Solar module angle control system and method with adaptive light conditioning

Abstract

This invention discloses an adaptive solar module angle control system and method, belonging to the field of solar module angle control technology. It includes a multi-dimensional illumination sensing module that uses sensors such as quantum dot spectrometers to collect environmental data; a geographic information processing module that calculates the sun's position through high-precision positioning and digital twins; an intelligent control decision-making module that generates adjustment strategies based on spatiotemporal graph neural networks and reinforcement learning; a module-driven execution module that uses composite driving to achieve angle adjustment and recover energy; and a fault diagnosis and self-healing module that uses fiber optic sensing and terahertz technology to detect and repair faults. This invention achieves adaptive angle adjustment of solar modules, accurately capturing changes in illumination through multi-dimensional sensing, saving energy and reducing consumption through composite driving, extending equipment life through intelligent diagnosis, and enhancing system anti-interference and security performance through bio-inspired collaboration and quantum encryption technologies, thereby reducing operation and maintenance costs and improving the overall efficiency of photovoltaic power plants.

Description

Technical Field

[0001] This invention relates to the field of solar module angle control technology, and in particular to a solar module angle control system and method for adaptive illumination adjustment. Background Technology

[0002] With the surge in global demand for clean energy, solar photovoltaic (PV) power generation has become a crucial direction for energy transition. However, traditional fixed solar modules struggle to adapt to dynamic changes in sunlight conditions, resulting in limited power generation efficiency. The sun's altitude and azimuth angles constantly change throughout the day, and seasonal climate variations, differences in latitude, and shading caused by complex terrain all prevent modules from consistently maintaining the optimal angle for sunlight exposure. Furthermore, frequent extreme weather events, such as dust storms causing dust accumulation on module surfaces and strong winds damaging module structures, further exacerbate power generation efficiency losses. Manual inspection and adjustment are not only costly but also far too slow to keep pace with the rapid changes in sunlight conditions, making it difficult to meet the high-efficiency operation requirements of large-scale PV power plants.

[0003] Existing solar module angle adjustment technologies have significant drawbacks. Most systems rely on a single light sensor and a simple PID control algorithm, enabling only fixed-routine adjustments based on time or sun position. They cannot detect changes in light intensity, spectral distribution, and atmospheric interference in complex environments in real time. For example, in cloudy weather, rapid cloud movement causes instantaneous fluctuations in light intensity, which traditional systems cannot respond to in time, resulting in significant power generation losses. Furthermore, traditional drive devices often use a single motor, resulting in low angle adjustment accuracy, high energy consumption, and a lack of monitoring of module structural health, making it impossible to detect potential faults early. Once the equipment jams or suffers mechanical damage, it can easily lead to power generation interruptions and high maintenance costs.

[0004] Furthermore, the coordinated control of modules in large-scale photovoltaic power plants remains a challenge for the industry. Traditional systems typically employ independent regulation modes, lacking information exchange and coordination strategies between modules. In partially shaded scenarios, the shadows cast by the front-row modules can severely impact the power generation efficiency of the rear-row modules, resulting in "mismatch losses." Moreover, existing technologies do not fully utilize cutting-edge technologies such as quantum computing, artificial intelligence, and bio-inspired algorithms, lacking the capabilities for predicting complex lighting environments and making multi-objective optimization decisions, making it difficult to achieve a comprehensive improvement in solar power generation efficiency, equipment lifespan, and operating costs. Summary of the Invention

[0005] The present invention proposes an adaptive light-adjusting solar module angle control system and method to solve the problems mentioned in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an adaptive light-adjusting solar module angle control system, comprising the following modules: Multi-dimensional illumination sensing module: Deploys a distributed quantum dot spectral sensor array, and the system innovatively designs a polarization light analysis unit. Combined with the Rayleigh scattering model, it inverts atmospheric aerosol concentration; the calculation formula is as follows: , denoted as atmospheric aerosol concentration; k is an empirical coefficient. To actually measure the degree of polarization of sunlight; This represents the theoretical degree of solar polarization. Geographic Information Processing Module: This module fuses inertial navigation and satellite positioning data using Kalman filtering, innovatively introduces digital twin technology, and calculates solar radiation on slopes using a terrain radiation model. The formula is as follows: , Solar radiation intensity; This refers to the intensity of direct solar radiation. The angle between the sunlight and the slope normal; The intensity of scattered solar radiation, The slope angle; To reflect solar radiation intensity; Intelligent control decision-making module: Construct a spatiotemporal graph neural network based on attention mechanism, innovatively design a multi-objective reinforcement learning framework, adopt Pareto optimal strategy to balance conflicting objectives, and introduce optical flow field analysis technology in cloud movement scenario to predict cloud movement speed and direction; Component-driven execution module: Develop a composite actuator of shape memory alloy and electromagnetic drive. SMA wire provides angle adjustment, electromagnetic drive mechanism provides response. The actuator adopts a biomimetic joint design. The system innovatively designs an energy recovery braking system. During component adjustment, when the motor is in the power generation state, the electrical energy is stored in the supercapacitor through a bidirectional DC / DC converter. Fault diagnosis and self-healing module: Deploy a distributed strain sensing network based on photonic crystal fiber. By measuring the Brillouin divergence frequency shift of the optical signal in the fiber, the system can monitor the structural health of the components. The system innovatively applies deep adversarial generative network to simulate faults. By training a discriminator, it can distinguish between normal and abnormal states. When a slight displacement of the component is detected, the system will activate the correction mechanism.

[0007] Furthermore, it also includes a bio-inspired cluster control module: this module mimics the pheromone mechanism of the ant colony algorithm, establishes virtual pheromone trajectories between adjacent components, and when a component detects a change in light, it releases "attractants" to neighboring components via wireless communication. The components adjust their own angles according to the pheromone concentration gradient, forming a self-organizing and coordinated regulation mode.

[0008] Furthermore, it also includes a quantum encrypted communication module: it adopts quantum key distribution technology based on entangled photon pairs to realize communication between the control center and component nodes, uses the BB84 protocol for key negotiation, detects eavesdropping behavior through decoy state method, and automatically compensates for fiber temperature drift during the establishment of quantum channel.

[0009] Furthermore, in the multi-dimensional illumination sensing module: an innovative solar-blind ultraviolet detection unit is designed, utilizing the absorption characteristics of the ozone layer in this band, and inverting atmospheric ozone content by measuring the intensity of solar-blind ultraviolet radiation, thus establishing a model relating ozone concentration to spectral attenuation. in, wavelength The intensity of ultraviolet radiation; The theoretical ultraviolet radiation intensity at this wavelength under conditions of no ozone absorption; For ozone absorption cross section, is the ozone concentration in the atmosphere; L is the optical path length of light in the atmosphere.

[0010] Furthermore, in the intelligent control decision-making module: a meta-learning framework is introduced. When encountering extreme weather, the system adjusts the control strategy through a few-shot learning mechanism. An innovative cognitive reasoning engine is designed, combining symbolic reasoning and neural networks to achieve an interpretable decision-making process.

[0011] Furthermore, in the component drive execution module: an ion polymer metal composite micro-drive array is developed, and a million-level micro-drive is integrated on the component surface to realize local deformation adjustment of the component surface. When local occlusion is detected, the micro-drive is controlled to reflect light to the unoccluded area. The system uses dielectric elastomer artificial muscle as an auxiliary drive.

[0012] Furthermore, in the fault diagnosis and self-healing module: terahertz time-domain spectroscopy technology is innovatively applied to detect internal defects of the component. By analyzing the propagation characteristics of electromagnetic waves in the component, microcracks in the encapsulation material are identified. The system establishes a database of correspondences between terahertz spectral features and defect types. The transfer learning method is used to transfer laboratory data to actual application scenarios. When microcracks are detected inside the battery cell, the system uses laser repair technology to locally weld the cracks.

[0013] Furthermore, the method for an adaptive light-adjusting solar module angle control system includes the following steps: Spatiotemporal illumination field reconstruction steps: Based on multi-dimensional illumination sensing data, a spatiotemporal illumination field model is constructed using Gaussian process regression. The system divides the space into a three-dimensional grid, and the illumination intensity of each grid point is represented as a function of time and space. in, In three-dimensional coordinates At point t, the light intensity at time t; Let be a function of the mean light intensity at that location and time. To address the noise term that follows a Gaussian distribution, the correlation between grid points is calculated using a kernel function. The system also introduces a Kalman filter to perform real-time correction of the illumination field prediction. Component-based collaborative optimization steps: Construct a multi-agent optimization model based on alliance game theory, where each component acts as an agent, and the optimal alliance structure is formed through negotiation. The system defines the complementarity functions between components. in, Let i be the total revenue of alliance S, and j be the component numbers within alliance S. Let i and j be the complementarity coefficients. To determine the power gain resulting from the coordinated angle adjustment of components i and j, the core stability algorithm is used to find the optimal coalition. Real-time energy flow scheduling steps: The system adopts a model predictive control algorithm to predict the light and load demand for the next 24 hours. During periods of sufficient light, the system prioritizes storing energy in solid-state batteries. When light is insufficient, the stored energy is converted into AC power for the load through a DC / AC converter. The system also supports V2G function, which feeds the stored energy back to the grid when the grid needs peak shaving. Biological-machine co-evolution steps: combining genetic algorithms with deep reinforcement learning to achieve continuous evolution of control strategies, innovatively introducing a biologically inspired fitness function, considering power generation efficiency, and also including system robustness, resource utilization and environmental adaptability indicators.

[0014] Furthermore, it also includes a quantum illumination prediction step: the system transforms the atmospheric radiative transfer equation into a quantum algorithm, solves the integral equation through a quantum phase estimation algorithm, and completes the computation in 1 second that would take a traditional algorithm 1 hour when dealing with complex atmospheric conditions. The system also develops a quantum-classical hybrid architecture, which allocates key computational tasks to the quantum processor and handles the remaining tasks by the classical computer.

[0015] Furthermore, it also includes digital ecosystem symbiosis steps: connecting physical power plants, digital twin models, operation and maintenance personnel, and users into a symbiotic network. The system uses blockchain technology to record all operations and data. Operation and maintenance personnel interact with the digital twin model through augmented reality glasses, and users participate in power plant management through mobile applications. The system establishes an ecological credit mechanism to distribute benefits according to the contributions of each participant, forming a sustainable photovoltaic ecosystem.

[0016] Compared with existing technologies, the beneficial effects of this invention are: The system uses multi-dimensional sensing devices such as quantum dot spectral sensors and polarization analysis units to accurately capture light intensity, spectral distribution, and atmospheric environmental parameters. Even in complex weather conditions such as smog and cloudy skies, it can quickly identify light change characteristics and provide precise basis for component angle adjustment.

[0017] In terms of control strategy, by combining spatiotemporal graph neural networks and multi-objective reinforcement learning algorithms, the system can dynamically predict illumination trends and adjust component angles in advance. Compared with traditional methods, the response speed is several times faster, significantly reducing power generation losses caused by sudden changes in illumination. The bio-inspired cluster control module enables self-organizing collaboration among components, automatically forming a "wave-like" adjustment mode in partially shading scenarios, effectively reducing "mismatch losses" caused by shadow shading and improving overall power generation efficiency.

[0018] The component-driven execution module employs a hybrid drive and energy recovery technology, achieving high-precision angle adjustment while recovering and reusing energy during component adjustment, significantly reducing system energy consumption. The fault diagnosis and self-healing module, utilizing photonic crystal fiber sensing and terahertz detection technologies, can predict potential faults up to one week in advance and automatically perform nanometer-level repairs, drastically reducing annual equipment downtime and significantly lowering maintenance costs. Furthermore, quantum encrypted communication ensures data transmission security, and quantum computing accelerates illumination prediction, enabling the system to maintain high efficiency even under extreme weather conditions, comprehensively improving the intelligence, reliability, and economy of photovoltaic power plants. Attached Figure Description

[0019] Figure 1 This is a schematic block diagram of the adaptive light-adjusting solar module angle control system proposed in this invention; Figure 2 This is a schematic block diagram of the adaptive light regulation solar module angle control method proposed in this invention; Figure 3 A comparison chart of measurement error rates between quantum dot sensors and traditional silicon photovoltaic cells under different weather conditions; Figure 4 Line graph comparing the calculation errors of illumination for mountain photovoltaic power stations using different models; Figure 5 A bar chart comparing the computation time of quantum and classical algorithms under different cloud thicknesses. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0022] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly; for example, they may refer to a fixed connection, a detachable connection, or an integral connection; they may refer to a mechanical connection or an electrical connection; they may refer to a direct connection or an indirect connection through an intermediate medium; and they may refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The invention will now be described in further detail with reference to the accompanying drawings.

[0023] Reference Figures 1 to 5 An adaptive light-adjusting solar module angle control system includes the following modules: Multi-dimensional illumination sensing module: In actual deployment, 320 composite sensor nodes are evenly distributed within a 10MW photovoltaic power plant. Each node integrates a quantum dot spectral sensor, a four-quadrant light intensity sensor, a polarization light analysis unit, and a solar-blind ultraviolet detection unit.

[0024] The quantum dot spectrometer selected is the QDS-780 model, based on the core-shell quantum dot material CdSe / ZnS. A spectral response of 300-1800 nm is achieved by adjusting the quantum dot size. An internal microfluidic cooling channel circulates deionized water at a flow rate of 0.5 L / min, and a TEC1-12706 semiconductor cooler is used to stably maintain the detector unit temperature at -10℃, suppressing thermal noise interference and improving the signal-to-noise ratio to 65 dB under low-light conditions. In clear weather, its spectral resolution reaches 0.8 nm, with a signal-to-noise ratio of 72 dB and a dynamic range of 120 dB. This sensor utilizes the quantum confinement effect to achieve high-precision detection (resolution ≤1 nm) in the 300-1800 nm spectral range and can identify spectral distortion characteristics under different atmospheric conditions.

[0025] The PSM-4Q four-quadrant light intensity sensor consists of four single-crystal silicon photodiodes arranged in a cross shape with a spacing of 2mm. It measures light intensity differences through a differential amplifier, achieving a resolution of 0.08W / m². 2 The angular response range is ±60°. The polarization analysis unit utilizes calcite crystals for spectral dispersion, combined with a PIN photodiode to measure the orthogonal polarization components. In the 400-700nm wavelength range, the polarization degree measurement accuracy is ±0.5%, and the polarization angle accuracy is ±1°. By measuring the degree of polarization (DOP) and angle of polarization (AOP) of sunlight, and combining this with the Rayleigh scattering model to invert atmospheric aerosol concentration, the calculation formula is as follows: .in, This represents the atmospheric aerosol concentration; k is an empirical coefficient related to the aerosol type. Different types of aerosols (such as dust and industrial particles) correspond to different k values, which are obtained through experimental calibration. This represents the actual measured degree of solar polarization. This is the theoretical degree of solar polarization without aerosol interference. This parameter is used to correct for lighting prediction errors caused by air pollution.

[0026] The solar-blind ultraviolet (240-280nm) detector unit utilizes the ozone layer's absorption characteristics in this wavelength range to retrieve atmospheric ozone content by measuring solar-blind ultraviolet radiation intensity. The system establishes a model relating ozone concentration to spectral attenuation. .in, The intensity of ultraviolet radiation at a specific wavelength (within the range of 240-280 nm) was measured. The theoretical ultraviolet radiation intensity at this wavelength under conditions of no ozone absorption; ozone for wavelength The absorption cross section, which characterizes the ability of ozone to absorb ultraviolet light of a specific wavelength, is determined experimentally. denoted as ozone concentration in the atmosphere; L is the optical path length of light traveling through the atmosphere. The sensor nodes operate in the 868MHz band using the LoRaWAN protocol, transmitting data to the aggregation node at a transmit power of 14dBm over a distance of 3.2km. The system collects data every 100ms, and after Kalman filtering, the light intensity measurement error is controlled within ±1.8W / m². 2 .

[0027] Geographic Information Processing Module: This module integrates a high-precision INS / GNSS deeply coupled positioning system with digital twin technology. It utilizes the NovAtelSPAN-CPT7 module, supporting GPS L1 / L2 / L5, GLONA SSL1 / L2, and BeiDou B1 / B2 / B3 tri-frequency signal reception. By fusing inertial navigation and satellite positioning data through Kalman filtering, horizontal positioning accuracy reaches ±1.2cm and vertical accuracy ±2.5cm in RTK mode. The built-in MEMS inertial measurement unit has a dynamic response frequency of 200Hz, maintaining centimeter-level positioning accuracy even in dynamic environments such as those with vehicle bumps, achieving precise positioning in dynamic conditions. When a slight component offset is detected, a nanometer positioning correction mechanism is activated, using a piezoelectric ceramic micro-displacement device (resolution ≥10nm) for sub-micrometer-level calibration.

[0028] In the application of digital twin technology, a virtual model mapping 1:1 to the physical power station is constructed based on the Unity3D engine. Terrain data utilizes 1:500 scale LiDAR point cloud data, with a point density of no less than 10 points per square meter, ensuring millimeter-level accuracy for the component model. Lighting simulation employs the Radiance algorithm, comprehensively considering factors such as atmospheric scattering and ground reflection, and synchronizing component position, angle, and environmental parameters in real time.

[0029] In photovoltaic power stations located in complex terrain areas, such as mountainous regions, a terrain radiation model (TRM) is used to calculate solar radiation on the slope, fully considering the effects of terrain shading, reflection, and scattering. The calculation formula is as follows: .in, It represents the solar radiation intensity on the slope, which is the total radiation on the slope after comprehensive calculation; Direct solar radiation intensity refers to solar radiation that arrives directly without scattering or reflection, and is a part of the source of slope radiation. The angle between the sunlight and the slope normal affects the intensity of the direct radiation projected onto the slope. β represents the intensity of scattered solar radiation, which is the radiation that reaches the surface after sunlight has been scattered by atmospheric molecules, aerosols, etc., reflecting the contribution of atmospheric scattering to slope radiation; β is the slope angle, which affects the weight of scattered and reflected radiation in the calculation of slope radiation. The intensity of reflected solar radiation refers to the solar radiation that reaches the slope after being reflected by the ground and other surfaces. In addition, the system acquires the latest solar calendar data via satellite daily at 00:00 and updates astronomical parameters accordingly.

[0030] Intelligent control decision-making module: It constructs a spatiotemporal graph neural network and reinforcement learning fusion architecture, and introduces the meta-learning (MAML) framework and innovatively designs a cognitive reasoning engine to achieve efficient and interpretable decision-making.

[0031] First, in the spatiotemporal graph neural network (STGNN) part, the input layer of the STGNN model receives 128-dimensional features such as illumination, location, and time. The hidden layers consist of three graph convolutional layers and two temporal convolutional layers. In the graph structure, each node represents a sensor or component, and the edge weights are dynamically adjusted based on spatial distance and historical relevance. This model is deployed on the NVIDIA Jetson AGXXavier edge computing platform, with an inference latency of no more than 30ms, enabling rapid processing and analysis of input information.

[0032] For the reinforcement learning part, an improved ParetoDDPG algorithm is used, with optimizations applied to power generation efficiency (weight 0.5), equipment loss (weight 0.3), and carbon footprint (weight 0.2). The state space encompasses 1024-dimensional features, and the action space consists of the component's pitch angle (0-90°) and azimuth angle (0-360°). The agent's experience replay buffer has a capacity of 1 million records, and the target network is updated every 1000 training steps.

[0033] Building upon this foundation, a Meta-Learning (MAML) framework is introduced, enabling the system to rapidly adapt to new environments. In extreme weather conditions (such as sandstorms), a few-shot learning mechanism allows for rapid adjustment of control strategies with only 5-10 samples, ensuring the system continues to operate well under harsh conditions. Furthermore, an innovatively designed cognitive reasoning engine combines symbolic reasoning with neural networks. When the system detects faults such as bird droppings obscuring component surfaces, it can not only accurately identify the fault type but also perform reasoning using a knowledge base. For example, it can infer that the fault occurs under light intensity >800W / m². 2 The best time to clean is when solar power can be used, which makes the decision-making process more interpretable and the decisions more scientific and reasonable.

[0034] Component-driven execution module: It integrates SMA / electromagnetic composite drive and energy recovery technology, and also integrates ion polymer metal composite (IPMC) micro-drive array and dielectric elastomer (DE) artificial muscle to realize component angle adjustment and deal with local occlusion.

[0035] In the SMA / electromagnetic composite drive, the core NiTi shape memory alloy wire has a diameter of 0.4mm, a length of 150mm, and a rated driving force of 250N, paired with a DC24V brushless servo motor AKM2G-AN06. The SMA wire is heated by a 3A peak current and a 50ms pulse width, enabling a fine angle adjustment of 0.5° / s to meet the component angle fine-tuning requirements; while the motor provides 2.5° / s... 2 Its rapid response allows for quick adjustment of component angles when needed.

[0036] The energy recovery system is equipped with a synchronous rectified Buck-Boost converter LM5175 and a capacitor bank consisting of 10 series and 5 parallel 100F / 2.7V supercapacitors. When the components fall back due to gravity, the DC power generated by the motor is stored in the capacitors via the converter, effectively improving energy utilization efficiency. Furthermore, the actuator has an IP68 protection rating, capable of withstanding ambient temperatures from -40℃ to +85℃.

[0037] Building upon this foundation, the module developed an ion-polymer-metal composite (IPMC) micro-actuator array, integrating millions of micro-actuators (≤100μm × 100μm in size) onto the module surface. When local shading is detected, these micro-actuators can be controlled to create a "wrinkled" structure on the module surface, cleverly reflecting light to the unshaded area and reducing power generation losses caused by local shading. Simultaneously, the system employs dielectric elastomer (DE) artificial muscles as auxiliary actuators, achieving an energy density of 3.4 J / cm². 3 It can complete the emergency adjustment of component angles within 0.1 seconds, further enhancing the system's ability to cope with emergencies.

[0038] Fault diagnosis and self-healing module: integrates multimodal monitoring and intelligent repair technology, covering multiple methods such as fiber optic monitoring, terahertz detection, nano-positioning correction and laser repair.

[0039] For multimodal monitoring, a 125μm diameter photonic crystal fiber is first embedded within the component frame, and Brillouin scattering signals are detected using an optical time-domain reflectometer. A 355nm pulsed laser with a pulse width of 10ns and a repetition frequency of 10kHz is employed, achieving a spatial resolution of 0.5m and a strain measurement accuracy of ±8με for monitoring the strain of the component. Simultaneously, terahertz detection technology is used. On one hand, a quantum cascade laser emits 0.6THz electromagnetic waves, and the reflected signals are received by a Schottky diode detector, enabling the detection of microcracks ≥50μm inside the component at a detection speed of 50mm. 2 / s; On the other hand, the system innovatively applies terahertz time-domain spectroscopy (THz-TDS) technology to analyze the propagation characteristics of electromagnetic waves in the 0.1-3THz frequency band within components, identifying microcracks with a precision ≤50μm in the packaging material. The system also establishes a database of correspondences between terahertz spectral features and defect types, and employs transfer learning methods to transfer laboratory data to real-world application scenarios, improving the accuracy and applicability of the detection.

[0040] In terms of intelligent repair, the nano-positioning correction uses the PZT piezoelectric ceramic micro-displacement device P-601.1CD, with a stroke range of ±40μm and a resolution of 8nm. It is combined with a high-precision capacitive displacement sensor with a resolution of 1nm to form a closed-loop control system to achieve precise positioning. When a microcrack is detected inside the battery cell, the system uses laser repair technology to locally weld the crack.

[0041] The invention also includes a bio-inspired cluster control module: the bio-inspired cluster control module realizes component self-organization and collaboration based on the ant colony algorithm, and imitates the pheromone mechanism of the ant colony algorithm to optimize the angle adjustment of the solar module and improve power generation efficiency.

[0042] The module uses virtual pheromone concentration The formula for measuring the attraction strength of component i to component j is as follows: .in, =0.05 is the evaporation coefficient, representing the rate at which pheromones evaporate over time, indicating that the pheromone concentration will naturally decrease over time; β=0.2 is the gain coefficient, used to measure the effect of synergistic power generation gain on the pheromone concentration; Δt is the time interval; The power generation gain of components i and j reflects the change in power generation efficiency brought about by the collaborative work between components. This parameter correlates the pheromone concentration with the actual power generation efficiency.

[0043] Each component broadcasts its own pheromone concentration every 10 seconds and receives information from neighboring components within a 15m radius. When a component detects a change in illumination, it releases "attractant" to neighboring components via wireless communication; the attractant intensity decreases with distance (attenuation coefficient = 0.9 / m). Then, based on a gradient climbing algorithm, the component adjusts its angle according to the received pheromone concentration gradient, achieving a self-organized collaborative adjustment mode. When partial shading is detected, the system automatically forms a "wave-like" adjustment mode, with peak heights between 0.5-1.5m and wavelengths between 3-8m. In this mode, rear-row components can capture more sunlight through the gaps between front-row components.

[0044] This invention also includes a quantum encrypted communication module: the quantum encrypted communication module achieves secure communication based on quantum key distribution. Utilizing a periodically polarized lithium niobate crystal, entangled photon pairs with a wavelength of 780 nm are generated through a parametric down-conversion process, achieving a generation rate of 10-1. 6 The system operates at a rate of 12kbps per second. Single-mode fiber is used as the quantum channel, with an attenuation coefficient not exceeding 0.3dB / km. Quantum repeaters based on atomic ensemble quantum memories are deployed every 50km. Key negotiation uses the BB84 protocol, employing a decoy state method to detect eavesdropping. The system generates a new 256-bit key every 10 minutes at a generation rate of 12kbps. This module enables the system's communication security to reach the upper limit of quantum mechanics theory, meeting financial-grade security requirements.

[0045] This invention includes the following steps: Spatiotemporal illumination field reconstruction steps: Based on multi-dimensional illumination sensing data, a spatiotemporal illumination field model is constructed using Gaussian process regression (GPR). The system divides the space into a three-dimensional grid (resolution ≤ 0.5m × 0.5m × 1m), and the illumination intensity at each grid point is represented as a function of time and space. in, Represents coordinates in three-dimensional space Light intensity at location and time t; This is a mean function of the light intensity at that location and time, reflecting the average trend of light intensity variation; The noise term, following a Gaussian distribution, characterizes the random fluctuations of actual illumination intensity relative to the mean. A kernel function is used to calculate the correlation between grid points, achieving smooth interpolation of the illumination field. This model can predict the illumination distribution over the next 30 minutes with a prediction error ≤ ±5%. The system also incorporates a Kalman filter for real-time correction of the illumination field prediction, improving prediction accuracy in dynamic environments.

[0046] Component-based collaborative optimization steps: Construct a multi-agent optimization model based on alliance game theory, where each component acts as an agent, negotiating to form the optimal alliance structure. The system defines complementarity functions between components: in, represents the total revenue of alliance S, which measures the overall benefit brought about by the collaboration of components within the alliance; i,j represent the component numbers within alliance S. The complementarity coefficients of components i and j reflect the degree of complementarity between the two components in terms of angle adjustment, light reception, etc., and are obtained through statistical analysis of historical data. This represents the power gain resulting from the coordinated angle adjustment of components i and j. A core stability algorithm is employed to find the optimal coalition, maximizing the overall system power generation efficiency.

[0047] Real-time energy flow scheduling steps: A power electronic multi-port energy router is developed to enable flexible energy exchange between photovoltaic modules, energy storage systems, loads, and the grid. The system employs a model predictive control (MPC) algorithm to predict solar irradiance and load demand for the next 24 hours and optimize energy storage charging and discharging strategies. During periods of sufficient sunlight, the system prioritizes storing energy in solid-state batteries (energy density ≥400Wh / kg); when sunlight is insufficient, the stored energy is converted into AC power for the load using a DC / AC converter. The system also supports V2G (Vehicle-to-Grid) functionality, allowing the stored energy to be fed back to the grid when peak shaving is required.

[0048] Biological-Machine Co-evolution Steps: Combining genetic algorithms with deep reinforcement learning enables continuous evolution of control strategies. The system performs a "strategy evolution" quarterly, generating an initial population through genetic algorithms and then optimizing the strategy using reinforcement learning. An innovative biologically inspired fitness function is introduced, considering not only power generation efficiency but also system robustness, resource utilization, and environmental adaptability. For example, in a desert environment, the system will evolve control strategies more resistant to sandstorms, such as reducing wind speed on component surfaces to minimize dust accumulation.

[0049] This invention also includes a quantum illumination prediction step: utilizing quantum computing to accelerate the solution process of the illumination prediction model. The system transforms the atmospheric radiative transfer equation into a quantum algorithm, and solves the integral equation using the quantum phase estimation (QPE) algorithm, achieving a computational speed improvement compared to classical algorithms. The quantum algorithm can perform in one second the computations that would take a traditional algorithm an hour when dealing with complex atmospheric conditions (such as multi-layered clouds). The system also features a quantum-classical hybrid architecture, allocating critical computational tasks to quantum processors while handling the remaining tasks with classical computers, achieving optimal resource allocation.

[0050] This invention also includes a digital ecosystem symbiosis step: constructing a digital ecosystem for photovoltaic power plants, connecting physical power plants, digital twin models, operation and maintenance personnel, and users into a symbiotic network. The system uses blockchain technology to record all operations and data, ensuring data immutability and traceability. Operation and maintenance personnel interact with the digital twin model through augmented reality (AR) glasses to obtain real-time equipment status information; users participate in power plant management through mobile applications, such as setting power generation priorities and viewing carbon emission reductions. The system establishes an ecological credit mechanism, distributing benefits based on the contributions of each participant, forming a sustainable photovoltaic ecosystem.

[0051] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An adaptive light-adjusting solar module angle control system, characterized in that, Includes the following modules: Multi-dimensional illumination sensing module: Deploys a distributed quantum dot spectral sensor array, and the system innovatively designs a polarization light analysis unit. Combined with the Rayleigh scattering model, it inverts atmospheric aerosol concentration; the calculation formula is as follows: , denoted as atmospheric aerosol concentration; k is an empirical coefficient. To actually measure the degree of polarization of sunlight; This represents the theoretical degree of solar polarization. Geographic Information Processing Module: This module fuses inertial navigation and satellite positioning data using Kalman filtering, innovatively introduces digital twin technology, and calculates solar radiation on slopes using a terrain radiation model. The formula is as follows: , Solar radiation intensity; This refers to the intensity of direct solar radiation. The angle between the sunlight and the slope normal; The intensity of scattered solar radiation, The slope angle; To reflect solar radiation intensity; Intelligent control decision-making module: Construct a spatiotemporal graph neural network based on attention mechanism, innovatively design a multi-objective reinforcement learning framework, adopt Pareto optimal strategy to balance conflicting objectives, and introduce optical flow field analysis technology in cloud movement scenario to predict cloud movement speed and direction; Component-driven execution module: Develop a composite actuator of shape memory alloy and electromagnetic drive. SMA wire provides angle adjustment, electromagnetic drive mechanism provides response. The actuator adopts a biomimetic joint design. The system innovatively designs an energy recovery braking system. During component adjustment, when the motor is in the power generation state, the electrical energy is stored in the supercapacitor through a bidirectional DC / DC converter. Fault diagnosis and self-healing module: Deploy a distributed strain sensing network based on photonic crystal fiber. By measuring the Brillouin divergence frequency shift of the optical signal in the fiber, the system can monitor the structural health of the components. The system innovatively applies deep adversarial generative network to simulate faults. By training a discriminator, it can distinguish between normal and abnormal states. When a slight displacement of the component is detected, the system will activate the correction mechanism.

2. The adaptive light-adjusting solar module angle control system according to claim 1, characterized in that, It also includes a bio-inspired cluster control module: This module mimics the pheromone mechanism of the ant colony algorithm, establishes virtual pheromone trajectories between adjacent components, and when a component detects a change in light, it releases "attractants" to neighboring components via wireless communication. The components adjust their own angles according to the pheromone concentration gradient, forming a self-organizing and coordinated regulation mode.

3. The adaptive light-adjusting solar module angle control system according to claim 1, characterized in that, It also includes a quantum encrypted communication module: it uses quantum key distribution technology based on entangled photon pairs to realize communication between the control center and component nodes, uses the BB84 protocol for key negotiation, detects eavesdropping behavior through decoy state method, and automatically compensates for fiber temperature drift during the establishment of quantum channel.

4. The adaptive light-adjusting solar module angle control system according to claim 1, characterized in that, In the multi-dimensional illumination sensing module: an innovative solar-blind ultraviolet detection unit is designed, utilizing the absorption characteristics of the ozone layer in this band. By measuring the intensity of solar-blind ultraviolet radiation, atmospheric ozone content is retrieved, and a model relating ozone concentration to spectral attenuation is established. in, wavelength The intensity of ultraviolet radiation; The theoretical ultraviolet radiation intensity at this wavelength under conditions of no ozone absorption; For ozone absorption cross section, is the ozone concentration in the atmosphere; L is the optical path length of light in the atmosphere.

5. The adaptive light-adjusting solar module angle control system according to claim 1, characterized in that, In the intelligent control decision-making module: a meta-learning framework is introduced. When encountering extreme weather, the system adjusts the control strategy through a few-shot learning mechanism. An innovative cognitive reasoning engine is designed, combining symbolic reasoning and neural networks to achieve an interpretable decision-making process.

6. The adaptive light-adjusting solar module angle control system according to claim 1, characterized in that, In the component drive execution module: an ion polymer metal composite micro-drive array is developed, and a million-level micro-drive is integrated on the component surface to realize local deformation adjustment of the component surface. When local occlusion is detected, the micro-drive is controlled to reflect light to the unoccluded area. The system uses dielectric elastomer artificial muscle as auxiliary drive.

7. The adaptive light-adjusting solar module angle control system according to claim 1, characterized in that, In the fault diagnosis and self-healing module: terahertz time-domain spectroscopy technology is innovatively applied to detect internal defects of the component. By analyzing the propagation characteristics of electromagnetic waves in the component, microcracks in the encapsulation material are identified. The system establishes a database of correspondence between terahertz spectral features and defect types. The transfer learning method is used to transfer laboratory data to actual application scenarios. When microcracks are detected inside the battery cell, the system uses laser repair technology to locally weld the cracks.

8. A method for an adaptive light-adjusting solar module angle control system based on claim 1, characterized in that, Includes the following steps: Spatiotemporal illumination field reconstruction steps: Based on multi-dimensional illumination sensing data, a spatiotemporal illumination field model is constructed using Gaussian process regression. The system divides the space into a three-dimensional grid, and the illumination intensity of each grid point is represented as a function of time and space. in, In three-dimensional coordinates At point t, the light intensity at time t; Let be a function of the mean light intensity at that location and time. To address the noise term that follows a Gaussian distribution, the correlation between grid points is calculated using a kernel function. The system also introduces a Kalman filter to perform real-time correction of the illumination field prediction. Component-based collaborative optimization steps: Construct a multi-agent optimization model based on alliance game theory, where each component acts as an agent, and the optimal alliance structure is formed through negotiation. The system defines the complementarity functions between components. in, Let i be the total revenue of alliance S, and j be the component numbers within alliance S. Let i and j be the complementarity coefficients. To determine the power gain resulting from the coordinated angle adjustment of components i and j, the core stability algorithm is used to find the optimal coalition. Real-time energy flow scheduling steps: The system adopts a model predictive control algorithm to predict the light and load demand for the next 24 hours. During periods of sufficient light, the system prioritizes storing energy in solid-state batteries. When light is insufficient, the stored energy is converted into AC power for the load through a DC / AC converter. The system also supports V2G function, which feeds the stored energy back to the grid when the grid needs peak shaving. Biological-machine co-evolution steps: combining genetic algorithms with deep reinforcement learning to achieve continuous evolution of control strategies, innovatively introducing a biologically inspired fitness function, considering power generation efficiency, and also including system robustness, resource utilization and environmental adaptability indicators.

9. The method for an adaptive light-adjusting solar module angle control system according to claim 8, characterized in that, It also includes a quantum illumination prediction step: the system transforms the atmospheric radiative transfer equation into a quantum algorithm, solves the integral equation through a quantum phase estimation algorithm, and completes the computation in 1 second that would take a traditional algorithm 1 hour when dealing with complex atmospheric conditions. The system also develops a quantum-classical hybrid architecture, which allocates key computational tasks to the quantum processor and handles the remaining tasks by the classical computer.

10. The method for an adaptive light-adjusting solar module angle control system according to claim 8, characterized in that, It also includes digital ecosystem symbiosis steps: connecting physical power plants, digital twin models, operation and maintenance personnel, and users into a symbiotic network. The system uses blockchain technology to record all operations and data. Operation and maintenance personnel interact with the digital twin model through augmented reality glasses, and users participate in power plant management through mobile applications. The system establishes an ecological credit mechanism to distribute benefits according to the contributions of each participant, forming a sustainable photovoltaic ecosystem.

Images