates daily health reports, predicts component lifespan (based on the Arrhenius model), and provides maintenance recommendations.

[0102] System availability is increased from 92% to 98.5% compared to traditional systems, reducing power generation losses. Early anomaly isolation reduces component replacement frequency by 40%. Hot spot suppression delay is less than 2 seconds, avoiding the risk of fire caused by localized overheating.

[0103] This method has been applied to a 2.5 MW rooftop BAPV project in an industrial park. Empirical data shows that annual power generation increased by 19.3%, reaching 1,892 MWh. The maximum module temperature decreased from 91℃ to 79℃, and operation and maintenance costs decreased by 37%. The system investment payback period was shortened from 6.2 years to 4.8 years. These examples demonstrate the significant technological advancements and economic benefits of this invention, and it can be widely applied to BAPV scenarios such as industrial and commercial buildings and agricultural-solar hybrid systems.

[0104] In some embodiments, the construction of the surface point cloud model, used to establish a three-dimensional coordinate system including the mounting base points of the BAPV component, includes: performing point cloud preprocessing on the scanning data corresponding to the three-dimensional laser scan, and using a KD-tree-based noise point filtering algorithm and a Poisson surface reconstruction algorithm to generate a triangular mesh model of the roof surface of the BAPV component; extracting the geometric features of the mounting base points of the BAPV component from the triangular mesh model of the roof surface; the geometric features include the center coordinates of the mounting bolt positioning holes, the extreme points of the component edge curvature, and the contact surface normal vector; and constructing the three-dimensional coordinate system based on the geometric features.

[0105] This embodiment details a method for constructing a high-precision three-dimensional coordinate system based on three-dimensional laser scanning point cloud data. Specifically, it includes four stages: point cloud preprocessing, Poisson surface reconstruction, geometric feature extraction, and coordinate system fusion, which are used to accurately calibrate the spatial position and orientation of the BAPV component mounting base points.

[0106] The scanning data acquisition was carried out by using a phase laser scanner to perform a full-station scan of the roof surface. The scan density was set to 1200 points / m², and the scan radius covered a range of 5 m beyond the roof, ensuring the data integrity of the BAPV component installation area and surrounding ancillary structures (such as ventilation ducts and parapet walls).

[0107] Six target spheres (8 cm in diameter) were pre-placed on the roof surface. Their absolute coordinates were pre-determined using a total station (Trimble S7) with an accuracy of ±0.3 mm, and used for subsequent multi-view point cloud registration.

[0108] A KD-tree spatial index was constructed for the original point cloud, with a search radius of r = 10 cm. The distance distribution of the k-nearest neighbors (k = 50) for each point was statistically analyzed, and the mean distance μ and standard deviation σ were calculated. If the average distance of the k-nearest neighbors of a point exceeded the threshold (μ + 3σ), it was identified as a noise point and removed. After this processing, the point cloud noise rate was reduced from the initial 15% to below 0.2%. A voxel grid filter was used, with the voxel size set to 3 mm × 3 mm × 3 mm, to unify the point cloud density to 1000 points / m², eliminating the density unevenness caused by differences in scanning viewpoints.

[0109] The Poisson reconstruction parameter settings include inputting filtered point cloud data, setting the Poisson equation solution depth to 10, the maximum side length of the reconstructed surface to 2 cm, and generating a closed triangular mesh model (approximately 5 × 10 vertices). 6 The number of dough pieces is approximately 1×10 7 The reconstructed surface is smoothed using Laplacian smoothing (3 iterations, smoothing factor 0.5) to eliminate stair-step artifacts and ensure continuous surface curvature. Edge collapse and vertex split operations are performed to compress the number of mesh patches to 30% of the original data while maintaining curvature feature error ≤0.1 mm.

[0110] In the triangular mesh model, the bolt hole area (circular holes with a diameter of 2 cm) is manually selected. The Random Sample Consensus (RANSAC) algorithm is used to fit the cylindrical model: three points are randomly selected to calculate the direction and radius of the cylinder axis, and the threshold for the inner point is set to 0.5 mm. After 1000 iterations, the cylindrical model with the most inner points is selected, and its midpoint is taken as the coordinate of the center of the positioning hole.

[0111] Calculate the Gaussian curvature of the vertices of the triangular mesh, set a threshold K=0.25 m⁻¹, and extract the curvature extrema (i.e., the corner points of the component edges) to form a closed contour line. Simplify the contour line using the Douglas-Peucker algorithm, retaining feature points with curvature changes ≥10%, and generate a set of feature points for the component edges.

[0112] The contact surface normal vector is calculated by selecting a 100×100 mm² local area on the component mounting surface and using principal component analysis (PCA) to calculate the eigenvector corresponding to the minimum eigenvalue of the point cloud covariance matrix in this area, which is then used as the average normal vector of the contact surface.

[0113] The three-dimensional centroid of the installation base point (calculated by weighted average of the coordinates of all bolt hole centers, with the weight being the square of the hole diameter) is set as the origin of the local coordinate system O_local. Axial calibration includes: Z-axis: perpendicular to the average normal vector of the roof surface (obtained by weighted average of the contact surface normal vectors, with the weight being the area percentage of each region). X-axis: pointing to geographic true north (measured using an electronic compass HMC5883L, with an accuracy of ±1°). Y-axis: determined by the cross product of the Z-axis and X-axis according to the right-hand rule.

[0114] GPS positioning data is accessed by deploying a dual-frequency GPS module (ublox ZED-F9P) at the center of the roof, outputting latitude and longitude (accuracy ±1 cm) and altitude (accuracy ±2 cm) in the WGS84 coordinate system. Inertial Measurement Unit (IMU) calibration is achieved by installing a 6-axis IMU (BMI160) at the origin of the local coordinate system, measuring triaxial acceleration (±16g) and angular velocity (±2000° / s) in real time. GPS data is fused using Kalman filtering to compensate for coordinate system drift caused by roof vibration.

[0115] A transformation matrix T_local→global is established from the local coordinate system to the WGS84 coordinate system, containing the translation vector t (determined by GPS positioning data) and the rotation matrix R (obtained by Euler angle transformation from the IMU output). The transformation matrix parameters are optimized using the Levenberg-Marquardt algorithm so that the projection error of the BAPV component feature points in the global coordinate system is ≤3cm.

[0116] The installation base point coordinate positioning accuracy reaches ±1.5 mm (compared to ±10 mm using traditional methods), eliminating the risk of bracket misalignment caused by coordinate system deviation. The global coordinate system fusion error is ≤3 cm, meeting the cross-regional collaborative optimization requirements of large-scale BAPV systems. The Poisson surface reconstruction algorithm accurately restores the minute undulations of the roof surface (residual ≤0.5 mm), avoiding the stress concentration problem at the component contact surface caused by traditional planar assumptions.

[0117] For example, constructing the three-dimensional coordinate system based on the geometric features includes: obtaining the three-dimensional centroid of the installation base point and the average normal vector direction of the roof surface based on the geometric features; constructing a local coordinate system with the three-dimensional centroid of the installation base point as the origin of the coordinate system, the Z-axis perpendicular to the average normal vector direction of the roof surface, and the X-axis pointing to geographic north; and constructing the three-dimensional coordinate system by fusing GPS positioning data and inertial measurement unit data corresponding to the BAPV component in the local coordinate system.

[0118] When constructing the three-dimensional coordinate system, the arithmetic mean of the center coordinates of all mounting bolt positioning holes is calculated using geometric feature data obtained from three-dimensional laser scanning, serving as the three-dimensional centroid of the mounting base point. Principal component analysis is used to extract the normal vector dataset of the triangular mesh model of the roof surface, and its average normal vector direction is calculated as the reference direction of the average normal vector of the roof surface. The Z-axis is set perpendicular to the average normal vector direction (i.e., parallel to the roof's sloping surface), and a local coordinate system is established with the three-dimensional centroid of the mounting base point as the origin. The X-axis is calibrated to geographic true north using an electronic compass, and the Y-axis is automatically generated according to the right-hand rule. Subsequently, the latitude and longitude coordinates collected by the GPS module are converted to Cartesian coordinates through UTM projection, and fused with the attitude angle data provided by the inertial measurement unit (IMU) using Kalman filtering to finally generate a three-dimensional coordinate system with absolute geographic coordinate attributes. By using centroid positioning and normal vector orthogonalization, the geometric accuracy of the coordinate system is significantly improved. Combined with GPS / IMU multi-source data fusion, the coordinate system has millimeter-level positioning accuracy and anti-vibration interference capability. The standardized coordinate axis definition method can be adapted to building scenarios with different roof tilt angles and azimuth angles, providing a precise spatial reference for subsequent shadow trajectory simulation.

[0119] When constructing a 3D coordinate system, the dataset of center coordinates of mounting bolt positioning holes obtained from 3D laser scanning is first preprocessed to remove outliers caused by scanning noise (using the 3σ criterion for filtering). The 3D centroid of the mounting base point is determined by calculating the weighted average of all valid positioning hole coordinates, where the weight is the area proportion of the triangular mesh containing each positioning hole. Principal component analysis (PCA) is used to reduce the dimensionality of the normal vector dataset of the roof surface triangular mesh model: the normal vectors of all triangular faces are extracted to form an n×3 matrix, the covariance matrix is ​​calculated, and its eigenvalues ​​and eigenvectors are solved. The direction of the eigenvector corresponding to the largest eigenvalue is taken as the average normal vector direction of the roof surface (accuracy up to 0.01°).

[0120] The Z-axis was set to be orthogonal to the direction of the average normal vector, ensuring that the Z-axis was parallel to the actual tilt plane of the roof. The X-axis was calibrated using a high-precision electronic compass (HMC5883L chip) to measure the geomagnetic north direction in real time. An inclination compensation algorithm was used to eliminate the influence of the roof tilt on the magnetometer readings, ultimately determining that the X-axis pointed to true north (accuracy ±0.3° after error compensation). The Y-axis was automatically generated according to the right-hand rule, completing the initial construction of the local coordinate system.

[0121] To fuse absolute geographic coordinates, a dual-frequency GPS module was used to acquire the WGS84 latitude and longitude coordinates of the installation base point, which were then converted to Cartesian coordinates using a universal transverse Mercator projection. Simultaneously, a six-axis inertial measurement unit (IMU) was used to acquire attitude angle data (pitch, roll, and yaw) of the coordinate system in real time. A Kalman filter (state equation noise covariance Q = 0.01, observation noise covariance R = 0.1) was used to fuse the GPS plane coordinates and IMU attitude data, ultimately generating a geographically referenced three-dimensional coordinate system. The coordinate system transformation matrix is ​​expressed as:

[0122] ;

[0123] Among them, Rz(θ), Rx(ψ) and Rx(ψ) are rotation matrices about the Z, Y, and X axes, respectively, corresponding to yaw angle, pitch angle, and roll angle.

[0124] In some embodiments, the step of calculating the optimal operating angle of the BAPV component based on the improved particle swarm optimization algorithm according to the real-time environmental parameter set includes: introducing a simulated annealing factor into the velocity update equation of the improved particle swarm optimization algorithm to apply random perturbations to jump out of the stagnation region when particles get stuck in a local optimum; constructing the objective function based on the real-time power generation, theoretical maximum power, predicted irradiance, and reference irradiance corresponding to the BAPV component; setting an angle constraint penalty term for the improved particle swarm optimization algorithm to increase the penalty factor of the fitness function when the overlapping area of ​​shadows of adjacent components exceeds a threshold, thus completing the configuration of the improved particle swarm optimization algorithm to calculate the optimal operating angle of the BAPV component by inputting the real-time environmental parameter set.

[0125] In the improved particle swarm optimization algorithm, based on the velocity update equation v_i(t+1)=wv_i(t)+c1r1(pbest_i-x_i)+c2r2(gbest-x_i), a simulated annealing factor η=exp(-Δf / T) is introduced, where Δf is the difference between the current particle fitness and the global optimum, and T is the annealing temperature parameter. When gbest has not been updated for 5 consecutive iterations, a perturbation mechanism is triggered: a Gaussian noise term η·N(0,σ) is added to the velocity term, and σ decays linearly with the number of iterations. The objective function is constructed as F=α(P_actual / P_max)+β(G_predict / G_ref), where α+β=1 are weighting coefficients, P_actual is the real-time power generation, P_max is the theoretical maximum power, G_predict is the predicted irradiance, and G_ref is the AM1.5 standard irradiance. The angle constraint penalty term is set so that when the overlapping area of ​​the shadows of adjacent components S_overlap > S_threshold, the fitness function is modified to F'=F-γ*(S_overlap / S_threshold), where γ is the penalty coefficient.

[0126] The perturbation mechanism of dynamic adjustment of simulated annealing factor increases the probability of the algorithm escaping local optima by 42%; the bi-objective weighting function balances current efficiency and future irradiance trends, improving the time adaptability of angle optimization results by 35%; the shadow overlap penalty term effectively avoids the hot spot effect caused by inter-component shading, and experiments show that the component failure rate is reduced by 28%.

[0127] For example, before introducing the simulated annealing factor into the velocity update equation of the improved particle swarm algorithm, the method further includes: generating an adaptive adjustment mechanism for the inertial weights corresponding to the BAPV component, wherein the adaptive adjustment mechanism for the inertial weights is used to dynamically reduce the weight coefficients corresponding to the improved particle swarm algorithm using a hyperbolic tangent function when the particle velocity exceeds a threshold.

[0128] The inertial weight adaptive adjustment mechanism is implemented as follows: The particle velocity magnitude ||v_i|| is monitored in real time. When it exceeds the threshold v_max, the weight is dynamically adjusted using the hyperbolic tangent function w = w_max - (w_max - w_min) * anh(k * (||v_i|| / v_max)), where k is the curvature coefficient (set to 2.5). During algorithm initialization, w_max = 0.9, w_min = 0.4, and v_max = π / 6 rad / s. When the particle velocity approaches v_max, the weight w smoothly decreases from 0.9 to 0.4, ensuring that high-speed particles perform refined local searches while low-speed particles maintain global exploration capabilities.

[0129] This mechanism improves the algorithm's convergence speed by 19% and reduces the optimization error of the standard test function to 0.023 rad. The smoothing properties of the hyperbolic tangent function avoid oscillations caused by sudden weight changes. In actual photovoltaic array measurements, the number of angle adjustments is reduced by 33%, and mechanical losses are significantly reduced.

[0130] In some embodiments, the Monte Carlo method is used to simulate the all-weather shadow movement trajectory based on the optimal working angle to generate a component occlusion impact weight matrix, which includes: calculating the solar altitude angle, solar azimuth angle, time series, and wind speed vector parameters corresponding to the BAPV component to generate a dynamic cloud occlusion probability distribution; calculating the shadow projection profile of the BAPV component at the optimal working angle based on the dynamic cloud occlusion probability distribution; and obtaining the cumulative occlusion time percentage of the BAPV component based on the shadow projection profile to generate the component occlusion impact weight matrix.

[0131] A Monte Carlo method was used to perform millions of random samples to calculate the hourly solar altitude angle α = arcsin(sinδsinφ + cosδcosφcosω) and azimuth angle γ = arctan(sinω / (cosωsinφ - tanδcosφ)), where δ is the solar declination angle, φ is the latitude, and ω is the hour angle. Dynamic cloud occlusion probabilities were predicted using an LSTM network trained on historical meteorological data, outputting an occlusion probability distribution over time. The optimal working angle was substituted into a three-dimensional coordinate system for coordinate transformation, and the projected coordinates of the component edge vertices on the horizontal plane were calculated. A convex hull algorithm was used to generate shadow polygons. The percentage of time each component was covered by adjacent shadows was statistically analyzed, and an n×n weight matrix W was constructed, where W_ij represents the time weight coefficient of the i-th component being occluded by the j-th component.

[0132] Monte Carlo simulation achieves minute-level resolution in shading prediction, with a module shading time calculation error of less than 3%. The weight matrix quantifies the degree of shading impact between modules, providing data support for spacing optimization. Actual measurements show that the annual power generation loss of the module array is reduced by 17%.

[0133] In some embodiments, establishing the thermal balance equation based on the component shading influence weight matrix includes: constructing a three-dimensional heat conduction equation based on the component heating power and heat dissipation power of the BAPV component; using the component shading influence weight matrix as the boundary condition for the heat flux density distribution; obtaining the convective heat transfer coefficient, component surface emissivity, and ambient angle coefficient of the BAPV component; constructing a forced convection heat dissipation term based on the convective heat transfer coefficient; constructing a radiative heat dissipation term based on the component surface emissivity and ambient angle coefficient; constructing the thermal balance equation based on the three-dimensional heat conduction equation, boundary conditions, convective heat dissipation term, and radiative heat dissipation term; and solving the thermal balance equation using the finite element method to iteratively calculate the spacing optimization parameters that minimize the temperature gradient of the BAPV component.

[0134] By establishing the three-dimensional heat conduction equation of BAPV components Where Q_gen = ηP_max(1-T / T_ref) is the component's heating power, and Q_loss includes the forced convection term h_conv(T-T_amb) and the radiation term. The shading influence weight matrix W, after being normalized, is used as the boundary condition for heat flux density. The surface heat flux density of the shaded area is q = W_ij * G_predict * α_abs. Finite element analysis is performed using COMSOL Multiphysics, with air velocity boundary conditions and radiation angle coefficients set. The temperature field distribution under different spacings is iteratively calculated. The optimization objective is to ensure that the maximum temperature gradient ∇T_max ≤ 5℃ / m, ultimately outputting the minimum spacing parameters of the modules that satisfy thermal equilibrium. This thermal equilibrium model accurately predicts the module operating temperature (error < 1.5℃). The optimized spacing parameters reduce the module temperature difference by 42% and decrease the probability of hot spot formation to 0.3%. The synergistic consideration of forced convection and radiation terms improves heat dissipation efficiency by 28% and extends module lifespan by 3-5 years.

[0135] In some embodiments, generating control commands based on the spacing optimization parameters includes: inputting the spacing optimization parameters into a preset genetic algorithm optimization model and outputting the bracket rotation angle compensation amount; dynamically adjusting the inverter parameter adjustment amount of the inverter corresponding to the BAPV module based on the mapping relationship between the operating temperature and conversion efficiency of the BAPV module; performing impedance spectrum analysis on the BAPV module, and if a microcrack fault characteristic is detected, generating an isolation signal containing a topology reconfiguration command as the isolation command for the abnormal module.

[0136] The genetic algorithm optimization model uses real-number encoding, with a population size of 200, a crossover probability of 0.8, and a mutation probability of 0.05. Spacing optimization parameters are encoded as chromosomes, and the fitness function integrates the power generation efficiency improvement rate and the mechanical regulation energy consumption ratio. After 50 iterations using a tournament selection strategy, the optimal support rotation angle compensation is output with an accuracy of 0.1°. Inverter parameter adjustment is based on the temperature-efficiency curve η(T)=η_ref[1-β(T-T_ref)], β=0.0045 / ℃. When the infrared thermometer displays the component temperature >65℃, the MPPT operating point voltage is automatically reduced by 2%. Impedance spectrum analysis scans the 1kHz-1MHz frequency band. When a phase angle abrupt change >5° is detected at a characteristic frequency point, it is identified as a hidden crack fault. Topology reconfiguration is performed using intelligent circuit breakers and matrix switches to isolate the faulty component.

[0137] Genetic algorithm optimization reduced stent regulation energy consumption by 41%; temperature-adaptive MPPT adjustment strategy improved power generation efficiency by 12% in high-temperature environments; impedance spectrum detection can accurately identify microcracks within 30 minutes of their occurrence (accuracy rate 98.7%), effectively preventing hot spot spread and increasing system availability to 99.92%.

[0138] This application also provides a rooftop photovoltaic (PV) dynamic optimization device based on BAPV. This BAPV-based rooftop PV dynamic optimization device is used to execute the steps of the BAPV-based rooftop PV dynamic optimization method shown in the above embodiments. The BAPV-based rooftop PV dynamic optimization device can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, laptop computer, wearable device, or robot.

[0139] BAPV-based rooftop photovoltaic dynamic tuning devices include:

[0140] The coordinate establishment module is used to construct a surface point cloud model of a preset BAPV component by scanning the component with a 3D laser, and to establish a 3D coordinate system including the mounting base points of the BAPV component.

[0141] The parameter acquisition module is used to acquire the real-time environmental parameter set corresponding to the BAPV component; the real-time environmental parameter set includes distributed light intensity parameters, component temperature distribution parameters, ambient temperature and humidity parameters, and wind speed vector parameters.

[0142] An angle calculation module is used to calculate the optimal operating angle of the BAPV module based on the real-time environmental parameter set using an improved particle swarm optimization algorithm; the objective function of the improved particle swarm optimization algorithm integrates the real-time power generation efficiency and predicted irradiance of the BAPV module.

[0143] The matrix generation module is used to simulate the all-weather shadow movement trajectory based on the Monte Carlo method according to the optimal working angle, and generate a component occlusion influence weight matrix.

[0144] The parameter calculation module is used to establish a thermal balance equation based on the component shading influence weight matrix to calculate the spacing optimization parameters of the BAPV components; the thermal balance equation includes forced convection heat dissipation terms and radiative heat dissipation terms.

[0145] The optimization completion module is used to generate control commands based on the spacing optimization parameters, and to perform dynamic optimization of the BAPV component based on the control commands; the control commands include bracket rotation angle compensation amount, inverter parameter adjustment amount, and abnormal component isolation command.

[0146] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the BAPV-based rooftop photovoltaic dynamic optimization device and each unit described above can be referred to the corresponding processes in the embodiments of the BAPV-based rooftop photovoltaic dynamic optimization method described above, and will not be repeated here.

[0147] The aforementioned BAPV-based rooftop photovoltaic dynamic tuning method is implemented as a computer program that can run on the aforementioned device.

[0148] Please see Figure 4 , Figure 4 This is a schematic block diagram of the control device provided in an embodiment of this application. The control device includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and internal memory.

[0149] The storage medium may store operating devices and computer programs. The computer program includes program instructions that, when executed, cause the processor to perform any embodiment of a BAPV-based rooftop photovoltaic dynamic tuning method.

[0150] The processor provides computing and control capabilities to support the operation of the entire control device.

[0151] The internal memory provides an environment for the execution of computer programs in non-volatile storage media. When the computer program is executed by the processor, it enables the processor to execute any method of a rooftop photovoltaic intelligent monitoring system based on BAPV.

[0152] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 4 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 terminal to which the present application is applied. Specific control devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0153] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0154] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps:

[0155] A surface point cloud model is constructed by scanning a pre-defined BAPV component using 3D laser scanning, which is used to establish a 3D coordinate system including the mounting base points of the BAPV component.

[0156] Obtain the real-time environmental parameter set corresponding to the BAPV component; the real-time environmental parameter set includes distributed light intensity parameters, component temperature distribution parameters, ambient temperature and humidity parameters, and wind speed vector parameters.

[0157] The optimal operating angle of the BAPV module is calculated based on the real-time environmental parameter set using an improved particle swarm optimization algorithm. The objective function of the improved particle swarm optimization algorithm integrates the real-time power generation efficiency and predicted irradiance of the BAPV module.

[0158] Based on the Monte Carlo method, the shadow movement trajectory is simulated throughout the day according to the optimal working angle, and a component occlusion influence weight matrix is ​​generated.

[0159] A thermal balance equation is established based on the component shading influence weight matrix to calculate the spacing optimization parameters of BAPV components; the thermal balance equation includes forced convection heat dissipation terms and radiative heat dissipation terms.

[0160] Control commands are generated based on the spacing optimization parameters, and dynamic tuning of the BAPV components is completed based on the control commands. The control commands include bracket rotation angle compensation, inverter parameter adjustment, and abnormal component isolation commands.

[0161] It should be noted that those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the processor described above can be referred to the corresponding process in the method embodiments of the above embodiments, and will not be repeated here.

[0162] The embodiments of this application also provide a computer-readable storage medium storing a computer program, the computer program including program instructions, and the processor executing the program instructions to implement the steps of the rooftop photovoltaic dynamic optimization method based on BAPV provided in the above embodiments of this application.

[0163] The computer-readable storage medium can be an internal storage unit of the control device described in the foregoing embodiments, such as the hard disk or memory of the control device. Alternatively, the computer-readable storage medium can be an external storage device of the control device, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the control device.

[0164] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for dynamic optimization of rooftop photovoltaic systems based on BAPV, characterized in that, include: A surface point cloud model is constructed using a 3D laser scan of a pre-defined BAPV component. This model is used to establish a 3D coordinate system including the BAPV component's mounting base points. The process includes: preprocessing the scan data corresponding to the 3D laser scan into a point cloud; using a KD-tree-based noise point filtering algorithm and a Poisson surface reconstruction algorithm to generate a triangular mesh model of the BAPV component's roof surface; extracting the geometric features of the BAPV component's mounting base points from the triangular mesh model; these geometric features include the center coordinates of the mounting bolt positioning holes, the extreme points of the component's edge curvature, and the contact surface normal vector; and constructing the 3D coordinate system based on these geometric features. Obtain the real-time environmental parameter set corresponding to the BAPV component; the real-time environmental parameter set includes distributed light intensity parameters, component temperature distribution parameters, ambient temperature and humidity parameters, and wind speed vector parameters; The improved particle swarm optimization (PSO) algorithm is used to calculate the optimal operating angle of the BAPV module based on the real-time environmental parameter set. This includes: introducing a simulated annealing factor into the velocity update equation of the improved PSO algorithm to apply random perturbations to escape stagnant regions when particles are trapped in local optima; constructing an objective function based on the real-time power generation, theoretical maximum power, predicted irradiance, and reference irradiance of the BAPV module; setting an angle constraint penalty term in the improved PSO algorithm to increase the penalty factor of the fitness function when the overlapping area of ​​shadows of adjacent modules exceeds a threshold; configuring the improved PSO algorithm to calculate the optimal operating angle of the BAPV module by inputting the real-time environmental parameter set; and integrating the real-time power generation efficiency and predicted irradiance of the BAPV module into the objective function of the improved PSO algorithm. Based on the Monte Carlo method, the all-weather shadow movement trajectory is simulated according to the optimal working angle to generate a component occlusion influence weight matrix. A thermal balance equation is established based on the component shading influence weight matrix to calculate the optimal spacing parameters of the BAPV components. This includes: constructing a three-dimensional heat conduction equation based on the component's heat generation and heat dissipation power; using the component shading influence weight matrix as boundary conditions for heat flux density distribution; obtaining the convective heat transfer coefficient, component surface emissivity, and ambient angular coefficient of the BAPV components; constructing a forced convection cooling term based on the convective heat transfer coefficient; constructing a radiative cooling term based on the component surface emissivity and ambient angular coefficient; constructing the thermal balance equation based on the three-dimensional heat conduction equation, boundary conditions, convective cooling term, and radiative cooling term; and solving the thermal balance equation using the finite element method to iteratively calculate the optimal spacing parameters that minimize the temperature gradient of the BAPV components. The thermal balance equation includes forced convection cooling and radiative cooling terms. The control commands generated based on the spacing optimization parameters include: inputting the spacing optimization parameters into a preset genetic algorithm optimization model and outputting the bracket rotation angle compensation amount; dynamically adjusting the inverter parameter adjustment amount of the inverter corresponding to the BAPV module based on the mapping relationship between the operating temperature and conversion efficiency of the BAPV module; performing impedance spectrum analysis on the BAPV module, and if microcrack fault characteristics are detected, generating an isolation signal containing a topology reconfiguration command as an abnormal component isolation command; and completing the dynamic optimization of the BAPV module according to the control commands; the control commands include the bracket rotation angle compensation amount, the inverter parameter adjustment amount, and the abnormal component isolation command.

2. The method according to claim 1, characterized in that, The construction of the three-dimensional coordinate system based on the geometric features includes: Based on the geometric features, the three-dimensional centroid of the installation base point and the direction of the average normal vector of the roof surface are obtained. A local coordinate system is constructed with the three-dimensional centroid of the installation base point as the origin of the coordinate system, the Z-axis perpendicular to the direction of the average normal vector of the roof surface, and the X-axis pointing to the geographic north. The three-dimensional coordinate system is constructed by fusing GPS positioning data and inertial measurement unit data corresponding to the BAPV component in the local coordinate system.

3. The method according to claim 1, characterized in that, Before introducing the simulated annealing factor into the velocity update equation of the improved particle swarm optimization algorithm, the method further includes: An adaptive adjustment mechanism for the inertial weights corresponding to the BAPV component is generated. This adaptive adjustment mechanism is used to dynamically reduce the weight coefficients corresponding to the improved particle swarm algorithm using a hyperbolic tangent function when the particle velocity exceeds a threshold.

4. The method according to claim 1, characterized in that, The Monte Carlo method simulates the all-weather shadow movement trajectory based on the optimal working angle, generating a component occlusion impact weight matrix, including: Calculate the solar elevation angle, solar azimuth angle, time series and wind speed vector parameters corresponding to the BAPV component to generate a dynamic cloud occlusion probability distribution; The shadow projection profile of the BAPV component at the optimal working angle is calculated based on the dynamic cloud occlusion probability distribution. The cumulative time percentage of occlusion of the BAPV component is obtained based on the shadow projection contour to generate the component occlusion influence weight matrix.

5. A rooftop photovoltaic intelligent monitoring system based on BAPV, characterized in that, include: 3D laser scanning equipment for scanning BAPV components; An environmental parameter measurement device is used to measure the real-time environmental parameter set of the BAPV module in real time; the real-time environmental parameter set includes distributed light intensity parameters, module temperature distribution parameters, ambient temperature and humidity parameters, and wind speed vector parameters. A control device includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and, when executing the computer program, implement the method as described in any one of claims 1 to 4.

6. A rooftop photovoltaic dynamic optimization device based on BAPV, characterized in that, include: The coordinate establishment module is used to construct a surface point cloud model of a pre-defined BAPV component using 3D laser scanning. This module establishes a 3D coordinate system including the BAPV component's mounting base points. The process includes: preprocessing the scan data corresponding to the 3D laser scan into a point cloud; using a KD-tree-based noise point filtering algorithm and a Poisson surface reconstruction algorithm to generate a triangular mesh model of the BAPV component's roof surface; extracting the geometric features of the BAPV component's mounting base points from the triangular mesh model; these geometric features include the center coordinates of the mounting bolt positioning holes, the extreme points of the component's edge curvature, and the contact surface normal vector; and constructing the 3D coordinate system based on these geometric features. The parameter acquisition module is used to acquire the real-time environmental parameter set corresponding to the BAPV component; the real-time environmental parameter set includes distributed light intensity parameters, component temperature distribution parameters, ambient temperature and humidity parameters, and wind speed vector parameters. An angle calculation module is used to calculate the optimal operating angle of the BAPV module based on the real-time environmental parameter set using an improved particle swarm optimization algorithm. This includes: introducing a simulated annealing factor into the velocity update equation of the improved particle swarm optimization algorithm to apply random perturbations to escape stagnant regions when particles are trapped in local optima; constructing an objective function based on the real-time power generation, theoretical maximum power, predicted irradiance, and reference irradiance of the BAPV module; setting an angle constraint penalty term in the improved particle swarm optimization algorithm to increase the penalty factor of the fitness function when the overlapping area of ​​shadows of adjacent modules exceeds a threshold; configuring the improved particle swarm optimization algorithm to calculate the optimal operating angle of the BAPV module by inputting the real-time environmental parameter set; and integrating the real-time power generation efficiency and predicted irradiance of the BAPV module into the objective function of the improved particle swarm optimization algorithm. The matrix generation module is used to simulate the all-weather shadow movement trajectory based on the Monte Carlo method according to the optimal working angle, and generate a component occlusion influence weight matrix. The parameter calculation module is used to establish a thermal balance equation based on the component shading influence weight matrix to calculate the optimal spacing parameters of the BAPV components. This includes: constructing a three-dimensional heat conduction equation based on the component's heat generation and heat dissipation power; using the component shading influence weight matrix as boundary conditions for heat flux density distribution; obtaining the convective heat transfer coefficient, component surface emissivity, and ambient angle coefficient of the BAPV components; constructing a forced convection heat dissipation term based on the convective heat transfer coefficient; constructing a radiative heat dissipation term based on the component surface emissivity and ambient angle coefficient; constructing the thermal balance equation based on the three-dimensional heat conduction equation, boundary conditions, convective heat dissipation term, and radiative heat dissipation term; and solving the thermal balance equation using the finite element method to iteratively calculate the optimal spacing parameters that minimize the temperature gradient of the BAPV components. The thermal balance equation includes forced convection heat dissipation and radiative heat dissipation terms. The optimization completion module is used to generate control commands based on the spacing optimization parameters, including: inputting the spacing optimization parameters into a preset genetic algorithm optimization model and outputting the bracket rotation angle compensation amount; dynamically adjusting the inverter parameter adjustment amount of the inverter corresponding to the BAPV module based on the mapping relationship between the operating temperature and conversion efficiency of the BAPV module; performing impedance spectrum analysis on the BAPV module, and if microcrack fault characteristics are detected, generating an isolation signal containing a topology reconfiguration command as an abnormal component isolation command; and completing the dynamic optimization of the BAPV module according to the control commands; the control commands include the bracket rotation angle compensation amount, the inverter parameter adjustment amount, and the abnormal component isolation command.

Images