An adaptive retractable photovoltaic support multi-objective optimization intelligent control system

Abstract

The application relates to the technical field of photovoltaic power generation supports, in particular to a self-adaptive retractable photovoltaic support multi-target optimization intelligent regulation and control system, which comprises a sensing unit, a retractable mechanism unit, a dynamics analysis unit, a multi-target optimization control unit, a digital twin unit and a power supply guarantee unit; each unit realizes data interaction through an industrial-grade wireless communication link; the frequency band of the communication link matches electromagnetic environment characteristics of a photovoltaic power station; and signal transmission timing matches support contraction and expansion action response characteristics; the output data dimension of the sensing unit covers the core features of photovoltaic support structure safety monitoring, meteorological environment adaptation and power generation efficiency optimization; through the screw theory and Lie group Lie algebra, the bottom configuration synthesis of a reversible deformation mechanism is realized; combined with multi-scale topological optimization and composite material design, the mechanism contraction storage efficiency and the structure lightweight level are greatly improved, and the load influence under extreme weather is effectively reduced.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power generation support technology, and in particular to an adaptive retractable photovoltaic support multi-objective optimization intelligent control system. Background Technology

[0002] Traditional photovoltaic (PV) support structures are mostly fixed or simple single-axis / dual-axis tracking structures, with limited functionality, poor environmental adaptability, and weak resistance to extreme weather. Existing movable PV support structures often focus only on power generation tracking, neglecting the ability to quickly and safely retract and maintain their posture under harsh conditions such as strong winds and heavy snow, and lacking a dynamic balance mechanism between power generation efficiency and structural safety. Their design methods mostly rely on static analysis, insufficiently considering the multibody dynamics during the deployment / retraction process, the nonlinear vibration of gapped mechanisms, and the structure-control coupling effect. This results in low system reliability, conservative control strategies, and difficulty in meeting the needs of large-scale, refined operation and maintenance of PV power plants under complex meteorological environments.

[0003] In summary, existing technologies suffer from problems such as unreasonable design of the retractable mechanism, lack of dynamic analysis, single control strategy, and lack of closed-loop regulation throughout the entire process. These issues result in insufficient environmental adaptability and operational reliability of photovoltaic brackets, making it difficult to balance power generation efficiency and structural safety. There is an urgent need for an integrated system that combines a highly efficient retractable mechanism, precise dynamic analysis, and multi-objective intelligent regulation. Summary of the Invention

[0004] The purpose of this invention is to solve the technical problems of insufficient environmental adaptability and operational reliability caused by unreasonable design of the shrinkable mechanism, lack of dynamic analysis, single control strategy, and lack of full-process closed-loop control in the existing technology. The embodiments of this invention provide an adaptive shrinkable photovoltaic support multi-objective optimization intelligent control system.

[0005] This system achieves a dynamic balance between photovoltaic bracket power generation efficiency and structural safety through the coordinated operation of sensing units, retractable mechanism units, dynamic analysis units, multi-objective optimization control units, digital twin units, power supply guarantee units, and closed-loop feedback correction units, thereby improving the system's environmental adaptability and operational reliability.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an adaptive retractable photovoltaic support multi-objective optimization intelligent control system, comprising a sensing unit, a retractable mechanism unit, a dynamic analysis unit, a multi-objective optimization control unit, a digital twin unit, and a power supply guarantee unit;

[0007] Each unit interacts with data through an industrial-grade wireless communication link. The frequency band of the communication link matches the electromagnetic environment characteristics of the photovoltaic power station, and the signal transmission timing matches the response characteristics of the support retraction and deployment actions.

[0008] The output data dimensions of the sensing unit cover the core features of photovoltaic support structure safety monitoring, meteorological environment adaptation, and power generation efficiency optimization.

[0009] The retractable mechanism unit is synthesized using spinor theory and Lie group Lie algebra to form a reversible deformation structure. The structure body is a composite material structure optimized by multi-scale topology, and the volume in the contracted state is significantly reduced compared to the unfolded state.

[0010] The dynamic analysis unit is equipped with a multibody dynamics model that includes gaps and structural flexibility. The model can simulate the nonlinear dynamic behavior of the stent's deployment and contraction process.

[0011] The multi-objective optimization control unit adopts a model predictive control algorithm to construct a multi-objective optimization function that integrates maximizing power generation efficiency and minimizing structural safety risks. The output end establishes a bidirectional hardware link with the drive module of the retractable mechanism unit.

[0012] The digital twin unit constructs a high-fidelity digital twin model of the support system, integrates a hardware-in-the-loop test platform based on dSPACE, and connects the model calibration link to the real-time data output of the sensing unit; each unit forms a linkage with the sensing decision-making action logic through hardware signal links.

[0013] Preferably, the sensing unit includes a meteorological environment sensing component, a structural state sensing module, and a data integration module; the monitoring dimensions of the meteorological environment sensing component cover the core meteorological parameters required for the photovoltaic support to resist wind and snow loads.

[0014] The response characteristics of the sensing elements in the structural state sensing module match the dynamic behavior of stress, deformation, and vibration of the support structure.

[0015] The output of the data integration module is physically connected to the model input interface of the dynamic analysis unit through a state calibration link, and the output fused data format is compatible with the analytical logic of the multibody dynamics model.

[0016] Preferably, the meteorological environment sensing component includes a wind speed and direction sensor, a snowfall sensor, a light intensity sensor, and an ambient temperature sensor.

[0017] The detection frequency band of the wind speed and direction sensor covers the wind speed range of the critical working condition of the photovoltaic support, and the sensing threshold of the snowfall sensor matches the critical value of the snow load on the support.

[0018] The sampling frequency of the light intensity sensor matches the time scale for optimizing photovoltaic tracking power generation efficiency, and the response range of the ambient temperature sensor covers the temperature range of changes in the mechanical properties of the support material.

[0019] The output of this component is connected to the data integration module together with the output of the structure state perception module. The integrated meteorological structure correlation data provides the basic input for multi-objective optimization.

[0020] Preferably, the structural state sensing module includes stress sensors, displacement sensors, and vibration sensors deployed at key nodes of the support, and also includes a gap state sensor integrated at the retractable joint.

[0021] The response range of the stress sensor covers the allowable stress range of the composite material structure, and the measurement accuracy of the displacement sensor matches the accuracy requirements of the bracket deformation monitoring.

[0022] The detection parameters of the vibration sensor are matched with the frequency characteristics of the nonlinear vibration of the bracket, and the identification criterion of the gap state sensor is the difference in physical characteristics of the joint gap of the retractable mechanism.

[0023] The signal transmission carrier of each sensing element is an insulating structure compatible with the electromagnetic environment of the photovoltaic power station, and its dielectric properties match the electric field distribution characteristics inside the support array.

[0024] Preferably, the retractable mechanism unit includes a reversible deformable structure body, a drive module, and a timing coordination module; the reversible deformable structure body completes configuration synthesis through spinor theory and Lie group Lie algebra, and achieves multi-scale topology optimization through a multi-disciplinary optimization platform;

[0025] The drive module is driven by a servo motor, and the output torque matches the load characteristics of the bracket's retraction and expansion. The drive signal is synchronized with the command timing of the multi-objective optimization control unit.

[0026] The synchronization benchmark of the timing coordination module is the operation time scale of the segmented contraction and expansion of the support. The timing coordination data output by the module provides time dimension support for the action allocation of the retractable mechanism unit through the timestamp encapsulation board.

[0027] Preferably, the dynamic analysis unit includes a multibody dynamics modeling component, a nonlinear behavior simulation component, and a load coupling analysis component;

[0028] The multibody dynamics modeling component is equipped with a dynamic model that includes gaps and structural flexibility, and the model parameters are calibrated by mechanical test data of the support structure; the simulation logic of the nonlinear behavior simulation component matches the vibration and impact characteristics of the support deployment and contraction process.

[0029] The load coupling analysis component integrates the coupling link between structural mechanics analysis and fluid dynamics wind load calculation. The analysis result output end is physically connected to the decision input interface of the multi-objective optimization control unit, and the output data is compatible with the constraint logic of the multi-objective optimization function.

[0030] Preferably, the multi-objective optimization control unit includes an MPC controller, a multi-objective function construction component, and a decision output module;

[0031] The algorithm parameters of the MPC controller are optimized through the Isight multidisciplinary optimization platform. The optimization strategy adopts a hybrid strategy combining sequential quadratic programming and genetic algorithm. The multi-objective optimization function is composed of the product of power generation efficiency weight coefficient and power generation efficiency, minus the product of structural safety risk value weight coefficient and structural safety risk value.

[0032] The output commands of the decision output module include support angle adjustment commands and contraction / expansion trigger commands. The command parameters are matched with the motion characteristics of the retractable mechanism unit and the analysis results of the dynamic analysis unit.

[0033] Preferably, the digital twin unit includes a high-fidelity model building component, a HIL testing component, and a model calibration module;

[0034] The model dimensions of the high-fidelity model building component match the physical characteristics of the photovoltaic support and environmental coupling system, and the model accuracy is calibrated by real-time data from the sensing unit.

[0035] The HIL test component is built on the dSPACE platform, and the test logic matches the real-time and reliability verification requirements of the multi-objective optimization control strategy. The calibration link of the model calibration module interacts in real time with the structural state data and meteorological environment data of the sensing unit, and maintains the consistency between the digital twin model and the physical support through iterative correction.

[0036] Preferably, the power supply guarantee unit adopts a dual-mode power supply mode that combines photovoltaic self-generation and self-consumption with energy storage backup, and has a built-in power management chip that can dynamically adjust the output voltage and power according to the instructions of the multi-objective optimization control unit.

[0037] When the support structure is in a retracted safety state, the power supply unit automatically reduces the power supply to non-core modules; when the support structure is in an extended power generation state, priority is given to ensuring the power supply stability of the sensing unit, multi-objective optimization control unit, and digital twin unit; this unit has overvoltage, overcurrent, and overheat protection functions, and the protection threshold matches the safety characteristics of the photovoltaic support structure's electrical system.

[0038] Preferably, the system further includes a closed-loop feedback correction unit; the input end of the closed-loop feedback correction unit establishes a data connection with the output ends of the sensing unit, the dynamic analysis unit, and the digital twin unit to collect the actual action parameters of the support, structural status data, and power generation efficiency data;

[0039] Based on the deviation between the output of the multi-objective optimization function and the actual operating data, the action parameters of the retractable mechanism unit, the model coefficients of the dynamic analysis unit, and the algorithm parameters of the MPC controller are corrected in reverse; the dynamic balance requirements of the power generation efficiency and structural safety of the correction direction matching support are met, thus forming a complete control process.

[0040] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0041] By using spinor theory and Lie group Lie algebra to achieve the underlying configuration synthesis of reversible deformation mechanisms, and combining multi-scale topology optimization and composite material design, the mechanism shrinkage and storage efficiency and structural lightweight level are greatly improved, effectively reducing the load impact under extreme weather conditions.

[0042] By leveraging a multibody dynamics model with gaps and structural flexibility, and load coupling analysis, nonlinear dynamic behavior is accurately simulated, providing high-fidelity data support for regulation and control, thus solving the reliability problem caused by traditional static design. Through model predictive control algorithms and multi-objective optimization functions, real-time meteorological and structural state information is deeply integrated to achieve a dynamic balance between power generation efficiency and structural safety.

[0043] Relying on a high-fidelity digital twin model and a hardware-in-the-loop testing platform, the effectiveness of control strategies is verified in advance and the model is continuously calibrated, shortening the R&D cycle and improving the control accuracy. Each unit forms a closed-loop control architecture for the entire process, and the operating performance is continuously optimized through multi-dimensional parameter correction, significantly enhancing the system's environmental adaptability, operational reliability, and full life cycle operation and maintenance adaptability, providing reliable technical support for the large-scale and refined operation of photovoltaic power plants. Attached Figure Description

[0044] Figure 1 This is a system flowchart of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0046] In the description of this invention, it should be understood that the terms length, width, up, down, front, back, left, right, vertical, horizontal, top, bottom, inside, outside, etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the 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, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "multiple" means two or more, unless otherwise explicitly specified.

[0047] according to Figure 1 As shown, this invention proposes an adaptive retractable photovoltaic support multi-objective optimization intelligent control system, including a sensing unit, a retractable mechanism unit, a dynamic analysis unit, a multi-objective optimization control unit, a digital twin unit, and a power supply guarantee unit;

[0048] Each unit interacts with data through an industrial-grade wireless communication link. The frequency band of the communication link matches the electromagnetic environment characteristics of the photovoltaic power station, and the signal transmission timing matches the response characteristics of the support retraction and deployment actions.

[0049] The output data dimensions of the sensing unit cover the core features of photovoltaic support structure safety monitoring, meteorological environment adaptation, and power generation efficiency optimization.

[0050] The retractable mechanism unit is synthesized using spinor theory and Lie group Lie algebra to form a reversible deformation structure. The structure body is a composite material structure optimized by multi-scale topology, and the volume in the contracted state is significantly reduced compared to the unfolded state.

[0051] The dynamic analysis unit is equipped with a multibody dynamics model that includes gaps and structural flexibility. The model can simulate the nonlinear dynamic behavior of the stent's deployment and contraction process.

[0052] The multi-objective optimization control unit adopts a model predictive control algorithm to construct a multi-objective optimization function that integrates maximizing power generation efficiency and minimizing structural safety risks. The output end establishes a bidirectional hardware link with the drive module of the retractable mechanism unit.

[0053] The digital twin unit constructs a high-fidelity digital twin model of the support system, integrates a hardware-in-the-loop test platform based on dSPACE, and connects the model calibration link to the real-time data output of the sensing unit; each unit forms a linkage with the sensing decision-making action logic through hardware signal links.

[0054] As an optional embodiment, the sensing unit includes a meteorological environment sensing component, a structural state sensing module, and a data integration module; the monitoring dimensions of the meteorological environment sensing component cover the core meteorological parameters required for the photovoltaic support to resist wind and snow loads.

[0055] The response characteristics of the sensing elements in the structural state sensing module match the dynamic behavior of stress, deformation, and vibration of the support structure.

[0056] The output of the data integration module is physically connected to the model input interface of the dynamic analysis unit through a state calibration link, and the output fused data format is compatible with the analytical logic of the multibody dynamics model.

[0057] As an optional embodiment, the meteorological environment sensing component includes a wind speed and direction sensor, a snowfall sensor, a light intensity sensor, and an ambient temperature sensor.

[0058] The detection frequency band of the wind speed and direction sensor covers the wind speed range of the critical working condition of the photovoltaic support, and the sensing threshold of the snowfall sensor matches the critical value of the snow load on the support.

[0059] The sampling frequency of the light intensity sensor matches the time scale for optimizing photovoltaic tracking power generation efficiency, and the response range of the ambient temperature sensor covers the temperature range of changes in the mechanical properties of the support material.

[0060] The output of this component is connected to the data integration module together with the output of the structure state perception module. The integrated meteorological structure correlation data provides the basic input for multi-objective optimization.

[0061] As an optional embodiment, the structural state sensing module includes stress sensors, displacement sensors, and vibration sensors deployed at key nodes of the support, and also includes a gap state sensor integrated at the retractable joint.

[0062] The response range of the stress sensor covers the allowable stress range of the composite material structure, and the measurement accuracy of the displacement sensor matches the accuracy requirements of the bracket deformation monitoring.

[0063] The detection parameters of the vibration sensor are matched with the frequency characteristics of the nonlinear vibration of the bracket, and the identification criterion of the gap state sensor is the difference in physical characteristics of the joint gap of the retractable mechanism.

[0064] The signal transmission carrier of each sensing element is an insulating structure compatible with the electromagnetic environment of the photovoltaic power station, and its dielectric properties match the electric field distribution characteristics inside the support array.

[0065] As an optional embodiment, the retractable mechanism unit includes a reversible deformable structure body, a drive module, and a timing coordination module; the reversible deformable structure body completes configuration synthesis through spinor theory and Lie group Lie algebra, and achieves multi-scale topology optimization through a multi-disciplinary optimization platform;

[0066] The drive module is driven by a servo motor, and the output torque matches the load characteristics of the bracket's retraction and expansion. The drive signal is synchronized with the command timing of the multi-objective optimization control unit.

[0067] The synchronization benchmark of the timing coordination module is the operation time scale of the segmented contraction and expansion of the support. The timing coordination data output by the module provides time dimension support for the action allocation of the retractable mechanism unit through the timestamp encapsulation board.

[0068] As an optional embodiment, the dynamic analysis unit includes a multibody dynamics modeling component, a nonlinear behavior simulation component, and a load coupling analysis component;

[0069] The multibody dynamics modeling component is equipped with a dynamic model that includes gaps and structural flexibility, and the model parameters are calibrated by mechanical test data of the support structure; the simulation logic of the nonlinear behavior simulation component matches the vibration and impact characteristics of the support deployment and contraction process.

[0070] The load coupling analysis component integrates the coupling link between structural mechanics analysis and fluid dynamics wind load calculation. The analysis result output end is physically connected to the decision input interface of the multi-objective optimization control unit, and the output data is compatible with the constraint logic of the multi-objective optimization function.

[0071] As an optional embodiment, the multi-objective optimization control unit includes an MPC controller, a multi-objective function construction component, and a decision output module;

[0072] The algorithm parameters of the MPC controller are optimized through the Isight multidisciplinary optimization platform. The optimization strategy adopts a hybrid strategy combining sequential quadratic programming and genetic algorithm. The multi-objective optimization function is composed of the product of power generation efficiency weight coefficient and power generation efficiency, minus the product of structural safety risk value weight coefficient and structural safety risk value.

[0073] The output commands of the decision output module include support angle adjustment commands and contraction / expansion trigger commands. The command parameters are matched with the motion characteristics of the retractable mechanism unit and the analysis results of the dynamic analysis unit.

[0074] As an optional embodiment, the digital twin unit includes a high-fidelity model building component, a HIL testing component, and a model calibration module;

[0075] The model dimensions of the high-fidelity model building component match the physical characteristics of the photovoltaic support and environmental coupling system, and the model accuracy is calibrated by real-time data from the sensing unit.

[0076] The HIL test component is built on the dSPACE platform, and the test logic matches the real-time and reliability verification requirements of the multi-objective optimization control strategy. The calibration link of the model calibration module interacts in real time with the structural state data and meteorological environment data of the sensing unit, and maintains the consistency between the digital twin model and the physical support through iterative correction.

[0077] As an optional embodiment, the power supply guarantee unit adopts a dual-mode power supply mode that combines photovoltaic self-generation and self-consumption with energy storage backup, and has a built-in power management chip that can dynamically adjust the output voltage and power according to the instructions of the multi-objective optimization control unit.

[0078] When the support structure is in a retracted safety state, the power supply unit automatically reduces the power supply to non-core modules; when the support structure is in an extended power generation state, priority is given to ensuring the power supply stability of the sensing unit, multi-objective optimization control unit, and digital twin unit; this unit has overvoltage, overcurrent, and overheat protection functions, and the protection threshold matches the safety characteristics of the photovoltaic support structure's electrical system.

[0079] As an optional embodiment, the system further includes a closed-loop feedback correction unit; the input end of the closed-loop feedback correction unit establishes a data connection with the output ends of the sensing unit, the dynamic analysis unit, and the digital twin unit to collect the actual action parameters of the support, structural status data, and power generation efficiency data;

[0080] Based on the deviation between the output of the multi-objective optimization function and the actual operating data, the action parameters of the retractable mechanism unit, the model coefficients of the dynamic analysis unit, and the algorithm parameters of the MPC controller are corrected in reverse; the dynamic balance requirements of the power generation efficiency and structural safety of the correction direction matching support are met, thus forming a complete control process.

[0081] according to Figure 1 The diagram shown is a structural schematic of the adaptive retractable photovoltaic support multi-objective optimization intelligent control system provided in this application embodiment. The system includes a sensing unit, a retractable mechanism unit, a dynamic analysis unit, a multi-objective optimization control unit, a digital twin unit, a power supply guarantee unit, and a closed-loop feedback correction unit. Each unit achieves data interaction through an industrial-grade wireless communication link. The signal transmission timing is matched with the support retraction and deployment action response characteristics to ensure the real-time performance and reliability of command transmission.

[0082] I. Perceptual Unit;

[0083] The sensing unit includes a meteorological environment sensing component, a structural state sensing module, and a data integration module (using an STM32H743 processor), which is used to collect multi-dimensional data required for the operation of the photovoltaic support, providing basic input for subsequent analysis and control.

[0084] (a) Meteorological environment sensing components;

[0085] The components include a wind speed and direction sensor (model: RS485 wind speed and direction transmitter), a snowfall sensor (model: optical snowfall monitor), a light intensity sensor (model: BH1750FVI), and an ambient temperature sensor (model: DS18B20). The installation positions of each sensor are precisely matched to its monitoring function.

[0086] The wind speed and direction sensor is installed on the top of the support column. The detection frequency band is 0-60m / s, covering the wind speed range of the critical wind resistance conditions of the photovoltaic support. The measurement accuracy is ±0.1m / s, and it can collect wind speed and direction data in real time.

[0087] The snowfall sensor is installed 10cm above the edge of the photovoltaic panel. The sensing threshold matches the critical value of the snow load on the bracket. The measurement range is 0-50mm / h, the accuracy is ±0.1mm, and it can identify the snowfall intensity and snow thickness.

[0088] The light intensity sensor is uniformly deployed on the surface of the photovoltaic panel, with a sampling frequency of 1Hz, matching the time scale for optimizing photovoltaic power generation efficiency. The measurement range is 0-65535 lux, with an accuracy of ±1 lux, capturing changes in light intensity in real time.

[0089] An ambient temperature sensor is embedded inside the crossbeam of the support frame, with a response range of -40℃ to 85℃, covering the temperature range of changes in the mechanical properties of the support frame material, and a measurement accuracy of ±0.5℃, to monitor the ambient temperature of the support frame during operation.

[0090] (ii) Structural state sensing module

[0091] The module includes stress sensors (model: HBM1-C2A / 100N), displacement sensors (model: KEYENCEGT2-H12K), vibration sensors (model: PCB352C65) deployed at key nodes of the support, and a gap status sensor (model: Micro-EpsilonoptoNCDT1420) integrated at the retractable joint.

[0092] The stress sensor is attached to key stress-bearing parts such as the connection between the support column and the crossbeam, and near the retractable joint. The response range is 0-100MPa, covering the allowable stress range of composite material structures. The measurement accuracy is ±0.5%FS, and the stress changes of the structure are monitored in real time.

[0093] The displacement sensor is installed at the connection of the expansion joint of the bracket, with a measurement range of 0-500mm and an accuracy of ±0.01mm, matching the accuracy requirements of bracket deformation monitoring and capturing the amount of structural displacement.

[0094] The vibration sensor is fixed in the middle of the crossbeam and detects vibration signals in the frequency range of 0-10kHz. It matches the frequency characteristics of the nonlinear vibration of the support and has a sensitivity of 100mV / g to collect structural vibration data.

[0095] The gap status sensor is deployed on the inside of the retractable joint, with a measurement range of 0-5mm and an accuracy of ±0.001mm. It uses the difference in physical characteristics of the joint gap as the identification benchmark to monitor joint wear and gap changes.

[0096] The signal transmission carrier of each sensing element is a polytetrafluoroethylene insulated structure, with dielectric properties matching the electric field distribution characteristics inside the support array. It also has an internal independent signal microchannel to ensure signal transmission anti-interference capability.

[0097] (iii) Data integration module;

[0098] The module receives raw data from the meteorological environment sensing component and the structural state sensing module, performs noise reduction processing through Kalman filtering (process noise variance Q=0.01, observation noise variance R=0.1) and median filtering (window size 5), integrates the data in the format of "meteorological parameters-structural parameters-timestamp", and outputs fused data format compatible with the multibody dynamics model analytical logic of the dynamics analysis unit. It is transmitted to the dynamics analysis unit through the state calibration link (shielded twisted pair transmission).

[0099] II. Retractable mechanism unit;

[0100] The retractable mechanism unit includes a reversible deformable structure body, a drive module, and a timing coordination module. Based on spinor theory and Lie group Lie algebra, it performs underlying mechanical configuration synthesis to achieve efficient contraction and stable deployment.

[0101] (a) Reversible deformation structure body;

[0102] The structural body abandons the traditional approach of splicing rods and uses spinor theory and Lie group Lie algebra to achieve configuration synthesis, generating an optimized configuration that can realize a predetermined folding / unfolding trajectory. The structural body is made of high-strength steel and aluminum alloy composite material, and multi-scale topology optimization is performed through the Isight multidisciplinary optimization platform.

[0103] The optimization process integrates structural mechanics analysis (using ANSYS software) and fluid dynamics wind load calculation (using Fluent software), and employs a hybrid strategy combining sequential quadratic programming and genetic algorithms to simultaneously optimize structural topology and material distribution.

[0104] In the retracted state, the photovoltaic panels are stacked, and the beams and columns are folded and stored, significantly reducing the volume compared to the unfolded state, effectively reducing the windward area and snow load under extreme weather conditions;

[0105] High-strength steel components with composite material structures are used for key load-bearing nodes (such as columns and joint shafts), while aluminum alloy components are used for telescopic beams and auxiliary supports, balancing structural strength and lightweight requirements.

[0106] (ii) Drive module;

[0107] The drive module is driven by a servo motor (model: Panasonic MSMF042L1U2M) equipped with a planetary gear reducer (reduction ratio 1:50) with an output torque of 50 N·m, matching the load characteristics of the bracket retraction and expansion; the motor has a built-in absolute encoder (17-bit resolution) to provide real-time feedback on motor rotation angle and speed data; the drive signal is transmitted via pulse width modulation (PWM) and synchronized with the command timing of the multi-objective optimization control unit, with a response delay of ≤10ms.

[0108] (iii) Timing Coordination Module;

[0109] The module has a built-in time-stamped calibration crystal oscillator (frequency stability ±1ppm), and the synchronization reference is the operation time scale of the segmented retraction and expansion of the support (each segment action duration 0.5-2s). The output timing coordination data provides time dimension support for the action allocation of the retractable mechanism unit through the timestamp encapsulation board (supporting millisecond-level time synchronization), ensuring that the multi-segment retraction and expansion actions are connected smoothly and avoiding action conflicts.

[0110] III. Dynamics Analysis Unit;

[0111] The dynamic analysis unit includes a multibody dynamics modeling component, a nonlinear behavior simulation component, and a load coupling analysis component. It is equipped with a multibody dynamics model that includes gaps and structural flexibility to accurately simulate the dynamic behavior of the support structure.

[0112] (a) Multibody dynamics modeling components;

[0113] The component was built using ADAMS software, which included a multibody dynamics model with gaps and structural flexibility. The model parameters were calibrated through mechanical tests of the support structure.

[0114] The gap parameters are measured using a laser interferometer to determine the actual gap size of the retractable joint (measurement accuracy ±0.001mm), and then input into the model as the gap excitation.

[0115] The structural flexibility parameters, such as the elastic modulus and Poisson's ratio of the composite material, are obtained through tensile testing (using an MTS universal testing machine). The natural frequencies and mode shapes of the structure are obtained through modal testing (using the hammer impact method) and incorporated into the model to simulate the elastic deformation of the structure.

[0116] The model can accurately simulate the nonlinear dynamics of the scaffold during its deployment and retraction, including impact vibrations caused by joint gaps and deformation coupling caused by structural flexibility.

[0117] (ii) Nonlinear behavior simulation components;

[0118] The simulation logic of the component matches the vibration and impact characteristics of the support during the deployment and retraction process. The simulation step size is set to 1ms, and the simulation duration covers the complete retraction / deployment cycle (5-10s). By outputting the displacement, velocity, acceleration, and stress time history curves of key nodes of the support through simulation, the peak dynamic response and dangerous working conditions are identified, providing data support for the design of control strategies.

[0119] (iii) Load coupling analysis component;

[0120] The component integrates a coupling link between structural mechanics analysis and fluid dynamics wind load calculation. It converts real-time wind speed and direction data collected by the meteorological environment sensing component into wind loads. Combined with snow load and structural self-weight, it calculates the instantaneous total load on the support through a load coupling algorithm (based on the Newton-Lagrange equation). The analysis result output is physically connected to the decision input interface of the multi-objective optimization control unit (using a CAN bus interface), and the output data is compatible with the constraint logic of the multi-objective optimization function.

[0121] IV. Multi-objective optimization control unit;

[0122] The multi-objective optimization control unit includes an MPC controller (based on the TITMS320C6748 chip), a multi-objective function construction component, and a decision output module. It uses a model predictive control algorithm to achieve multi-objective optimization regulation.

[0123] (a) MPC controller;

[0124] The controller's algorithm parameters were optimized using the Isight multidisciplinary optimization platform. The optimization strategy employed a hybrid approach combining sequential quadratic programming and genetic algorithms.

[0125] The prediction time domain is set to 50 steps, the control time domain is set to 10 steps, and the sampling time is 100ms.

[0126] Constraints include the bracket angle range (0°-90°), motor output torque limit (0-50 N·m), and structural stress threshold (≤80 MPa) to ensure that control commands are within safe boundaries.

[0127] (ii) Components for constructing multi-objective functions;

[0128] The multi-objective optimization function is composed of the product of the power generation efficiency weighting coefficient and the power generation efficiency, minus the product of the structural safety risk value weighting coefficient and the structural safety risk value.

[0129] The power generation efficiency is calculated by combining real-time irradiance and temperature data of the photovoltaic panel with the photovoltaic module efficiency model.

[0130] The structural safety risk value is calculated by weighting the stress data output by the dynamic analysis unit and the ratio of peak vibration to allowable value.

[0131] The weighting coefficients are determined by the analytic hierarchy process. The power generation efficiency weighting coefficient ω1 ranges from 0.3 to 0.7, and the structural safety risk value weighting coefficient ω2 ranges from 0.3 to 0.7. These values ​​can be dynamically adjusted according to the season and meteorological environment (e.g., ω2 is 0.7 during extreme weather seasons and ω1 is 0.7 during sunny seasons).

[0132] (iii) Decision Output Module;

[0133] The module's output commands include support angle adjustment commands and retraction / expansion trigger commands. The command parameters are simultaneously matched with the motion characteristics of the retractable mechanism unit and the analysis results of the dynamic analysis unit.

[0134] The angle adjustment command controls the rotation angle of the servo motor through pulse signals, with an adjustment accuracy of ±0.1°, enabling real-time tracking of the photovoltaic panel.

[0135] When the structural safety risk value exceeds the threshold (≥0.8) or the meteorological parameters reach the extreme threshold (wind speed ≥25m / s or snow thickness ≥10cm), a contraction command is triggered, and the support is controlled to fold and contract according to the predetermined sequence.

[0136] The instructions are transmitted to the drive module of the retractable mechanism unit via industrial Ethernet, with a transmission delay of ≤10ms.

[0137] V. Digital Twin Unit;

[0138] The digital twin unit includes a high-fidelity model building component, a HIL testing component, and a model calibration module, which constructs a virtual mapping of the entire lifecycle of the support system to support the verification and optimization of control strategies.

[0139] (a) High-fidelity model building components;

[0140] A high-fidelity digital twin model was jointly constructed using Unity3D and MATLAB / Simulink, with the model dimensions matching the physical characteristics of the photovoltaic support system coupled with the environment.

[0141] The geometric model reproduces the dimensions of the support structure, details of the retractable joints, and material properties at a 1:1 scale.

[0142] The physical model integrates multibody dynamics, photovoltaic power generation, and meteorological environment models, and can simulate the dynamic behavior of the support structure, power generation performance, and the interaction between the environment.

[0143] The initial parameters of the model were calibrated using physical prototype test data to ensure consistency between the virtual model and the physical support.

[0144] (ii) HIL test components;

[0145] A HIL test platform was built based on the dSPACE platform (model: DS1202) to test the real-time performance and reliability verification requirements of multi-objective optimization control strategies.

[0146] The platform includes a real-time processor, I / O interface boards, and a load simulator, which can simulate the input and output signals of the bracket drive module and sensors.

[0147] The hardware circuit board of the control unit is connected to the platform, and the simulation data generated by the virtual model interacts with the control unit to verify the real-time performance, stability and robustness of the control algorithm;

[0148] The test scenarios include normal light tracking, strong wind contraction, heavy snow protection, and fault simulation, covering typical operating conditions of the support structure.

[0149] (iii) Model calibration module;

[0150] The module's calibration link interacts in real time with the structural state data and meteorological environment data of the sensing unit, maintaining consistency between the digital twin model and the physical support through iterative correction.

[0151] The physical support status data (angle, stress, vibration, etc.) is collected every 10 seconds.

[0152] Calculate the deviation between the virtual model output and the physical data. If the deviation exceeds the threshold (≥5%), correct the model parameters (such as joint friction coefficient and structural stiffness coefficient) using the least squares method to ensure model accuracy.

[0153] VI. Power Supply Guarantee Unit;

[0154] The power supply backup unit adopts a dual-mode power supply combining photovoltaic self-generation and self-consumption with energy storage backup. It has a built-in power management chip (model: TIBQ76952) that can dynamically adjust the output voltage and power according to the instructions of the multi-objective optimization control unit.

[0155] The self-consumption portion of the photovoltaic system generates electricity through photovoltaic panels (power ≥ 50W) that are matched with the support frame, and then supplies power after being regulated by a DC-DC converter (input voltage 12-24V, output voltage 12V);

[0156] The energy storage backup section uses a lithium battery pack (capacity 100Ah, voltage 12V), which automatically switches to power supply when the photovoltaic panel power generation is insufficient or at night, with a battery life of ≥24 hours.

[0157] The output voltage is stable at 12V±0.5V, and the output power can be dynamically adjusted within the range of 10-50W: when the bracket is in the retracted safety state, the power supply is reduced to 10-20W to reduce the power supply power of non-core modules; when in the deployed power generation state, the power supply is increased to 30-50W to prioritize the power supply stability of the sensing unit, multi-target optimization control unit and digital twin unit.

[0158] The unit has overvoltage (≥15V), overcurrent (≥5A), and overheat (≥60℃) protection functions. The protection threshold matches the safety characteristics of the photovoltaic bracket electrical system. After the protection is triggered, the power supply is automatically cut off and a warning signal is sent.

[0159] VII. Closed-loop feedback correction unit;

[0160] The input of the closed-loop feedback correction unit establishes a data connection with the outputs of the sensing unit, dynamic analysis unit, and digital twin unit to collect actual action parameters of the support structure, structural status data, and power generation efficiency data.

[0161] Based on the deviation between the output of the multi-objective optimization function and the actual operating data, the action parameters of the retractable mechanism unit (such as motor speed and retraction / expansion sequence), the model coefficients of the dynamic analysis unit (such as gap damping coefficient and structural flexibility parameters), and the algorithm parameters of the MPC controller (such as prediction time domain and weighting coefficient) are corrected in reverse.

[0162] The dynamic balance between the power generation efficiency and structural safety of the support structure is adjusted to meet the requirements of the correct direction. For example, when the actual power generation efficiency is more than 5% lower than the optimized value, the power generation efficiency weighting coefficient is increased and the support tracking angle parameter is adjusted; when the actual structural stress exceeds the model prediction value by more than 10%, the stress calculation coefficient of the dynamic model is corrected.

[0163] The correction cycle is 1 minute, forming a complete control process to ensure continuous optimization of system performance.

[0164] VIII. System Full-Process Workflow

[0165] The specific steps for this application are as follows:

[0166] Data acquisition phase: The meteorological environment sensing component and structural state sensing module of the sensing unit collect multi-dimensional data in real time. After noise reduction and integration by the data integration module, the data is transmitted to the dynamic analysis unit and the digital twin unit.

[0167] Dynamics analysis phase: The dynamics analysis unit is based on a multibody dynamics model with gaps and structural flexibility. It combines sensing data to perform nonlinear behavior simulation and load coupling analysis, and outputs dynamic response data and load data.

[0168] Multi-objective optimization decision-making stage: The multi-objective optimization control unit receives dynamic analysis data and sensing data, and generates support angle adjustment commands or contraction and expansion trigger commands through MPC controller and multi-objective function calculation;

[0169] Action execution phase: The retractable mechanism unit receives instructions and performs angle adjustment or retraction / expansion actions according to the time-coordinated module's plan to achieve light tracking or safety protection;

[0170] Digital twin verification phase: The digital twin unit simulates the action and operating status of the support in real time through a high-fidelity model, and the HIL test component verifies the effectiveness of the control strategy;

[0171] Closed-loop correction stage: The closed-loop feedback correction unit collects actual operating data and virtual model data, calculates the deviation and corrects the parameters of each unit in reverse to form closed-loop control.

[0172] The above embodiments detail the technical solution of the present invention. Through the coordinated linkage and deep integration of various units, a dynamic balance between the power generation efficiency and structural safety of the photovoltaic support system is achieved, improving the system's environmental adaptability and operational reliability. This provides reliable technical support for the efficient and stable operation of photovoltaic power plants under complex meteorological environments. Those skilled in the art can implement the technical solution of the present invention based on the above embodiments; related details can be flexibly adjusted, all of which are within the scope of protection of the present invention.

[0173] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An adaptive retractable photovoltaic support multi-objective optimization intelligent control system, characterized in that, It includes a sensing unit, a retractable mechanism unit, a dynamic analysis unit, a multi-objective optimization control unit, a digital twin unit, and a power supply guarantee unit; Each unit interacts with data through an industrial-grade wireless communication link. The frequency band of the communication link matches the electromagnetic environment characteristics of the photovoltaic power station, and the signal transmission timing matches the response characteristics of the support retraction and deployment actions. The output data dimensions of the sensing unit cover the core features of photovoltaic support structure safety monitoring, meteorological environment adaptation, and power generation efficiency optimization. The retractable mechanism unit is synthesized using spinor theory and Lie group Lie algebra to form a reversible deformation structure. The structure body is a composite material structure optimized by multi-scale topology. The dynamic analysis unit is equipped with a multibody dynamics model that includes gaps and structural flexibility. The model can simulate the nonlinear dynamic behavior of the stent's deployment and contraction process. The multi-objective optimization control unit adopts a model predictive control algorithm to construct a multi-objective optimization function that integrates maximizing power generation efficiency and minimizing structural safety risks. The output end establishes a bidirectional hardware link with the drive module of the retractable mechanism unit. The digital twin unit constructs a high-fidelity digital twin model of the support system, integrates a hardware-in-the-loop test platform based on dSPACE, and connects the model calibration link to the real-time data output of the sensing unit. The dynamic analysis unit includes a multibody dynamics modeling component, a nonlinear behavior simulation component, and a load coupling analysis component; The multibody dynamics modeling component is equipped with a dynamic model that includes gaps and structural flexibility, and the model parameters are calibrated by mechanical test data of the support structure; the simulation logic of the nonlinear behavior simulation component matches the vibration and impact characteristics of the support deployment and contraction process. The load coupling analysis component integrates the coupling link between structural mechanics analysis and fluid dynamics wind load calculation. The analysis result output end is physically connected to the decision input interface of the multi-objective optimization control unit, and the output data is compatible with the constraint logic of the multi-objective optimization function.

2. The adaptive retractable photovoltaic support multi-objective optimization intelligent control system according to claim 1, characterized in that, The sensing unit includes a meteorological environment sensing component, a structural state sensing module, and a data integration module; the monitoring dimensions of the meteorological environment sensing component cover the core meteorological parameters required for the photovoltaic support to resist wind and snow loads. The response characteristics of the sensing elements in the structural state sensing module match the dynamic behavior of stress, deformation, and vibration of the support structure. The output of the data integration module is physically connected to the model input interface of the dynamic analysis unit through a state calibration link, and the output fused data format is compatible with the analytical logic of the multibody dynamics model.

3. The adaptive retractable photovoltaic support multi-objective optimization intelligent control system according to claim 2, characterized in that, The meteorological environment sensing components include wind speed and direction sensors, snowfall sensors, light intensity sensors, and ambient temperature sensors. The detection frequency band of the wind speed and direction sensor covers the wind speed range of the critical working condition of the photovoltaic support, and the sensing threshold of the snowfall sensor matches the critical value of the snow load on the support. The sampling frequency of the light intensity sensor matches the time scale for optimizing photovoltaic tracking power generation efficiency, and the response range of the ambient temperature sensor covers the temperature range of changes in the mechanical properties of the support material. The output of this component is connected to the data integration module together with the output of the structure state perception module. The integrated meteorological structure correlation data provides the basic input for multi-objective optimization.

4. The adaptive retractable photovoltaic support multi-objective optimization intelligent control system according to claim 2, characterized in that, The structural state sensing module includes stress sensors, displacement sensors, and vibration sensors deployed at key nodes of the support, as well as a gap state sensor integrated at the retractable joint. The response range of the stress sensor covers the allowable stress range of the composite material structure, and the measurement accuracy of the displacement sensor matches the accuracy requirements of the bracket deformation monitoring. The detection parameters of the vibration sensor are matched with the frequency characteristics of the nonlinear vibration of the bracket, and the identification criterion of the gap state sensor is the difference in physical characteristics of the joint gap of the retractable mechanism. The signal transmission carrier of each sensing element is an insulating structure compatible with the electromagnetic environment of the photovoltaic power station, and its dielectric properties match the electric field distribution characteristics inside the support array.

5. The adaptive retractable photovoltaic support multi-objective optimization intelligent control system according to claim 1, characterized in that, The retractable mechanism unit includes a reversible deformable structure body, a drive module, and a timing coordination module; the reversible deformable structure body completes configuration synthesis through spinor theory and Lie group Lie algebra, and achieves multi-scale topology optimization through a multi-disciplinary optimization platform; The drive module is driven by a servo motor, and the output torque matches the load characteristics of the bracket's retraction and expansion. The drive signal is synchronized with the command timing of the multi-objective optimization control unit. The synchronization benchmark of the timing coordination module is the operation time scale of the segmented contraction and expansion of the support. The timing coordination data output by the module provides time dimension support for the action allocation of the retractable mechanism unit through the timestamp encapsulation board.

6. The adaptive retractable photovoltaic support multi-objective optimization intelligent control system according to claim 1, characterized in that, The multi-objective optimization control unit includes an MPC controller, a multi-objective function construction component, and a decision output module; The algorithm parameters of the MPC controller are optimized through the Isight multidisciplinary optimization platform. The optimization strategy adopts a hybrid strategy combining sequential quadratic programming and genetic algorithm. The multi-objective optimization function is composed of the product of power generation efficiency weight coefficient and power generation efficiency, minus the product of structural safety risk value weight coefficient and structural safety risk value. The output commands of the decision output module include support angle adjustment commands and contraction / expansion trigger commands. The command parameters are matched with the motion characteristics of the retractable mechanism unit and the analysis results of the dynamic analysis unit.

7. The adaptive retractable photovoltaic support multi-objective optimization intelligent control system according to claim 1, characterized in that, The digital twin unit includes a high-fidelity model building component, a HIL testing component, and a model calibration module; The model dimensions of the high-fidelity model building component match the physical characteristics of the photovoltaic support and environmental coupling system, and the model accuracy is calibrated by real-time data from the sensing unit. The HIL test component is built on the dSPACE platform, and the test logic matches the real-time and reliability verification requirements of the multi-objective optimization control strategy. The calibration link of the model calibration module interacts in real time with the structural state data and meteorological environment data of the sensing unit, and maintains the consistency between the digital twin model and the physical support through iterative correction.

8. The adaptive retractable photovoltaic support multi-objective optimization intelligent control system according to claim 1, characterized in that, The power supply guarantee unit adopts a dual-mode power supply mode that combines photovoltaic self-generation and self-consumption with energy storage backup. It has a built-in power management chip that can dynamically adjust the output voltage and power according to the instructions of the multi-objective optimization control unit. When the support is in a retracted and safe state, the power supply unit automatically reduces the power supply to non-core modules; When the support structure is in the deployed power generation state, priority is given to ensuring the power supply stability of the sensing unit, multi-objective optimization control unit, and digital twin unit; this unit has overvoltage, overcurrent, and overheat protection functions, and the protection threshold matches the safety characteristics of the photovoltaic support structure's electrical system.

9. The adaptive retractable photovoltaic support multi-objective optimization intelligent control system according to claim 1, characterized in that, The system also includes a closed-loop feedback correction unit; the input of the closed-loop feedback correction unit establishes a data connection with the output of the sensing unit, the dynamic analysis unit, and the digital twin unit to collect the actual action parameters of the support, structural status data, and power generation efficiency data. Based on the deviation between the output of the multi-objective optimization function and the actual operating data, the action parameters of the contractible mechanism unit, the model coefficients of the dynamic analysis unit, and the algorithm parameters of the MPC controller are corrected in reverse.

Images